August 2021

Reference Manual

Fusion
Welcome
Welcome to Fusion for Mac, Linux and Windows!

Fusion is the world’s most advanced compositing software for visual effects artists,
broadcast and motion graphic designers and 3D animators. With over 30 years of
development, Fusion has been used on over 1000 major Hollywood blockbuster feature
films! Fusion features an easy and powerful node based interface so you can construct
complex effects simply by connecting various types of processing together. That’s super
easy and extremely fast! You get a massive range of features and effects included,
so you can create exciting broadcast graphics, television commercials, dramatic title
sequences and even major feature film visual effects!

Fusion Studio customers can also use DaVinci Resolve Studio to get a complete set of
editing, advanced color correction and professional Fairlight audio post production tools.
Clips in the DaVinci Resolve timeline can be shared with Fusion so you can collaborate
on your complex compositions within your VFX team and then render the result back
directly into the DaVinci Resolve timeline. We hope you enjoy reading this manual and
we can’t wait to see the work you produce with Fusion.

The Fusion Engineering Team

Grant Petty
CEO Blackmagic Design

Fusion 17 Welcome 2
Contents
Menu Descriptions 6

PART 1
Fusion Fundamentals
1 Introduction to Compositing in Fusion 8
2 Exploring the Fusion Interface 13
3 Getting Clips into Fusion 59
4 Rendering Using Saver Nodes 80
5 Working in the Node Editor 103
6 Node Groups, Macros, and Fusion Templates 147
7 Using Viewers 167
8 Editing Parameters in the Inspector 205
9 Animating in Fusion’s Keyframes Editor 228
10 Animating in Fusion’s Spline Editor 245
11 Animating with Motion Paths 274
12 Using Modifiers, Expressions, and Custom Controls 291
13 Bins 303
14 Fusion Connect 322
15 Preferences 335

PART 2
2D Compositing
16 Controlling Image Processing and Resolution 381
17 Managing Color for Visual Effects 390
18 Understanding Image Channels 402
19 Compositing Layers in Fusion 436
20 Rotoscoping with Masks 458
21 Paint 480
22 Using the Tracker Node 504
23 Planar Tracking 536
24 Using Open FX, Resolve FX, and Fuse Plug-Ins 542

PART 3
3D Compositing
25 3D Compositing Basics 546
26 3D Camera Tracking 598
27 Particle Systems 615

PART 4
Advanced Compositing Techniques
28 Optical Flow and Stereoscopic Nodes 624

Fusion 17 Contents 3
PART 5
Fusion Page Effects
29 3D Nodes 638
30 3D Light Nodes 748
31 3D Material Nodes 761
32 3D Texture Nodes 788
33 Blur Nodes 813
34 Color Nodes 837
35 Composite Nodes 892
36 Deep Pixel Nodes 906
37 Effect Nodes 922
38 Film Nodes 953
39 Filter Nodes 970
40 Flow Nodes 986
41 Flow Organizational Nodes 989
42 Fuses 994
43 Generator Nodes 996
44 I/O Nodes 1033
45 LUT Nodes 1056
46 Mask Nodes 1065
47 Matte Nodes 1102
48 Metadata Nodes 1155
49 Miscellaneous Nodes 1162
50 Optical Flow 1197
51 Paint Node 1210
52 Particle Nodes 1219
53 Position Nodes 1276
54 Resolve Connect 1294
55 Shape Nodes 1300
56 Stereo Nodes 1331
57 Tracker Nodes 1361
58 Transform Nodes 1406
59 VR Nodes 1431
60 Warp Nodes 1441
61 Modifiers 1469

PART 6
Other Information
62 Regulatory Notices, Safety Information and Warranty 1506

Fusion 17 Contents 4
Navigation Guide

Chapter 1

Introduction
to Compositing
in Fusion
This introduction is designed explicitly to help users who are new to Fusion get
started learning this exceptionally powerful environment for creating and editing
visual effects and motion graphics right from within DaVinci Resolve or using the
stand-alone Fusion Studio application.

This documentation covers both the Fusion Page inside DaVinci Resolve and the
stand-alone Fusion Studio application.

Contents By clicking the Contents heading,

you will be taken to the Contents page
What Is Fusion?
The Fusion Page within DaVinci Resolve
 9
 9
at the beginning of this manual.
The Fusion Studio Stand-Alone Application  10
What Kinds of Effects Does Fusion Offer?  11
How Hard Will This Be to Learn?  12 By clicking each title or the
page number, you will be taken
to the appropriate page.

By clicking the footer on each page,

you will be taken to the Contents
Chapter 1 Introduction to Compositing in Fusion 8
page of that particular Chapter.

Fusion 17

Menu Descriptions
For ease of use navigating this manual, each menu item is listed here, and by clicking
on the name of the menu function, you will be taken to the appropriate part of the
manual that describes that function.

Fusion
Show Toolbar – Page 33
Toggles the Fusion toolbar on or off.

Fusion Settings – Page 54

Opens the Fusion Settings window.
By clicking on the name of the
Reset Composition menu function, you will be taken
to the appropriate part of the manual
Resets a Fusion composition to its initial state.

Macro Editor – Page 150

Opens the Macro Editor.

Import
that describes that function.
Specific file format import for Fusion.
> Alembic Scene – Page 639
> FBX Scene – Page 586
> PSD – Page 78
> Shapes – Page 74
> SVG – Page 74
> Tracks – Page 73

Render All Savers – Page 84

Renders OpenEXR files.

By clicking the footer, you will be

taken to the Contents page.
Fusion 17 Menu Descriptions 6

Fusion 17 Navigation Guide 5

Fusion 17

Menu Descriptions
For ease of use navigating this manual, each menu item is listed here, and by clicking
on the name of the menu function, you will be taken to the appropriate part of the
manual that describes that function.

Fusion
Show Toolbar –  Page 33
Toggles the Fusion toolbar on or off.

Fusion Settings – Page 54

Opens the Fusion Settings window.

Reset Composition
Resets a Fusion composition to its initial state.

Macro Editor – Page 150

Opens the Macro Editor.

Import
Specific file format import for Fusion.
> Alembic Scene –
  Page 639
> FBX Scene –
  Page 586
> PSD – Page 78
> Shapes –
  Page 74
> SVG – Page 74
> Tracks –
  Page 73

Render All Savers –

  Page 84
Renders OpenEXR files.

Fusion 17 Menu Descriptions 6

 PART 1

Fusion
Fundamentals
Chapter 1

Introduction
to Compositing
in Fusion
This introduction is designed explicitly to help users who are new to Fusion get
started learning this exceptionally powerful environment for creating and editing
visual effects and motion graphics right from within DaVinci Resolve or using the
stand-alone Fusion Studio application.

This documentation covers both the Fusion Page inside DaVinci Resolve and the
stand-alone Fusion Studio application.

Contents
What Is Fusion?  9
The Fusion Page within DaVinci Resolve  9
The Fusion Studio Stand-Alone Application  10
What Kinds of Effects Does Fusion Offer?  11
How Hard Will This Be to Learn?  12

Chapter 1 Introduction to Compositing in Fusion 8

What Is Fusion?
Blackmagic Design Fusion is powerful 2D and 3D visual effects compositing software with over thirty
years of evolution serving the motion picture and broadcast industry, creating effects seen in
countless films and television series. It is available as a stand-alone application as well as a page
within DaVinci Resolve.
In its purest form, Fusion is a collection of image-processing engines called nodes. These nodes
represent effects like blurs and color correctors, as well as images, 3D models, and spline masks.
Similar to effects you may be familiar with, each node includes a set of parameters that can be
adjusted and animated over time. Stringing different nodes together in a graphical user interface
called a node tree allows you to create sophisticated visual effects. The nodes, node trees, and all
settings you create are saved in a document called a Composition, or “comp” for short.

The Fusion Page within DaVinci Resolve

Merged right into DaVinci Resolve, the Fusion page makes it possible to jump immediately from editing
right into compositing, with no need to export media, relink files, or launch another application to get
your work done. Now everything you need lives right inside DaVinci Resolve.

The Fusion page in DaVinci Resolve, showing viewers, the Node Editor, and the Inspector

How Do I Use the Fusion Page?

The relationship between the Edit page and the Fusion page is similar to the relationship between the
Edit page and the Color page. Each clip can have a grade applied to it in the Color page, and similarly
every clip can have a composition applied to it in the Fusion page.
If you use the Fusion page to create any kind of effect or composite, a badge appears on that clip in
the Timeline to show that clip has a composition applied to it.

Clips with Fusion page compositions have

a Fusion badge to the right of the name.

Chapter 1 Introduction to Compositing in Fusion 9

To create an effect in the Fusion page of DaVinci Resolve, you need only park the playhead over a clip
in the Edit or Cut page and then click the Fusion page button. Your clip is immediately available as a
MediaIn node in the Fusion page, ready for you to add a variety of stylistic effects. You can paint out
an unwanted blemish or feature, build a quick composite to add graphics or text, or accomplish any
other visual effect you can imagine, built from the Fusion page’s toolkit of effects.
Alternatively, in DaVinci Resolve, you have the option of editing together all the clips you want to use,
superimposing and lining up every piece of media you’ll need with the correct timing, before selecting
them and creating a Fusion clip. A Fusion clip functions as a single item in the Edit or Cut page
timeline, but once in the Fusion page, each piece of media you’ve assembled is revealed in a fully built
Fusion composition, ready for you to start adding nodes to customize for whatever effect you need
to create.
Whichever way you want to work, all this happens on the very same timeline as editing, grading, and
audio post, for a seamless back and forth as you edit, refine, and finish your projects.

How Do Fusion Effects Differ from Edit Page Effects?

When using DaVinci Resolve, you can create numerous effects in the Edit page. Transitions, fades,
superimpositions, over-the-shoulder picture-in-picture effects, time remapping, and lower third titles
are some of the effects that can be quickly and more efficiently created in the Edit or Cut page
timeline. However, the Fusion page’s node-based interface lets you go deep into the details of a
composition to create sophisticated 2D and 3D effects with precise control and endless customization.
Effects that include more than two or three layers can be much more manageable in Fusion. Green or
blue screen composites, sky replacements, and object removal are all effects better suited for Fusion’s
more advanced toolset.

How Do Fusion Effects Differ from Color Page Effects?

The Color page in DaVinci Resolve can also handle some visual effects work. Effects that blur the line
between color grading and finishing can be very fast and intuitive in the Color Page, especially for
people already familiar with the Color page toolset. Beauty work and small image repairs can make
efficient use of the Color page’s straightforward Tracking tool, Face Refinement, and Patch Replacer
effects. However, when it comes to more challenging blue/green screen compositing, the tools built
around Fusion’s powerful Delta keyer are more capable of handling these shots. Integrating 3D
objects into live-action scenes, split-screen effects, motion graphics, and precise keyframing are all
better suited to the Fusion page.

The Fusion Studio Stand-Alone Application

Creating visual effects with the stand-alone Fusion Studio software begins with opening Fusion,
creating a new composition, importing some clips via Loader nodes, and building out your composite
with effects. Just like the Fusion Page in DaVinci Resolve, you add effects using different nodes from
the Effects Library, and you combine multiple layers of imagery using Merge nodes. Once you’ve
created the desired result, add a Saver node to the end of the tree of nodes you’ve created to render
your final result.

Rendering Out Your Final Result

Unlike the Fusion Page in DaVinci Resolve, which renders directly back into the Edit or Cut page
timeline, the final step in Fusion Studio is to render the finished effect to disk as a movie file or image
sequence. The last node in every node tree is a Saver node. Saver nodes configure the output file
format and render the file to disk. You can use as many Saver nodes in a composite as you need. For
instance, you might use multiple Saver nodes to render out intermediate areas of a composite or to
output a composite in multiple formats.

Chapter 1 Introduction to Compositing in Fusion 10

What Kinds of Effects Does Fusion Offer?
In addition to the kinds of robust compositing, paint, rotoscoping, and keying effects you’d expect from
a fully-featured 2D compositing environment, Fusion offers much more.

3D Compositing
Fusion has powerful 3D nodes that include extruded 3D text, simple geometry, and the ability to
import 3D models. Once you’ve assembled a 3D scene, you can add cameras, lighting, and material
shaders, and then render the result with depth-of-field effects and auxiliary channels to integrate with
more conventional layers of 2D compositing, for a sophisticated blending of 3D and 2D operations in
the very same node tree.

A 3D scene with textured 3D text, created entirely within Fusion

Particles
Fusion also has an extensive set of nodes for creating particle systems that have been used in major
motion pictures, with particle generators capable of spawning other generators, 3D particle
generation, complex simulation behaviors that interact with 3D objects, and endless options for
experimentation and customization. You can create particle system simulations for VFX or more
abstract particle effects for motion graphics.

A 3D particle system, also created entirely within Fusion

Chapter 1 Introduction to Compositing in Fusion 11

Text
The Text tools in Fusion are exceptional, giving you layout and animation options in both 2D and 3D.
Furthermore, within DaVinci Resolve, these Text tools have been incorporated into the Edit and Cut
pages as Fusion Titles. These title templates are compositions saved from Fusion as macros with
published controls that are visible in the Edit or Cut page Inspector for easy customization, even if
you’re working with people who don’t know Fusion.

A multi-layered text composite integrating video clips and Fusion-generated elements

And Lots More

The list goes on. With Stereo and VR adjustment nodes, Planar Tracking, Deep Pixel nodes for
re‑compositing rendered 3D scenes using Auxiliary Channel data, powerful Masking and Rotoscoping
nodes, and Warping effects, Fusion is an impressively featured environment for building worlds, fixing
problems, and flying multi-layered motion graphics animations through your programs.

How Hard Will This Be to Learn?

That depends on what you want to do, but honestly it’s not so bad with this PDF at your side, helping
guide the way. It’s worth repeating that this Fusion documentation was developed specifically to help
users who’ve never before worked with Fusion learn the core concepts needed to perform the basics,
in preparation for learning the rest of the application on your own.
Fusion is a deep, production-driven product that’s had decades of development, so its feature set is
deep and comprehensive. You won’t learn it in an hour, but much of what you’ll find won’t be so very
different from other compositing applications you may have used. And if you’ve familiarized yourself
with the node-based grading workflow of the DaVinci Resolve Color page, you’ve already got a leg up
on understanding the central operational concept of compositing in Fusion.

Chapter 1 Introduction to Compositing in Fusion 12

Chapter 2

Exploring the
Fusion Interface
This chapter provides an orientation on the Fusion user interface, providing a
quick tour of what tools are available, where to find things, and how the different
panels fit together to help you build and refine compositions in this powerful
node‑based environment.

Contents
The Fusion User Interface  15
The Work Area  16
Interface Toolbar  16
Choosing Which Panel Has Focus  17
Viewers  17
Zooming and Panning into Viewers  19
Loading Nodes Into Viewers  19
Clearing Viewers  20
Viewer Controls  20
Time Ruler and Transport Controls  22
Time Ruler Controls in the Fusion Page  22
Time Ruler Controls in Fusion Studio  23
The Playhead  23
Zoom and Scroll Bar  24
Transport Controls in the Fusion Page  24
Audio Monitoring  26
Transport Controls in Fusion Studio  27
Changing the Time Display Format  31
Keyframe Display in the Time Ruler  31
The Fusion RAM Cache for Playback  31

Chapter 2 Exploring the Fusion Interface 13

Toolbar  33
Customizing the Toolbar  33
Node Editor  35
Adding Nodes to Your Composition  35
Removing Nodes from Your Composition  36
Identifying Node Inputs and Node Outputs  36
Node Editing Essentials  36
Navigating the Node Editor  38
Vertical Node Editor Layouts  38
Keeping Organized  39
Status Bar  39
Effects Library  40
The Inspector  41
The Tools and Modifiers Panels  41
Parameter Header Controls  42
Parameter Tabs  42
Keyframes Editor  43
Keyframes Editor Control Summary  43
Adjusting Clip Timings  44
Adjusting Effect Timings  44
Adjusting Keyframe Timings  44
Spline Editor  45
Spline Editor Control Summary  46
Choosing Which Parameters to Show  46
Essential Spline Editing  47
Essential Spline Editing Tools and Modes  47
Thumbnail Timeline in the Fusion Page  49
The Media Pool in the Fusion Page  50
Importing Media Into the Media Pool on the Fusion Page  50
Bins in Fusion Studio  51
Bins Interface  51
The Bin Studio Player  53
The Console  53
Customizing Fusion  54
The Fusion Settings Window  54
Showing and Hiding Panels  55
Resizing Panels  56
Fusion Studio Floating Frame  56
Fusion Studio Keyboard Remapping  56
Undo and Redo  57

Chapter 2 Exploring the Fusion Interface 14

The Fusion User Interface
If you open up everything at once, Fusion is divided into four principal regions designed to help you
make fast work of node-based compositing. The viewer(s) are at the top, the work area is at the
bottom, the Inspector is at the right, and the Effects Library is the area found at the left. In
DaVinci Resolve’s Fusion page, the Effects Library shares space with the Media Pool. All these panels
work together to let you add effects, paint to correct issues, create motion graphics or title sequences,
or build sophisticated 3D and multi-layered composites.

The Fusion user interface shown completely

However, Fusion doesn’t have to be that complicated, and in truth, you can work very nicely with only
the viewer, Node Editor, and Inspector open for a simplified experience.

A simplified set of Fusion controls for everyday working

Chapter 2 Exploring the Fusion Interface 15

The Work Area
You probably won’t see the term “the work area” used much, in favor of the specific panels within the
work area that you’ll be using. Still, the area referred to as the work area is the region at the bottom
half of the Fusion user interface, within which you can expose the three main panels used to construct
compositions and edit animations in Fusion. These are the Node Editor, the Spline Editor, and the
Keyframes Editor. By default, the Node Editor is the first thing you’ll see, and the main area you’ll be
working within, but it can sit side-by-side with the Spline Editor and Keyframes Editor as necessary.
You can make more horizontal room on your display for these three panels by putting the Effects
Library and Inspector into a half-height mode, if necessary.

The work area showing the Node Editor, the Spline Editor, and the Keyframes Editor

Interface Toolbar
At the very top of Fusion is a toolbar with buttons that let you show and hide different parts of the user
interface (UI). Buttons with labels identify which parts of the UI can be shown or hidden. In
DaVinci Resolve’s Fusion page, if you right-click anywhere within this toolbar, you have the option of
displaying this bar with or without text labels.

The UI toolbar of the Fusion page

The UI toolbar of Fusion Studio

These buttons are as follows, from left to right:

– Media Pool/Effects Library Full Height: Lets you set the area used by the Media Pool
(DaVinci Resolve only) and/or Effects Library to take up the full height of your display, giving you
more area for browsing at the expense of a narrower Node Editor and viewer area. At half-height,
the Media Pool/Templates/Effects Library are restricted to the top half of the UI along with the
viewers (you can only show one at a time), and the Node Editor takes up the full width of your
display.
– Media Pool: (DaVinci Resolve only): Shows and hides the Media Pool, from which you can drag
additional clips into the Node Editor to use them in your Fusion page composition.
– Effects Library: Opens or hides the repository of all node tools available to use in Fusion. From
here, you can click nodes to add them after the currently selected node in the Node Editor, or you
can drag and drop nodes to any part of the node tree you like.
– Clips: (DaVinci Resolve only): Opens and closes the Thumbnail timeline, which lets you navigate
your program, create and manage multiple versions of compositions, and reset the current
composition.
– Nodes: Opens and closes the Node Editor, where you build and edit your compositions.

Chapter 2 Exploring the Fusion Interface 16

 – Console (Fusion Studio only): The Console is a window in which you can see the error, log,
script, and input messages that may explain something Fusion is trying to do in greater detail. The
Console is also where you can read FusionScript outputs, or input FusionScripts directly.
– Spline: Opens and closes the Spline Editor, where you can edit the curves that interpolate
keyframe animations to customize and perfect their timing. Each keyframed parameter appears
hierarchically within the effect in which it appears in a list to the left.
– Keyframes: Opens and closes the Keyframes Editor, which shows each clip and effects node in
your Fusion composition as a layer. You can use the Keyframes Editor to edit and adjust the timing
of keyframes that have been added to various effects in your composition. You can also use the
Keyframes Editor to slide the relative timing of clips that have been added to Fusion, as well as
to trim their In and Out points. A spreadsheet can be shown and hidden within which you can
numerically edit keyframe values for selected effects.
– Metadata (DaVinci Resolve only): Hides or shows the Metadata Editor, which lets you read
and edit the available clip and project metadata associated with any piece of media within a
composite.
– Inspector: Shows or hides the Inspector, which shows you all the editable parameters and controls
that correspond to selected nodes in the Node Editor. You can show the parameters for multiple
nodes at once, and even pin the parameters of nodes you need to continue editing so that they’re
displayed even if those nodes aren’t selected.
– Inspector Height: Lets you open the Inspector to be half height (the height of the viewer area)
or full height (the height of your entire display). Half height allows more room for the Node Editor,
Spline Editor, and/or Keyframes Editor, but full height lets you simultaneously edit more node
parameters or have enough room to display the parameters of multiple nodes at once.

Choosing Which Panel Has Focus

Whenever you click somewhere on the Fusion interface using the pointer or using a keyboard shortcut
to “select” a particular panel, you give that panel of the user interface “focus.” A panel with focus
captures specific keyboard shortcuts to do something within that panel, as opposed to doing
something elsewhere in the interface.
To make it easier to keep track of which panel has focus, a highlight appears at the top edge of
whichever panel has focus. In DaVinci Resolve, you must turn on “Show focus indicators in the User
Interface” in the UI Settings panel of the User Preferences to see the highlight.

The focus indicator shown at the top edge of the Media

Pool, shown next to a viewer that doesn’t have focus

Viewers
The viewer area displays either one or two viewers at the top of the Fusion page, and this is
determined via the Viewer button at the far right of the Viewer title bar. Each viewer can show a single
node’s output from anywhere in the node tree. You assign which node is displayed in which viewer.
This makes it easy to load separate nodes into each viewer for comparison. For example, you can load
a Keyer node into the left viewer and the final composite into the right viewer, so you can see the
image you’re adjusting and the final result at the same time.

Chapter 2 Exploring the Fusion Interface 17

Dual viewers let you edit an upstream node in one while seeing its effect on the overall composition in the other.

Ordinarily, each viewer shows 2D nodes from your composition as a single image. However, when
you’re viewing a 3D node, you have the option to set that viewer to one of several 3D views. A
perspective view gives you a repositionable stage on which to arrange the elements of the world
you’re creating. Alternatively, a quad view lets you see your composition from four angles, making it
easier to arrange and edit objects and layers within the XYZ axes of the 3D space in which
you’re working.

Loading a 3D node into a viewer switches on a Perspective view

TIP: In Perspective view, you can hold down the Option key and drag in the viewer to pivot
the view around the center of the world. All other methods of navigating viewers
work the same.

The viewers have a variety of capabilities you can use to compare and evaluate images. This section
provides a short overview of viewer capabilities to get you started.

Chapter 2 Exploring the Fusion Interface 18

Zooming and Panning into Viewers
There are standardized methods of zooming into and panning around viewers when you need a closer
look at the situation. These methods also work with the Node Editor, Spline Editor, and
Keyframes Editor.

Methods of panning viewers:

– Middle-click and drag to pan around the viewer.
– Hold down Shift and Command and drag the viewer to pan.
– Drag with two fingers on a track pad to pan.

Methods of scaling viewers:

– Click a viewer, and press the equals key (=) to zoom in, and the minus key (-) to zoom out.
– Press the middle and left mouse buttons simultaneously and drag left or right to resize the viewer.
– Hold down the Command key and use your pointer’s scroll control to zoom in
and out of the viewer.
– Hold down the middle mouse button, and then click the left mouse button to zoom in, or click the
right button to zoom out. The scaling uses a fixed amount, centered on the position of the cursor.
– Click a viewer and press Command-1 to resize the image in the viewer to 100 percent.
– Click a viewer and press Command-F to reset the image in the viewer to fit the viewer.
– Click the Scale viewer menu and choose Fit or a percentage.
– Right-click on a viewer and choose an option from the Scale submenu of the contextual menu.
This includes a Custom Scale command that lets you type your own scale percentage.
– Hold down the Command key and drag with two fingers on a track pad to zoom in and
out of the viewer.

Methods of spinning 3D viewers:

– In 3D Perspective view, hold down the Shift key and drag to spin the stage around.
– In 3D Perspective view, hold down the Shift key and drag with two fingers on a track pad to spin
the stage around.

Loading Nodes Into Viewers

When you first open the Fusion page in DaVinci Resolve, the output of the current empty composition
(the MediaOut1 node) is usually showing in viewer 2. If you’re in Dual-viewer mode, viewer 1 remains
empty until you assign a node to one of them.
When using Fusion Studio, nothing is loaded into either of the viewers until you assign a node to
one of them.

To load specific nodes into specific viewers:

– Hover the pointer over a node, and click one of two buttons that appear at the bottom
left of the node.
– Click once to select a node, and press 1 (for the left viewer) or 2 (for the right viewer).
– Right-click a node and choose View On > None/Left View/Right View in the contextual menu.
– Right-click the control header of a node in the Inspector, and choose View On > None/Left View/
Right View from the contextual menu.
– Drag a node and drop it over the viewer you’d like to load it into (this is great for tablet users).

Chapter 2 Exploring the Fusion Interface 19

When a node is being viewed, a View Indicator button appears at the bottom left. This is the same
control that appears when you hover the pointer over a node. Not only does this control let you know
which nodes are loaded into which viewer, but they also expose little round buttons for switching
between viewers.

Viewer assignment buttons at

the bottom of nodes indicate
when they’re being viewed.

Clearing Viewers
To clear an image from a viewer, click in the viewer to make it active; a thin red highlight is displayed at
the top of the active viewer. With the viewer active, press the Tilde (~) key. This key is usually found to
the left of the 1 key on U.S. keyboards. The fastest way to remove all the images from all the viewers is
to make sure none of the viewers is the active panel, and then press the Tilde key.

Viewer Controls
A series of buttons and pop-up menus in the viewer’s title bar provides several quick ways of
customizing the viewer display.

Controls in the viewer title bar

– Zoom menu: Lets you zoom in on the image in the viewer to get a closer look, or zoom out to get
more room around the edges of the frame for rotoscoping or positioning different layers. Choose
Fit to automatically fit the overall image to the available dimensions of the viewer.
– Split Wipe button and A/B Buffer menu: You can actually load two nodes into a single viewer
using that viewer’s A/B buffers by choosing a buffer from the menu and loading a node into the
viewer. Turning on the Split Wipe button (press Forward Slash) shows a split wipe between the
two buffers, which can be dragged left or right via the handle of the onscreen control, or rotated
by dragging anywhere on the dividing line on the onscreen control. Alternatively, you can switch
between each full-screen buffer to compare them (or to dismiss a split-screen) by pressing Comma
(A buffer) and Period (B buffer).
– SubView type: (These aren’t available in 3D viewers.) Clicking the icon itself enables or disables
the current “SubView” option you’ve selected, while using the menu lets you choose which
SubView is enabled. This menu serves one of two purposes. When displaying ordinary 2D nodes,
it lets you open up SubViews, which are viewer “accessories” within a little pane that can be used
to evaluate images in different ways. These include an Image Navigator (for navigating when
zoomed far into an image), Magnifier, 2D viewer (a mini-view of the image), 3D Histogram scope,
Color Inspector, Histogram scope, Image Info tooltip, Metadata tooltip, Vectorscope, or Waveform
scope. The Swap option (Shift-V) lets you switch what’s displayed in the viewer with what’s being
displayed in the Accessory pane. When displaying 3D nodes, this button lets you turn on the
quad-paned 3D viewer.
– Node name: The name of the currently viewed node is displayed at the center of the
viewer’s title bar.

Chapter 2 Exploring the Fusion Interface 20

 – RoI controls: Clicking the icon itself enables or disables RoI (Region of Interest) limiting in the
viewer, while using the menu lets you choose the region of the RoI. RoI lets you define the region
of the viewer in which pixels actually need to be updated. When a node renders, it intersects
the current RoI with the current Domain of Definition (DoD) to determine what pixels should be
affected. When enabled, you can position a rectangle to restrict rendering to a small region of the
image, which can significantly speed up performance when you’re working on very high resolution
or complex compositions. Auto (the default) sets the region to whatever is visible at the current
zoom/pan level in the viewer. Choosing Set lets you draw a custom region within the frame by
dragging a rectangle that defaults to the size of the viewer, which is resizable by dragging the
corners or sides of the onscreen control. Choosing Lock prevents changes from being made to
the current RoI. Choosing Reset resets the RoI to the whole viewer.
– Color controls: Lets you choose which color and/or image channels to display in the viewer.
Clicking the icon itself toggles between Color (RGB) and Alpha, the two most common things you
want to see (pressing C or A also toggles between Color and Alpha). Opening the menu displays
every possible channel that can be displayed for the currently viewed node, commonly including
RGB, Red, Green, Blue, and Alpha (available from the keyboard by pressing R, G, B, or A). For
certain media and nodes, additional auxiliary channels are available to be viewed, including
Z-depth, Object ID, Material ID, XYZ Normals, and so on.
– Viewer LUT: Clicking the icon itself toggles LUT (LookUp Table) display on or off, while the menu lets
you choose which of the many available color space conversions to apply to the viewer. The top
five options let you choose Fusion controls that can be customized via the Edit item at the bottom
of this menu. The rest of this menu shows all LUTs installed in the LUT directory to use for viewing.

By default, when using DaVinci Resolve, the viewers in the Fusion page show you the image prior to
any grading done in the Color page, since the Fusion page comes before the Color page in the
DaVinci Resolve image processing pipeline. When you’re working on clips that have been converted
to linear color space for compositing, it is desirable to composite and make adjustments to the image
relative to a normalized version of the image that appears close to what the final will be. Enabling the
LUT display lets you do this as a preview, without permanently applying color adjustments to
the image.
– Option menu: This menu contains various settings that pertain to the viewers in Fusion.
– Snap to Pixel: When drawing or adjusting a polyline mask or spline, the control points will snap
to pixel locations.
– Show Controls: Toggles whatever onscreen controls are visible for the currently selected node.
– Region: Provides all the settings for the Region of Interest in the viewer.
– Smooth Resize: This option uses a smoother bilinear interpolated resizing method when
zooming into an image in the viewer; otherwise, scaling uses the nearest neighbor method and
shows noticeable aliasing artifacts. However, this is more useful when you zoom in at a pixel
level since there is no interpolation.
– Show Square Pixels: Overrides the auto aspect correction when using formats with
non-square pixels.
– Normalized Color Range: Allows for the visualization of brightness values outside of the normal
viewing range, particularly when working with floating-point images or auxiliary channels.
– Checker Underlay: Toggles a checkerboard underlay that makes it easy to see areas of
transparency.
– Gain/Gamma: Exposes a simple pair of Gain and Gamma sliders that let you adjust the viewer’s
brightness.
– 360 View: Used to properly display spherical imagery in a variety of formats, selectable from
this submenu.
– Stereo: Used to properly display stereoscopic imagery in a variety of formats, selectable from
this submenu.

Chapter 2 Exploring the Fusion Interface 21

Time Ruler and Transport Controls
The Time Ruler, located beneath the viewer area, is based on the total duration of the composition.
What it represents depends on which version of Fusion you’re using:
– For DaVinci Resolve users, the duration displayed in the Time Ruler range depends on what’s
currently selected in the Edit or Cut page timeline.
– In Fusion Studio, the Time Ruler depends on the Global Start and End values set in the Fusion
Studio Preferences > Defaults.

The transport controls under the Time Ruler include playback controls, audio monitoring, as well as
number fields for the composition duration and playback range. Additional controls enable motion blur
and proxy settings.

Time Ruler Controls in the Fusion Page

If you’ve selected a single clip in the Edit or Cut page Timeline, then the global range shown in the
Time Ruler is based on the total source duration for that clip. You cannot move the playhead outside
the global range. The yellow lines, called the render range, identify the current In and Out points for
the clip and are the only frames visible in the Fusion page. All frames outside of this range constitute
the unused head and tail handles of that source clip.

The Time Ruler displaying ranges for a clip in the Timeline via yellow marks (the playhead is red)

If you’ve created a Fusion clip or a compound clip, then the “working range” reflects the entire
duration of that clip.

The Time Ruler displaying ranges for a Fusion clip in the Timeline

Render Range
The render range determines the range of frames that are visible in the Fusion page and that are used
for interactive playback, disk caches, and previews. Frames outside the default render range are not
visible in the Fusion page and are not rendered or played.
You can modify the duration of the render range for preview and playback only. Making the range
shorter or longer does not trim the clip in the Edit or Cut page Timelines.

You can change the render range in the Time Ruler by doing one of the following:
– Hold down the Command key and drag a new range within the Time Ruler.
– Drag either the start or end yellow line to modify the start or end of the range.
– Right-click within the Time Ruler and choose Set Render Range from the contextual menu.
– Enter new ranges in the Range In and Out fields to the left of the transport controls.
– Drag a node from the Node Editor to the Time Ruler to set the range to the duration of that node.

Chapter 2 Exploring the Fusion Interface 22

You can return the render range to the In and Out points of the timeline clip by
doing one of the following.
– Right-click within the Time Ruler and choose Auto Render Range.
– Click back in the Edit or Cut page, and then return to the Fusion page.

Time Ruler Controls in Fusion Studio

The Time Ruler, located beneath the viewer area, displays two different frame ranges: one for the
entire composition, called the global range, and the other called the render range, which determines
what to render and what to cache in memory for previews. The global start and end range takes up the
entire Time Ruler and sets the total duration of a composition. You cannot move the playhead outside
the global range.

The Time Ruler displaying ranges for a clip in the Timeline via yellow marks (the playhead is red)

Global Start and End Range

The global start and end range is simply the total duration of the current composition.

You can change the global range by doing one of the following:
– To change the global range for all new compositions, choose Fusion Studio > Preferences on
macOS or File > Preferences on Windows or Linux. In the Global and Default Settings panel, enter
a new range in the Global range fields.
– To change the Global range for the current composition, enter a new range in the Global Start and
End fields to the left of the transport controls.
– Dragging a node from the Node Editor to the Time Ruler automatically sets the Global and Render
Range to the extent of the node.

Render Range
The render range determines the range of frames used for interactive playback, disk caches, and
previews. Frames outside the render range are not rendered or played, although you can still drag the
playhead to these frames to see the unused frames.
To preview or render a specific range of a composition, you can modify the render range in a
variety of ways.

You can set the render range in the Time Ruler by doing one of the following:
– Hold down the Command key and drag a new range within the Time Ruler.
– Right-click within the Time Ruler and choose Set Render Range from the contextual menu to set
the Render Range based on the selected Node’s duration.
– Enter new ranges in the Range In and Out fields to the left of the transport controls.
– Drag a node from the Node Editor to the Time Ruler to set the range to the duration of that node.

The Playhead
A red playhead within the Time Ruler indicates the currently viewed frame. Clicking anywhere within
the Time Ruler jumps the playhead to that frame, and dragging within the Time Ruler drags the
playhead within the available duration of that clip or composition.

Chapter 2 Exploring the Fusion Interface 23

Zoom and Scroll Bar
A two-handled gray scroll bar lets you zoom into the range shown by the Time Ruler, which is useful if
you’re using a very large Global range such that the Render range is a tiny sliver in the Time Ruler.
Dragging the left or right handles of this bar zooms relative to the opposite handle, enlarging the width
of each displayed frame. Once you’ve zoomed in, you can drag the scroll bar left or right to scroll
through the composition.

TIP: Holding the middle mouse button and dragging in the Time Ruler lets you scroll the
visible range.

Transport Controls in the Fusion Page

The Transport Controls in DaVinci Resolve’s Fusion page include buttons that control playback as well
as time fields on the left side for setting the render range and the current time on the right side.
Additional controls are available in the right-click menu.

Controlling Playback
There are six transport controls underneath the Time Ruler in the Fusion page. These buttons include
Composition First Frame, Play Reverse, Stop, Play Forward, Composition Last Frame, and Loop.

The Fusion page controls for playback

Navigation Shortcuts
Many standard transport control keyboard shortcuts you may be familiar with work in Fusion, but some
are specific to Fusion’s particular needs.

To move the playhead in the Time Ruler using the keyboard, do one of the following:
– Spacebar: Toggles forward playback on and off.
– JKL: Basic JKL playback is supported, including J to play backward, K to stop,
and L to play forward.
– Back Arrow: Moves 1 frame backward.
– Forward Arrow: Moves 1 frame forward.
– Shift-Back Arrow: Moves to the clip’s Global End frame.
– Shift-Forward Arrow: Moves to the clip’s Global Start frame.
– Command-Back Arrow: Jumps to the Render Range In point.
– Command-Forward Arrow: Jumps to the Render Range Out point.

Real-Time Playback Not Guaranteed

Because many of the effects you can create in the Fusion page are processor-intensive, there is no
guarantee of real-time playback at your project’s full frame rate unless you’ve cached your
composition first (discussed later).

Chapter 2 Exploring the Fusion Interface 24

Frame Increment Options
Right-clicking either the Play Reverse or Play Forward buttons opens a contextual menu. This menu
contains options to set a frame increment value, which lets you use a keyboard shortcut to move the
playhead in sub-frame or multi-frame increments.
Moving the playhead in multi-frame increments can be useful when rotoscoping. Moving the playhead
in sub-frame increments can be useful when rotoscoping or inspecting interlaced frames one field at a
time (0.5 of a frame).

Right-click the Play Forward or Play Backward buttons to

choose a frame increment in which to move the playhead.

Looping Options
The Loop button can be toggled to enable or disable looping during playback. You can right-click this
button to choose the looping method that’s used:
– Playback Loop: The playhead plays to the end of the Time Ruler and starts from the
beginning again.
– Ping-pong Loop: When the playhead reaches the end of the Time Ruler, playback reverses
until the playhead reaches the beginning of the Time Ruler, and then continues to ping-pong
back and forth.

Render Range Fields

The two time fields on the left side of the transport controls are used to modify the Render Range. You
can enter time values in frames to modify the In and Out of the render range for previews and caching.

The Render Start and Render End time fields

Chapter 2 Exploring the Fusion Interface 25

Audio Monitoring
Playing a composition in DaVinci Resolve’s Fusion page will play the audio from the Edit or Cut page
Timeline. You can choose to hear the audio or mute it using the Audio toolbar button to the left of the
transport controls. The audio waveforms are displayed in the Keyframes Editor to assist in the timing of
your animations.

TIP: If the Mute button is enabled on any Timeline tracks, audio from those tracks will not be
heard in Fusion.

For Fusion Studio, audio can be loaded using the Loader node’s Audio tab. The audio functionality is
included in Fusion Studio for scratch track (aligning effects to audio and clip timing) purposes. Final
renders should almost always be performed without audio. Audio can be heard if it is brought in
through a Loader node.

To hear the audio from a specific Loader node:

– Right-click over the Speaker icon and choose the file name that contains the audio
you want to hear.

Audio Toolbar Button

The Audio button in the toolbar is a toggle that can be used to enable or mute audio playback
associated with the clip. Additionally, right-clicking this button displays a contextual menu that can be
used to select a MediaIn node in the Fusion page or an external WAV file in Fusion Studio.

The Current Time Field

The Current Time field at the right of the transport controls displays the frame number for the playhead
position, which corresponds to the frame seen in the viewer. However, you can also enter time values
into this field to move the playhead by specific amounts.
When setting ranges and entering frame numbers to move to a specific frame, numbers can be
entered in sub-frame increments. You can set a range to be –145.6 to 451.75 or set the playhead to
115.22. This can be very helpful when animating parameters because you can set keyframes where
they actually need to occur, rather than on a frame boundary, so you get more natural animation.
Having sub-frame time lets you use time remapping nodes or just scale keyframes in the Spline view
and maintain precision.

The Fusion Page Viewer Quality and Proxy Options

Right-clicking anywhere in the transport control area other than over the Play Forward/Play Reverse
buttons lets you turn on and off Fusion quality controls. You can either enable high-quality playback at
the expense of more significant processing times or enter various proxy modes that temporarily lower
the display quality of your composition to speed processing as you work.
Rendering for final output is always done at the highest quality, regardless of these settings.

High Quality
As you build a composition, often the quality of the displayed image is less important than the
speed at which you can work. The High Quality setting gives you the option to either display
images with faster interactivity or at final render quality. When you turn off High Quality, complex
and time-consuming operations such as area sampling, anti-aliasing, and interpolation are skipped
to render the image to the viewer more quickly. Enabling High Quality forces a full-quality render
to the viewer that’s identical to what is output during final delivery.

Motion Blur
The Motion Blur button is a global setting. Turning off Motion Blur temporarily disables motion blur
throughout the composition, regardless of any individual nodes for which it’s enabled. This can

Chapter 2 Exploring the Fusion Interface 26

 significantly speed up renders to the viewer. Individual nodes must first have motion blur enabled
before this button has any effect.

Proxy
The Proxy setting is a draft mode used to speed processing while you’re building your composite.
Turning on Proxy reduces the resolution of the images that are rendered to the viewer, speeding
render times by causing only one out of every x pixels to be processed, rather than processing
every pixel. The value of x is decided by adjusting a slider in the General panel of the Fusion
Settings, found in the Fusion menu.

Auto Proxy
The Auto Proxy setting is a draft mode used to speed processing while you’re building your
composite. Turning on Auto Proxy reduces the resolution of the image while you click and drag to
adjust a parameter. Once you release that control, the image snaps back to its original resolution.
This lets you adjust processor-intensive operations more smoothly, without the wait for every
frame to render at full quality causing jerkiness. You can set the auto proxy ratio by adjusting a
slider in the General panel of the Fusion Settings found in the Fusion menu.

Selective Updates
When working in Fusion, only the tools needed to display the images in the viewer are updated. The
Selective Update options select the mode used during previews and final renders.
The options are available in the Fusion Preferences General panel. The three options are:
– Update All (All): Forces all the nodes in the current node tree to render. This is primarily used
when you want to update all the thumbnails displayed in the Node Editor.
– Selective (Some): Causes only nodes that directly contribute to the current image to be rendered.
So named because only selective nodes are rendered. This is the default setting.
– No Update (None): Prevents rendering altogether, which can be handy for making many changes
to a slow-to-render composition.

Transport Controls in Fusion Studio

The transport controls in Fusion Studio include buttons to control playback, time fields on the left side
for setting the global range and render range, and a Render button to initiate rendering of the
composite. There are also controls on the right side for proxy and motion blur. The time field on the far
right is used for the current time.

Fusion Studio transport controls

Controlling Playback
There are eight transport controls underneath the Time Ruler in Fusion Studio. These buttons include
Composition First Frame, Step Backward, Play Reverse, Stop, Play Forward, Step Forward,
Composition Last Frame, and Loop.

Fusion Studio transport controls

Chapter 2 Exploring the Fusion Interface 27

Navigation Shortcuts
Many standard transport control keyboard shortcuts you may be familiar with work in Fusion, but there
are some keyboard shortcuts specific to Fusion’s particular needs.

To move the playhead in the Time Ruler using the keyboard, do one of the following:
– Spacebar: Toggles forward playback on and off.
– JKL: Basic JKL playback is supported, including J to play backward, K to stop,
and L to play forward.
– Back Arrow: Moves 1 frame backward.
– Forward Arrow: Moves 1 frame forward.
– Shift-Back Arrow: Moves to the clip’s Global End frame.
– Shift-Forward Arrow: Moves to the clip’s Global Start frame.
– Command-Back Arrow: Jumps to the Render Range In point.
– Command-Forward Arrow: Jumps to the Render Range Out point.

Real-Time Playback Not Guaranteed

Because many of the effects you can create in the Fusion page are processor-intensive, there is no
guarantee of real-time playback at your project’s full frame rate, unless you’ve cached your
composition first. For more information, see the “Fusion RAM Cache for Playback” section later in
this chapter.

Frame Increment Options

Right-clicking the Step Backward, Play Reverse, Play Forward, or Step Forward buttons opens a
drop-down menu with options to set a frame increment value. Selecting a frame number from the
menu lets you move the playhead in sub-frame or multi-frame increments whenever you use a
keyboard shortcut or press the Step Forward/Backward buttons.
Moving the playhead in multi-frame increments can be useful when rotoscoping. Moving the playhead
in sub-frame increments can be useful when rotoscoping or inspecting interlaced frames one field at a
time (0.5 of a frame).

Right-click the Step Forward or Step

Backward buttons to choose a frame
increment in which to move the playhead.

Chapter 2 Exploring the Fusion Interface 28

Looping Options
The Loop button can be toggled to enable or disable looping during playback. You can right-click this
button to choose the looping method that’s used:
– Playback Loop: The playhead plays to the end of the Time Ruler and starts from the
beginning again.
– Ping-pong Loop: When the playhead reaches the end of the Time Ruler, playback reverses
until the playhead reaches the beginning of the Time Ruler, and then continues to ping-pong back
and forth.

Range Fields
The four time fields on the left side of the transport controls are used to quickly modify the global
range and render range in Fusion Studio.

Global Time Fields to the left of the transport controls

Audio
The Audio button is a toggle that mutes or enables any audio associated with the clip. Additionally,
right-clicking on this button displays a drop-down menu that can be used to select a WAV file, which
can be played along with the composition, and to assign an offset to the audio playback.

Render
Clicking the Render button in the transport controls displays the composition’s Render Settings dialog.
This dialog is used to configure the render options and initiate rendering of any Saver nodes in the
composition. Shift-clicking on the button skips the dialog, using default render values (full resolution,
high quality, motion blur enabled).

The Current Time

The Current Time field at the right of the transport controls shows the frame at the position of the
playhead, which corresponds to the frame seen in the viewer. However, you can also enter time values
into this field to move the playhead by specific amounts.
When setting ranges and entering frame numbers to move to a specific frame, numbers can be
entered in sub-frame increments. You can set a range to be –145.6 to 451.75 or set the playhead to
115.22. This can be very helpful when animating parameters because you can set keyframes where
they actually need to occur, rather than on a frame boundary, so you get more natural animation.
Having sub-frame time lets you use time remapping nodes or just scale keyframes in the Spline view
and maintain precision.

NOTE: Many fields in Fusion can evaluate mathematical expressions that you type into them.
For example, typing 2 + 4 into most fields results in the value 6.0 being entered. Because
Feet + Frames uses the + symbol as a separator symbol rather than a mathematical symbol,
the Current Time field will not correctly evaluate mathematical expressions that use the +
symbol, even when the display format is set to Frames mode.

Fusion Studio Viewer Quality and Proxy Options

Five buttons along the right side of the transport controls let you either enable high-quality playback at
the expense of greater processing times, or enter various proxy modes that temporarily lower the
display quality in order to speed processing as you work.

Chapter 2 Exploring the Fusion Interface 29

Rendering for final output is always done at the highest quality, regardless of these settings.

Five buttons control the viewer quality, motion blur,

proxy options, and image-processing update settings.

HiQ
As you build a composition, often the quality of the displayed image is less important than the speed
at which you can work. The High Quality setting gives you the option to either display images with
faster interactivity or at final render quality. When you turn off High Quality, complex and time-
consuming operations such as area sampling, anti-aliasing, and interpolation are skipped to render the
image to the viewer more quickly. Enabling High Quality forces a full-quality render to the viewer that’s
identical to what will be output during final delivery.

MB
The Motion Blur button is a global setting. Turning off Motion Blur temporarily disables motion blur
throughout the composition, regardless of any individual nodes for which it’s enabled. This can
significantly speed up renders to the viewer. Individual nodes must first have motion blur enabled
before this button has any effect.

Prx
A draft mode to speed processing while you’re building your composite. Turning on Proxy reduces the
resolution of the images that are rendered to the viewer, speeding render times by causing only one
out of every x pixels to be processed, rather than processing every pixel. The value of x is decided by
adjusting a slider in the General panel of the Fusion Preferences, found under the Fusion menu on
macOS or the File menu on Windows and Linux.

Aprx
A draft mode to speed processing while you’re building your composite. Turning on Auto Proxy
reduces the resolution of the image while you click and drag to adjust a parameter. Once you release
that control, the image snaps back to its original resolution. This lets you adjust processor-intensive
operations more smoothly, without the wait for every frame to render at full quality causing jerkiness.
You can set the auto proxy ratio by adjusting a slider in the General panel of the Fusion Preferences,
found under the Fusion menu on macOS or the File menu on Windows and Linux.

Selective Updates
The last of the five buttons on the right of the transport controls is a three-way toggle that determines
when nodes update images in the viewer. By default, when working in Fusion, any node needed to
display the image in the viewer is updated. The Selective Update button can change this behavior
during previews and final renders.
The three options are:
– Update All (All): Forces all the nodes in the current node tree to render. This is primarily used
when you want to update all the thumbnails displayed in the Node Editor.
– Selective (Some): Causes only nodes that directly contribute to the current image to be rendered.
So named because only selective nodes are rendered. This is the default setting.
– No Update (None): Prevents rendering altogether, which can be handy for making a lot of changes
to a slow-to-render composition.

The options are also available in the Fusion Preferences General panel.

Chapter 2 Exploring the Fusion Interface 30

Changing the Time Display Format
By default, all time fields and markers in Fusion count in frames, but you can also set the time display
to SMPTE timecode or Feet + Frames.

To change the time display format:

1 Choose Fusion > Fusion Settings in DaVinci Resolve, choose Fusion Studio > Preferences in
Fusion Studio on macOS, or choose File > Preferences in Fusion Studio on Windows or Linux.
2 When the Fusion settings dialog opens, select the Global and Default settings > Defaults panel
and choose a Timecode option.
3 Select the Frame Format panel. If you’re using timecode, choose a frame rate and turn on the
“has fields” checkbox if your project is interlaced. If you’re using feet and frames, set the Film Size
value to match the number of frames found in a foot of film in the format used in your project.
4 Click Save.

Keyframe Display in the Time Ruler

When you select a node with keyframed parameters, those keyframes appear in the Time Ruler as
little white tic marks, letting you navigate among and edit keyframes without having to open the
Keyframes Editor or Spline Editor to see them.

The Time Ruler displaying keyframe marks

To move the playhead in the Time Ruler among keyframes:

– Press Option-Left Bracket ([) to jump to the next keyframe to the left.
– Press Option-Right Bracket (]) to jump to the next keyframe to the right.

The Fusion RAM Cache for Playback

When assembling a node tree, all image processing operations are rendered live to display the final
result in the viewers. However, as each frame is rendered, and especially as you initiate playback
forward or backward, these images are automatically stored to a RAM cache as they’re processed so
you can replay those frames in real time. The actual frame rate achieved during playback is displayed
in the Status bar at the bottom of the Fusion window during playback. Of course, when you play
beyond the cached area of the Time Ruler, uncached frames need to be rendered before being added
to the cache.
Priority is given to caching nodes that are currently being displayed, based on which nodes are loaded
to which viewers. However, other nodes may also be cached, depending on available memory and on
how processor-intensive those nodes happen to be, among other factors.

Memory Limits of the RAM Cache

There is a single setting in DaVinci Resolve for limiting the RAM used for caching. This setting is
located in the DaVinci Resolve Preferences Memory and GPU panel.
– Limit Fusion Memory Cache To: This slider sets the maximum amount of RAM that Fusion can
access for caching. It is a subset of the RAM allocated to DaVinci Resolve. You can assign a
maximum of 75% to Fusion from DaVinci Resolve’s total RAM allocation. When not using the Fusion
page, the RAM is released for other pages in DaVinci Resolve.

Chapter 2 Exploring the Fusion Interface 31

There are two settings in Fusion Studio for limiting the RAM used for caching. These settings are
located in the Preferences Memory panel.
– Limit Caching To: This slider sets the maximum amount of RAM used for caching. The 60% default
setting on a 32-GB system limits the cache to 19.2 GB. The maximum amount you can assign to
Fusion Studio is limited to 80% of the total system memory. This leaves a minimum amount of
memory for other applications and the operating system.
– Leave at least # MBytes: This number field further limits caching in cases where the system’s
available free RAM drops below the entered value. For instance, setting this to 200 MB attempts
to keep 200 MB of RAM free for the OS or other applications. Setting the number field to 0 allows
Fusion Studio to use the full amount of RAM specified by the Limit Caching To setting, ignoring
other apps.

When the size of the cache reaches the Fusion Caching/Memory Limits setting found in the Memory
panel of the Preferences, then lower-priority cache frames are automatically discarded to make room
for new caching. You can keep track of the RAM cache usage via a percentage indicator on the far
right of the Status bar at the bottom of the Fusion window.

Displaying Cached Frames

All cached frames for the currently viewed node are indicated by a green line at the bottom of the
Time Ruler. Any green section of the Time Ruler should play back in real time.

The green lines indicate frames that have been cached for playback.

Temporarily Preserving the Cache When Changing Quality or Proxy Settings

If you toggle the composition’s quality settings or proxy options, the cache is not immediately
discarded. The green line instead turns red to let you know the cache is being preserved and can be
used again when you go back to the original level of quality or disable proxy mode. However, if you
play through those frames at the new quality or proxy settings, this preserved cache is overwritten
with a new cache at the current quality or proxy setting.

A red line indicates that cached frames from a different

quality or proxy setting are being preserved.

There’s one exception to this, however. When you cache frames at the High Quality setting, and you
then turn off High Quality, the green frames won’t turn red. Instead, the High Quality cached frames
are used even though the HiQ setting has been disabled.

Chapter 2 Exploring the Fusion Interface 32

Toolbar
The toolbar, located underneath the Time Ruler, contains buttons that let you quickly add commonly
used nodes to the Node Editor. Clicking any of these buttons adds that node after the currently
selected node in the node tree, or adds an unconnected instance of that node if no nodes are
selected. The Toolbar can be customized and saved for specific tasks.

The toolbar has buttons for adding commonly used nodes to the Node Editor.

The default toolbar is divided into sections that group commonly used nodes together. As you hover
the pointer over any button, a tooltip shows you that node’s name.
– Loader/Saver nodes (Fusion Studio Only): The Loader node is the primary node used to
select and load clips from the hard drive. The Saver node is used to write or render your
composition to disk.
– Generator/Title/Paint nodes: The Background and FastNoise generators are commonly used to
create all kinds of effects, and the Title generator is obviously a ubiquitous tool, as is Paint.
– Color/Blur nodes: ColorCorrector, ColorCurves, HueCurves, and BrightnessContrast are the four
most commonly used color adjustment nodes, while the Blur node is ubiquitous.
– Compositing/Transform nodes: The Merge node is the primary node used to composite one
image against another. ChannelBooleans and MatteControl are both essential for reassigning
channels from one node to another. Resize alters the resolution of the image, permanently altering
the available resolution, while Transform applies pan/tilt/rotate/zoom effects in a resolution-
independent fashion that traces back to the original resolution available to the source image.
– Mask nodes: Rectangle, Ellipse, Polygon, and BSpline mask nodes let you create shapes to use
for rotoscoping, creating garbage masks, or other uses.
– Particle system nodes: Three particle nodes let you create complete particle systems when you
click them from left to right. pEmitter emits particles in 3D space, while pMerge lets you merge
multiple emitters and particle effects to create more complex systems. pRender renders a 2D
result that can be composited against other 2D images.
– 3D nodes: Seven 3D nodes let you build sophisticated 3D scenes. These nodes auto attach to
one another to create a quick 3D template when you click from left to right. ImagePlane3D lets you
connect 2D stills and movies for compositing into 3D scenes. Shape3D lets you create geometric
primitives of different kinds. Text3D lets you build 3D text objects. Merge3D lets you composite
multiple 3D image planes, primitive shapes, and 3D text together to create complex scenes, while
SpotLight lets you light the scenes in different ways, and Camera3D lets you frame the scene in
whatever ways you like. Renderer3D renders the final scene and outputs 2D images and auxiliary
channels that can be used to composite 3D output against other 2D layers.

When you’re first learning to use Fusion, these nodes are really all you need to build most common
composites. Once you’ve become a more advanced user, you’ll still find that these are truly the most
common operations you’ll use.

Customizing the Toolbar

You can add and remove tools from the Fusion toolbar and then save the custom toolbar as a preset.
New tools can be added by dragging them from the Effects Library or the Node Editor, and dividers
can be added to group tool sets together.

Chapter 2 Exploring the Fusion Interface 33

To create a new toolbar, do the following:
1 Right click in an empty area of the toolbar and choose Customize > Create Toolbar from the
contextual menu.
2 Enter a name for the toolbar in the dialog box, and click OK.

To rearrange the tools in the toolbar, do the following:

1 Create a new custom toolbar or select an existing custom toolbar.
2 Drag a node in the toolbar to a new location.

To add a tool to the toolbar:

1 Create a new custom toolbar or select an existing custom toolbar.
2 Do one of the following:
– Drag a node from the Effects Library to the location on the toolbar where you want it added.
– Drag a node from the Node Editor to the location on the toolbar where you want it added.

To add a divider to a toolbar, do the following:

1 Create a new custom toolbar or select an existing custom toolbar.
2 Right-click over any tool and choose Customize > Add Divider. A divider is added to the right
of the tool.

To remove a tool from a toolbar, do the following:

1 Create a new custom toolbar or select an existing custom toolbar.
2 Right-click over any tool and choose Remove [name of tool].

To remove a group of tools between two dividers, do the following:

1 Create a new custom toolbar or select an existing custom toolbar.
2 Right-click over any tool in a group and choose Remove Group.

To prevent a custom toolbar from being modified:

– Right-click over the toolbar and choose Lock from the contextual menu.

To switch between toolbars:

– Right-click over the toolbar and choose the custom toolbar name or choose Default to return to
Fusion’s default toolbar.

To delete a custom toolbar, do the following:

1 Right-click over the toolbar and choose the name of the custom toolbar you want to delete.
2 Right-click over the toolbar again and choose Customize > Remove [toolbar name].

Chapter 2 Exploring the Fusion Interface 34

Node Editor
The Node Editor is the heart of Fusion because it’s where you build the tree of nodes that makes up
each composition. Each node you add to the node tree adds a specific operation that creates one
effect, whether it’s blurring the image, adjusting color, painting strokes, drawing and adding a mask,
extracting a key, creating text, or compositing two images into one.
You can think of each node as a layer in an effects stack, except that you have the freedom to route
image data in any direction to branch and merge different segments of your composite in completely
nonlinear ways. This makes it easy to build complex effects, but it also makes it easy to see what’s
happening, since the node tree doubles as a flowchart that clearly shows you everything that’s
happening, once you learn to read it.

The Node Editor displaying a node tree creating a composition

Adding Nodes to Your Composition

Depending on your mood, there are a few ways you can add nodes from the Effects Library to your
composition. For most of these methods, if there’s a single selected node in the Node Editor, new
nodes are automatically added after it, but if there are no selected nodes or multiple selected nodes,
then new nodes are added as disconnected from anything else.

Methods of adding nodes include:

– Click a button in the toolbar.
– Open the Effects Library, find the node you want in the relevant category, and click once on a
node you’d like to add.
– Right-click on a node and choose Insert Tool from the drop-down menu to add it after the node
you’ve right-clicked on. Or, you can right-click on the background of the Node Editor to use that
submenu to add a disconnected node.
– Press Shift-Spacebar to open a Select Tool dialog, type characters corresponding to the name of
the node you’re looking for, and press the Return key (or click OK) when it’s found. Once you learn
this method, it’ll probably become one of your most frequently used ways of adding nodes.

Chapter 2 Exploring the Fusion Interface 35

 The Select Tool dialog lets you find any
node quickly if you know its name.

Removing Nodes from Your Composition

Removing nodes is as simple as selecting one or more nodes, and then pressing the Delete or
Backspace keys.

Identifying Node Inputs and Node Outputs

Each node displays small colored connections around the edges. One or more arrows represent
inputs, and the square connection represents the tool’s processed output, of which there is always
only one. If you hover the pointer over any of a node’s inputs or output, the name of that input or
output immediately appears in the Status bar. If you wait for a few more moments, a floating tooltip
displays the same name right over the node.

Node Editing Essentials

Each node has inputs and outputs that are “wired together” using connections. The inputs are
represented by arrows that indicate the flow of image data from one node to the next, as each node
applies its effect and feeds the result (via the square output) to the next node in the tree. In this way,
you can quickly build complex results from a series of relatively simple operations.

Three nodes connected together

Chapter 2 Exploring the Fusion Interface 36

You can connect a single node’s output to the inputs of multiple nodes (called “branching”).

One node branching to two to split the image to two operations

You can then composite images together by connecting the output from multiple nodes to certain
nodes such as the Merge node that combines multiple inputs into a single output.

Two nodes being merged together into one to create a composite

By default, new nodes are added from left to right in the Node Editor, but they can also flow from top
to bottom, right to left, bottom to top, or in all directions simultaneously. Connections automatically
reorient themselves along all four sides of each node to maintain the cleanest possible presentation as
you rearrange other connected nodes.

Nodes can be oriented in any direction; the input arrows let you follow the flow of image data.

Chapter 2 Exploring the Fusion Interface 37

Navigating the Node Editor
As your Node tree gets larger, parts inevitably go offscreen. When a portion of the node tree is
offscreen, a resizable Navigator pane appears in the upper-right corner. The Navigator is a miniature
representation of the entire node tree that you can drag within to pan to different parts of your
composition quickly. You can resize the navigator using a handle in the lower-left corner, and you can
choose to show or hide the Navigator by right-clicking the Node Editor to access the Options submenu
of the contextual menu.

The Navigator pane for accessing offscreen parameters or tools

There are other standard methods of panning and zooming around the Node Editor.

Methods of navigating the Node Editor:

– Middle-click and drag to pan around the Node Editor.
– Hold down Shift and Command and drag the Node Editor to pan.
– Press the Middle and Left mouse buttons simultaneously and drag to resize the Node Editor.
– Hold down the Command key, and use your mouse’s scroll control to resize the Node Editor.
– Right-click the Node Editor and choose an option from the Scale submenu of the contextual menu.
– Press Command-1 to reset the Node Editor to its default size.
– Drag two fingers on a track pad to pan.
– Hold the Command key down and drag two fingers on a track pad to resize the Node Editor.

Vertical Node Editor Layouts

Alternative Node View Layout presets located in the Fusion page allow for positioning the Node Editor
vertically, either alongside the Inspector, or along the left side of the screen. This can be very helpful
when animating in the Spline Editor or Keyframes Editor.
When in the Fusion page, you can choose the layouts from Workspace > Layout Presets. Choosing a
vertical layout allows the node tree to flow from top to bottom, leaving much more room along the
lower half of the screen for the Spline Editor or Keyframes Editor.

Chapter 2 Exploring the Fusion Interface 38

The Mid Flow Vertical layout preset used with the Vertical Flow direction setting.

When using the vertical layouts, enabling the Flow > Vertical Build Direction option in the Fusion
settings will cause all new Node trees to build vertically, leaving maximum room for Fusion’s
animation tools.
You can then save alternative layouts based on these two vertical presets using the Workspace >
Layout Presets submenu.
When you want to return to the default horizontal Node Editor layout, just choose Workspace > Layout
Presets > Fusion Presets > Default.
These Layout options are not available in Fusion Studio, however, you can use the Floating Frame to
position the Node Editor wherever you like.

Keeping Organized
As you work, it’s important to keep the node trees that you create tidy to facilitate a clear
understanding of what’s happening. Fortunately, the Fusion Node Editor provides a variety of methods
and options to help you with this, found within the Options and Arrange Tools submenus of the Node
Editor contextual menu.

Status Bar
The Status bar in the lower-left corner of the Fusion window shows you a variety of up-to-date
information about things you’re selecting and what’s happening in Fusion. For example, hovering the
pointer over a node displays information about that node in the Status bar. Additionally, the currently
achieved frame rate appears whenever you initiate playback, and the percentage of the RAM cache
that’s used appears at all times. Other information, updates, and warnings appear in this area
as you work.

Chapter 2 Exploring the Fusion Interface 39

The Status bar under the Node Editor showing you information about a node under the pointer

Occasionally the Status bar will display a badge to let you know there’s a message in the console you
might be interested in. The message could be a log, script message, or error.

A notification that there’s a message in the Console

Effects Library
The Effects Library in Fusion shows all the nodes and effects available in Fusion, including third-party
OFX plug-ins, if installed. If you are using DaVinci Resolve, Resolve FX also appear in the OFX
category. While the toolbar shows many of the most common nodes you’ll use in any composite, the
Effects Library contains every single tool available in Fusion, organized by category, with each node
ready to be quickly added to the Node Editor. Suffice it to say that there are many, many more nodes
available in the Effects Library than on the toolbar, spanning a wide range of uses.

The Effects Library with Tools open

The hierarchical category browser of the Effects Library is divided into several sections depending on
whether you are using Fusion Studio or the Fusion page within DaVinci Resolve. The Tools section is
the most often used since it contains every node that represents an elemental image-processing
operation in Fusion. The OpenFX section contains third-party plug-ins, and if you are using the Fusion
page, it also contains ResolveFX, which are included with DaVinci Resolve. A third section, only visible

Chapter 2 Exploring the Fusion Interface 40

when using the Fusion page in DaVinci Resolve, is the Templates section. The Template section
contains a variety of additional content including templates for Lens Flares, Backgrounds, Generators,
Particle Systems, Shaders (for texturing 3D objects), and other resources for use in your composites.

The Templates section of the Effects Library

The Effects Library’s list can be made full height or half height using a button at the far left of
the UI toolbar.

The Inspector
The Inspector is a panel on the right side of the Fusion window that you use to display and manipulate
the parameters of one or more selected nodes. When a node is selected in the Node Editor, its
parameters and settings appear in the Inspector.

The Inspector shows parameters from

one or more selected nodes.

The Tools and Modifiers Panels

The Fusion Inspector is divided into two panels. The Tools panel is the main panel that shows you the
parameters of selected nodes. The Modifiers panel shows optional extensions to the standard
parameters of a tool. In the following image, a Perturb modifier has been added to a parameter to add

Chapter 2 Exploring the Fusion Interface 41

random animation to that parameter, and the controls found on the Modifier panel let you customize
what kind of randomness is being added.

The Modifier panel showing a Perturb modifier

Other nodes display node-specific items here. For example, Paint nodes show each brush stroke as an
individual set of controls in the Modifiers panel, available for further editing or animating.

Parameter Header Controls

A cluster of controls appears at the top of every node’s controls in the Inspector.

Common Inspector controls

– Set Color: A pop-up menu that lets you assign one of 16 colors to a node, overriding a
node’s own color.
– Versions: Clicking Versions reveals another toolbar with six buttons. Each button can hold an
individual set of adjustments for that node that you can use to store multiple versions of an effect.
– Pin: The Inspector is also capable of simultaneously displaying all parameters for multiple nodes
you’ve selected in the Node Editor. Furthermore, a Pin button in the title bar of each node’s
parameters lets you “pin” that node’s parameters into the Inspector so that they remain there even
when that node is deselected, which is valuable for key nodes that you need to adjust even while
inspecting other nodes of your composition.
– Lock: Locks that node so that no changes can be made to it.
– Reset: Resets all parameters within that node.

Parameter Tabs
Many nodes expose multiple tabs’ worth of controls in the Inspector, seen as icons at the top of the
parameter section for each node. Click any tab to expose that set of controls.

Nodes with several tabs’ worth of parameters

Chapter 2 Exploring the Fusion Interface 42

Keyframes Editor
The Keyframes Editor displays each node in the current composition as a stack of layers within a
miniature timeline. The order of the layers is largely irrelevant as the order and flow of connections in
the node tree dictate the order of image-processing operations. You use the Keyframes Editor to trim,
extend, or slide Loader, MediaIn, and effects nodes, or to adjust the timing of keyframes, which appear
superimposed over each effect node unless you open them up into their editable track.

The Keyframes Editor is used to adjust the timing of clips, effects, and keyframes.

Keyframes Editor Control Summary

At the top, a series of zoom and framing controls let you adjust the work area containing the layers.
– A Horizontal zoom control lets you scale the size of the editor.
– A Zoom to Fit button fits the width of all layers to the current width of the Keyframes Editor.
– A Zoom to Rect tool lets you draw a rectangle to define an area of the Keyframes
Editor to zoom into.
– A Sort pop-up menu lets you sort or filter the tracks in various ways.
– An Option menu provides access to many other ways of filtering tracks and
controlling visible options.

A timeline ruler provides a time reference, as well as a place in which you can scrub the playhead.
At the left, a track header contains the name of each layer, as well as controls governing that layer.
– A lock button lets you prevent a particular layer from being changed.
– Nodes that have been keyframed have a disclosure control, which when opened displays a
keyframe track for each animated parameter.

In the middle, the actual editing area displays all layers and keyframe tracks available in the current
composition.
At the bottom-left, Time Stretch and Spreadsheet mode controls provide additional ways to manipulate
keyframes.
At the bottom right, the Time/TOffset/TScale drop-down menu and value fields let you numerically
alter the position of selected keyframes either absolutely, relatively, or based on their distance from
the playhead.

Chapter 2 Exploring the Fusion Interface 43

Adjusting Clip Timings
Each Loader or MediaIn node that represents a clip used in a composition is represented as a layer in
this miniature timeline. You can edit a layer’s In or Out points by positioning the pointer over the
beginning or end of a segment and using the resize cursor to drag that point to a new location. You
can slide a layer by dragging it to the left or right, to better line up with the timing of other elements in
your composition.
The Keyframes Editor also lets you adjust the timing of elements that you’ve added from directly
within Fusion.

Adjusting Effect Timings

Each effect node also appears as a layer, just like clips. You can resize the In and Out points of an
effect layer, and slide the entire layer forward or backward in time, just like a Loader or MediaIn layers.
If you trim an effect layer to be shorter than the duration of the composition, the effect cuts in at
whichever frame the layer begins, and cuts out after the last frame of that layer, just like a clip on
a timeline.

Adjusting Keyframe Timings

When you’ve animated an effect by adding keyframes to a parameter in the Inspector, the Keyframes
Editor is used to edit the timing of keyframes easily. By default, all keyframes applied to parameters
within a particular node’s layer appear superimposed in one flat track over the top of that layer.
To edit keyframes, you can click the disclosure control to the left of any animated layer’s name in the
track header, which opens up keyframe tracks for every keyframed parameter within that layer.

Keyframe tracks exposed

Keyframe Editing Essentials

Here’s a short list of keyframe editing methods that will get you started.

Methods of adjusting keyframes:

– You can click on a single keyframe to select it.
– You can drag a bounding box over a series of keyframes to select them all.
– You can drag keyframes left and right to reposition them in time.
– You can right-click one or more selected keyframes and use commands from the drop-down menu
to change keyframe interpolation, copy/paste keyframes, or even create new keyframes.
– You can Option-drag one or more selected keyframes to drag a duplicate of them to another
position in the keyframe track.

To change the position of a keyframe using the toolbar, do one of the following:
– Select a keyframe, and then enter a new frame number in the Time Edit box.
– Choose T Offset from the Time Editor pop-up, select one or more keyframes, and enter a
frame offset.

Chapter 2 Exploring the Fusion Interface 44

 – Choose T Scale from the Time Editor pop-up, select one or more keyframes, and enter a multiplier
added to the current playhead frame position. For instance, if the playhead is on frame 10 and the
keyframe is on frame 30, entering the TScale value of 2 will position the keyframe on frame 50.
The distance between the playhead and original keyframe is 20, so (20 x 2) = 40, which is then
added to the playhead position.

Time Stretching Keyframes

If you select a range of keyframes in a keyframe track, you can turn on the Time Stretch tool to show a
box used to squeeze and stretch the entire range of keyframes relative to one another. The Time
Stretcher changes the overall timing of a sequence of keyframes without losing the relative timing from
one keyframe to the next. Alternatively, you can turn on Time Stretch and draw a bounding box around
the keyframes you want to adjust to create a time-stretching boundary that way. Click the Time Stretch
tool again to turn it off.

Time stretching keyframes

The Keyframe Spreadsheet

If you turn on the Spreadsheet and then click on the name of a layer in the keyframe track, the numeric
time position and value (or values if it’s a multi-dimensional parameter) of each keyframe appear as
entries in the cells of the Spreadsheet. Each column represents one keyframe, while each row
represents a single aspect of each keyframe.

Editing keyframes in the Spreadsheet

For example, if you’re animating a motion path, then the “Key Frame” row shows the frame each
keyframe is positioned at, and the “Path1Displacement” row shows the position along the path at each
keyframe. If you change the Key Frame value of any keyframe, you’ll move that keyframe to a new
frame of the Timeline.

Spline Editor
The Spline Editor provides a more detailed environment for editing the timing, value, and interpolation
of keyframes. Using control points at each keyframe connected by splines (also called curves), you can
adjust how animated values change over time. The Spline Editor has four main areas: the Zoom and
Framing controls at the top, the Parameter list at the left, the Graph Editor in the middle, and the
toolbar at the bottom.

Chapter 2 Exploring the Fusion Interface 45

The Spline Editor is divided into the Zoom controls at
top, Parameter list at left, Graph Editor, and toolbar.

Spline Editor Control Summary

At the top, a series of Zoom and Framing controls let you adjust the work area containing the curves.
– Vertical and horizontal zoom controls let you scale the size of the editor.
– A Zoom to Fit button fits the width of all curves to the current width of the Spline Editor.
– A Zoom to Rect tool lets you draw a rectangle to define an area of the Spline Editor to zoom into.
– A Sort drop-down menu lets you sort or filter the curves in various ways.
– An Option menu provides access to many other ways of filtering curves and controlling
visible options.

A Timeline ruler provides a time reference, as well as a place in which you can scrub the playhead.
The Parameter list at the left is where you decide which splines are visible in the Graph view. By
default, the Parameter list shows every parameter of every node in a hierarchical list. Checkboxes
beside each name are used to show or hide the curves for different keyframed parameters. Color
controls let you customize each spline’s tint to make splines easier to see in a crowded situation.
The Graph view that takes up most of this panel shows the animation spline along two axes. The
horizontal axis represents time, and the vertical axis represents the spline’s value. Selected control
points show their values in the edit fields at the bottom of the graph.
Lastly, the toolbar at the bottom of the Spline Editor provides controls to set control point interpolation,
spline looping, or choose Spline editing tools for different purposes.

Choosing Which Parameters to Show

Before you start editing splines to customize or create animation, you need to choose which
parameter’s splines you want to work on.

To show every parameter in every node:

– Click the Splines Editor Option menu and choose Expose All Controls. Toggle this control off again
to go back to viewing what you were looking at before.

To show splines for the currently selected node:

– Click the Splines Editor Option menu and choose Show Only Selected Tool.

Chapter 2 Exploring the Fusion Interface 46

Essential Spline Editing
The Spline Editor is a deep and sophisticated environment for keyframe and spline editing
and retiming, but the following overview will get you started using this tool for creating and
refining animation.

To select one or more control points:

– Click any control point to select it.
– Command-click multiple control points to select them.
– Drag a bounding box around multiple control points to select them as a group.

To edit control points and splines:

– Click anywhere on a spline to add a control point.
– Drag one or more selected control points to reshape the spline.
– Shift-drag a control point to constrain its motion vertically or horizontally.

To edit Bézier curves:

– Select any control point to make its Bézier handles visible, and drag the Bézier handles.
– Command-drag a Bézier handle to break the angle between the left and right handles.

To delete control points:

– Select one or more control points and press the Delete or Backspace key.

Essential Spline Editing Tools and Modes

The Spline Editor toolbar at the bottom contains a mix of control point interpolation buttons, Spline
loop modes, and Spline editing tools.

Control Point Interpolation

The first six buttons let you adjust the interpolation of one or more selected control points.

Control point interpolation controls

– Smooth: Creates automatically adjusted Bézier curves to create smoothly interpolating animation.
– Flat: Creates linear interpolation between control points.
– Invert: Inverts the vertical position of non-animated LUT splines. This does not operate on
animation splines.
– Step In: For each keyframe, creates sudden changes in value at the next keyframe to the right.
Similar to a hold keyframe in After Effects® or a static keyframe in the DaVinci Resolve Color page.
– Step Out: Creates sudden changes in value at every keyframe for which there’s a change in value
at the next keyframe to the right. Similar to a hold keyframe in After Effects or a static keyframe in
the DaVinci Resolve Color page.
– Reverse: Reverses the horizontal position of selected keyframes in time, so the
keyframes are backward.

Chapter 2 Exploring the Fusion Interface 47

Spline Loop Modes
The next three buttons let you set up spline looping after the last control point on a parameter’s spline,
enabling a limited pattern of keyframes to animate over a far longer duration. Only the control points
you’ve selected are looped.

Spline Loop modes

– Set Loop: Repeats the same pattern of keyframes over and over.
– Set Ping Pong: Repeats a reversed set of the selected keyframes and then a duplicate set of the
selected keyframes to create a more seamless pattern of animation.
– Set Relative: Repeats the same pattern of selected keyframes but with the values of each
repeated pattern of keyframes being incremented or decremented by the trend of all keyframes
in the selection. This results in a loop of keyframes where the value either steadily increases or
decreases with each subsequent loop.

Spline Editing Tools

The next five buttons provide specialized Spline editing tools.

Spline editing controls

– Select All: Selects every keyframe currently available in the Spline Editor.
– Click Append: Click once to select this tool and click again to de-select it. This tool lets you add
or adjust keyframes and spline segments (sections of splines between two keyframes) depending
on the keyframe mode you’re in. With Smooth or Linear keyframes, clicking anywhere above or
below a spline segment adds a new keyframe to the segment at the location where you clicked.
With Step In or Step Out keyframes, clicking anywhere above or below a line segment moves that
segment to where you’ve clicked.
– Time Stretch: If you select a range of keyframes, you can turn on the Time Stretch tool to show a
box you can use to squeeze and stretch the entire range of keyframes relative to one another, to
change the overall timing of a sequence of keyframes without losing the relative timing from one
keyframe to the next. Alternatively, you can turn on Time Stretch and draw a bounding box around
the keyframes you want to adjust to create a time-stretching boundary that way. Click Time Stretch
a second time to turn it off.
– Shape Box: Turn on the Shape Box to draw a bounding box around a group of control points you
want to adjust in order to horizontally squish and stretch (using the top/bottom/left/right handles),
corner pin (using the corner handles), move (dragging on the box boundary), or corner stretch
(Command-drag the corner handles).
– Show Key Markers: Turning on this control shows keyframes in the top ruler that correspond to the
frame at which each visible control point appears. The colors of these keyframes correspond to
the color of the control points they’re indicating.

Chapter 2 Exploring the Fusion Interface 48

Thumbnail Timeline in the Fusion Page
In the Fusion page of DaVinci Resolve, the Thumbnail timeline (hidden by default) can be opened by
clicking the Clips button in the UI toolbar and appears underneath the Node Editor when it’s open. The
Thumbnail timeline shows every clip in the current Timeline, giving you a way to navigate from one clip
to another. Each thumbnail has a pop-up menu for creating and switching among multiple versions of
compositions, and resetting the current composition, when necessary.

The Thumbnail timeline lets you navigate the Timeline and manage versions of compositions.

Right-clicking on any thumbnail exposes a contextual menu.

The contextual menu for the Thumbnail timeline

To open another clip:

– Click any thumbnail to jump to that clip’s composition. The current clip is outlined in orange.

To create and manage versions of compositions:

– To create a new version of a composition: Right-click the current thumbnail, and choose Create
New Composition from the contextual menu.
– To load a different composition: Right-click the current thumbnail, and choose “NameOfVersion”
> Load from the contextual menu.
– To delete a composition: Right-click the current thumbnail, and choose “NameOfVersion” > Delete
from the contextual menu.

To reset the current composition:

– Right-click the current thumbnail, and choose Reset Current Composition from the
contextual menu.

To change how thumbnails are identified:

– Double-click the area underneath any thumbnail to toggle among clip format, clip name, and
a mystery that shall someday be solved by an intrepid team of adventurers embarking on a
dangerous quest.

Chapter 2 Exploring the Fusion Interface 49

The Media Pool in the Fusion Page
In DaVinci Resolve’s Fusion page, the Media Pool continues to serve its purpose as the repository of
all media you’ve imported into your project. This makes it easy to add additional clips to your
compositions simply by dragging the clip you want from the Media Pool into the Node Editor. The
media you add appears as a new MediaIn node in your composition, ready to be integrated into your
node tree however you need it.

The Media Pool in Thumbnail mode showing video clips

TIP: If you drag one or more clips from the Media Pool onto a connection line between two
nodes in the Node Editor, the clips are automatically connected to that line via enough Merge
nodes to connect them all.

For more information on using the myriad features of the Media Pool, see Chapter 17, “Adding and
Organizing Media with the Media Pool” in the DaVinci Resolve Reference Manual.

Importing Media Into the Media Pool on the Fusion Page

If you find yourself in the Fusion page and you need to quickly import a few clips for immediate use,
you can do so in a couple of different ways.

To add media by dragging one or more clips from the Finder to the Fusion page Media Pool
(macOS only):
1 Select one or more clips in the Finder.
2 Drag those clips into the Media Pool of DaVinci Resolve, or to a bin in the Bin list. Those clips are
added to the Media Pool of your project.

Chapter 2 Exploring the Fusion Interface 50

To use the Import Media command in the Fusion page Media Pool:
1 With the Fusion page open, right-click anywhere in the Media Pool, and choose Import Media.
2 Use the Import dialog to select one or more clips to import, and click Open. Those clips are added
to the Media Pool of your project.

For more information on importing media using the myriad features of the Media page, see Chapter 17,
“Adding and Organizing Media with the Media Pool” in the DaVinci Resolve Reference Manual.

Bins in Fusion Studio

Bins in Fusion Studio are similar to the Media Pool in DaVinci Resolve. Bins are organizational panels
that provide an easy way of accessing commonly used tools, settings, macros, compositions, and
media content. They can keep all your custom content and resources close at hand, so you can use
them without searching through your hard drives. Bins can also be shared over a network to improve a
collaborative workflow with other Fusion Studio artists.

The Bins window

To open the Bins window:

– Choose File > Bins from the menu bar.

Similar to the Media Pool in DaVinci Resolve, when adding an item to the Fusion bins, a link is created
between the item on disk and the bins. Fusion does not copy the file into its own cache or hard drive
space. The file remains in its original format and in its original location.

Bins Interface
The Bins window is actually a separate application used to save content you may want to reuse at a
later time. The Bins window is separated into two panels. The sidebar on the left is a bin list where
items are placed into categories, while the panel on the right displays the selected bin’s content.

Chapter 2 Exploring the Fusion Interface 51

 The Bin list sidebar

The Bin list organizes content into bins or folders using a hierarchical List view. These folders can be
organized to suit your workflow, but standard folders are provided for Clips, Compositions, Favorites,
Settings, and Tools. Parent folders contain subfolders that hold the content. For instance, the Tools bin
is a parent folder to all the categories of tools. To access subfolders, click the disclosure arrow to the
left of the parent folder’s name.
When you select a bin from the bin list, the contents of the folder are displayed in the Contents panel
as thumbnail icons.

The Bins icon view

A toolbar along the bottom of the bin provides access to organization, playback, and editing controls.

The Bins toolbar

– New Folder: Creates a new folder in the current window.

– New Reel: Creates an empty reel that can contain multiple clips edited together into a timeline.
– New Clip: Opens a dialog to link a new media file into a bin.
– Studio Player: Opens a playback viewer for a selected clip.

Chapter 2 Exploring the Fusion Interface 52

 – Icon/List view: This button toggles between showing contents of a bin in thumbnail view
and list view.
– Checkerboard: Shows a checkerboard pattern in a clip thumbnail to signify transparency.
– Thumbnail size: Provides a few preset sizes for thumbnail icons.
– Right-clicking in any area of the bin window displays a pop-up menu to access most of a bin’s
features. Right-clicking on an item in a bin shows the same menu with additional options for
renaming, playing, or deleting the item.

The Bin Studio Player

Selecting a clip in a bin and clicking the Studio Player button or double-clicking the clip opens the
Studio Player. The Studio Player can be used to view clips, view metadata, and add notes.

The Bin Studio Player

For more information on Bins and the Studio Player, see Chapter 74, “Bins” in the DaVinci Resolve
Reference Manual or Chapter 13 in the Fusion Reference Manual.

The Console
The Console is a window in which you can see the error, log, script, and input messages that may
explain something Fusion is trying to do in greater detail. The Console is also where you can read
FusionScript outputs, or input FusionScripts directly. In DaVinci Resolve, the Console is available by
choosing Workspace > Console or choosing View > Console in Fusion Studio. There is also a Console
button in the Fusion Studio User Interface toolbar.
Occasionally the Status bar displays a badge to let you know there’s a message in the Console you
might be interested in.

Chapter 2 Exploring the Fusion Interface 53

 The Console window

A toolbar at the top of the Console contains controls governing what the Console shows. At the top
left, the Clear Screen button clears the contents of the Console. The next four buttons toggle the
visibility of error messages, log messages, script messages, and input echoing. Showing only a
particular kind of message can help you find what you’re looking for when you’re under the gun at
3:00 in the morning. The next three buttons let you choose the input script language. Lua 5.1 is the
default and is installed with Fusion. Python 2.7 and Python 3.6 require that you install the appropriate
Python environment on your computer. Because scripts in the Console are executed immediately, you
can switch between input languages at any time.
At the bottom of the Console is an Entry field. You can type scripting commands here for execution in
the current comp context. Scripts are entered one line at a time, and are executed immediately. For
more information on scripting, see the Fusion Scripting Manual.

Customizing Fusion
This section explains how you can customize Fusion to accommodate whatever workflow
you’re pursuing.

The Fusion Settings Window

Fusion has its own settings window, accessible by choosing Fusion > Fusion Settings in
DaVinci Resolve, or in Fusion Studio by choosing Fusion > Preferences on macOS or File > Preferences
on Windows or Linux. This window has a variety of options for customizing the Fusion experience.
For more information, see Chapter 76, “Preferences” in the DaVinci Resolve Reference Manual or
Chapter 15 in the Fusion Reference Manual.

Chapter 2 Exploring the Fusion Interface 54

 The Fusion Settings window set to the General panel

Saving Fusion Layouts

It is possible to customize the layout and configuration of panels to suit the size of the desktop and
monitor, or to match personal preferences.

In DaVinci Resolve, configure and resize the panels you want displayed and then:
– Choose Workspace > Layout Presets > Save Layout Presets.

In Fusion Studio, configure and resize the panels you want displayed and then:
– Click the Grab Document Layout button in the Preferences > Layout panel to save the layout for all
new Compositions.
– Click the Grab Program Layout button to remember the size and position of any floating views, and
enable the Create Floating Views checkbox to automatically create the floating windows when
Fusion restarts.

Showing and Hiding Panels

The UI toolbar at the top of the screen lets you open panels you need and hide those you don’t. It’s
the simplest way to create a layout for your particular needs at the moment.

The UI toolbar of the Fusion page

Chapter 2 Exploring the Fusion Interface 55

Resizing Panels
You can change the overall size of each panel using preset configurations, or you can adjust them
manually. The viewers and Work panel are inverse of each other. The more space used to display the
Work panel, the less space available for the viewers. To resize a panel, manually drag anywhere along
the raised border surrounding the edges of the panel.

Dragging the edge between

two viewers to resize it

Fusion Studio Floating Frame

Fusion Studio includes a Floating Frame window that can be used to house any panel.

To place a panel in the Floating Frame, do the following:

1 In Fusion Studio, choose Window > New Floating Frame.
2 Right-click in the Floating Frame and choose the panel from Add View submenu.

When using multiple monitors, you can choose to have floating panels spread across your displays for
greater flexibility.

Fusion Studio Keyboard Remapping

When using Fusion Studio, functions and tools can be mapped to hot keys on your keyboard by
choosing Views > Customize Hotkeys.

The Fusion Studio Hotkey Manager window

Chapter 2 Exploring the Fusion Interface 56

The Fusion Hotkey Manager dialog is divided into two sections. The left is where you select the
functional area where you want to assign a keyboard shortcut. The right side displays the keyboard
shortcut if one exists. You can use the New button at the bottom of the dialog to add a new
keyboard shortcut.
For instance, if you want to add a shortcut for a specific node:
1 Open the Keyboard Hotkey Manager.
2 Select Views > Effect from the Target area of the Hotkey Manager.
3 Below the Key/Action area, click the New button to create a new keyboard shortcut for the Node.
4 In the Edit Hotkey window, click the Tools disclosure arrow, and then select Blur to display all the
Blur-related nodes.
5 Select Glow in the Action panel.
6 At the top of the Edit Hotkey window, type G as the shortcut for the Glow node, and then click OK.
7 Glow and the G hotkey will now appear in the Key/Action area on the right.
8 Click OK to close the Hotkey Manager.
9 Click in the Node Editor and press G to add a Glow node.

Undo and Redo

Undo and Redo commands let you back out of steps you’ve taken or commands you’ve executed and
reapply them if you change your mind. Fusion is capable of undoing the entire history of things you’ve
done since creating or opening a particular project. When you close a project, its entire undo history is
purged. The next time you begin work on a project, its undo history starts anew.
There is no practical limit to the number of steps that are undoable (although there may be a limit to
what you can remember).

To simply undo or redo changes you’ve made one at a time:

– Choose Edit > Undo (Command-Z) to undo the previous change.
– Choose Edit > Redo (Shift-Command-Z) to redo to the next change.

Undo History in DaVinci Resolve

When using the Fusion page in DaVinci Resolve, you can also undo several steps at a time using the
History submenu and window. At the time of this writing, this only works for multiple undo steps in the
Media, Cut, Edit, and Fairlight pages.

To undo and redo using the History submenu:

1 Open the Edit > History submenu, which shows (up to) the last 20 things you’ve done.
2 Choose an item on the list to undo back to that point. The most recent thing you’ve done appears
at the top of this list, and the change you’ve just made appears with a check next to it. Steps
that have been undone but that can still be redone remain in this menu, so you can see what’s
possible. However, if you’ve undone several changes at once and then you make a new change,
you cannot undo any more and those steps disappear from the menu.

Chapter 2 Exploring the Fusion Interface 57

 The History submenu, which lets you undo several steps at once

Once you’ve selected a step to undo to, the menu closes and the project updates to show you its
current state.

To undo and redo using the Undo window:

1 Choose Edit > History > Open History Window.
2 When the History dialog appears, click an item on the list to undo back to that point. Unlike
the menu, in this window the most recent thing you’ve done appears at the bottom of this list.
Selecting a change here grays out changes that can still be redone, as the project updates to
show you its current state.

The Undo History window lets you browse the entire available undo stack of the current page.

3 When you’re done, close the History window.

Sharing Memory with Other DaVinci Resolve Pages

When using the Fusion page within DaVinci Resolve, the Memory and GPU panel in the Preferences
window controls the overall amount of system memory used by the application and a subset allocated
to the Fusion page. The Fusion page is limited to roughly 65% of the total memory allocated to
DaVinci Resolve. The memory allocated to the Fusion page is released when you are not actively
working on the page.

Chapter 2 Exploring the Fusion Interface 58

Chapter 3

Getting Clips
into Fusion
This chapter details the various ways you can move clips into Fusion
as you build your compositions.

Contents
Preparing Compositions in the Fusion Page  60
Working on Single Clips in the Fusion Page  60
Turning One or More Clips into Fusion Clips  61
Adding Fusion Composition Generators  62
Creating a Fusion Composition Clip in a Bin  63
Using Fusion Transitions  63
Adding Clips from the Media Pool  64
Adding Clips from the File System  65
Using MediaIn Nodes  65
MediaIn Node Inputs  65
Inspector Properties of MediaIn Nodes  65
Using Loader and Saver Nodes in the Fusion Page  68
Preparing Compositions in Fusion Studio  70
Setting Up a Composition  72
Reading Clips into Fusion Studio  73
Aligning Clips in a Fusion Studio Composition  74
Loader Node Inputs  75
Using Proxies for Better Performance  75
Presetting Proxy Quality  76
File Format Options  77
Loading Audio WAV Files in Fusion Studio  79

Chapter 3 Getting Clips into Fusion 59

Preparing Compositions in the Fusion Page
Ordinarily, clips come into the Fusion page from the Edit or Cut page Timeline as either a single clip, or
as multiple layers contained within a Fusion clip. You can also add clips to a composition directly from
DaVinci Resolve’s Media Pool. How clips find their way into a Fusion composition can determine how
they function within that composition and what resolution that composition outputs to the rest of
DaVinci Resolve.

Working on Single Clips in the Fusion Page

Positioning the playhead over a clip in the Edit page or Cut page Timeline and clicking the Fusion
page button causes that clip to appear in the Fusion page as a single MediaIn node connected to a
MediaOut node. Only the topmost visible clip is taken into Fusion. Clips that aren’t visible because
they’re on lower tracks with clips above them are ignored, unless you disable the clips or tracks that
appear above. These very simple default compositions are referred to unofficially in this manual as
“single-clip compositions.”
The MediaIn node represents the image that’s fed to the Fusion page for further work, and the
MediaOut node represents the final output that’s fed onward to the Color page for grading.

The default node tree that appears when you first open the
Fusion page while the playhead is parked on a clip.

This initial node structure makes it easy to quickly use the Fusion page to create relatively simple
effects using the procedural flexibility of node-based compositing.
For example, if you have a clip that’s an establishing shot, with no camera motion, that needs some
fast paint to cover up a bit of garbage in the background, you can open the Fusion page, add a Paint
node, and use the Clone mode of the Stroke tool to paint it out quickly.

A simple paint effect applied to a shot with no camera motion.

TIP: The resolution of a single clip brought into Fusion via the Edit or Cut page Timeline is the
resolution of the source clip, not the Timeline resolution.

Once you’ve finished, simply go back to the Edit or Cut page and continue editing, because the entire
Fusion composition is encapsulated within that clip, similarly to how grades in the Color page are also
encapsulated within a clip. However you slip, slide, ripple, roll, or resize that clip, the Fusion effects
you’ve created and the Color page grades you’ve made follow that clip’s journey through your
edited Timeline.

Chapter 3 Getting Clips into Fusion 60

 TIP: While you’ll likely want to do all the compositing for a greenscreen style effect in the
Fusion page, it’s also possible to add a keyer, such as the excellent DeltaKeyer node,
between the MediaIn and MediaOut nodes, all by itself. When you pull a key this way, the
alpha channel is added to the MediaOut node, so your clip on the Edit page has transparency,
letting you add a background clip on a lower track of your Edit page Timeline.

How Nodes Are Named

While the documentation refers to nodes by their regular name, such as “MediaIn,” the actual names of
nodes in the Fusion Node Editor have a number appended to them, to indicate which node is which
when you have multiple instances of a particular type of node.

Turning One or More Clips into Fusion Clips

For situations where you know you’re creating a more ambitious composited effect that requires
multiple layers edited together with very specific timings, you can create a “Fusion clip” right from the
Timeline. For example, if you have a foreground greenscreen clip, a background clip, and an additional
graphic clip, you can stack them all on the Timeline as superimposed clips. You can then use the Edit
page slip and slide features to align their timings so they work together as necessary. You can also
edit multiple consecutive clips together that you want to use in a composition as a series of clips. Once
that’s done, you can select every clip in the stack to create a Fusion clip, so you can easily use all
these superimposed layers within a Fusion composite.

To create a Fusion clip:

1 Edit all the clips you want to use in the Edit page Timeline.
2 Select all clips you want to be in the same composite at once.
3 Right-click one of the selected clips and choose New Fusion Clip from the contextual menu.
4 A new clip, named “Fusion Clip X” (where X is an automatically incrementing number) appears
in the currently selected bin of the Media Pool and in the Timeline to replace the previously
selected clips.
5 With the playhead parked over that clip, open the Fusion page to see the new arrangement of
those clips in the Fusion page Node Editor.

Chapter 3 Getting Clips into Fusion 61

 A stack of clips to use in a composite (Top), and turning that stack into a Fusion clip in the Edit page (Bottom).

The nice thing about creating a Fusion clip is that every superimposed clip in a stack is automatically
connected into a cascading series of Merge nodes that create the desired arrangement of clips. Note
that whatever clips were in the bottom of the stack in the Edit page appear at the top of the Node
Editor in the Fusion page, but the arrangement of background and foreground input connections is
appropriate to recreate the same compositional order.

The initial node tree of the three clips we turned into a Fusion clip.

TIP: Fusion clips change the working resolution of the individual clips to match the Timeline
resolution. For instance, if two 4K clips are stacked one on top of the other in an HD Timeline,
creating a Fusion clip resizes the clips to HD. The full resolution of the individual 4K clips is
not available in Fusion. To maintain the full resolution of course clips, bring only one clip into
the Fusion composition from the Edit or Cut page Timeline, and then bring other clips into the
Fusion composition using the Media Pool.

Adding Fusion Composition Generators

The Generator category of the Edit page Effects Library has a Fusion Composition generator. It’s
useful for creating an empty placeholder in the Timeline that you later want to work on in the Fusion
page to create a more fully-featured Fusion composition.

To create a blank Fusion clip in the Edit page:

1 Open the Effects Library, and select the Effects category.

Chapter 3 Getting Clips into Fusion 62

2 Edit a Fusion Composition clip into the Timeline in whichever way is most convenient.
a. You can drag a Fusion Composition clip into the Timeline, which will result in a clip that’s the
length of the “Standard generator duration” preference, which is 5 seconds by default.
b. You can set In and Out points in the Timeline, and drag the Fusion Composition clip onto any
of the editing overlays of the Timeline viewer to perform that sort of edit to insert, overwrite,
“place on top,” or ripple overwrite it into a specific place in the Timeline, for a specific duration.
3 A new clip named “Fusion Composition” appears in the Timeline. It initially displays only black in
the Timeline viewer, since it’s a blank composition with no contents.
4 With the playhead parked over that clip, open the Fusion page. Since this composition is blank,
there’s only a single MediaOut node in the Node Editor. At this point, you can add whatever media,
generators, and other effects you require.

NOTE: Audio on a track beneath a Fusion Composition effect cannot be heard in the
Fusion page.

Creating a Fusion Composition Clip in a Bin

You can create an empty Fusion Composition clip in any bin in the Media Pool without creating a
Timeline. This method can be useful for creating motion graphics or titles when there is no Timeline or
when you plan on using the clip in multiple Timelines.
To create a blank Fusion Composition clip in a bin:
1 Select the bin in the Media Pool where you want to save the Fusion Composition.
2 Right click in an empty area of the bin and choose New Fusion Composition.
3 In the New Fusion Composition clip dialog, enter a Name for the clip, a duration, and a frame rate,
and then click Create.
4 The clip will appear in the bin. To open it in Fusion, do one of the following:
– Double-click the Fusion Composition
– Right-click over the Fusion Composition clip and choose Open in Fusion Page

Using Fusion Transitions

Specific Fusion transitions are available in the Edit page Effects Library. You can use these transitions
to bring two clips and the transition between them into the Fusion Page. These transitions can then be
modified and saved back to the Edit page Timelines or saved as a new, reusable Fusion Transition that
appears in the Edit Page Effects lIbrary.

To apply and open a Fusion Transition:

1 Open the Effects Library, and select the Video Transitions category.
2 Scroll to the bottom of the transition list and drag one of the Fusion Transitions onto a cut in
the Timeline.
3 Right-click over the Fusion Transition in the Timeline and choose Open in Fusion Page.
4 The Fusion page opens with two MediaIn nodes representing the two sides of the transitions.
The MediaIn nodes connect to a cross dissolve or a group of nodes used to create the transition.
At this point, you can modify the transition using masks or other nodes and return to the Edit page
to see the results.

Chapter 3 Getting Clips into Fusion 63

 A Noise Dissolve Transition opened in the Fusion page.

To learn about creating custom Fusion Transitions that appear in the Effects Library, go to Chapter 67,
“Node Groups, Macros, and Fusion Templates” in the DaVinci Resolve Reference Manual or Chapter 6
in the Fusion Reference Manual.

Adding Clips from the Media Pool

You can open the Media Pool on the Fusion page and drag clips directly to the Node Editor to add
them to your node tree.

Dragging a clip from the Media Pool (Left), and dropping it onto your composition (Right).

When you add a clip by dragging it into an empty area of the Node Editor, it becomes another MediaIn
node, disconnected, that’s ready for you to merge into your current composite in any one of a
variety of ways.

TIP: Dragging a clip from the Media Pool on top of a connection line between two other
nodes in the Node Editor adds that clip as the foreground clip to a Merge node.

When you add additional clips from the Media Pool, those clips becomes a part of the composition,
similar to how Ext Matte nodes you add to the Color page Node Editor become part of that
clip’s grade.

Audio with Media Pool Clips

Audio from a clip brought in through the Media Pool is muted by default. Hearing the audio from a
Media Pool clip is a two step process.

To hear audio from a clip brought in through the Media Pool, do the following:
1 Select the clip in the Node Editor.
2 In the Inspector, click the Audio tab and select the clip name from the Audio Track
drop-down menu.
3 Right-click the speaker icon in the toolbar, then choose the MediaIn for the Media Pool
clip to solo its audio.

Chapter 3 Getting Clips into Fusion 64

You can now use the speaker icon contextual menu to switch back and forth between all the
MediaIn nodes.

Adding Clips from the File System

You also have the option of dragging clips from the file system directly into the Node Editor. When you
do this, they’ll be added to the currently selected bin of the Media Pool automatically. So, if you have a
library of stock animated background textures and you’ve just found the one you want to use using
your file system’s search tools, you can simply drag it straight into the Node Editor to use it right away.

Using MediaIn Nodes

The MediaIn nodes in the Fusion Page are the foundation of every composition you create. This
section provides more detail about the controls available for adjusting MediaIn and Loader nodes.

MediaIn Node Inputs

MediaIn nodes have one Effects mask input and one output. In the case of the Effects mask input,
connecting a mask node such as a Polygon or B-Spline node creates an alpha channel in the
MediaIn node.

TIP: If you connect a mask node without any shapes drawn, that mask outputs full
transparency, with the result that the image output by the MediaIn node is uselessly blank. If
you want to rotoscope over a MediaIn node, first create a disconnected mask node, and with
the mask node selected (exposing its controls in the Inspector) and the MediaIn node loaded
into the viewer, draw your mask. Once the shape you’re drawing has been closed, you can
connect the mask node to the MediaIn node’s input, and you’re good to go.

Inspector Properties of MediaIn Nodes

Which Inspector options are available for MediaIn nodes of your composition depends on how you
imported the media.

MediaIn Node Parameters for Clips in a Timeline

When you create a composition using clips from the Edit page Timeline, the MediaIn nodes for those
clips display fewer parameters than those imported directly from the Media Pool, because the timing
of clips that have already been edited into a Timeline is already set.

The Inspector Image tab parameters

for a clip from the Timeline

Chapter 3 Getting Clips into Fusion 65

Image Tab
– Clip Name: Displays the name of that clip.
– Process Mode: Lets you choose whether the clip represented by that node will be processed as
Full Frames, or via one of the specified interlaced methods.
– MediaID: A unique ID assigned by DaVinci Resolve for that clip.
– Layer: In the case of Photoshop PSD files, this selects the layer of the PSD file to use.

Source Color Space

The Color Space Type menu sets the color space of the media to help achieve a linear workflow.
Unlike the Gamut tool, this doesn‘t perform any actual color space conversion but rather adds the
source space data into the metadata, if that metadata doesn‘t already exist. The metadata can then be
used downstream by a Gamut tool with the From Image option, or in a Saver if explicit output spaces
are defined there.
– Auto: Passes along any metadata that might be in the incoming image.
– Space: Allows the user to set the color space from a variety of options.

Source Gamma Space

The Curve Type menu automatically determines or allows you to choose the Gamma setting for the
image and allows the option to remove the Gamma curve to help achieve a linear workflow.
– Auto: Passes along any metadata that might be in the incoming image.
– Space: Allows you to choose a specific setting from a Gamma Space drop-down menu, while a
visual graph lets you see a representation of the gamma setting you’ve selected.
– Log: Similar to the Log-Lin node, this option reveals specific log-encoded gamma profiles so that
you can select the one that matches your content. A visual graph shows a representation of the
log setting you’ve selected. When Cineon is selected from the Log Type menu, additional Lock
RGB, Level, Soft Clip, Film Stock Gamma, Conversion Gamma, and Conversion table options are
presented to finesse the gamma output.
– Remove Curve: Depending on the selected gamma space or on the gamma space found in Auto
mode, the associated gamma curve is removed from the material, effectively converting it to
output in a linear color space.
– Pre-Divide/Post-Mujltiply: Lets you convert “straight” alpha channels into pre-multiplied alpha
channels, when necessary.

TIP: All content in the DaVinci Resolve Fusion page is processed using 32-bit floating-point
bit depth, regardless of the content’s actual bit depth.

Audio Tab
The Inspector for the MediaIn node contains an Audio tab, where you can choose to solo the audio
from the clip or hear all the audio tracks in the Timeline.

Chapter 3 Getting Clips into Fusion 66

 The MediaIn Audio tab

The Audio tab in the MediaIn node is used to select the track for playback, slip the audio timing, and
reset the audio cache
If the audio is out of sync when playing back in Fusion, the Audio tab’s Sound Offset wheel allows you
to slip the audio in subframe frame increments. The slipped audio is only modified in the Fusion page.
All other pages retain the original audio placement.

Purging the Audio Cache

The audio and its settings are cached for faster performance. If you change which audio tracks you
want to play back in Fusion, or you use the Sound Offset wheel to slip the audio tracks, you need to
purge the audio cache. Also, if you return to the Edit, Cut, or Fairlight page and modify the audio
levels, you need to purge the audio cache.

To purge the audio cache after any change to the audio playback:
– Click the Purge Audio Cache button in the Inspector.

The audio will be updated when you next play back the composition.

Aligning Clips from the Media Pool

When you add a clip from the Media Pool or your file system directly into a composition, the resulting
MediaIn node has more options than clips from the Edit page Timeline. These additional options make
it easier to align the Media Pool clips with other clips from the Edit or Cut page Timeline. You can also
trim clips, hold the first or last frame for a longer duration than the original media, and reverse or loop
the clip to get more range for your composition.

The Inspector parameters for a clip

imported from the Media Pool.

Chapter 3 Getting Clips into Fusion 67

Below is a list of controls that are added beyond the controls that appear when a clip is added from
the Edit or Cut page Timeline.
– Global In/Out: Use this control to specify the position of this node within the composition. For
instance, when the clip is added to the comp from the Media Pool, it is added at frame 0. However,
the MediaIn node from the Edit page Timeline may not start until a much later frame, based on
where it is edited into the Timeline. Use Global In to specify the frame on which that the clip starts
so that it aligned with media from the Edit page Timeline. It is easiest to view and change the
alignment of different clips in the comp while viewing the Keyframes Editor.
To slide the clip in time or align it to other clips without changing its length, place the mouse
pointer in the middle of the range control and drag it to the new location, or enter the value
manually in the Global In value control.
If the Global In and Out values are decreased to the point where the range between the In and
Out values is smaller than the n number of available frames in the clip, Fusion automatically trims
the clip by adjusting the Clip Time range control. If the Global In/Out values are increased to the
point where the range between the In and Out values is larger than the number of available
frames in the clip, Fusion automatically lengthens the clip by adjusting the Hold First/Last Frame
controls. Extended frames are visually represented in the range control by changing the color of
the held frames to purple in the control.
– Trim: The Trim range control is used to trim frames from the start or end of a clip. Adjust the Trim
In to remove frames from the start and set Trim Out to specify the last frame of the clip. The values
used here are offsets. A value of 5 in Trim In would use the 5th frame in the sequence as the start,
ignoring the first four frames. A Trim Out value of 95 would stop loading frames after the 95th.
– Hold First Frame/Hold Last Frame: The Hold First Frame and Hold Last Frame controls will hold
the first or last frame of the clip for the specified amount of frames. Held frames are included in a
loop if the footage is looped.
– Reverse: Select this checkbox to reverse the footage so that the last frame is played first and the
first frame is played last.
– Loop: Select this checkbox to loop the footage until the end of the project. Any lengthening of the
clip using Hold First/Last Frame or shortening using Trim In/Out is included in the looped clip.

Using Loader and Saver Nodes

in the Fusion Page
The Loader and Saver nodes in Fusion Page are specifically used for workflows that center around
multi-channel EXR files. OpenEXR media can contain high-quality floating-point image data, multiple
matte channels, as well as auxiliary channels rendered from 3D software. The Loader node lets you
add OpenEXR files to a composition directly from the file system, retaining all the channels that are
embedded within that file. Saver nodes enable you to render either all or part of a composition as EXR
files directly to disk, bypassing the DaVinci Resolve Deliver page.

To import a multi-channel EXR image sequence into the Fusion page:

1 Open the Fusion page.
2 Open the Effects Library, and select the Tools > I/O category, and click the Loader node.
3 In the OS navigation window that opens, select the EXR image sequence you want to import, and
click Open.
4 A Loader node linked to the EXR file will appear in the Fusion page, although a clip will not appear
in the Media Pool.

Chapter 3 Getting Clips into Fusion 68

Loader Node Parameters
The image tab of Loader nodes shares parameters with MediaIn nodes, as described previously in this
chapter. However, when using a Loader node for EXR files, the Format tab selectively enables and
disables the use of specific auxiliary channels contained in the file.

The Format tab in a Loader node Inspector

displays Aux channels in EXR files.

Outputting Images Using Saver Nodes

Saver nodes render OpenEXR image sequences to disk directly from the Fusion page. Saver nodes
can be added to any branch of a node tree, allowing you to export one or more subsets of nodes in a
composition. You can add as many Saver nodes as you like to whichever branches of your
composition’s node tree you need to output, to export multiple parts of a composition. For example, if
a particular composition is made up of one branch of nodes working together to create a complex
background, and another branch of nodes creating a foreground character with transparency, you
could export the background and foreground branches as separate OpenEXR files using two
Saver nodes.
To do this, simply create a Saver node after any set of nodes you want to output, and then open the
Inspector and click Browse to choose a name and location for the rendered result.
When naming the file, you must add the .exr file extension. Fusion sets the output format accordingly.
A four-digit frame number is automatically added before the filename extension. However, you can
specify the frame padding by adding several zeroes to indicate the number of digits. For example, 000
signifies 001.
Once the Saver node is set up, output one or more Saver nodes, and choose Fusion > Render
All Savers.

Chapter 3 Getting Clips into Fusion 69

 The Inspector parameters for a Saver node.

Manual Disk Caching Using Loader and Saver Nodes

The Loader and Saver nodes in the Fusion page are also useful for optimizing extremely complex and
processor-intensive compositions. For example, you can render out specific branches of a node tree
that no longer require frequent adjustment to OpenEXR via a Saver node, then reimport the result to
take the place of the original branch of nodes in order to improve the performance of your
composition. Used this way, Loader and Saver nodes provide a bulletproof manual workflow for
caching using media files that will never be automatically purged unless you specifically delete them.
As long as you retain the original branch of nodes, you can always readjust and re-render these
manually cached parts of a composition, if necessary.

Preparing Compositions in Fusion Studio

The next few sections in this chapter cover preparing a project and adding clips into a composition
when using Fusion Studio. The term composition, or comp, is used to refer to the Fusion project file. By
default, opening the Fusion Studio application creates a new empty composition when it’s launched. A
composition can contain single frames, image sequences, or movie files at various resolutions and bit
depths. Knowing which files you can load in, how to set up a composition to handle them, and finally,
reading those files in are the first steps in beginning to composite.

Opening, Closing, and Saving Compositions

As soon as you open Fusion Studio, a new empty composition is created. If necessary, you can also
create or open multiple compositions at once. Each additional composition is opened as a tab to the
main Fusion Studio window.

Three compositions opened as tabs in Fusion Studio.

To create a second new composition:

– Choose File > New.

To open an existing composition, do one of the following:

– Choose File > Open.
– Choose File > Open Recent, and choose from the list of recently opened comps.
– Drag a composition file from an OS file browser into the tabbed composition area at the top of the
Fusion Studio window.
– Double-click on a composition file in the OS file browser.

Chapter 3 Getting Clips into Fusion 70

The following methods can be used to close the current composition:
– Choose File > Close from the menu at the top of the Fusion window.
– Click the Close X icon on the right of the composition’s tab.

If the composition has unsaved changes, a dialog box appears allowing you to save before closing.

TIP: Compositions that have unsaved changes will display an asterisk (*) next to the
composition’s name in the Fusion Studio title bar and in the composition’s tab.

To save the current composition, you can do the following:

– Choose File > Save and enter a name if the comp has yet to be named.
– Choose File > Save As to save under a new name.
– Choose File > Save Version to save the current composition with an added three-digit version
number at the end of the name. Each time you save a version, the number automatically
increments and the comp file is saved in the same location as the first version.

Auto Save
Auto save automatically saves the composition to a temporary file at preset intervals. Auto saves help
to protect you from loss of work due to power loss, software issues, or accidental closure.
To enable auto save for new compositions, choose Fusion Studio > Preferences, and then locate
Global > General > Auto Save in the Preferences dialog.
An auto-save file does not overwrite the current composition in the file system. A file with the same
name is created in the same folder as the composition but with the extension .autosave instead of .
comp. Unsaved compositions will place the autosave file in the default folder specified by the Comp:
path in the Paths panel of the Global Preferences.
If an auto-save file is present when Fusion Studio loads a composition, a dialog will appear asking to
load the auto-saved or original version of the composition.

The Composition File Format

Composition files are saved as readable plain text files. Using plain text files to describe a composition
makes it easier to integrate Fusion into structured, visual effects pipeline and asset management
solutions.
Composition files can be opened and edited using any standard text editing program. However, it is
never a good idea to open the file using a word processor, such as Microsoft Word or Apple Pages, as
these will generally save additional formatting information which will make the composition unreadable
to Fusion.

Importing and Exporting a Composition from DaVinci Resolve

Although the compositions created in DaVinci Resolve’s Fusion page are saved in the DaVinci Resolve
database as .drp project files, you can import and export Fusion composition files when in the Fusion
page. This makes it very easy to share Fusion compositions between the different applications.

To export a Fusion composition from DaVinci Resolve:

1 From within DaVinci Resolve, switch to the Fusion page with the composition you want to export.
2 Choose File > Export Fusion Composition.
3 A Save dialog appears in which you can enter a name and location from the exported
Fusion composition.

Chapter 3 Getting Clips into Fusion 71

A .comp extension is added to the end of the filename. Only the node tree created in the Fusion page
is exported. Clips not added to the Node Editor will not appear in the Fusion Studio bins. ResolveFX
added to the comp will also not translate from the Fusion page to Fusion Studio.
MediaIn nodes from DaVinci Resolve are automatically converted to Loader nodes, and if the file path
remains identical, the media is automatically relinked.
MediaOut nodes are converted to Saver nodes.
The return trip can also be performed, saving a composition file from Fusion Studio and importing it
into the Fusion page within DaVinci Resolve.

To import a composition from Fusion Studio into the Fusion page within DaVinci Resolve:
1 From within Fusion Studio, open the composition you want to move into the Fusion page.
2 From within DaVinci Resolve, switch to the Fusion page with an empty composition. The
composition you import will completely replace the existing composition in the Fusion page
Node Editor.
3 Choose File > Import Fusion Composition.
4 In the Open dialog, navigate to the Fusion comp and click Open.
The new comp is loaded into the Node Editor, replacing the previously existing composition.

TIP: To keep an existing comp in the Fusion page and merge a new comp from Fusion Studio,
open Fusion Studio, select all the nodes in the Node Editor, and press Command-C to copy
the selected nodes. Then, open DaVinci Resolve and switch the Fusion page with the
composition you want, click in an empty location in the Node Editor, and press Command-V to
paste the Fusion Studio nodes. Proceed to connect the pasted node tree into the existing
one using a Merge or Merge 3D node.

Setting Up a Composition
Source media can come in a variety of formats, including HD, UHD, and 4K or larger. Often you will
have different formats within a single comp. Each format has different properties, from resolution to
color depth and gamma curve. Fusion can mix and match material of different formats together in a
single composite, but it is important to note how Fusion Studio configures and combines materials of
different formats when loading and merging them together.
When you open Fusion Studio, an empty composition is created. The first thing you do when starting
on a new composition is to set the preferences to match the intended final output format. The
preferences are organized into separate groups: one for global preferences, and one for the
preferences of the currently opened composition.
Although the final output resolution is determined in the Node Editor, the Frame Format preferences
are used to determine the default resolution used for new Creator tools (i.e., text, background, fractals,
etc.), aspect ratio, as well as the frame rate used for playback.

Chapter 3 Getting Clips into Fusion 72

 Frame Format preferences for images generated in Fusion.

If the same frame format is used day after day, the global Frame Format preferences should match the
most commonly used footage. For example, on a project where the majority of the source content will
be 1080p high definition, it makes sense to set up the global preferences to match the frame format of
the HD source content you typically use.

To set up the default Frame Format for new compositions, do the following:
1 Choose Fusion Studio > Preferences.
2 Click the Global and Default Settings disclosure triangle in the sidebar to open the Globals group.
3 Select the Frame Format category to display its options.

When you set options in the Global Frame Format category, they determine the default frame format
for any new composition you create. They do not affect existing compositions or the composition
currently open. If you want to make changes to existing compositions, you must open the comp. You
can then select the Frame Format controls listed under the comp’s name in the sidebar.
For more information on preferences, see Chapter 76, “Preferences” in the DaVinci Resolve Reference
Manual or Chapter 15 in the Fusion Reference Manual.

Reading Clips into Fusion Studio

Once the Frame Format preferences are set, you usually begin to create a composite by reading in
source media. When Fusion reads in media, it doesn’t convert or move the original files; it simply reads
the files in place, from whichever storage volume they’re on. You are always dealing with the original
source files in their original location.
Source media is read into a comp using a Loader tool. Although there are other tools within Fusion
Studio that you can use to generate images like gradients, fractals, or text, each still image, image
sequence, or movie file must be added to your comp using a Loader tool.

Loader and Saver tools are used to add

media to Fusion Studio and render it out.

Chapter 3 Getting Clips into Fusion 73

To add media to your comp, do one of the following:
– Click Effects to open the Effects Library, and then select Tools > I/O > Loader.
– Click the Loader icon in the toolbar.
– Right-click over the Node Editor, and then choose Add Tool > I/O > Loader.
– Drag a file from an OS file browser window into the Node Editor.

If multiple files are dragged into the Node Editor, a separate Loader is added for each file. However, if
you drag a single frame from an image sequence, the entire series of the image sequence is read into
the comp using one Loader, as long as the numbers are sequential.

To add only one frame of an image sequence to your comp:

– Hold Shift while you drag a single frame from an image sequence into the Node Editor.
This comes in handy when you want to read in photographs from a digital camera that are
numbered sequentially.
A Loader represents any clip, image file, or graphic that you bring into Fusion. However, other types of
media can also be brought into Fusion Studio. Photoshop PSD files, SVG splines, and 3D models in the
Alembic, FBX, and OBJ format can be imported using the File > Import menu.

TIP: Using File > Import > Footage creates a new composition along with a Loader node for
the footage. The selected media is automatically used for the name of the composition.

For more information about the Loader node, see Chapter 105, “I/O Nodes” in the DaVinci Resolve
Reference Manual or Chapter 44 in the Fusion Reference Manual.

Aligning Clips in a Fusion Studio Composition

When you add a clip into a composition, the resulting Loader node is added at frame 0 of the
composition. However, the vital portion of the clip you are interested in may not start until a few frames
or even seconds later. To ensure you can align the timing of each piece of media, each Loader
includes timing and trimming options in the Inspector. You can also hold the first or last frame for a
longer duration than the original media, and reverse or loop the clip to get more range for
your composition.
At the top of the Inspector are the Global In and Global Out settings. This range slider determines
when in your composition the clip begins and ends. It is the equivalent of sliding a clip along a track in
a Timeline. The Hold First Frame and Hold Last Frame dials at the bottom of the Inspector allow you to
freeze frames in case the clip is shorter than the composition’s global time.

Use the Global In/Out range slider to slide a clip in

time to have it appear at the correct time in a comp.

Below the filename in the Inspector is a Trim In and Out range slider. This range slider determines the
start frame and end frame of the clip. Dragging the Trim In will remove frames from the start of the clip,
and dragging the Trim Out will remove frames from the end of the clip.

Chapter 3 Getting Clips into Fusion 74

 Use the Trim In/Out to remove unnecessary
frames from the start or end of a clip.

Although you may remove frames from the start of a clip, the Global In always determines where in
time the clip begins in your comp. For instance, if the Loader has a Global In starting on frame 0, and
you trim the clip to start on frame 10, then frame 10 of the source clip will appear at the comp’s starting
point on frame 0. 
Instead of using the Inspector to adjust timing, it is visually more obvious if you use the Keyframes
Editor. For more information on the Keyframes Editor and adjusting a clip’s time, see Chapter 70,
“Animating in Fusion’s Keyframe Editor” in the DaVinci Resolve Reference Manual or Chapter 9 in the
Fusion Reference Manual.

Loader Node Inputs

Loader nodes have one Effects mask input and one output. In the case of the Effects mask input,
connecting a mask node such as a Polygon or B-Spline node automatically creates an alpha channel in
the Loader node.

TIP: If you connect a Mask node without any shapes drawn, that mask outputs full
transparency, so the result is that the image output by the MediaIn node is blank. If you want
to rotoscope over a MediaIn node, first create a disconnected Mask node, and with the Mask
node selected and the Media In node loaded into the viewer, draw your mask. Once the
shape you’re drawing has been closed, connect the Mask node to the MediaIn node’s input,
and you’re good to go.

Using Proxies for Better Performance

For increased performance, you can do one of two things:
– Generate smaller media files and write them to disk using Optimized Media in DaVinci Resolve
– Render out proxy files using Saver nodes in Fusion Studio
Both applications also allow you to generate proxies on-the-fly without rendering new files to
disk using the Proxy and Auto Proxy options in the transport controls area.
To enable the Proxy and Auto Proxy options, you can do one of two things, depending on the
version of Fusion you are using:
– In the Fusion page, right-click the empty area behind the transport controls to enable the
Proxy option.

Chapter 3 Getting Clips into Fusion 75

 Proxy and Auto Proxy options in the
transport controls right-click menu.

– In Fusion Studio, click the Proxy (Prx) button in the transport area to enable the usage of proxies.
The Proxy option reduces the resolution of the images as you view and work with them. Instead of
displaying every pixel, the Proxy option processes one out of every x pixels interactively. In Fusion
Studio, the value of x is determined by right-clicking the Prx button and selecting a proxy ratio
from the drop-down menu. For instance, choosing 5 from the menu sets the ratio at 5:1. In the
Fusion page, the proxy ratio is set by choosing Fusion > Fusion Settings and setting the Proxy
slider in the General panel.

The Proxy menu sets the ratio for

skipping pixels when processing.

The Auto Proxy button enables Fusion to interactively degrade the image only while adjustments are
made. The image returns to normal resolution when the control is released. Similar to the Prx button in
Fusion Studio, you can set the Auto Proxy ratio by right-clicking the APrx button and choosing a ratio
from the menu.
When a Loader node is selected, the Inspector includes a Proxy Filename field where you can specify
a clip that will be loaded when the Proxy mode is enabled. This allows smaller versions of the image to
be loaded to speed up file I/O from disk and processing. This is particularly useful when working with
high resolution files like EXR that might be stored on a remote server. Lower resolution versions of the
elements can be stored locally, reducing network bandwidth, interactive render times, and
memory usage.
The proxy clip that you create must have the same number of frames as the original clip, and if using
image sequences, the sequence numbers for the clip must start and end on the same frame numbers.
If the proxies are the same format as the original files, the proxies will use the same format options in
the Inspector as the originals.

Presetting Proxy Quality

When using Fusion Studio, rather than right-clicking over the Proxy button to set the proxy quality, you
can preset the standard and Auto Proxy quality in the Fusion Preferences window. The General pane
in the Preferences window includes sliders for both standard Proxy files and Auto Proxy files.
These sliders designate the default ratio used to create proxies when the Proxy and Auto Proxy modes
are turned on. These settings do not affect the final render quality.

Chapter 3 Getting Clips into Fusion 76

 TIP: Even though the proxies are being processed smaller than their original size, the viewers
scale the images so they refer to original resolutions.

File Format Options

The Fusion interface in DaVinci Resolve and Fusion Studio display specific options for various file
formats in slightly different ways. Where Fusion Studio displays most file-specific options in the
Loader’s Format tab, the Fusion page in most cases displays these options in the main Image tab of
the MediaIn node. The only exception being the OpenEXR format. Its extensive options are displayed
in a separate tab even in the MediaIn node of the Fusion page. Not all file formats have options. Only
the DPX, OpenEXR, PSD, and QuickTime formats provide additional options when loaded.

DPX
The Format tab in Fusion Studio’s Loader node for DPX files is used to convert image data from
logarithmic to linear. These settings are often left in bypass mode, and the Log to Linear conversion is
handled using a Cineon Log node.

OpenEXR
The OpenEXR format provides a compact and flexible high dynamic range (float) format. The format
supports a variety of extra non-RGBA channels and metadata. These channels can be viewed and
enabled in the Format tab of the Inspector.

The Format tab in a Loader node Inspector

displays Aux channels in EXR files.

Chapter 3 Getting Clips into Fusion 77

Photoshop PSD
There are two methods for importing Photoshop PSD files. You can either import the PSD file and have
it represented as a single node in the Node Editor or import the PSD and have each layer represented
as a node in the Node Editor. If you do not need independent control over each layer or blend modes
are not used when creating the PSD file, then importing the PSD file as a single node will make for a
more manageable experience. If you do need control over each layer or Blend modes used in the PSD
file are critical, then you should import the file so each layer becomes a node in the Node Editor. Each
method is explained below for the Fusion page and Fusion Studio.
– Using the Media Pool in DaVinci Resolve’s Fusion page: Any PSD file added to the Media Pool
in DaVinci Resolve can be accessed from the Fusion page. After dragging the PSD file from the
Media Pool into the Node Editor, the image appears as a MediaIn node. From there, you can select
which layer to use from the PSD file from the Layer drop-down menu in the Inspector.
– Using a Loader node in Fusion Studio: This lets you read in Photoshop PSD files with the
ability to select the layer in the PSD file that is used in the comp. Fusion can load any one of the
individual layers stored in the PSD file, or the completed image with all layers. Transformation and
adjustment layers are not supported.

The Format tab in the Inspector displays specific

controls for a Photoshop PSD file.

To load all layers individually from a PSD file, with appropriate blend modes, do one of the following:
– In DaVinci Resolve, switch to the Fusion page and choose Fusion > Import > PSD.
– In Fusion Studio, choose File > Import > PSD.

Using either of the methods above creates a node tree where each PSD layer is represented by a
node and one or more Merge nodes are used to combine the layers. The Merge nodes are set to the
Apply mode used in the PSD file and automatically named based on the Apply mode setting.

The two layers of a Photoshop PSD file are imported and

connected to a Merge node set to a Screen Apply mode.

QuickTime
QuickTime files can potentially contain multiple tracks. You can use the Format tab in the Inspector to
select one of the tracks.

Chapter 3 Getting Clips into Fusion 78

Loading Audio WAV Files in Fusion Studio
You can load WAV format audio-only files into Fusion Studio. The entire WAV file is loaded into RAM in
order to quickly display the waveform in the Keyframe Editor. That being the case, it is best to use the
shortest possible audio file you need for the Comp, so as not to use up more memory than necessary.

TIP: AIFF files can be loaded on macOS.

You can either load the audio file independent of any nodes, or load an audio file into the Saver node.
The benefit of using a Saver node to load the audio is that you can view the audio waveforms in the
Keyframes Editor.

To load a WAV audio file, do the following:

1 Right-click over the speaker icon and select Choose from the contextual menu.
2 In the file browser window, select the audio WAV file track to be used.

To load a WAV audio file using a Saver node, do the following:

1 Add a Saver node to the Node Editor.
2 In the Inspector, click the Audio tab and click the Browse button.
3 In the file browser window, select the audio WAV file track to be used.

To view the Audio Waveform in Fusion Studio, do the following:

1 Open the Keyframes Editor.
2 Expand the Saver track to view the audio waveform.

When you want to find the precise location of an audio beat, transient, or cue, you can slowly drag
over the audio waveform to hear the audio.

Chapter 3 Getting Clips into Fusion 79

Chapter 4

Rendering Using
Saver Nodes
This chapter covers how to render compositions using Saver nodes in Fusion Studio
and the Fusion page in DaVinci Resolve. It also covers how to render using multiple
computers over a network when using Fusion Studio.

Contents
Rendering Overview  81
Rendering in the Fusion Page  81
Rendering in Fusion Studio  82
Rendering with the Saver Node  82
Setting Filenames for Export  83
Using the Render Settings Dialog  84
Render Settings Dialog Options  85
Rendering Previews  86
Setting Up Network Rendering in Fusion Studio  86
Licensing for Network Rendering  87
Configuring the Render Master and Render Nodes  88
Setting Up the Render Manager  90
Submitting Comps to Network Render  91
Using the Render Settings Dialog for Network Rendering  91
Using the Render Manager Window for Network Rendering  92
Working with Node Groups  93
Viewing the Render Log  94
Using Third-Party Render Managers with Fusion Studio  94

Chapter 4 Rendering Using Saver Nodes 80

Preparing Compositions for Network Rendering  96
Using Relative Paths  96
Using Mapped Drives  97
Installing All Fonts on Render Nodes  98
Installing Third-Party Plug-Ins on Render Nodes  98
Other Uses of Network Rendering  98
Flipbook Previews  98
Disk Cache  98
When Renders Fail  99
Automatic Rejoining of the Queue  99
Relaunching Render Nodes with Fusion Server  99
Frame Timeouts  99
Heartbeats  100
Managing Memory Use  100
Override Composition Settings  100
Render Several Frames at Once  100
Simultaneous Branching  101
Limitations of Render Nodes  101
Time Stretching  101
Linear Tools  101
Saving to Multi-Frame Formats  101
Troubleshooting  101
Checking the Render Log  102
Check the Composition  102

Rendering Overview
When you have finished creating a composition in Fusion, you need to render the files out to disk for
playback and integration into a larger timeline. Fusion Studio and the Fusion page in DaVinci Resolve
use very different workflows for rendering. To finish a composite in the Fusion Page, you use a
MediaOut node to cache the results into the Edit or Cut page Timeline. The DaVinci Resolve Deliver
page handles the final render of the entire Timeline. To get completed composites out of Fusion
Studio, you configure and render them starting with a Saver node in the Node Editor. Fusion Studio is
also capable of distributing a variety of rendering tasks to other machines on a network.

Rendering in the Fusion Page

In the Fusion page, a MediaOut node is required for getting your composite from the Fusion page
back into the Edit or Cut page Timeline. Whatever the MediaOut node displays when you see it in the
viewer is what gets rendered back into the Edit or Cut page. This process is semi-automatic in
DaVinci Resolve, where the Smart Render Cache setting begins caching the MediaOut node almost
immediately when you return to the Edit or Cut page Timeline. The cache file format and any resolution
scaling to fit the composition into the Timeline Resolution is handled in the DaVinci Resolve
Project Settings.

Chapter 4 Rendering Using Saver Nodes 81

Rendering in Fusion Studio
In Fusion Studio, all rendering goes through Saver nodes. Similar to MediaOut nodes in the Fusion
page, Saver nodes are most often appended to the end of a node tree to render the final composite.
The Saver node determines the name, format, and location of the rendered files.

Rendering with the Saver Node

To begin rendering in Fusion Studio, you must add at least one Saver node to the node tree. Most of
the time, you will place at least one Saver node at the very end of your tree to render the final image.

A single Saver node is added to the end of a node tree to render the final composite.

You can attach multiple Saver nodes anywhere along the node tree to render out different parts of a
composite. In the example below, three Saver nodes are added at different points in the node tree.
The top two render out each half of the composite while the bottom renders the results of the entire
composite.

Multiple Saver nodes can be added to different parts of a node tree.

You can also use multiple Saver nodes stemming from the same node in order to create several output
formats. The below example uses the three Savers to export different formats of the same shot.

Chapter 4 Rendering Using Saver Nodes 82

 Multiple Saver nodes can be added to create different formats for output.

Adding a Saver node to a node tree automatically opens a Save dialog where you name the file and
navigate to where the exported file is saved. You can then use the Inspector to configure the
output format.
For more information on the Saver node, see Chapter 105, “I/O Nodes” in the DaVinci Resolve
Reference Manual or Chapter 44 in the Fusion Reference Manual.

Setting Filenames for Export

If you use a file extension when naming the file, Fusion will set the output format accordingly. For
example, naming your file image_name.exr will set the Inspector to output an EXR file or naming a file
image_name.mov will set the Inspector for an H264 QuickTime movie. If you decide to change or
modify the setting of the file type, the Saver’s format tab in the Inspector contains the specific
parameters for the selected format.

The Saver’s Format tab with controls for QuickTime

If you decide to output an image sequence, a four-digit frame number is automatically added before
the filename extension. For example, naming your file image_name.exr results in files named image_
name0000.exr, image_name0001.exr, and so on. You can specify the frame padding by adding
several zeroes to indicate the number of digits. For example, entering a filename as image_
name_000.exr results in a sequence of images with the names Image_name_000.exr, Image_
name_001.exr, Image_name_002.exr, and so on.

NOTE: The starting frame number always uses the Time Ruler start frame number.

Chapter 4 Rendering Using Saver Nodes 83

Using the Render Settings Dialog
In Fusion Studio, you initiate rendering by clicking the Render button and opening the Render Settings
dialog. This dialog configures the quality, frame range, and network usage of the rendering.
To Render a comp in Fusion Studio:
1 Connect a Saver node at the end of your composition.
2 Enter a name and a location for the saved file(s) in the Save window.
3 Set the format using the Format tab in the Inspector, if necessary.
4 Click the Render button in the transport controls area or choose File > Render All Savers.

The Render Settings dialog opens providing options for the rendered output.

The Render Settings dialog options

Ensure that the frame range and other parameters are correct and click Start Render.

Saver nodes in the DaVinci Resolve Fusion page

The Fusion page also includes a Saver node, although it is used for a different purpose than
the Saver node in Fusion Studio. Rendering from the Fusion page is handled primarily through
the MediaOut node. There is no Render Settings dialog since the rendering location and
format is predetermined by DaVinci Resolve’s cache settings. Saver nodes in the Fusion page
are a special case, and are used to render OpenEXR files only. Once you add a Saver node,
you enter the filename with the extension. exr. Click the Browse button to select a location for
the EXR sequence, and then choose Fusion > Render All Savers.
Using the Saver node is useful for optimizing extremely complex and processor-intensive
compositions. For example, you can render out specific branches of a node tree that no
longer requires frequent adjustment to OpenEXR via a Saver node, and then reimport the
result to take the place of the original branch of nodes in order to improve the performance of
your composition.
Alternatively, you can render out multi-channel mattes or EXR images containing Arbitrary
Output Variables (AOVs) to bring into other applications.

Chapter 4 Rendering Using Saver Nodes 84

Render Settings Dialog Options
Many of the options in the Fusion Studio Render Settings dialog are used when you need to create a
quick preview or a test render. The options in this dialog allow you to increase performance by
disabling some of the image-processing operations that are time consuming but deliver higher quality
results. Often the first settings you set are those in the Configurations section. This section determines
whether you want to produce a final high quality render or a faster preview render. Selecting Final
prevents you from modifying the options that will limit the quality.

Settings
When the Configuration section is set to Preview, the Settings section of the Render dialog includes
three options that determine the overall quality and appearance of your final output. These buttons
also have a significant impact on render times. When the Configurations setting is set to Final, these
options cannot be disabled
– HiQ: When enabled, this setting renders in full image quality. If you need to see what the final
output of a node would look like, then you would enable the HiQ setting. If you are producing a
rough preview to test animation, you can save yourself time by disabling this setting.
– MB: The MB in this setting stands for Motion Blur. When enabled, this setting renders with motion
blur applied if any node is set to produce motion blur. If you are generating a rough preview and
you aren’t concerned with the motion blur for animated elements, then you can save yourself time
by disabling this setting.
– Some: When Some is enabled, only the nodes specifically needed to produce the image of the
node you’re previewing are rendered.

Size
When the Configurations section is set to Preview, you can use the Size options to render out frame
sizes lower than full resolution. This is helpful when using the Render dialog to create proxies or just
creating a smaller file size.

Network
The Network setting controls the distribution of rendering to multiple computers. For more information,
see the network rendering section in this chapter.

Shoot On
Again, this option is only available when Configurations is set to Preview. The Shoot On setting allows
you to skip frames when rendering. You can choose to render every second, third, or fourth to save
render time and get faster feedback. You can use the Step parameter to determine the interval at
which frames are rendered.

Frame Range
Regardless of whether the Configurations is set to Final or Preview, this option defaults to the current
Render In/Out Range set in the Time Ruler to determine the start and end frames for rendering. You
can modify the range to render more or fewer frames.

Configurations
When set to Final, the Render Settings are set to deliver the highest quality results, and you cannot
modify most of the options in this dialog. When set to Preview, you can set the options to gain faster
rendering performance. Once you’ve created a useful preview configuration, you can save it for later
use by clicking the Add button, giving it a name, and clicking OK.

Chapter 4 Rendering Using Saver Nodes 85

Rendering Previews
You can render Flipbook previews into a viewer. These Flipbook previews exist entirely within RAM.
They are created by right-clicking over a node in the Node Editor and choosing Create > Play/Preview
on > Left viewer/Right viewer from the drop-down menu. The Render Settings dialog appears where
you can configure the preview and initiate the rendering. You can also Option-drag a node directly
from the Node Editor into a viewer. The Render Settings dialog will be displayed, and the preview will
appear on the viewer you target.
For more information on rendering RAM previews, see Chapter 68 “Using Viewers” in the
DaVinci Resolve Reference Manual or Chapter 7 in the Fusion Reference Manual.

TIP: Option-Shift-dragging a node into a viewer will skip the Render dialog and
previously used settings.

Setting Up Network Rendering

in Fusion Studio
Fusion Studio is capable of distributing a variety of rendering tasks to an unlimited number of
computers on a network, allowing multiple computers to assist with creating network-rendered
previews, disk caches, and final renders.
Using the Render Settings dialog or the built-in Render Manager, you can submit compositions to be
rendered by other copies of Fusion Studio, as well as to one or more Fusion Render nodes. Rendering
can also be controlled through the command line for integration with third-party render managers like
Deadline, Rush, and Smedge.
Render nodes are computers that do not have the full Fusion application installed but do have Fusion
Render Node software installed. The Render Node software is not installed by default when you install
Fusion Studio, but it can be installed at any time using the Fusion Render Node Installer. The installer is
located in the Blackmagic Fusion Studio installer.dmg on macOS and the Blackmagic Fusion Studio.zip
on Linux and Windows. Fusion Studio is licensed for an unlimited number of Render nodes, so you can
install the Render Node software on as many macOS, Windows, and Linux computers that you want
involved in network rendering.

To install a Render node:

1 Download and unzip the Blackmagic Fusion Studio archive. Then locate the Render Node Installer.
2 Copy the Install Fusion Render Node [version] to each computer on the network that you want to
perform rendering operations.
3 Install the Render node.

By default, the Render Node application will be added to the Start Menu on Windows under
Blackmagic Design. On macOS, it is added to the menu bar, and on Linux it appears in the app
launcher. Each time you log in to the computer, the Render Node application will run automatically.
To disable the Render Node application from starting up automatically, choose Quit from the Render
Node icon in the macOS menu. On Linux, right-click over the icon and choose Kill Process, and on
Windows, delete the shortcut from the Windows Startup directory.

Chapter 4 Rendering Using Saver Nodes 86

Licensing for Network Rendering
Most versions of Fusion Studio are licensed by connecting a single-seat hardware key (dongle) to the
same computer where Fusion is installed. Each dongle includes an unlimited number of cross-platform
Render Node licenses, which you can install on as many macOS, Windows, and Linux computers as
you need. For Fusion Studio to access the Render Nodes, the computer with the Fusion Studio dongle
needs to be on the same local network subnet as the Render Nodes. The network licensing does not
require individual license files; instead, the Render Nodes automatically search for the dongle on the
subnet, making it easy to set up. Single-seat dongles do not “float” over a network; they must be
connected to the same computer where Fusion Studio operates.

Multi-License Dongles
Using a multi-license dongle, you can license 10 copies of Fusion Studio by connecting the dongle to
any computer on the same subnet. Since these licenses “float” over a network, Fusion Studio does not
have to be running on the same computer where the dongle is connected. As long as Fusion Studio is
on the same subnet, it can automatically find the license server and check out an available license.
Multi-seat dongles can be combined together to tailor the number of Fusion seats in a larger facility.
For example, three dongles each licensed for 10 Fusion Studios would serve up 30 licenses. This also
allows for redundancy. For instance, in the example above, three computers can act as license
servers. If the first server fails for some reason, Fusion Studio will automatically try the next server.
Alternatively, multiple dongles can also be plugged into a single computer.

Setting Up a License Server

Setting up the license for network rendering begins by connecting either a single-seat or multi-seat
dongle to a computer that will act as the host for the license server. The Render Node installer installs
Fusion Server, which acts as the license server, although the Render Node does not have to run on
that same computer. The Fusion Server is set up to launch at startup and run unobtrusively in the
background as a service/daemon, ready to serve licenses and Fusion bins. The Fusion Server is used
to serve up multiple licenses so it must be running whenever you want to operate Fusion Studio or use
the Render Nodes for network rendering. Its default is to start up automatically and remain running as
long as a license is being used by another computer or a Render Node is operating. If nothing is using
the Fusion Server, it will quit after 30 seconds.
You will need your network administrator to set firewall rules allowing the Fusion Server, FusionScript,
and the Fusion Render Node applications to communicate and confirm licensing with the computer
that has the Fusion Studio dongle.
If for some reason you remove a dongle or the network drops out, the licenses of any connected
Fusion Studio application will also drop. Upon losing its license, Fusion Studio will start searching for
another license, locally or on a different machine. If no license is found, Fusion pauses rendering and
displays a dialog with options to retry the search or autosave the comp and quit. Render Nodes only
check for a license on the network once during startup, so they are not affected by removing the
dongle or network issues.

Setting Up a License Server with Environment Variables

Environment variables provide a way to specify flexible or “variable” configuration options. When
network rendering with Fusion, environment variables can be useful for temporarily setting a location
or choosing a preference file. Using the FUSION_LICENSE_SERVER environment variable, you can
set different locations for the File Server.
Instead of looking in a single location for the Fusion Server, you can set up multiple license servers
separated by semicolons. For instance, fu:SetPrefs(“Global.EnvironmentVars.FUSION_LICENSE_
SERVER”, “192.168.1.12; 192.168.10.55;*”)
You can also use the environment variable to scan for license servers within a subnet—for example,
“bobs-mac.local;10.0.0.23;*;license.mystudio.com”. Including an asterisk (*) indicates a broadcast
search of the local subnet.

Chapter 4 Rendering Using Saver Nodes 87

Like most environment variables, you can put the license server in the Global Preferences via the Prefs
text file. EnvironmentVars: fu:SetPrefs(“Global.EnvironmentVars. FUSION_LICENSE_SERVER”,
“10.0.0.23;*”) fu:SavePrefs()
See Chapter 76, “Preferences,” in the DaVinci Resolve Reference Manual or Chapter 15 in the Fusion
Reference Manual for more information on using environment variables.

NOTE: The use of straight quotes (" ") in the environment variables above are intentional and
should not be replaced with typographer’s, or curly, quotes (“ ”).

Configuring the Render Master and Render Nodes

There are two roles played by computers involved in network rendering.
– The Render Master manages the list of compositions to be rendered (the queue) and allocates
frames to Render Nodes for rendering. Metaphorically speaking, the Render Master is the traffic
cop of this process.
– The Render Nodes are the main computers used for the rendering process. All computers
involved in network rendering must be on the same network subnet, and they all must have
access to the various files (including Fonts and third-party plug-ins) used to create the composite.
The path to the files must be the same for each computer involved in rendering.

Preparing the Render Master

The Render Master manages the list of compositions to be rendered (the queue) and allocates frames
to Render Nodes for rendering. The Render Master is also used to maintain the Render Node list and
to push updates to the various Render Nodes when needed. At least one computer in the render farm
must be configured to act as the Render Master.
Any copy of Fusion can act as a Render Master by setting up the Fusion Network Preferences.

The Network panel sets up computers as Render Masters.

Chapter 4 Rendering Using Saver Nodes 88

Acting as a Render Master has no significant impact on render performance. The system resources
consumed are insignificant. However, there are specific steps you must take to make one of your
computers a Render Master.
To set up the Render Master:
1 Install a copy of Fusion Studio on the computer you want to be the Render Master.
2 In Fusion Studio, choose Fusion Studio > Preferences on macOS or File > Preferences on
Windows and Linux
3 In the Preferences dialog, select the Global > Network Preferences panel.
4 Enter the name of the Render Master in the Name field and enter the IP address.
5 Enable the Make This Machine a Render Master checkbox.
6 If you want to use this computer as part of the render farm, enable the Allow This Machine to Be
Used as a Network Slave checkbox as well.

To have the Render Node act as the Render Manager:

– Select a node in the Render Manager and choose Set Default Master from the contextual menu.

Once a computer is enabled to act as the master, use the Render Manager to add the Render Nodes it
will manage. The Render Manager dialog is described in detail later in this chapter.

Preparing Render Nodes

Before you can begin rendering on the network, the Render Nodes must be set up to accept
instructions from the Render Master.

In Fusion Studio, you can enable the computer to be used as a Render Node in two ways:
– Choose File > Allow Network Renders.
– Enable the Allow This Machine to Be Used as a Network Slave in the Global >
Network Preferences.

On a macOS Render Node computer:

– Click the Render Node icon in the menu bar and choose Allow Network Renders.

The Render Node menu accessed

in the macOS menu bar

On a Linux Render Node computer:

– Right-click the Fusion Render Node icon in the App Launcher and choose Allow Network Renders.

On a Windows Render Node computer:

– Right-click the Fusion Render Node icon in the taskbar Notification area and choose Allow
Network Renders.

Chapter 4 Rendering Using Saver Nodes 89

Setting Up the Render Manager
The Render Manager window is used to monitor the progress of rendering. It can be used to reorder,
add, or remove compositions from a queue, and to manage the list of Render Nodes used for
rendering. To open the Render Manager window in Fusion Studio, choose File > Render Manager.

The Render Manager is used to reorder, add, and remove compositions from a render queue.

The Render Master is always listed as the first computer in the Slave list along the right side. This
allows the Render Manager to render local queues without using the network. For the Render Master
to control additional Render Nodes, the nodes must be added to the Slave list.
Right-clicking in the Slave list allows you to add Render Nodes by entering the Render Node’s name or
IP address. You can also choose Scan to have the Render Manager look for Render Nodes on the
local network.

The Render Manager is used to add Render Nodes.

Scanning for Render Nodes

With the Render Manager open, you can scan for Render Nodes by choosing Slave > Scan for Slaves
in the menu bar or by right-clicking in the Render Manager’s Slave list and choosing Scan for Slaves
from the drop-down menu.

Scanning searches IP addresses on the

subnet for active Render Nodes.

Chapter 4 Rendering Using Saver Nodes 90

Scanning looks through all IP addresses on the network subnet to determine whether any other
computers in the local network are actively responding on the port Fusion uses for network rendering.
A copy of the Fusion Render Node must be running on the remote computer in order for it to be
detected by the scan.

Manually Adding Render Nodes

To manually add a Render Node to the Slave list, select Add Slave from the Slave menu or right-click in
the Render Manager’s Slave list and choose Add Slave from the drop-down menu.

The Add Slave dialog allows you to

manually enter a Render Node name or
IP address to locate it on the subnet.

In the Add Slave dialog that opens, enter the name or the IP address of the remote Render Node.
The Render Manager will attempt to resolve names into IP addresses and IP addresses into names
automatically. You can use this method to add Render Nodes to the list when they are not currently
available on the network.

Removing Render Nodes

To remove a computer from the Slave list, select the Render Node in the list and choose Slave >
Remove Slave(s) or right-click over the Render Node in the Slave list and choose Remove Slave(s) from
the pop-up menu. You can use Command on macOS or Ctrl on Windows and Linux to select multiple
Render Nodes for removal.

Loading and Saving Render Node Lists

The list of Render Nodes is automatically saved in the Documents > Blackmagic Design > Fusion >
Queue folder when you quit the Render Manager. You can save and reload alternative lists of Render
Nodes by choosing Slaves > Save Slave List and Load Slave list from the menu.

Submitting Comps to Network Render

To submit a comp to render on the network, you can use the Render Manager, the Render Settings
dialog, or a third-party render farm application. The Render Settings dialog is quicker, while the the
Render Manager and third-party render farm applications can provide more feedback and control over
the process.

Using the Render Settings Dialog for Network Rendering

When starting a preview or a final render, selecting the Use network checkbox from the Render
Settings dialog and submitting the render adds a composition to the end of the current queue in the
Render Manager. The Render Master used is based on the Fusion preferences from the workstation
submitting the comp.

Chapter 4 Rendering Using Saver Nodes 91

 The Use Network checkbox
enables network rendering from
the Render Settings dialog.

NOTE: Distributed network rendering works for image sequences like EXR, TIFF, and DPX.
You cannot use network rendering for Quicktime, H264, ProRes, or MXF files.

Using the Render Manager Window for Network Rendering

The Render Manager uses a render queue that lets you batch render comps. Compositions are
rendered in the order in which they’re listed in the Render Manager, with the top entry rendered first,
followed by the next item down, and so on. Multiple comps in a queue may render simultaneously
depending on the group of Render Nodes they are using and the priority assigned to each comp.

To add a comp to the queue in the Render Manager:

– Click the Add Comp button and navigate to the comp on your hard drive.
– Right-click in the queue list and select Add Comp from the drop-down menu, and then navigate to
the comp on your hard drive.
– Drag a comp file from an OS window into the Render Manager’s queue list.

The Add Comp button in the Render Manager adds a

comp to the queue for batch rendering over the network.

Removing a Composition from the Queue

To remove a composition from the queue, select the composition in the queue list and press the
Backspace/Delete key or right-click over the comp in the queue list and choose Remove Composition
from the drop-down menu.

Saving and Reloading Queue Lists

It can be useful to save a queue list to reuse at a later time. The current queue list is saved in the
Documents > Blackmagic Design > Fusion > Queue folder. To save the current queue with a new
name, choose File > Save Queue As in the Render Manager menu bar. To reload a saved Queue,
choose File > Load Queue and navigate to the saved location.

Reordering the Queue Lists

In the middle of a job, priorities for finishing a composition may change. Shifting deadlines may require
that a composition further down in the queue be rendered sooner. You can move comps to a new
position in the queue by dragging them in the queue list. If a composition with a status set to Done is
moved lower in the queue, it does not re-render. To re-render a comp, right-click on the comp in the
queue and choose Clear Completed frames from the drop-down menu.

Chapter 4 Rendering Using Saver Nodes 92

Working with Node Groups
Render Nodes can be configured into Groups, which are then used when submitting compositions.
For example, imagine you have five Render Nodes. All the Render Nodes are members of the group
named All. Two of the Render Nodes include more memory and faster processors, so you create a
new group called Hi-Performance.
New Render Nodes are automatically added to All, but you can assign them to other groups as well.

To assign a Render Node to a Group:

1 Open the Render Manager and select the Render Nodes to assign to a group.
2 Choose Slave > Assign Group or right-click over the Render Nodes and choose Assign Group
from the drop-down menu.
3 In the Choose Group dialog, enter a name for the group.
4 To assign Render Nodes to multiple groups, separate the name of each group using a comma
(e.g., All, Local, or Hi-Performance). The order of the groups determines the priority. See “Using
Multiple Groups” below.

When a render is submitted to the network, it is automatically sent to the All group. However, you can
choose to submit it to other groups in the list.

To submit a comp to a group from the Render Manager:

1 Open the Render Manager.
2 Submit the comp.
3 Click the Pause Render button.
4 Right-click over the comp in the queue list and select Assign Group.
5 In the Assign Group dialog, select an existing group to render the comp and click OK.
6 Click the Resume Render button.

To submit a comp to a group from the Render Settings dialog:

1 Click the Render button in the transport controls area.
2 In the Render Settings dialog, enable the Use Network checkbox.
3 Select an existing group from the Available groups list.
4 Click the Start Render button.

Continuing with the group example above, five Render Nodes are contained in the All group,
and two of those Render Nodes are also in the Hi-Performance group. If you submit a render to the
Hi‑Performance group, only two of the computers on the network are used for rendering.
If a composition is then submitted to the All group, the remaining three machines will start rendering
the new composition. Once the two Render Nodes in the Hi_Performance group complete the first
job, they join the render in progress on the All group.
Groups are optional and do not have to be used. However, groups can make managing large networks
of Render Nodes easier and more efficient.

Chapter 4 Rendering Using Saver Nodes 93

Using Multiple Node Groups
A single Render Node can be a member of multiple groups. A single composition can also be
submitted to multiple groups. Submitting a composition to multiple groups results in it rendering on all
Render Nodes in the selected groups.
When a Render Node is a member of multiple groups, the order of the groups is important because
the order defines the priority for that Render Node.
For example, if groups are assigned to a Render Node as All, Hi-Performance, then renders submitted
to the All group take priority. Any renders in progress that were submitted to the Hi-Performance
group will be overridden. If the order is changed to Hi-Performance, All, then the priority is reversed.

Viewing the Render Log

The Render Log is displayed in the lower half of the Render Manager window, although it can also be
displayed in the console window. The text in the log displays the Render Manager activities, including
which frame is assigned to which Render Node, which Render Nodes have loaded the compositions in
the queue, and statistics for each render after completion.

To view the Render Log in the console:

– Open the Render Manager and choose Misc > Show Render Log.

There are two modes for the Render Log: a Verbose mode and a Brief mode. Verbose mode logs all
events from the Render Manager, while Brief mode logs only which frames are assigned to each
Render Node and when they are completed.

To disable Verbose mode:

– Choose Misc > Verbose Logging from the Render Manager’s menu bar.

Using Third-Party Render Managers with Fusion Studio

You can make use of third-party render manager software to control network rendering. This allows for
efficient sharing of your computer resources between the many applications that may make use of
them. Examples of such managers are Smedge from Uberware LLC, Rush from Seriss, and Deadline
from GetRender. Generally, these render managers use a command line renderer. By default, Fusion’s
Render Nodes operate as a service to the Fusion internal render manager. However, you can also run
the Render Nodes via the command line for third-party render managers.
Keep in mind that using a third-party render manager will prevent the use of some of Fusion’s network
rendering features, such as the ability to create network rendered Flipbook Previews and disk caches.

Command Line Rendering

For studios using third-party render farm managers like Deadline, Smedge, or Rush, the Fusion Render
Node can be called via command line passing arguments and file paths. In this Windows example, a
Render Node is called to load a composition called exampleV001, and render 10 frames:
//pathtoRN/FusionRenderNode.exe //pathtoProject/exampleV001.comp -render
-start 101 -end 110 -quit

This would start up, render frames from 101 to 110, and then quit.

Chapter 4 Rendering Using Saver Nodes 94

The following table lists additional command line features.

Command Description

Installs the license server as a service or daemon,

“Fusion Server -i”
launching it on startup before user login.

Causes the Fusion Server to run persistently in the

“Fusion Server -S” (capital S)
background, until force-quit.

Full path and comp name, like

<filename.comp>
/storage/project/episode/shot/filename.comp.

-render Tells the Render Node to render.

Passes a series of frame ranges to be rendered—

-frames <frameset>
e.g., 101..110,120,121,130..150.

-start <frame> Sets the start frame of the render.

-end <frame> Sets the last frame of the render.

Normally set to 1, step skip frames for rendering.

-step <step>
For instance, 2 would render every second frame

Causes the Render Node to quit after the render is

-quit
complete.

Prompts the node to connect to a manager at

-join <host>
<hostname,IP>, and (re)join any ongoing renders.

The node remains running and waits for incoming

-listen
requests from a manager.

Causes the Render Node to output information

-log <filename> about the render to a log file. This appends to the
end of an existing log file.

-cleanlog Clears existing text from a log file.

-verbose Outputs more detailed information into the log file.

Suppresses pop-ups and interface buttons from

-quiet
displaying and needing interaction.

-version Returns the Render Node version number.

Sets the node’s process priority to high, above,

-pri high|above|normal|below|idle
normal, below, or idle.

Allows storing custom values that can be fetched

-args <arg1> [, <arg2> ...] by calling the script function GetArgs(), which will
return a table of { <arg1>, <arg2>, ... }

TIP: An X11 virtual frame buffer is required to make a headless Linux command line
interface work.

Chapter 4 Rendering Using Saver Nodes 95

Preparing Compositions
for Network Rendering
The way you construct a composition in Fusion Studio can help or hinder network rendering.
The media you read in, where plug-ins are installed, and the mix of operating systems on your
networked computers all play a part in how smoothly your network rendering goes. Your setup must
include several essential parts before network rendering will work:
– License dongle, Render Master, and Render Nodes must be on the same local network (subnet).
– Fusion Server must be running as a background service on the same
computer where the dongle is installed.
– All source media from the comp should be placed on a network volume.
– The network volume must be mounted on each Render Node.
– Loaders must point to the media on the mounted volumes.
– Savers must write to a drive that is mounted on each render node.
– The Fusion comp must be saved to a volume that is mounted on each Render Node.
– All Render Nodes and Render Masters need read and write access to any volumes specified as a
source media location or render destination.
– Make sure all fonts used in the comp for Text+ and 3D text nodes are installed on all the
Render Nodes
– Make sure all Render Nodes have third-party OFX plug-ins installed if any are used in the comp.

Below are more details about some of these items.

Using Relative Paths

The file paths used to load a composition and its media, and to save the composition’s rendered
results, are critical to the operation of network rendering. Each computer used for network rendering
must be able to access the media location for each Loader in the comp. Savers must be set up to save
to folders that all Render Nodes can see and to which all Render Nodes have write permission. Even
the composition must be saved in a folder accessible to all Render Nodes, and it should be added to
the queue list using a path that’s visible to all Render Nodes.
For example, if you open a composition located at c:\compositions\test1.comp in Fusion Studio and
add the composition to the network rendering queue, the Render Manager sends a message to each
Render Node to load the composition and render it. The problem is that each computer is likely to
have its own c:\drive that does not contain the comp you created. In most cases, the Render Nodes
will be unable to load the composition, causing it to fail.
Path Maps located in Fusion Preferences are virtual paths used to replace segments of file paths. They
can change the absolute paths used by Loader and Saver nodes to relative paths. There are a number
of Path Maps already in Fusion, but you can also create your own. The most common path to use is the
Comp:\ path.
Comp:\ is a shortcut for the folder where the actual composition is saved. So, using Comp:\ in a Loader
makes the path to the media file relative, based on the saved location of the comp file. As long as all
your source media is stored in the same folder or subfolder as your comp file, Fusion locates the
media regardless of the actual hard drive name.

Chapter 4 Rendering Using Saver Nodes 96

Here’s an example of a file structure that enables you to use relative file references. The composition
is stored in the following file path:
Volumes\Project\Shot0810\Fusion\Shot0810.comp

And your source media is stored here:

Volumes\Project\Shot0810\Fusion\Greenscreen\0810Green_0000.exr

This overall directory structure can be seen in the following screenshot:

File paths can use relative paths based on the location of the saved comp file.

In this situation, using the Comp:\ path means your media location starts from your comp file’s location.
The relative path set in the Loader node would then be:
Comp:\Greenscreen\0810Green_0000.exr

Replacing the Loader’s path to start with Comp:\

creates a relative path from the comp file’s location.

If your source media’s actual file path uses a subfolder in the same folder as the comp file’s folder:
Volumes\Project\Shot0810\Footage\Greenscreen\0810Green_0000.exr

The relative path set in the Loader node would then be:
Comp:\..\Footage\Greenscreen\0810Green_0000.exr

The two dots .. instruct the path to go up one folder.

TIP: Some Path Maps are not set up on a Fusion Render Node automatically. For instance,
you must manually add an entry for macros if you are using macros in your comp.

Using Mapped Drives

Having the comp and media set to relative paths solves only part of the problem. Each computer
involved in rendering needs to have access to the drive where the comp and source media are
located. Mapping a drive provides permanent access to a folder on another computer or network
storage device.
Windows assigns a new drive letter to the folder, and it can be accessed just like any other drive
connected to your computer. Mapped drives assign a letter of the alphabet to a shared network
resource. Your shared drives must be the same drive letters on all Render Nodes. For example, if your
media is on drive Z, then the network drive must appear as the letter Z on each of the Render Nodes.

Chapter 4 Rendering Using Saver Nodes 97

On macOS, you can map a network drive using Connect to Server from the Go menu. Entering the
smb:// path to the drive will mount it on the computer. Using Accounts > LogIn Items, you can have the
network drive auto-mount after a reboot as well.

Installing All Fonts on Render Nodes

All fonts used by Text tools in the composition must be available to all nodes participating in the
render. The render will otherwise fail on the slaves that do not have the font installed.

Installing Third-Party Plug-Ins on Render Nodes

All third-party plug-ins and tools used by a composition must be installed in the plug-ins directory of
each Render Node. A Render Node attempting to render a composition that uses a plug-in that’s not
installed will fail to render. Licensed plug-ins are required on each Render Node.

Other Uses of Network Rendering

Although you will probably set up network rendering for the purpose of accelerating the output of your
final renders, Fusion is capable of using the network for other purposes as well. You can use the
nerdier nodes on your network to accelerate the creation fo Flipbook Previews and disk
caching as well.

Flipbook Previews
Fusion Studio is able to use Render Nodes to accelerate the production of Flipbook Previews, allowing
for lightning fast previews. Frames for previews that are not network rendered are rendered directly
into memory. Select the Use Network checkbox and the Render Nodes to render the preview frames
to the folder set in the Preferences Global > Path > Preview Renders. This folder should be accessible
for all Render Nodes that will participate in the network render. The default value is Temp\, which is a
virtual path pointing to the system’s default temp folder. This will need to be changed before network
rendered previews can function. Once the preview render is completed, the frames that are produced
by each Render node are spooled into memory on the local workstation. As each frame is copied into
memory, it is deleted from disk.

Disk Cache
Right-clicking a node in the Node Editor and choosing Cache to Disk opens a dialog used to create
the disk cache. If you enabled the Use Network checkbox and click the Pre-Render button to submit
the disk cache, the network Render Nodes are used to accelerate the creation of the disk cache.

Render Nodes can be used for disk caching as well as final renders.

Chapter 4 Rendering Using Saver Nodes 98

When Renders Fail
It is a fact of life that render queues occasionally fail. The composition has an error, the power goes
out, or a computer is accidentally disconnected from the network are some causes for failure. If no one
is available to monitor the render, the risk that an entire queue may sit inactive for several hours may
become a serious problem.
Fusion Studio includes a variety of measures to protect the queue and ensure that the render
continues even under some of the worst conditions.

Automatic Rejoining of the Queue

If a Render Node becomes unavailable to the Render Master for any reason, frames assigned to that
Render Node are reassigned among the remaining Render nodes in the list.
When the Render Node becomes available for rendering again, it will signal the Render Master that it
is ready to render again, and new frames will be assigned to that Render Node.
This is why it is important to set the Render Master in the network preferences of the Render Nodes.
If the master is not set, the Render Node will not know what master to contact when it
becomes available.
In the Fusion Render Node Preferences, select the Tweaks panel. Using the Last Slave Restart
Timeout field, you can enter the number of seconds Fusion waits after the last Render Node goes
offline before aborting that queue and waiting for direct intervention.

Relaunching Render Nodes with Fusion Server

Fusion Server is a small utility installed with Fusion Studio and the Render Node. The application is
silently launched by each Fusion Render Node when started.
Fusion Server monitors the Render Node to ensure that the Render Node is still running during a
render. It consumes almost no CPU cycles and very little RAM. If the monitored Render Node
disappears from the system’s process list without issuing a proper shutdown signal, as can happen
after a crash, the Fusion Server relaunches the Render Node, allowing it to rejoin the render.
Fusion Server will only detect situations where the Render Node has exited abnormally. If the Render
Node is still in the process list but has become unresponsive for some reason, the Fusion Server
cannot detect the problem. Hung processes like this are detected and handled by frame timeouts, as
described below.

Frame Timeouts
Frame timeouts are a fail-safe method of canceling a Render Node’s render if a frame takes longer
than the specified time (with a default of 60 minutes, or one hour). The frame timeout ensures that an
overnight render will continue if a composition hangs or begins swapping excessively and fails to
complete its assigned frame.
The timeout is set per composition in the queue. To change the timeout value for a composition from
the default of 60 minutes, right-click on the composition in the Render Manager’s queue list and select
Set Frame Timeout from the contextual menu.

Chapter 4 Rendering Using Saver Nodes 99

Right-click over a comp in the Render Manager to set a timeout value.

To change the frame timeout value, choose Set Frame Time Out from the Render Manager’s Misc
menu and enter the number of seconds you want for the Time Out.

Heartbeats
Often, the network environment is made up of computers with a variety of CPU and memory
configurations. The memory settings used on the workstation that created a composition may not be
appropriate for all the Render Nodes in the network. The Render Node software offers the ability to
override the memory settings stored in the composition and use custom settings more suited to the
system configuration of a specific Render Node.
The number of heartbeats in a row that must be missed before a Render Node is removed from the list
by the manager, as well as the interval of time between heartbeats, can be configured in the Network
Preferences panel of the master. The default settings for these options are fine for 90% of cases.
If the compositions that are rendered tend to use more memory than is physically installed, this will
cause swapping of memory to disk. It may be preferable to increase these two settings somewhat to
compensate for the sluggish response time until more RAM can be added to the slave.

Managing Memory Use

Often, the network environment is made up of computers with a variety of CPU and memory
configurations. The memory settings used on the workstation that created a composition may not be
appropriate for all the Render Nodes in the network. The Render Node software offers the ability to
override the memory settings stored in the composition and use custom settings more suited to the
system configuration of a specific Render Node.
To access preferences for a node, right-click on the icon in the Windows Notification area or from the
macOS menu bar and choose Preferences. In the Preferences dialog, select the Memory panel.

Override Composition Settings

Enable this option to use the Render Node’s local settings to render any incoming compositions.
Disable it to use the default settings that are saved into the composition.

Render Several Frames at Once

Fusion has the ability to render multiple frames at once for increased render throughput. This slider
controls how many frames are rendered simultaneously. The value displayed multiplies the memory
usage (a setting of 3 requires three times as much memory as a setting of 1).
Normal values are 2 or 3, although machines with a lot of memory may benefit from higher values,
whereas machines with less memory may require the value to be 1.

Chapter 4 Rendering Using Saver Nodes 100

Simultaneous Branching
Enable this option to render every layer in parallel. This can offer substantial gains in throughput but
may also use considerably more memory, especially if many layers are used in the composition.
Machines with limited memory may need to have Simultaneous Branching disabled when rendering
compositions with many layers.

Limitations of Render Nodes

There are a few important limitations to remember while setting up compositions and
rendering over a network.

Time Stretching
Compositions using the Time Stretcher and Time Speed tools may encounter difficulties when
rendered over the network. Speeding up or slowing down compositions and clips requires fetching
multiple frames before and after the current frame that is being rendered, resulting in increased I/O to
the file server. This may worsen bottlenecks over the network and lead to inefficient rendering. If the
composition uses the Time Stretcher or Time Speed tools, make certain that the network is up to the
load or pre-render that part of the composition before network rendering.

Linear Tools
Certain tools cannot be network rendered properly. Particle systems from third-party vendors, such as
Genarts’s Smoke and Rain, and the Fusion Trails node cannot render properly over the network. These
tools generally store the previously rendered result and use it as part of the next frame’s render, so
every frame is dependent on the one rendered before it. This data is local to the tool, so these tools
do not render correctly over a network.

Saving to Multi-Frame Formats

Multiple machines cannot render a single QuickTime file. Always render to separate sequential file
formats like EXR, DPX, JPEG, and so on. Once the render is complete, a single workstation can load
the image sequence in order and save to the desired compiled format.

NOTE: The above does not apply to network rendered previews, which are previews created
over the network that employ spooling to allow multi-frame formats to render successfully.
Only final renders are affected by this limitation.

Troubleshooting
There are some common pitfalls when rendering across a network. Virtually all problems with network
rendering have to do with path names or plug-ins. Return to the “Preparing Compositions for Network
Rendering” section in this chapter to review some of the essential setup requirements. Verify that all
Render Nodes can load the compositions and the media, and that all Render Nodes have installed the
plug-ins used in the composition.
If some difficulties persist, contact Blackmagic Design’s technical support using the support section on
the Blackmagic Design website. Save a copy of the render.log file to send to technical support.

Chapter 4 Rendering Using Saver Nodes 101

Checking the Render Log
The log file shown in the Render Manager dialog displays messages that can assist with diagnosing
why a render or node has failed. The render log shows a step-by-step account of what happened (or
didn’t happen) during a render. If a Render Node cannot be found, fails to load a composition or render
a frame, or simply stops responding, it will be recorded here.

Check the Composition

The Render Manager’s Status field in the render log indicates if a composition fails to render. Some
possible causes of this are as follows:
– No Render Nodes Could Be Found: On the Preferences Network tab, make sure that there is at
least one Render Node available, running and enabled. If all Render Nodes are listed as Offline
when they are not, check the network.
– The Composition Could Not Be Loaded: Some Render Nodes may not be able to load a
composition while others can. This could be because the Render Node could not find the
composition (check that the path name of the composition is valid for that Render Node) or
because the composition uses plug-ins that the Render Node does not recognize.
– The Render Nodes Stop Responding: If a network link fails, or a Render Node goes down
for some reason, the Render Node will be removed from the active list and its frames will be
reassigned. If no more Render Nodes are available, the composition will fail after a short delay
(configurable in network preferences). If this happens, check the render log for clues as to which
Render Nodes failed and why.
– The Render Nodes Failed to Render a Frame: Sometimes a Render Node simply cannot render
a particular frame. This could be because the Render Node could not find all the source frames it
needed, or the disk it was saving to become full or because of any other reason for which Fusion
might normally be unable to render a frame. In this case, the Render Manager will attempt to
reassign that failed frame to a different Render Node. If no Render Node can render the frame, the
render will fail. Try manually rendering that frame on a single machine and observe what happens.
– Check the Render Nodes: Fusion’s Render Manager incorporates a number of methods to
ensure the reliability of network renders. Periodically, the Render Manager will send signals called
Heartbeats, generated at regular intervals, to detect network or machine failures. In this event, a
failed Render Node’s outstanding frames are reassigned to other Render Nodes where possible.
In rare cases, a Render Node may fail in a way that the heartbeat continues even though the
Render Node is no longer processing. If a Render Node failed (although the Render Master may
not have detected it) and you do not want to wait for the Frame Timeout, simply restart the Fusion
workstation or Fusion Render Node that has hung. This triggers the heartbeat check, reassigns
the frames on which that Render Node was working, and the render should continue. Heartbeats
may fail if the system that is performing the render is making extremely heavy use of the Swap
file or is spending an extraordinary amount of time waiting for images to become available over a
badly lagged network. The solution is to provide the Render Node with more RAM, adjust memory
settings for that node, or upgrade the network bandwidth.
– Check the Network: At the Render Master, bring up the Network tab of the Preferences dialog
box and click Scan. If a Render Node is not listed as running, the Render Master will not be able to
contact it for network rendering. Alternatively, bring up a command prompt and ping the Render
Nodes manually. If the remote systems do not respond when they are up and running, the network
is not functioning and should be examined further.

Chapter 4 Rendering Using Saver Nodes 102

Chapter 5

Working in the
Node Editor
This chapter discusses how to work in the Node Editor, including multiple ways to
add, connect, rearrange, and remove nodes to create any effect you can think of.

Contents
Learning to Use the Node Editor  105
Navigating within the Node Editor  106
Automatic Node Editor Navigation  106
Using the Node Navigator  106
Node View Bookmarks  107
Adding Nodes to a Composition  109
Adding, Inserting, and Replacing Nodes Using the Toolbar  109
Adding Nodes Quickly Using the Select Tool Window  110
Adding Nodes from the Effects Library  111
Adding, Inserting, and Replacing Nodes Using the Contextual Menu  114
Deleting Nodes  114
Disconnected Nodes  114
Selecting and Deselecting Nodes  115
Selecting Nodes  115
The Active Node  115
Deselecting Nodes  116
Loading Nodes into Viewers  116
Viewed Nodes When You First Open Fusion  117
Node View Indicators  117
Drag and Drop Nodes into a Viewer  117

Chapter 5 Working in the Node Editor 103

Using the Contextual Menu  118
Clearing Viewers  118
Create/Play Preview  118
Connecting and Disconnecting Nodes  118
Node Basics  118
How to Connect Nodes  119
Dropping Connections on Top of Nodes  119
Identifying Node Inputs  121
Node Order Matters  122
Upstream and Downstream Nodes  124
Disconnecting and Reconnecting Nodes  124
Tracing Connections Through the Node Tree  125
Branching  125
Connecting Merge Nodes  126
Automatically Creating a Merge Node When Adding Nodes  127
Automatically Creating a Merge Node by Connecting Two Outputs  127
Connection Options and Routers  128
Using Routers to Reshape and Branch Connections  129
Swapping Node Inputs  130
Extracting and Inserting Nodes  130
Cut, Copy, and Paste Nodes  131
Cut, Copy, and Paste in the Node Editor  131
Pasting Node Settings  132
Copying and Pasting Nodes to and from Any Text Editor  132
Instancing Nodes  133
Using Instanced Nodes  134
De-Instancing and Re-Instancing Specific Parameters  135
Keeping Node Trees Organized  135
Moving Nodes  135
Renaming Nodes  136
Changing Node Colors  137
Using Sticky Notes  137
Using Underlay Boxes  138
Node Thumbnails  139
Choosing Which Nodes Show Thumbnails  139
Switching Thumbnails between Images and Icons  140
Finding Nodes  141
Performing Simple Searches  141
Using Regular Expressions  142

Chapter 5 Working in the Node Editor 104

Custom Node Settings  143
Managing Saved Settings  143
Resetting Defaults  143
Saving and Loading Alternate Node Settings  144
Adding Saved Settings from the File System  144
Node Modes Including Disable and Lock  144
Node Editor Options  145
Node Tooltips and the Status Bar  146

Learning to Use the Node Editor

The Node Editor (formerly called the Flow or Flow Editor) is the heart of Fusion’s compositing interface.
It uses a flowchart structure called a node tree that lets you build a composition out of interconnected
nodes, as opposed to using layers in a layer list. Each clip you add to the composition, and each
image-processing operation you apply to those clips, is added as a node, all of which are joined
together with connections that propagate image data from one node to the next. Each individual node
performs a relatively simple operation, but collectively they combine to let you create wonderfully
complex results.

The Node Editor.

This chapter discusses how to work in the Node Editor in greater detail, showing you how to add,
connect, rearrange, and remove nodes to create any effect you can think of.

To display the Node Editor:

– Click the Nodes button on the UI toolbar.

The Nodes button in the UI toolbar.

Chapter 5 Working in the Node Editor 105

Navigating within the Node Editor
The Node Editor is the place where everything relating to nodes and the construction of your
composites happens. The more you learn about how to navigate within the Node Editor, the faster
you’ll be able to work. There are a variety of standard methods of panning and zooming around the
Node Editor, many of which are shared with other panels in Fusion.

Methods of panning the Node Editor:

– Middle-click and drag to pan around the Node Editor.
– Hold down Shift and Command, and then click and drag within the Node Editor to pan.
– Drag with two fingers on a track pad to pan in the Node Editor

Methods of zooming the Nod4e Editor:

– Press the Middle and Left buttons simultaneously and drag to resize the Node Editor.
– Hold down the Command key and use your pointer’s scroll control to resize the Node Editor.
– Right-click the Node Editor and choose an option from the Scale submenu of the contextual menu.
– Press Command-1 to reset the Node Editor to its default size.
– Hold down the Command key and drag with two fingers on a track pad to resize the Node Editor.

Automatic Node Editor Navigation

If a node that is not visible in the Node Editor becomes selected, either by using the Find command or
by selecting a node’s header in the Inspector, the Node Editor will automatically pan to display the
node in the visible area.

Using the Node Navigator

Another useful way to pan around the Node Editor is to use the Node Navigator. The Node Navigator
is a small rectangular overview in the upper-right corner of the Node Editor. It gives a bird’s eye view
of the entire composition, with an inner outline that indicates the portion of the composition that is
visible in the panel. You can use the Node Navigator when you are zoomed in on a node tree and
want to pan around a composition.

The Node Navigator.

To display or hide the Node Navigator, do one of the following:

– Right-click in an empty area of the Node Editor, and then choose Options > Show Navigator.
– Press the V key.

Chapter 5 Working in the Node Editor 106

To have the Node Navigator resume displaying automatically when needed after you’ve closed it:
– Right-click in an empty area of the Node Editor, and then choose Options > Auto Navigator.

To change the size of Node Navigator, do the following:

– Drag the lower-left corner of the Navigator to resize it.

Drag the corner to resize the Navigator.

To return to the default Node Navigator size, do the following:

– Right-click anywhere within the Node Navigator and choose Reset Size.

To pan the Node Editor using the Node Navigator, do the following:
– Drag within the Node Navigator to move around different parts of your node tree.
– Within the Navigator, drag with two fingers on a track pad to move around different
parts of your node tree.

Node View Bookmarks

Bookmarks are another way of navigating the Node Editor. Bookmarks save the position and scale of
the Node Editor, so you can quickly and precisely jump from viewing one group of nodes to
viewing another.

To add a bookmark, do the following:

1 Pan and scale in the Node Editor to view a group of nodes you are interested in.
2 From the Options menu in the upper right of the Node Editor, choose Add Bookmark, or
press Cmd-D.
3 In the Manage Bookmarks dialog that opens, enter a name for the bookmark and click the
Add button.

The first nine saved bookmarks are given keyboard shortcuts and listed in the Options menu. They are
also listed in the Go To Bookmarks dialog along with any saved bookmarks beyond the initial nine.

Chapter 5 Working in the Node Editor 107

 The Node Editor Options menu
with nine bookmarks added

TIP: You can return the Node Editor to the default scale by right-clicking in the Node Editor
and choosing Scale > Default Scale or pressing Cmd-1.

If your Node Tree changes and you want to update Bookmark names or delete bookmarks, those
tasks can be done in the Manage Bookmarks dialog.

To rename or delete a bookmark, do the following:

1 From the Options menu in the upper right of the Node Editor, choose Manage Bookmarks.
2 In the Manage Bookmarks dialog that opens, right-click over the bookmark and
choose Rename or Remove.
3 Click OK to close the Manage Bookmarks dialog.

Using Bookmarks
You can jump to a Bookmark view by selecting a bookmark listed in the Options menu or choosing
Go To Bookmarks to open the Go To Bookmarks dialog. The Go To Bookmarks dialog has all the
bookmarks listed in the order they were created in the current composition. Double-clicking on any
entry in the dialog will update the Node Editor to that view and close the Go To Bookmarks dialog.
If you have a long list of bookmarks, you can use the search field at the bottom of the dialog to enter
the name of the bookmark you want to find.

Changing the Sort Order and Assigning Keyboard Shortcuts

Bookmarks appear in the Options menu and in the Go To Bookmarks dialog in the order they were
created. The top nine bookmarks listed are assigned keyboard short cuts. If you want to change the
sort order of the list you can do that in the Manage Bookmarks dialog. This can be useful if you want
some bookmarks to have specific keyboard shortcuts, or you want a bookmark you made early in the
process to not have a keyboard shortcut assigned to it.

To change the order of the listed bookmarks, do the following:

1 From the Options menu in the upper right of the Node Editor, choose Manage Bookmarks.
2 In the Manage Bookmarks dialog drag a bookmark up or down in the list.
3 An insert line appears where the bookmark will be inserted. Release the mouse button when the
insert line is where you want the bookmark to be listed.
The keyboard short cuts will update to reflect the new order.

Chapter 5 Working in the Node Editor 108

 TIP: You can hold down the Shift key to select multiple bookmarks and move them
simultaneously up or down in the Manage Bookmark list.

Using Underlays as Bookmarks

Underlays added to the Node Editor are automatically added as bookmarks. They are listed in the
Options menu below the list of your custom bookmarks. You can omit Underlays from showing in the
list of bookmarks by opening the Go To Bookmarks dialog and disabling the checkbox to Show
Underlays. When the checkbox is disabled, Underlays will not show in the Go To Bookmarks dialog,
and they will not appear in the Options menu.

Adding Nodes to a Composition

– You can add nodes to the Node Editor in a variety of ways, depending on the type of node you’re
adding, and how much guidance you need to find what you’re looking for. Additionally, the way
you add nodes to a composition may also be dictated by how you need to attach that node to the
current node tree.

Make Sure You’re Adding Compatible Nodes

It’s a good rule of thumb to make sure that whenever you’re adding or inserting new nodes to
the node tree, that you’re adding nodes that are compatible with the nodes you’re trying to
attach to. For example, you’ll have no problem inserting a Blur, Color, Filter, Paint, or Position
node after almost any 2D operation. However, if you try to add a Merge3D node after a Glow
node, it won’t automatically connect, because those two nodes cannot be connected directly.

Adding, Inserting, and Replacing Nodes Using the Toolbar

The Fusion toolbar, located above the Node Editor, displays a selection of frequently-used nodes,
displayed as buttons with distinct icons. These buttons make it fast to add Merge, Background, Paint,
Mask, Transform, and many other commonly used nodes with the click of a button, or the drag of
your pointer.

The Fusion page toolbar.

TIP: If you don’t know which node a particular icon corresponds to, just hover your pointer
over any toolbar button and a tooltip will display the full name of that tool.

Methods of adding nodes by clicking toolbar buttons:

– To add a node after a selected node: Select a node in the Node Editor and then click a
toolbar button.
– To add a disconnected node to the Node Editor: Deselect all nodes in the Node Editor and then
click a toolbar button.

Chapter 5 Working in the Node Editor 109

Methods of adding nodes by dragging toolbar buttons:
– To insert a new node into the node tree: Drag a toolbar button into the Node Editor and onto the
connection line between any two compatible nodes. When the connection highlights as the node
is over it, drop the node and it’ll be inserted.
– To create a disconnected node: Drag a toolbar button into an empty part of the Node Editor.
Dragging a toolbar button into the Inspector also creates a disconnected node.
– To insert a new node after a node loaded into a viewer: Drag a toolbar button onto a viewer to
insert a new node after whichever node is viewed, regardless of whether any nodes are selected.

To replace a node in the Node Editor with a node from the toolbar:
1 Drag a button from the toolbar so that it’s directly over the node in the Node Editor that you want
replaced. When the node underneath is highlighted, drop the node.

Dragging a node from the toolbar to replace an existing tool.

2 Click OK in the dialog to confirm the replacement.

TIP: When you replace one node with another, any settings that are identical between the
two nodes are copied into the new node. For example, replacing a Transform node with a
Merge will copy the existing center and angle values from the Transform to the Merge.

Adding Nodes Quickly Using the Select Tool Window

The next fastest way of adding or inserting nodes to the Node Editor is using the Select Tool window,
which lets you search for any node available to Fusion by typing a few characters. Once you learn this
method, it’ll probably become one of your most frequently-used ways of adding nodes.

To use the Select Tool window to add nodes:

1 Do one of the following to determine if you want to insert a node or create a disconnected node:
– If you want to insert a node, select a node that’s compatible with the one you’ll be creating, and
the new node will be inserted after it.
– If you want to create a disconnected node, deselect all nodes.

2 Press Shift-Spacebar to open the Select Tool dialog.

3 When the window appears, type characters corresponding to the name of the node you’re looking
for. A list automatically appears with likely candidates, and you can use the Up and Down arrow
keys to select the correct node (if it’s not selected already).
4 When you’ve selected the correct node, press the Return key (or click OK), and that node will be
either inserted or added.

Chapter 5 Working in the Node Editor 110

 The Select Tool dialog lets you find any
node quickly if you know its name.

TIP: Whenever you use the Select Tool window, the text you entered is remembered the next
time you open it, so if you want to add another node of the same kind—for example, if you
want to add two Blur nodes in a row—you can just press Shift-Spacebar and then press
Return to add the second Blur node.

Adding Nodes from the Effects Library

While the toolbar shows many of the most common nodes you’ll be using in any composition, the
Effects Library contains every single tool available in Fusion, organized by category, with each node
ready to be quickly added to the Node Editor. If you need more guidance to find the node you’re
looking for, or if you just want to browse around and see what’s available, the Effects Library is the
perfect place to start.

To open the Effects Library:

– Click the Effects Library button in the UI toolbar at the top of the Fusion window.

The Effects Library appears at the upper-left corner of the Fusion window, and consists of two panels.
A category list at the left shows all categories of nodes and presets that are available, and a list at the
right shows the full contents of each selected category.

Chapter 5 Working in the Node Editor 111

 The Tools bin of the Effects Library exposing 3D nodes.

By default, the category list shows two primary sets of effects, Tools and Templates, with disclosure
controls to the left that hierarchically show all subcategories within each category. The top two
categories are:
– Tools: Tools consist of all the effects nodes that you use to build compositions, organized by
categories such as 3D, Blur, Filter, Mask, Particles, and so on. If you have third-party OFX plug-ins
on your workstation, those appear in here as well.
– Templates: When using the Fusion page in DaVinci Resolve, templates consist of presets, macros,
and utilities that have been created to get you started quickly. For example, Backgrounds consists
of a variety of customizable generators that have been created using a combination of Fusion
tools. Lens flares presents a wide variety of multi-element lens flares that you can add to any
composition. Particles has a selection of pre-made particle systems that you can customize for
your own use. Shaders has a variety of materials that you can use as texture maps for 3D text and
geometry that you create in Fusion. And there are many, many other categories’ worth of useful
presets and macros that you can learn from and use in your own projects.

Adding, Inserting, and Replacing Tools Using the Effects Library

Adding nodes to the Node Editor from the Tools category of the Effects Library is very similar to
adding nodes from the toolbar.

Methods of adding nodes by clicking in the Effects Library:

– To add a node after a selected node: Select a node in the Node Editor and then click a node in
the browser of the Effects Library.
– To add a disconnected node to the Node Editor: Deselect all nodes in the Node Editor and then
click a node in the browser of the Effects Library.

Methods of adding nodes by dragging from the Effects Library:

– To insert a new node into the node tree: Drag a node from the browser of the Effects Library
into the Node Editor and onto the connection line between any two compatible nodes. When the
connection highlights as the node is over it, drop the node and it’ll be inserted.

Chapter 5 Working in the Node Editor 112

 – To create a disconnected node: Drag a node from the browser of the Effects Library into an
empty part of the Node Editor. Dragging a toolbar button into the Inspector also creates a
disconnected node.
– To insert a new node after a node loaded into a viewer: Drag a node from the browser of the
Effects Library onto a viewer to insert a new node after whichever node is viewed, regardless of
whether any nodes are selected.

To replace a node in the Node Editor with a node from the Effects Library:
1 Drag a node from the browser of the Effects Library so it’s directly over the node in the Node
Editor that you want replaced. When that node is highlighted, drop it.
2 Click OK in the dialog to confirm the replacement.

Adding, Inserting, and Replacing Templates Using the Effects Library

Adding items from DaVinci Resolve’s Fusion page Templates category is often a bit different.
Sometimes, as when adding a Lens Flare, a single node can be added or inserted into the Node
Editor. When this is the case, adding nodes works the same as when adding from the Tools category.

Adding a Lens Flare effect.

Other times, such as when adding an item from the “How to” category, dragging a single item from the
Node Editor results in a whole node tree being added to the Node Editor. Fortunately, all nodes of the
incoming node tree are automatically selected when you do this, so it’s easy to drag the entire node
tree to another location in the Node Editor where there’s more room. When this happens, the nodes of
the incoming effect are exposed so you can reconnect and reconfigure it as necessary to integrate the
effect with the rest of your composition.

Adding a LightWrap effect from the “How to” bin of the Templates category of the Effects Library.

Chapter 5 Working in the Node Editor 113

Adding, Inserting, and Replacing Nodes
Using the Contextual Menu
Another way of adding, inserting, and replacing nodes is to use the Node Editor’s contextual menu,
which has dedicated submenus that let you create any kind of node available in Fusion. This can be a
convenient when the pointer is already in the Node Editor selecting, moving, or connecting nodes.

Methods of adding nodes using the contextual menu:

– To add a node: Right-click in an empty area of the Node Editor, and choose a node from the
Add Tool submenu.
– To insert a node: Right-click a node in the Node Editor, and choose a node from the Insert
Tool submenu.
– To replace a node: Right-click a node in the Node Editor, and choose a node from the
Replace Tool submenu.

TIP: When you replace one node with another, any settings that are identical between the
two nodes are copied into the new node. For example, replacing a Transform node with a
Merge will copy the existing center and angle values from the Transform to the Merge.

Deleting Nodes
To delete one or more selected nodes, press Delete (macOS) or Backspace (Windows), or right-click
one or more selected nodes and choose Delete from the contextual menu. The node is removed from
the Node Editor, and whichever nodes are connected to its primary input and output are now
connected together. Nodes connected to other inputs (such as mask inputs) become disconnected.

Before deleting a node from a node tree (top), and after upstream and
downstream nodes have automatically reconnected (bottom).

Disconnected Nodes
It’s perfectly fine to have disconnected nodes, or even entire disconnected branches of a node tree, in
the Node Editor alongside the rest of a composition. All disconnected nodes are simply ignored while
being saved for possible future use. This can be useful when you’re saving nodes that you’ve
customized but later decided you don’t need. It’s also useful for saving branches of trees that you’ve
since exported to be self-contained media that’s re-imported to take the place of the original effect,
but you want to save the original nodes just in case you need to redo your work.

Chapter 5 Working in the Node Editor 114

Selecting and Deselecting Nodes
In order to work with nodes in the Node Editor in any way, or modify node parameters in the Inspector,
you first need to learn to select the node or nodes you want to work with.

Selecting Nodes
Selecting nodes is one of the most fundamental things you can do to move nodes or target them for
different operations. There are a variety of methods you can use.

Methods of selecting nodes:

– To select a single node: Click any node in the Node Editor.
– To select multiple nodes one at a time: Command-click each node you want to select.
– To select a whole region of nodes: Drag a bounding box around all nodes you want to select.
– To select all upstream or downstream nodes: Right-click a node and choose Select >
Upstream Nodes/Downstream Nodes from the contextual menu.
– To select all nodes in the Node Editor: Press Command-A.
– To select a node from the Keyframe Editor: Click any layer in the Keyframe Editor to select the
corresponding node in the Node Editor.

The Active Node

When you select a single node using any of the methods described above, the selected node is
known as the active node, and is highlighted orange to indicate that its parameters are currently
editable in the Inspector (if the Inspector is open). This also indicates that node will be targeted for
specific operations (such as inserting new nodes).
While multiple nodes can be selected, only one node will be the active node. To indicate the
difference, the active node remains highlighted with orange, while all other selected nodes are
highlighted with white. Unselected nodes have simple black outlines.

The active node is highlighted orange, while other selected nodes are highlighted white.

To set the active node when there are multiple selected nodes:
– Option-click one of the selected nodes in the Node Editor to make that one the active node.
– Open the Inspector (if necessary), and click a node’s header bar to make it the active node.

Chapter 5 Working in the Node Editor 115

Deselecting Nodes
Deselecting nodes, when necessary, works pretty much as you would expect.

Methods of deselecting nodes:

– Click once in the background of the Node Editor to deselect all nodes.
– Press Command-Shift-A to deselect all nodes.
– Command-click to deselect multiple nodes one at a time.
– Command-drag a bounding box to deselect a group of selected nodes at one time.

Loading Nodes into Viewers

Once you’ve started building a composition, the next thing you need to learn is how to view specific
nodes that you want to work on. This is important because the combination of which node is being
viewed and which node is currently selected (these aren’t always the same node) often determines
which onscreen controls are available and how they appear.
In the following example, you’re set up to rotoscope an image using a Polygon node that’s attached to
the garbage mask input of a MatteControl node which is inserting the mask as an alpha channel.

A node tree for doing a simple rotoscoping job.

As seen in the screenshot above, you’ll want to load the upstream MediaIn or Loader node into a
viewer while the Polygon node is selected for editing in order to see the full image you’re rotoscoping
while keeping the Polygon node’s spline visible.

Chapter 5 Working in the Node Editor 116

Viewed Nodes When You First Open Fusion
When you first open the Fusion page in DaVinci Resolve, the output of the current empty composition
(the MediaOut1 node) is usually showing in viewer 2. If you’re in Dual-viewer mode, viewer 1 remains
empty until you assign a node to one of them.
When you first open Fusion Studio with an empty comp, both viewers remain empty even after reading
in media using a Loader node. The viewers only display content when you assign a node to
one of them.
There are several different ways to display a node in a viewer. Which ones you use depends on how
you like to work.

Node View Indicators

The View indicators are displayed under each node, and serve two purposes. First, they’re a clickable
interface for displaying a node’s output in one of the viewers. Second, they’re an indication of which
nodes in the Node Editor are being viewed. By default, there are two round indicators, representing
the two viewers. The left and right indicators correspond to the left and right viewers, regardless of
whether both viewers are visible or just one.

A viewer indicator enabled for the right

viewer and disabled for the left viewer.

To load a node into a viewer using the Node View indicators:

– Clicking an indicator turns it white to show that node is currently loaded in the corresponding
viewer. Clicking it again turns the indicator black and removes it from the viewer. Nodes only
display View indicators if they’re currently being viewed. If you want to view indicators, hovering
the pointer over the node makes the indicators visible and available for clicking.
– You can also use keyboard shortcuts to toggle each View indicator. The default two viewers are
assigned numeric keyboard shortcuts 1 and 2. Pressing the corresponding number once displays
the selected node in the appropriate display view, while pressing it again clears that display.

For complex compositions, you may need to open additional viewers. For example, one viewer may be
used to display the end result of the final comp, while another viewer displays the source, a third
viewer displays a mask, and a fourth viewer might be a broadcast monitor connected via a Blackmagic
DeckLink card or other display hardware. When you have more than two viewers, additional View
indicators are added and each one is assigned a consecutive number between 3 and 9.
The more viewers you add, the more you may need help remembering which viewer is represented by
which View indicator. Positioning the pointer over the View indicator in question will display a tooltip
with the name of the viewer it represents.

Drag and Drop Nodes into a Viewer

If the View indicators are too small of a target for you to click on reliably and you are not keyboard
oriented, another way to load a node into a viewer is to drag and drop it onto the viewer you want to
load it into. This offers a quick explicit way to assign a node to a viewer, especially for pen and tablet
users. Please note that as you drag, the node will appear to move at first, but it’ll snap back into its
original location once the pointer leaves the Node Editor.

Chapter 5 Working in the Node Editor 117

Using the Contextual Menu
You can also right-click a node, and then choose View On > Left or Right to display the node on the
appropriate viewer.

Clearing Viewers
Whenever you load a node into a viewer, you prompt that node, all upstream nodes, and other related
nodes to be rendered. If you load nodes into both viewers, this is doubly true. If you want to prevent
your computer from processing views that aren’t currently necessary, you can clear each viewer.

Methods of clearing viewers:

– Press 1 or 2 to empty the left or right viewers if they’re filled.
– Press ` (the Accent key) to empty both viewers.

Create/Play Preview
You can right-click a node, and choose an option from the Create/Preview Play On submenu of the
contextual menu to render and play a preview of any node’s output on one of the available viewers.
The Render Settings dialog is displayed, and after accepting the settings, the tool will be rendered and
the resulting frames stored in RAM for fast playback on that view.

TIP: Hold the Shift key when selecting the viewer from the menu to bypass the Render dialog
and to start creating the preview immediately using the default settings or the last settings
used to create a preview.

Connecting and Disconnecting Nodes

Once you’ve started to add nodes to your composition, you need to connect them to perform their
intended operations.

Node Basics
Each node displays small colored knots around the edges. One or more arrows represent inputs, and
the square represent the tool’s processed output, of which there is always only one. Outputs are white
if they’re connected properly, gray if they’re disconnected, or red to let you know that something’s
wrong and the node cannot process properly.

A Blur node with a Foreground Input, Mask Input, and Output.

Each node takes as its input the output of the node before it. By connecting a MediaIn node’s output
to a Blur node, you move image data from the MediaIn node to the Blur node, which does something
to process the image before the Blur node’s output is in turn passed to the next node in the tree.

Chapter 5 Working in the Node Editor 118

 Two nodes connected together.

How to Connect Nodes

To manually connect one node to another, click on one node’s output and drag a connection line out
to drop on another node’s input. The order in which you drag node connections is not important; you
can just as easily drag a connection from one node’s input to another node’s output and get the
same results.

Before (top), and after (bottom) dragging a connection

line and dropping it to connect two nodes.

Dropping Connections on Top of Nodes

To make your life a bit easier, you can also drag a connection line and drop it directly on top of the tile
of a node to automatically connect to the default input of that node, which is usually labeled
“background” or “input.” In the following example, a connection is dragged from the output of a
MediaIn node and dropped onto the tile of a Blur1 node, and the background input is connected first.

Before (top), and after (bottom) dragging a connection

line and dropping it on top of a node.

If you drop a connection on top of a node that already has the background input connected, then the
second most important connection will be attached, which for multi-input nodes is the foreground
input, and for other single-use nodes may be the Effects Mask input.

Chapter 5 Working in the Node Editor 119

 Before (top), and after (bottom) dragging a connection line and dropping it on
top of a node that has the background input already connected.

Some multi-input nodes are capable of adding inputs to accommodate many connections, such as
the Merge3D node. These nodes simply add another input whenever you drop a connection
onto them.

After dragging a connection line and dropping

it on top of a Merge3D node.

Attaching Connections to Specific Inputs

If you want to make sure you don’t attach a connection to the default input of a node, then you need to
drop it right on top of the specific node input you want to attach it to. If you can see the input’s label in
the tooltip bar, then you know you’re correctly positioned to make a good connection.
However, there’s an alternate method of connecting nodes together in instances where there are
several inputs to choose from and you want to make sure you’re choosing the correct one. Hold down
the Option key while dragging a connection from one node’s output and dropping it onto the body of
another node. This opens a pop-up menu from which you can choose the specific input you want to
connect to, by name. Please note that this menu only appears after you’ve dropped the connection on
the node and released your pointing device’s button.

Chapter 5 Working in the Node Editor 120

 Option-dragging a node connection to drop onto
another node exposes a node input menu.

Automatically and Manually Attaching Mask Nodes

Mask nodes, such as the Polygon, B-Spline, Ellipse, or Rectangle, have a different automatic behavior
when you connect them to other nodes. If you drag a connection from a Mask node onto the body of
another node, it will automatically connect itself to the default mask input, which is usually the effect
mask input. The assumption is that you’re using the mask to limit the node’s effect somehow. However,
this isn’t always the case, so you’ll need to be careful of this behavior to make sure you’re attaching
your mask to the input that will actually create the effect you need.

Before (left) and after (right) dragging a connection from a Mask node and dropping it on top of a MatteControl node.

Identifying Node Inputs

While you are still figuring out all the nodes and their inputs, hovering the pointer over any knot will
display a node tip with the knot’s name.

TIP: Rather than remembering the different knot types, press the right mouse button, hold
Option, and drag from the output of a node to the center of another tool. When you release
the mouse, a tooltip will appear allowing you to select the knot you want to connect to.

Chapter 5 Working in the Node Editor 121

Node Order Matters
The order in which nodes are attached defines the order in which each image-processing operation is
applied to the image.
In the following example, a MediaIn node adds a clip to the composition, while a Defocus node blurs
the image, and then a TV node adds scanlines and vertical distortion. Those effect nodes are then
connected to the MediaOut node in the Fusion page in DaVinci Resolve or a Saver node in
Fusion Studio.

Adding a Defocus effect first, then the TV node second.

As you can see above, connecting the Defocus node first, followed by the TV node, means that while
the initial image is softened, the TV effect is sharp. However, if you reverse the order of these two
nodes, then the TV effect distorts the image, but the Defocus node now blurs the overall result, so that
the TV effect is just as soft as the image it’s applied to. The explicit order of operations you apply
makes a big difference.

Chapter 5 Working in the Node Editor 122

 Adding a TV effect first, and a Defocus second.

As you can see, the node tree that comprises each composition is a schematic of operations with
tremendous flexibility. Additionally, the node tree structure facilitates compositing by giving you the
ability to direct each node’s output into separate branches, which can be independently processed
and later recombined in many different ways, to create increasingly complex composites while
eliminating the need to precompose, nest, or otherwise compound layers together, which would impair
the legibility of your composition.
In the following example, several graphics layers are individually transformed and combined with a
series of Merge nodes. The result of the last Merge node is then transformed, allowing you to move
the entire collection of previous layers around at once. Because each of these operations is clearly
represented via the node tree, it’s easy to see everything that’s happening, and why.

The output of five Text nodes being combined using

Merge nodes is modified by a single Transform node.

Chapter 5 Working in the Node Editor 123

Upstream and Downstream Nodes
Since nodes can be positioned anywhere in the Node Editor, and added in any direction, nodes are
referred to as being upstream and downstream of one another. Once you select a node, all other
nodes that directly or indirectly connect to its input are considered to be upstream. Any other nodes
that are directly or indirectly connected to the output are said to be downstream.
This is an important distinction to make because, unlike layer-based systems, the visual positioning of
nodes in your node tree has no bearing on the order of operations in that composition. The only thing
that matters is whether nodes are upstream or downstream of each other.

Tools upstream (left) and downstream (right) of the Merge node.

TIP: To help you stay organized, there are Select > Upstream/Downstream commands in the
Node Editor contextual menu for selecting all upstream or downstream nodes to move them,
group them, or perform other organizational tasks.

Disconnecting and Reconnecting Nodes

Node trees are a continuous work in progress, requiring constant revision and rearrangement as you
discover new details that need to be finessed, or things that you can do better once the overall
composition has taken shape. To facilitate quick changes, each connection between two nodes is
divided into two halves: the output half (connected to the upstream node’s output) and the input half
(connected to the downstream node’s input). This can only be seen when you hover the pointer over a
connection. The half your pointer is over is highlighted in blue.

The two halves of a connection line that are revealed

when you hover your pointer over it.

By clicking and/or dragging these two halves, it’s possible to quickly disconnect, reconnect, and
overwrite node connections, which is essential to rearranging your node tree quickly and efficiently.

To disconnect two nodes, do one of the following:

– Click once on the input half of the connection between two nodes.
– Click on the input arrow to which a connection is attached, then drag to pull the connection away
from the tool and drop it anywhere in an empty area of the Node Editor.

To overwrite a current connection:

– Drag the output or input half of a connection, and drop it directly onto another node’s input or
output. This simultaneously disconnects the previous connection and connects the one you’re
dragging.

Chapter 5 Working in the Node Editor 124

To reconnect a connection from one node to another:
– Drag the output or input half of a connection to disconnect it from one node, and drop the
connection directly on another node’s input or output.

Tracing Connections Through the Node Tree

Positioning the pointer over a node causes the connections attached to that node to become
highlighted, which makes it easier to see which nodes it’s attached to. Additionally, highlighted
connections display the color of the input they’re connected to, which makes it easy to see if they’re
connected to a foreground, a background, or a particular kind of mask.

Hovering the pointer over a node highlights the color of all connections,
telling you what kinds of inputs are connected.

Additionally, positioning the pointer over a connection causes a tooltip to appear that displays the
output and input that connection is attached to.

Hovering the pointer over a node highlights the

connection between it and other nodes.

Branching
A node’s input can only have one connection attached to it. However, a tool’s output can be
connected to inputs on as many nodes as you require. Splitting a node’s output to inputs on multiple
nodes is called branching. There are innumerable reasons why you might want to branch a node’s
output. A simple example is to process an image in several different ways before recombining these
results later on in the node tree.

A MediaIn node branched to two node operations

and then recombined using a Merge node.

Chapter 5 Working in the Node Editor 125

Alternatively, it lets you use one image in several different ways—for example, feeding the RGB to one
branch for keying and compositing, while feeding the A channel to the Effects Mask input of another
node to limit its effect, or feeding RGB to a tracker to extract motion information.

A MediaIn node branched to two different

kinds of inputs, used separately.

Connecting Merge Nodes

The Merge node is the primary tool available for compositing images together. Each Merge node is
capable of combining two inputs to create a third, using standard compositing methods and composite
modes. For more extensive information about the Merge node, see Chapter 105 “IO Nodes” in the
DaVinci Resolve Reference Manual or Chapter 44 in the Fusion Reference Manual. For this chapter, all
you need to know is that if you attach a background image to the Background input (such as a
landscape), and a foreground image with an alpha channel to the Foreground input (such as a graphic
with an alpha channel), the Merge node will combine them into a single image for further compositing.

Two MediaIn nodes and a DeltaKeyer node attached to a Merge node, creating a composite.

Each Merge node has three inputs:

– Background (orange): The default input. Whichever image is connected to this input defines the
output resolution of the Merge node.
– Foreground (green): The secondary input, meant for whichever image you want to be “on top.”
– Effect Mask (blue): An optional input you can use to attach a mask or matte with which to limit the
effect of the Merge node.
It’s important to make sure you’re attaching the correct nodes to the correct inputs to ensure you’re
getting the result you want, and it’s important to keep these inputs in mind when you connect to a
Merge node. Of course, you can always drag a connection to a specific input to make sure you’re
connecting things the way you need. However, if you’re in a hurry and you simply drag connections
right on top of a Merge node:
– The first connection will be made to the background input.
– The second connection will be made to the foreground input.
– The third connection will be made to the effect mask input.

Chapter 5 Working in the Node Editor 126

 TIP: When you add a Merge node after a selected node by clicking the Merge button on the
toolbar, by clicking on the Merge icon in the Effects Library, or by right-clicking a node in the
node tree and choosing Insert Tool > Composite > Merge from the contextual menu, the new
Merge node is always added with the background connected to the upstream node coming
before it.

Automatically Creating a Merge Node When Adding Nodes

There’s a nice shortcut for connecting Merge nodes if you want to connect the incoming clip
immediately to your node tree as the top layer of a composite, and that’s to drag a clip from an
Operating System window or a Generator from the Effects Library right on top of any connection line.
When you drop the resulting node, this automatically creates a Merge node, the background input of
which is connected to the next node to the left of the connection you dropped the clip onto, and the
foreground input of which is connected to the new node that represents the clip or Generator you’ve
just added.

Dragging a node from the Media Pool onto a connection (left),

and dropping it to create a Merge node composite (right).

Additionally, If you drag two or more nodes from an OS window into the Node Editor at the same time,
Merge nodes will be automatically created to connect them all, making this a fast way to initially
build a composite.

By dragging three nodes from an OS window to the Node Editor (left),

Merge nodes are automatically created to connect them all (right).

Automatically Creating a Merge Node

by Connecting Two Outputs
Here’s an endlessly useful shortcut for when you have a disconnected node that you want to
composite over another node. Drag a connection from the output of the node you want to be the
foreground layer, and drop it on top of the output of the node you want to be the background layer,
and a Merge node will be automatically created to build that composite.

Chapter 5 Working in the Node Editor 127

Dragging a connection from a disconnected node to another node’s
output (left), and dropping it to create a Merge node composite (right).

Connection Options and Routers

By default, the Node Editor uses linear connections that are drawn straight between any two
connected nodes. While efficient, this sometimes causes connection lines to overlap nodes, which
some people feel interferes with the view of the Node Editor.

Linear connections between nodes.

If you like, you can change how connections are drawn by enabling orthogonal connections, which
automatically draws lines with right angles to avoid having connections overlap nodes.

Optional orthogonal connections between nodes.

Functionally, there’s no difference to your composition; this only affects how your node tree appears.

To change how connections are drawn in the Node Editor:

– Right-click the Node Editor background and choose one of the following from the
contextual menu.
– Options > Direct Pipes
– Options > Orthogonal Pipes

Chapter 5 Working in the Node Editor 128

Using Routers to Reshape and Branch Connections
If you want to force a particular connection to be drawn at an angle to keep your node tree tidy, you
can add a router to either linear or orthogonal connections to force an angle so it will be drawn
however you like.

A router added to force a connection to be drawn at an angle.

Routers are tiny nodes with a single input and an output, but with no parameters except for a
comments field (available in the Inspector), which you can use to add notes about what’s
happening in that part of the composition.
Even more usefully, you can branch a router’s output to multiple nodes, which makes routers even
more useful for keeping node trees neat in situations where you want to branch the output of a
node in one part of your node tree to other nodes that are all the way on the opposite end of that
same node tree.

A router branching its output to multiple nodes.

Chapter 5 Working in the Node Editor 129

Methods of using routers:
– To add a router to a connection: Option-click anywhere on a connection.
– To move a router: Drag the router to a new location, and the connection will reshape
itself as necessary.
– To branch a router’s output: Drag a connection from the router output to the input of another
node. You can branch a router’s output as many times as you need to.
– To remove a router: Select any router and press the Delete key, or right-click a router and choose
Delete from the contextual menu.

Swapping Node Inputs

For multiple-input nodes such as the Merge, Merge 3D, and Dissolve nodes, there’s a quick method of
swapping the Primary and Secondary inputs, such as the foreground and background inputs of a
Merge tool, when you find you’ve accidentally connected them in the wrong order. If a node has more
than two of its inputs connected, only the foreground and background inputs will be swapped.

To swap the primary inputs of a multi-input node, do one of the following:

– Select a node and press Command-T to reverse its inputs.
– Right-click a node and choose Swap Inputs from the contextual menu.

Before swapping node inputs (left), and after swapping node inputs (right),
the connections don’t move but the colors change.

Inputs can move freely around the node, so swapping two inputs doesn’t move the connection lines;
instead, the inputs change color to indicate you’ve reversed the background (orange) and foreground
(green) connections.

Extracting and Inserting Nodes

When building a composition, you’ll often find that you need to rearrange nodes that you’ve already
added, in order to connect them in different ways to obtain a better result. Happily, this is easy to do
by extracting one or more nodes from one part of a node tree and inserting them at another part of
the node tree.
To extract one or more nodes from their position in the node tree:
– To extract a single node: Hold down the Shift key, drag a node from the node tree up or down to
disconnect it, and then drop the node before releasing the Shift key. That node is now detached,
and the output of the next upstream node is automatically connected to the input of the next
downstream node to fill the gap in the node tree.
– To extract multiple nodes: Select the nodes you want to extract, hold down the Shift key, drag
one of the selected nodes up or down to disconnect them, and then drop the node before
releasing the Shift key. Those nodes are now detached (although they remain connected to one
another), and the output of the next upstream node is automatically connected to the input of the
next downstream node to fill the gap in the node tree.

Chapter 5 Working in the Node Editor 130

 Before extracting a pair of nodes (left), and after extracting a pair of nodes (right).

After you’ve extracted a node, you can re-insert it into another connection somewhere else. You can
only insert one node at a time.

To insert a disconnected node in the Node Editor between two compatible nodes:
1 Hold down the Shift key and drag a disconnected node directly over a connection between two
other nodes.
2 Once the connection highlights, drop the node, and then release the Shift key. That node is now
attached to the nodes coming before and after it.

Before inserting a node (left), and after inserting a node (right).

TIP: If you hold down the Shift key, you can extract a node and re-insert it somewhere else
with a single drag.

Cut, Copy, and Paste Nodes

The standard operations of cut, copy, and paste are also available in the Node Editor. You can use
them to temporarily remove nodes from the Node Editor, create duplicate nodes, or even copy the
settings from one node and paste those settings into another node with compatible settings.

Cut, Copy, and Paste in the Node Editor

The standard commands all work, but with some special features specific to the Node Editor.

To copy one or more selected nodes, do one of the following:

– Right-click a node and choose Copy from the contextual menu.
– Choose Edit > Copy from the Edit menu (Command-C).

To cut one or more selected nodes, do one of the following:

– Right-click over the node and choose Cut from the contextual menu.
– Choose Edit > Cut from the Edit menu (Command-X).

Chapter 5 Working in the Node Editor 131

When you paste into the Node Editor, you create a copy of the last node or nodes you’ve cut or
copied. When pasting, there are a few different things you can do to control where pasted
nodes appear.

To paste one or more selected nodes, do one of the following:

– To paste nodes to be inserted after another node: Select the node in the node tree you want to
insert the pasted node(s) to, and choose Edit > Paste (Command-V).
– To paste nodes to be disconnected from the rest of the node tree: Deselect all nodes, and then
choose Edit > Paste (Command-V), or right-click anywhere in the Node Editor and choose Paste
from the contextual menu.
– To paste disconnected nodes in a specific area of the Node Editor: Deselect all nodes, and
then click the place in the Node Editor where you want pasted node(s) to appear, and choose
Edit > Paste (Command-V), or right-click anywhere in the Node Editor and choose Paste from the
contextual menu.
– To paste a node to replace an existing node in the Node Editor: Right-click a node in the Node
Editor that you want to replace, choose Paste from the contextual menu, and when a dialog
appears asking if you want to replace that node, click OK. This only works when you use the
contextual menu command.

TIP: When you paste a MediaIn, Loader, or Generator node so it will be inserted after a
selected node in the node tree, a Merge tool is automatically created and used to composite
the pasted node by connecting it to the foreground input. While this can save you a few
steps, some artists may prefer to perform these sorts of merges manually, so this can be
changed using the Default Preferences panel in the Global preferences.

Pasting Node Settings

Instead of pasting a node, you can choose to paste just the parameter settings that you copied from
another node. This can be useful if you’ve carefully set or animated parameters in one node that you
want to also use in another node.
Note that you can paste settings between two nodes of the same type, or between two entirely
different kinds of nodes that happen to have one or more of the same parameters in the Inspector.
When copying settings from one type of node to another, only the settings that match between two
nodes will be copied. A common example is to copy an animated Center parameter from a Transform
node to the Center parameter of a Mask node.

To Paste settings from one node to another:

1 Select a node that has settings you want to copy, and choose Copy from the
Edit menu (Command-C).
2 Right-click a node you want to paste those settings to, and choose Paste Settings
from the contextual menu.

Copying and Pasting Nodes to and from Any Text Editor

The format of nodes in the Node Editor is not binary, but is in fact a simple text format. The implications
of that may not be obvious, but one example benefit is clear when you start dealing with nodes.
One or more nodes can be copied from the Node Editor and pasted directly into a text editor or email.
This pastes the selection in text format, just as it’s saved internally in Fusion. For example, if you copy
the following set of three nodes:

Chapter 5 Working in the Node Editor 132

 A set of three nodes being copied.

And you then paste into a new text editing document, you get the following:

The same three nodes pasted into a text editor.

At this point, you have the option of editing the text (if you know what you’re doing), emailing it to
colleagues, or storing it in a digital notepad of some sort for future use. To use this script in Fusion
again, you need only copy it and paste it back into the Node Editor.

TIP: This is a very easy way to pass specific node settings back and forth between artists
who may not be in the same room, city, or country.

Instancing Nodes
Normally, when you use copy and paste to create a duplicate of a node, the new node is completely
independent from the original node, so that changes made to one aren’t rippled to the other. However,
there are times when two nodes must have identical settings at all times. For example, when you’re
making identical color corrections to two or more images, you don’t want to constantly have to adjust
one color correction node and then manually adjust the other to match. It’s a hassle, and you risk
forgetting to keep them in sync if you’re working in a hurry.
While there are ways to publish controls in one node and connect them to matching controls in
another node, this becomes prohibitively complex and time consuming for nodes in which you’re
making adjustments to several controls. In these cases, creating “instanced” nodes is a real time-saver,
as well as an obvious visual cue in your node tree as to what’s going on.

Chapter 5 Working in the Node Editor 133

Using Instanced Nodes
Instanced nodes are nodes that have been created using the Paste Instance command, and which
share settings with the original node so that a change made to one instanced node is also
automatically applied to all other instances of that node (as well as the original node you copied).
To create an Instance, do the following:
1 Select a node you want to instance, and copy it (Command-C).
2 Do one of the following:
– To create a disconnected instance of a node: Right-click in the background of the Node Editor,
and choose Paste Instance from the contextual menu (Command-Shift-V).
– To insert an instanced node between two other nodes: Select a node that’s upstream of
where you want to insert the instanced node, and press Command-Shift-V. Alternatively,
you can right-click directly on a connection line, and choose Paste Instance from the
contextual menu.

However you paste an instance, the name of that instanced node takes the form “Instance_
NameOfNode.” If you paste multiple instances, each instance is numbered “Instance_
NameOfNode_01.”

A green link line shows an instanced Blur node’s relationship to the original Blur node it was copied from.

When a node tree contains instanced nodes, a green line shows the link between the original node
and its instances. You have the option to hide these green link lines to reduce visual clutter in the
Node Editor.

To toggle the visibility of green instance link lines in the Node Editor:
1 Right-click anywhere in the background of the Node Editor.
2 Choose Options > Show Instance Links from the contextual menu.
If you’ve been using an instance of a node and you later discover you need to use it to apply separate
adjustments, you can “de-instance” the node.

Chapter 5 Working in the Node Editor 134

To de-instance a node, making it independent:
1 Right-click an instanced node.
2 Choose Deinstance from the contextual menu. That node is now independent from the original
node. Once you de-instance a node, you cannot re-instance it, but you can undo the operation.

NOTE: If you’ve de-instanced a node and you cannot undo the operation because you’ve
restarted DaVinci Resolve, you can only recreate an instance by copying the original and
pasting an instance again.

De-Instancing and Re-Instancing Specific Parameters

By default, every parameter in an instanced node is linked to the original node, so that any change you
make is rippled across. However, from time to time you’ll find the need to independently adjust just
one or two parameters while keeping the rest of that node’s parameters linked. For this reason,
instead of de-instancing the entire tool, you can de-instance individual parameters.

To de-instance a single parameter:

– Right-click on a parameter’s name or value in the Inspector, and choose Deinstance from the
contextual menu.
If you’ve only de-instanced individual parameters, you can re-instance those parameters later on if you
change your mind.

To re-instance a single parameter:

– Right-click on a parameter’s name or value in the Inspector, and choose Reinstance from the
contextual menu. That parameter immediately inherits the setting of the original node.

Keeping Node Trees Organized

Similar to working with files on your desktop, even the simplest of composites require you to do some
amount of organization. In this section we’ll look at some basic node operations, some of which you
may already be familiar with just from using your computer’s operating system or other applications.

Moving Nodes
Selecting one or more nodes and dragging them moves them to a new location, which is one of the
simplest ways of organizing a node tree, by grouping nodes spatially according to the role they play in
the overall composition.
Keep in mind that the location of nodes in the Node Editor is purely aesthetic, and does nothing to
impact the output of a composition. Node tree organization is purely for your own peace of mind, as
well as that of your collaborators.

TIP: Once you’ve arranged the nodes in a composition in some rational way, you can use the
Sticky Note and Underlay tools to add information about what’s going on and to visually
associate collections of nodes more definitively. These tools are covered later in this section.

Chapter 5 Working in the Node Editor 135

Snapping Nodes to the Grid
By default, you can position nodes freely wherever you want them to be. However, keeping nodes and
connection lines straight and aligned can make them easier to read. To help keep them aligned, you
can have nodes you’re dragging automatically snap to the grid.
To have nodes snap to the grid as they’re dragged:
– Right-click over an empty area of the Node Editor, and choose Arrange Tools > To Grid from the
contextual menu. All nodes you drag now snap to the nearest grid coordinate.
– Right-click over an empty area of the Node Editor, and choose Arrange Tools > To Connected
from the contextual menu. All nodes you drag now snap to the horizontal or vertical position of the
nodes they’re attached to.

TIP: You can set “Arrange to Grid” or “Arrange to Connected” as the default for new
compositions by choosing Fusion > Fusion Settings in DaVinci Resolve or File > Preferences
in Fusion Studio, and turning the Fusion > Node Editor > Arrange To Grid or Arrange to
Connected checkboxes on.

Commands to “Clean Up” a Node Tree

The grid in the background of the Node Editor can be used to align nodes, either by eye
or automatically.

To “clean up” an unruly node tree:

– Right-click in an empty section of the Node Editor, and choose Line Up All Tools to Grid from the
contextual menu. All nodes in the Node Editor will move to align and center themselves along the
nearest grid lines.

To “clean up” only one or more selected nodes:

– Right-click one of the selected nodes and choose Line Up to Grid from the contextual menu. All
selected nodes will move to align and center themselves along the nearest grid lines, while all
unselected nodes will be left as they are.

Renaming Nodes
Each node that’s created is automatically assigned a name (based on its function) and a number
(based on how many of that type of node have been created already). For example, the first Blur node
added to a composition will be called Blur1, the second will be Blur2, and so on. Although initially
helpful, larger compositions may benefit from important nodes having more descriptive names to make
it easier to identify what they’re actually doing, or to make it easier to reference those nodes in
expressions.

To rename a node:
1 Do one of the following:
– Right-click a node and choose Rename from the contextual menu.
– Select a node and press F2.

2 When the Rename dialog appears, type a new name, and then click OK or press Return.

NOTE: If multiple nodes are selected, multiple dialogs will appear asking for a name for each tool.

Since Fusion can be scripted and use expressions, the names of nodes must adhere to a scriptable
syntax. Only use alphanumeric characters (no special characters), and do not use any spaces.

Chapter 5 Working in the Node Editor 136

Also, you cannot start a node name with a number. If you accidentally create a name that doesn’t
exactly follow the guidelines, spaces and invalid characters will be automatically deleted.
If you want to see the original node types instead of the node names, press and hold
Command-Shift-E.

Changing Node Colors

You can change the color of any node by selecting it, opening the Inspector, and choosing a new color
from the Node Color pop-up in the Inspector header for that node. Alternatively, you can right-click a
node and choose a color from the Set Color submenu.
To return a node to its regular color, right-click it and choose Set Color > Clear Color from the
contextual menu, or open the Node Color pop-up for a node in the Inspector, and choose Clear Color.

Using Sticky Notes

A good way to add notes about different parts of a composition, client feedback about various details,
and other information you want to keep track of, is to add Sticky Notes to the Node Editor.

A Sticky Note in the Node Editor.

Sticky Notes are yellow boxes in which you can type whatever text you want. They can be resized,
moved, and collapsed when they’re not being edited, but once created they remain attached to the
background of the Node Editor where you placed them until you either move them or delete them.

Methods of working with Sticky Notes:

– To create a Sticky Note: Click somewhere in the Node Editor where you want a Sticky Note to
appear. Then, press Shift-Spacebar, type sticky, and press the Return key when the Sticky Note
appears in the Select Tool window. Alternatively, you can open the Effects Library, open the Tools
> Node Editor category, and click or drag the Sticky Notes node to create a new one.
– To open a Sticky Note to full size: Double-click a minimized Sticky Note and it expands to a larger,
resizable yellow box.
– To edit a Sticky Note: If necessary, double-click a Sticky Note to open it to full size, and then click
once in the body of the note to place a text cursor. You can edit text within the Sticky Note just like
any other text editor.
– To rename a Sticky Note: Right-click a Sticky Note, choose Rename, type a new name into the
Rename dialog, and click OK. Alternatively, you can select a Sticky Note, press F2 to open the
Rename dialog, and press Return to close it when you’re done.
– To resize a Sticky Note: Double-click a Sticky Note to open it to full size, and then drag any of the
edges or corners to make it larger or smaller.
– To minimize a Sticky Note: Click the close box at the upper left-hand corner of the Sticky Note,
and it collapses to a small tile.
– To delete a Sticky Note: Right-click any Sticky Note and choose Delete from the contextual menu
or select the Sticky note in the Node Editor and press the Delete key.

Chapter 5 Working in the Node Editor 137

Using Underlay Boxes
Underlay Boxes are a good way of associating a collection of nodes that work together to perform a
specific task in your composition. They’re simply colored rectangles that you can put nodes inside of.
Once you place nodes inside an Underlay, you can move the Underlay, and all the nodes within move
along with it.

An Underlay in the Node Editor.

Underlay Boxes can be named to identify the purpose of that collection of nodes, and they can be
colored to be distinct from other Underlay Boxes or to adhere to some sort of color code for your
compositions.

Methods of working with Underlay Boxes:

– To create an Underlay Box: Click somewhere in the Node Editor where you want the Underlay
Box to appear. Then, press Shift-Spacebar, type under, and press the Return key when the
Underlay Box appears in the Select Tool window. Alternatively, you can open the Effects Library,
open the Tools > Node Editor category, and click or drag the Underlay Box node to create a
new one.
– To create an Underlay Box around specific nodes: Select the nodes in the Node Editor that you
want surrounded by an Underlay Box. Then, press Shift-Spacebar, type under, and press the
Return key when the Underlay Box appears in the Select Tool window. Alternatively, you can open
the Effects Library, open the Tools > Node Editor category, and click the Underlay Box node to
have it added and sized to encompass all the selected nodes.
– To resize an Underlay Box: Drag any of the edges or corners to make it larger or smaller.
– To rename an Underlay Box: Option-click the Underlay Box to select just the box and not the
contents, and then right-click it and choose Rename (or press F2). Type a new name into the
Rename dialog and click OK or press Return.
– To change the color of an Underlay Box: Option-click the Underlay Box to select just the box and
not the contents, and then right-click it and choose a color from the Set Color submenu.
– To put nodes inside of an Underlay Box: Select the nodes you want to place inside an
Underlay Box, and then drag them to fit inside. The Underlay Box must be big enough to fit all the
nodes. Alternatively, you can place an Underlay Box near a collection of nodes you want to put
inside it, and then resize the Underlay Box to encompass all those nodes.
– To move an Underlay Box and all its nodes: Once nodes have been placed inside an Underlay
Box and have been deselected, you can move the entire collection of nodes together by dragging
the Underlay Box by its title bar.

Chapter 5 Working in the Node Editor 138

 – To remove nodes from an Underlay Box: There are two ways you can remove nodes from an
Underlay Box.
– With both the Underlay Box and nodes deselected, drag a bounding box or Command-click to
select all nodes in the box you want to remove, and drag them out.
– Resize the Underlay Box so that it’s smaller than the collection of nodes it originally
encompassed. Once an Underlay Box is so small that even the last node sticks out beyond its
edge, those nodes are automatically removed from the Underlay Box, and you can move or
delete the Underlay Box without moving those nodes.

– To delete an Underlay Box and all nodes within: Select an Underlay Box and press the Delete
key to delete both the Underlay Box and all nodes found inside it. If you don’t also want to delete
the nodes, first drag the nodes out of the box.
– To delete an Underlay Box but keep all nodes within: Option-click the Underlay Box to select it
and not the nodes, and then press the Delete key. The nodes within remain where they were.

Node Thumbnails
Once a source or an effect has been added to the Node Editor, it’s represented by a node. By default,
nodes are rectangular and thin, making it easier to fit reasonably complicated grades within a relatively
small area. However, if you like, you can also display node thumbnails.

A node in the Node Editor shown without and with a thumbnail.

Nodes can be displayed as a small rectangle or as a larger square. The rectangular form displays the
node’s name in the center, while the square form shows either the tool’s icon or a thumbnail of the
image it is outputting.

TIP: Even if you’re not displaying node thumbnails, you can quickly obtain detailed
information about a node and the data it’s processing by hovering your pointer over it in the
Node Editor and viewing the tooltip bar below.

Choosing Which Nodes Show Thumbnails

If you want to use node thumbnails to help visually identify media and operations in your node trees,
there are a variety options for which nodes should display thumbnails in the contextual menu that
appears when you right-click anywhere in the background of the Node Editor.

Chapter 5 Working in the Node Editor 139

Force All Tile Pictures
This option shows thumbnails for every single node in the Node Editor. This can make simple node
trees easier to read, but it’ll make all node trees take up considerably more room.

NOTE: If Show Thumbnails is enabled, nodes may not update until the playhead is moved in
the Time Ruler.

Force Active Tile Pictures

You may also choose to only show thumbnails for nodes that are currently selected, which can make it
easier to see which node you’re working on. When nodes become deselected, the thumbnails will be
hidden again.

Force Source Tile Pictures

This enables thumbnails for all MediaIn and Loader nodes in the Node Editor, as well as all generators,
and is a great way to be able to quickly see where all the clips are in a composition.

Force Mask Tile Pictures

This enables thumbnails for all Mask nodes in a composition, which can make them easier to
distinguish when you’re building complex shapes made from multiple Mask nodes.

Manually Showing Tile Pictures and Node Options

You also have the option of manually choosing which nodes you’d like to show thumbnails.
For example, there may be certain key points of the node tree where you’d like to see a small visual
representation of what’s happening in the composition.

To toggle thumbnails for one or more specific nodes:

1 Select one or more nodes in the Node Editor.
2 Right-click one of the selected nodes, and choose one of the following from the contextual menu:
– Show > Show Tile Pictures
– Show > Show Modes/Options

When you’ve manually enabled thumbnails for different nodes, they’ll remain visible whether or not
those nodes are selected.

Switching Thumbnails between Images and Icons

Whenever you enable node thumbnails, you have the choice of having these thumbnails either display
an image of the state of the image at that node, or you can instead choose to display the icon for that
particular node. The setting for this affects all nodes at once.

To display icons instead of thumbnails:

– Right-click anywhere in the background of the Node Editor and deselect Show Thumbnails in the
contextual menu.

Chapter 5 Working in the Node Editor 140

 Sometimes Nodes Only Show Icons
As you add more and more nodes to a composition, you’ll notice that some nodes never
display an image in their thumbnail. In these cases, the default icon for that node is displayed
instead of an image.
Most nodes in the Particle and 3D categories fall into this group. The exceptions are the
pRender node and the Render 3D node. These two nodes are capable of displaying a
rendered thumbnail if you have the menu options set for Thumbnails to be displayed.
In other cases, whether nodes display images in their thumbnail is more situational. Some
Transform nodes are able to concatenate their results with one another, passing the actual
processing downstream to another node later in the node tree. In this case, upstream
Transform nodes don’t actually process the image, so they don’t produce a thumbnail.
In other situations where the Loader is not reading in a clip or the clip is trimmed in the
Keyframes Editor to be out of range, it can cause the node not to process the image, so it will
not produce a rendered Thumbnail image. Also, nodes that have been set to Pass Through
mode are disabled and do not display a rendered Thumbnail image.

Finding Nodes
Modern visual effects require detailed work that often results in compositions with hundreds of nodes.
For such large node trees, finding things visually would have you panning around the Node Editor for a
long, long time. Happily, you can quickly locate nodes in the Node Editor using the Find dialog.

The Find dialog lets you quickly locate nodes

wherever they are in the Node Editor.

Performing Simple Searches

To do simple searches using node names is easy.

To search for a node in the Node Editor:

1 Press Command-F, or right-click in an empty area of the Node Editor and choose Find from the
contextual menu.
2 When the Find dialog appears, do the following:
– Enter a search term in the Find field.
– Choose search options, such as whether to match the whole phrase in the Find field, whether
to match the case, whether to use a sequence number, or whether to use a regular expression
in the Find field.
– Choose what to search. Options include tool name, tool type name, or tool type ID.

Chapter 5 Working in the Node Editor 141

3 To perform the find, do one of the following:
– Click Find Next to try to select a downstream node matching the criteria.
– Click Find Previous to try to select an upstream node matching the criteria.
– Click Find All to try to select all nodes in the Node Editor that match the criteria.

The Find window closes. If either the Find Next, Find Previous, or Find All operations are successful,
the found node or nodes are selected. If not, a dialog appears letting you know that the string could
not be found.

TIP: Finding all the nodes of a particular type can be very useful if you want, for example, to
disable all Resize nodes. Find All will select all the nodes based on the search term, and you
can temporarily disable them by pressing the shortcut for Bypass, Command-P.

Using Regular Expressions

If you need to do more complicated searches, you can turn on the Regular Expression checkbox,
which lets you enter some simple expressions with which to create more complex find operations.
Some useful examples of regular expressions that are valuable include the use of Character Sets.

Character Sets
Any characters typed between two brackets [ ] will be searched for. Here are some examples of
character set searches that work in Fusion.
[a-z]

Finds: Every node using a lower caps letter

[a-d]

Finds: Every lower caps letter from a to d, and will find nodes with a, b, c, or d
[Tt]

Finds: Every node with an upper case T or a lower case t

[aeiou]

Finds: Every vowel

[0-9]

Finds: Every numeral

[5-7]

Finds: Every numeral from five to seven, and will find nodes numbered with 5, 6, or 7

Chapter 5 Working in the Node Editor 142

Custom Node Settings
When a node is added to the Node Editor, its parameters are set to the default values for that type of
node. If you find yourself constantly readjusting the parameters of a node to a preferred starting point
as soon as it’s added to the node tree, you can override the default node settings with your own
custom settings.

To save new default settings for a particular type of node:

1 Create a new node.
2 Open the Inspector and customize that node’s settings to the new defaults you want it to have.
3 Right-click that node in the Node Editor, or right-click that node’s control header in the Inspector,
and choose Settings > Save Default from the contextual menu.

TIP: You can also save six different settings for a node in the Node Editor using the Version
buttons at the top of the Inspector. For more information, see Chapter 69, “Editing Parameters
in the Inspector” in the DaVinci Resolve Reference Manual or Chapter 8 in the Fusion
Reference Manual.

Managing Saved Settings

Custom node default settings are saved to a folder on your hard drive that’s based on the Path Map >
Defaults preference in the Fusion Settings. This path is customizable for facilities where multiple
compositing artists use a common set of facility defaults, stored somewhere commonly accessible.
The default paths are:
– For macOS systems, this path defaults to: /UserName/Library/Application Support/
Blackmagic Design/DaVinci Resolve/Fusion/Defaults.
– For Windows systems, this path defaults to C: \Users\<username>\AppData\Roaming\
Blackmagic Design\DaVinci Resolve\Fusion\Defaults.
– For Linux systems, this path defaults to: ~/.fusion/BlackmagicDesign/DaVinci Resolve/
Fusion/Defaults.

If you browse this directory, the settings for each node are saved using a name taking the form
INTERNALNAME_PUBLICNAME.settings, where INTERNALNAME is the internal name of the Fusion
tool, and PUBLICNAME is the name of the Node that’s derived from the internal Fusion tool. For
example, the default setting for a Blur node would be called Blur_Blur.setting. This naming convention
is partly to ensure that third-party plug-in nodes don’t overwrite the defaults for built-in Fusion nodes
that happen to have the same name.

Resetting Defaults
Even if you’ve created new default settings for new nodes, you can always reset individual parameters
to the original default setting. In addition, it’s easy to restore the original default settings for new nodes
you create.

To reset a single parameter to the original default settings:

1 Create a new node.
2 Open the Inspector and customize a parameter to the new default value you want it to have.
3 Right-click that parameter in the Inspector, and choose Set to Default from the contextual menu.

Chapter 5 Working in the Node Editor 143

To reset every parameter in a node to the original defaults, do one of the following:
– Right-click on the node and choose Settings > Reset Default.
– Right-click that node’s control header in the Inspector, and choose Settings > Reset Default.
– Delete the .setting file from the Defaults folder.

NOTE: When you use the Settings > Reset Default command, the default .setting file is
deleted. If you want to save a node’s settings as alternate settings, you should use the
Settings > Save As command.

Saving and Loading Alternate Node Settings

Once you change parameter values for a node using the Inspector, those values can also be saved as
an alternate setting for that node, which can be reused at a later time.

To save alternate settings for a node:

1 Right-click on a tool, and then choose Settings > Save As from the contextual menu.
2 When the Save File dialog appears, enter a name for the Setting and save it to your hard drive.
Unlike saved defaults, the .settings files can be saved anywhere on the file system. They do not
need to be in the Default Settings folder.

To load a saved setting for one or more nodes:

1 Right-click a node and choose Settings > Load from the contextual menu.
2 Use the Open File dialog to select the settings you want to load into that node, and then click
Open. Those settings are now applied to that node.

Adding Saved Settings from the File System

Saved settings in your File system can also be used to create new nodes by dragging the .setting file
into the Node Editor from a standard file browser. Once dropped, that setting turns into a new node.

TIP: If you drop a setting directly onto a connection line, the new node will be inserted onto
that connection.

Node Modes Including Disable and Lock

Right-clicking one or more nodes and opening the contextual menu reveals a series of commands in
the Modes submenu, some with accompanying keyboard shortcuts, that let you set control visibility,
disable, lock, update, and cache nodes.
– Show Controls: Sets whether that node reveals its parameters in the Inspector when it’s selected
and whether its onscreen controls appear in viewers. On by default.
– Pass Through: (Command-P) Identical to the toggle switch in the Inspector that turns nodes off
and on. Disabled nodes are ignored as image data is passed from the next previous upstream
node to the next downstream node. On by default.
– Locked: (Command-L) Identical to the lock button in the Inspector that prevents a node from being
edited in the Inspector. Off by default.

Chapter 5 Working in the Node Editor 144

 – Update: (Command-U) On by default. While this option is enabled, all changes to the node will
cause it to re-render. When Update is disabled, you can still change the node’s parameters, but
those changes will not process or update the image until Update is re-enabled. While disabled,
the last processed image for that node will be displayed as a freeze frame. One example of when
this is useful is when you have a large or processor-intensive composition (such as a particularly
intense particle system), and disabling this option temporarily will let you quickly make several
quick parameter adjustments to different nodes without forcing you to wait for the node tree to
re-render after every adjustment. Another example of when this is useful is when you want to
quickly see the effect of animated downstream nodes while keeping upstream nodes that are too
processor-intensive to play in real time from rendering additional frames.
– Force Cache: When enabled, this node’s output for the current frame has an extremely high cache
priority, essentially forcing it to stay cached in memory. Off by default.

Toggling any one of these node modes displays a badge within that node indicating its state.

Node Editor Options

Right-clicking in an empty area of the Node Editor will bring up the contextual menu and the Options
submenu. The Options submenu contains several choices that can be used to customize how the
Node Editor looks and behaves.
– Pipes Always Visible: Enabling this option causes a connection to cross over a node instead of
beneath it, sometimes making it easier to follow the connection’s path.
– Show Hidden Connections: When enabled, the Inspector option to Hide Incoming Connections in
every node is overridden and all connections are displayed in the Node Editor.
– Aspect Correct Tile Pictures: Aspect Correct Tile Pictures forces the display of thumbnails to be
aspect corrected, which is slower but visually more accurate. This option is enabled by default.
– Full Tile Render Indicators: Enabling this option causes the thumbnail to flash green
when rendering, which makes it easier to identify which node is processing in a large,
complex node tree.
– Show Grid: This option can be used to enable or disable the Node Editor’s background grid.
– Show Instance Links: When enabled, the Node Editor draws a green connection between an
instanced node and its parent.
– Auto Remove Routers: If routers are disconnected from a tool, they are automatically deleted
from the Node Editor. This option is enabled by default to eliminate the need to delete orphaned
routers.
– Show Navigator: Enabling this option displays a small overview window of the entire node tree in
the Node Editor’s top-right corner. For more information, see the Navigator section in this chapter.
– Auto Navigator: The Navigator only appears when one or more nodes is outside the visible area
of the Node Editor. For more information, see the Navigator section in this chapter.
– Build Flow Vertically/Horizontally: Node trees can be built either horizontally from left to right or
vertically from top to bottom. Enabling one of these options determines whether new nodes are
added beneath the current node or to the right of the current tool.
– Orthogonal/Direct Pipes: Use these two options to decide whether connections between nodes
are drawn as Direct (straight) lines or Orthogonal (bent) lines.

Chapter 5 Working in the Node Editor 145

Node Tooltips and the Status Bar
Even in simple node trees, it’s easy to forget some essential detail about the nodes in your comp.
To help you figure out what everything’s for, you can hover the pointer over any node in the Node
Editor to display information in the Status bar at the bottom of the Node Editor consisting of that node’s
name, frame size, pixel aspect, resolution, and color depth.

The Status bar located beneath the Node Editor.

If you wait a few moments later, a more elaborate presentation of the same information appears
within a floating tooltip in the Inspector. This tooltip gives you additional information about the
Domain (Image and DoD) and the data range used by that clip.

The floating tooltip showing node information

that appears within the Node Editor.

Chapter 5 Working in the Node Editor 146

Chapter 6

Node Groups,
Macros, and
Fusion Templates
This chapter reveals how to use groups, macros, and templates in Fusion so working
with complex effects becomes more organized, more efficient, and easier.

Contents
Groups  148
Creating Groups  148
Deleting Groups  149
Expanding and Collapsing Groups  149
Panning and Scaling within Open Group Windows  149
Ungrouping Nodes  149
Saving and Reusing Groups  149
Macros  150
Creating Macros  150
Using Macros  152
Re-Editing Macros  152
Other Macro Examples  152
Creating Fusion Templates  152
Getting Started with a Fusion Title Template  152
Saving a Title Macro  153
Using Your New Title Template  156
Getting Started with a Fusion Transition Template  157
Creating a Fusion Transition Template  158

Chapter 6 Node Groups, Macros, and Fusion Templates 147

Using Your New Transition Template  160
Getting Started with a Fusion Generator Template  160
Creating a Fusion Generator Template  161
Using Your New Generator Template  163
Creating a Fusion Effect Template  163
Creating a Fusion Effect Template for Two or More Layers  164
Changing Durations of a Template  164
Creating a Custom Template Icon  165
Using Fusion Template Bundles  165

Groups
When you work on complex visual effects, node trees can become sprawling and unwieldy, so
grouping tools together can help you better organize all the nodes and connections. Groups are
containers in your node tree that can hold multiple nodes, similar to the way a folder on your Desktop
holds multiple files. There is no limit to the number of nodes that can be contained within a group, and
you can even create subgroups within a group.

Creating Groups
Creating a group is as simple as selecting the nodes you want to group together and using the
Group command.

To create a group:
1 Select the nodes you want grouped together.
2 Right-click one of the selected nodes and choose Group from the contextual menu (Command-G).

Several nodes selected in preparation for making a group (left), and the resulting group (right).

The selected nodes are collapsed into a group, which is displayed as a single node in the Node Editor.
The Group node can have inputs and outputs, depending on the connections of the nodes within the
group. The Group node only displays inputs for nodes that are already connected to nodes outside
the group. Unconnected inputs inside the group will not have an Input knot displayed on the
Group node.

Chapter 6 Node Groups, Macros, and Fusion Templates 148

Deleting Groups
Deleting a group is no different from deleting any other node in the Node Editor. Select a group and
press Delete, Backspace, or Forward-Delete, and the group along with all nodes contained within it
are removed from the node tree.

Expanding and Collapsing Groups

A collapsed group is represented by a single “stack” node in the node tree. If you want to modify any
of the nodes inside the group, you can open the group by double-clicking it or by selecting the group
node and pressing Command-E.

An open group window showing

the minimize button.

When you open a group, a floating window shows the nodes within that group. This floating window is
its own Node Editor that can be resized, zoomed, and panned independently of the main Node Editor.
Within the group window, you can select and adjust any node you want to, and even add, insert, and
delete nodes while it is open. When you’re ready to collapse the group again, click the minimize button
at the top left corner of the floating window, or use the keyboard shortcut (Cmd-E).

Panning and Scaling within Open Group Windows

You can pan and scale an open group window using the same mouse buttons you use to pan and
scale the main Node Editor. However, when you’re working in an expanded group and simultaneously
making changes to the main node tree, you may want to prevent the expanded group from being
individually panned or scaled. Turning off the Position button at the right of the group title bar locks the
group nodes to the size of the nodes in the rest of the overall node tree. Turning on this Position
button lets you size group nodes independently of the rest of the node tree.

Ungrouping Nodes
If you decide you no longer need a particular group, or you simply find it easier to have constant
access to all the nodes in the group at once, you can decompose or “ungroup” the group without
deleting the nodes within it to eliminate the group but keep the contents in the Node Editor.

To ungroup nodes, do the following:

1 Right-click on the group.
2 Choose Ungroup from the contextual menu. The nodes inside the group are placed back in the
main node tree.

Saving and Reusing Groups

One of the best features of groups is that every group and its settings can be saved for later use in
other shots or projects. Groups and their settings can be recalled in various ways.
A good example of when you might want to Save and Load a group is in a studio with two or more
compositing artists. A lead artist in your studio can set up the master comp and create a group
specifically for keying greenscreen. That key group can then be passed to another artist who refines

Chapter 6 Node Groups, Macros, and Fusion Templates 149

the key, builds the mattes, and cleans up the clips. The setting can then be saved out and loaded back
into the master comp. As versions are improved, these settings can be reloaded, updating the
master comp.

Methods of saving and reusing groups:

– To save a group: Right-click a group and choose Settings > Save As from the contextual menu.
– To reuse a group: Drag it from your computer’s file browser directly into the Node Editor.
This creates a new group node in the node tree with all the same nodes as the group you saved.
– To load the settings from a saved group to another group with the same nodes: Right-click a
group in the Node Editor and choose Settings > Load from the contextual menu.

In Fusion Studio, you can also save and reuse groups from the Bins window:
– To save a group: Drag the group rom the Node Editor into the opened Bin window.
A dialog will appear to name the group setting file and the location where it should be saved
on disk. The .settings file will be saved in the specified location and placed in the bins for easy
access in the future.

Macros
Some effects aren’t built with one tool, but with an entire series of operations, sometimes in complex
branches with interconnected parameter controls. Fusion provides many individual effects nodes for
you to work with but gives users the ability to repackage them in different combinations as
self‑contained “bundles” that are either macros or groups. These “bundles” have several advantages:
– They reduce visual clutter in your node tree.
– They ensure proper user interaction by allowing you to restrict which controls from each node of
the macro are available to the user.
– They improve productivity by allowing artists to quickly leverage solutions to common compositing
challenges and creative adjustments that have already been built and saved.

Macros and groups are functionally similar, but they differ slightly in how they’re created and presented
to the user. Groups can be thought of as a quick way of organizing a composition by reducing the
visual complexity of a node tree. Macros, on the other hand, take longer to create because of how
customizable they are, but they’re easier to reuse in other comps.

Creating Macros
While macros let you save complex functions for future use in very customized ways, they’re actually
pretty easy to create.

To make a macro from nodes in the Node Editor:

1 Select the nodes you want to include in the macro you’re creating. Because the macro you’re
creating will be for a specific purpose, the nodes you select should be connected together to
produce a particular output from a specific set of inputs.

TIP: If you want to control the order in which each node’s controls will appear in the
macro you’re creating, Command-click each node in the order in which you want it
to appear.

2 Right-click one of the selected nodes and choose Macro > Create Macro from the
contextual menu.

Chapter 6 Node Groups, Macros, and Fusion Templates 150

 A Macro Editor window appears, showing each node you selected as a list, in the order in which
each node was selected.

The macro editor with a Blur node and Color Corrector node.

3 First, enter a name for the macro in the field at the top of the Macro Editor. This name should
be short but descriptive of the macro’s purpose. No spaces are allowed, and you should
avoid special characters.
4 Next, open the disclosure control to the left of each node that has controls you want to expose
to the user and click the checkbox to the right of each node output, node input, and node control
that you want to expose.
The controls you check will be exposed to users in the order in which they appear in this list, so
you can see how controlling the order in which you select nodes in Step 1, before you start editing
your macro, is useful. Additionally, the inputs and outputs that were connected in your node tree
are already checked, so if you like these becoming the inputs and outputs of the macro you’re
creating, that part is done for you.
For each control’s checkbox that you turn on, a series of fields to the left of that control’s row lets
you edit the default value of that control as well as the minimum and maximum values that control
will initially allow.
5 When you’re finished choosing controls, click Close.
6 A dialog prompts you to save the macro. Click Yes.
7 A Save Macro As dialog appears in which you can re-edit the Macro Name (if necessary), and
choose a location for your macro.
To have a macro appear in the Fusion page Effects Library Tools > Macros category, save it in the
following locations:
– On macOS: Macintosh HD/Users/username/Library/Application Support/Blackmagic Design/
DaVinci Resolve/Fusion/Macros/
– On Windows: C:\Users\username\AppData\Roaming\Blackmagic Design\DaVinci Resolve\
Support\Fusion\Macros
– On Linux: home/username/.local/share/DaVinciResolve/Fusion/Macros

Chapter 6 Node Groups, Macros, and Fusion Templates 151

 To have a macro appear in the Fusion Studio Effects Library Tools > Macros category, save it in the
following locations:
– On macOS: Macintosh HD/Users/username/Library/Application Support/Blackmagic Design/
Fusion/Macros/
– On Windows: C:\Users\username\AppData\Roaming\Blackmagic Design\Fusion\Macros
– On Linux: home/username/.fusion/BlackmagicDesign/Fusion/Macros

8 When you’re done, click Save.

Using Macros
Macros can be added to a node tree using the Add Tool > Macros or Replace Tool > Macros submenus
of the Node Editor contextual menu.

Re-Editing Macros
To re-edit an existing macro, just right-click anywhere within the Node Editor and choose the macro
you want to edit from the Macro submenu of the same contextual menu. The Macro Editor appears,
and you can make your changes and save the result.

Groups Can Be Accessed Like Macros

Groups can also be loaded from the Insert Tool > Macros submenu if you save a group’s .
setting file to the Macros folder in your file system. For example, on macOS, the Macintosh
HD/Library/Application Support/Blackmagic Design/DaVinci Resolve/Fusion/Macros/ directory.

Other Macro Examples

A macro can also be used as a custom LUT. Just copy the macro’s .setting file to the LUTs: folder, and
the macro will be selectable in the viewers as a LUT. These LUT macros can be used for more than just
a color adjustment; you could make a macro that does YUV 4:2:2 resampling, a resize, a sharpening
filter, or just watermarking.

Creating Fusion Templates

The integration of Fusion into DaVinci Resolve has enabled the ability to create Fusion Titles,
Transitions, Effects, and Generators templates for use in the Edit page. You can create these templates
in the Fusion page or within Fusion Studio and then copy them into DaVinci Resolve. Fusion Titles,
Generators, and Transition templates are essentially comps created in Fusion but editable in the
Timeline of the Edit page with custom controls. This section shows you how it’s done.

Getting Started with a Fusion Title Template

The first part of creating a Fusion title template is to create a Fusion composition consisting of Fusion-
generated objects assembled to create nearly any kind of title or generator you can imagine. If you’re
really ambitious, it can include animation. In this example, 3D titles and 2D titles have been combined
into a show opener.

Chapter 6 Node Groups, Macros, and Fusion Templates 152

Building a composition to turn into a title template.

Saving a Title Macro

Macros are basically Fusion compositions that have been turned into self-contained nodes. Ordinarily,
these nodes are used as building blocks inside of Fusion so that you can turn frequently-made
compositing tricks that you use all the time into your own nodes. However, you can also use this macro
functionality to build title templates for the Edit page.
Having built your composition, select every single node you want to include in that template except for
the MediaIn and MediaOut nodes in DaVinci Resolve or Loader and Saver nodes in Fusion Studio.

Selecting the nodes you want to turn into a title template.

TIP: If you want to control the order in which node controls will be displayed later on, you can
Command-click each node you want to include in the macro, one by one, in the order in
which you want controls from those nodes to appear. This is an extra step, but it keeps things
better organized later on.

Chapter 6 Node Groups, Macros, and Fusion Templates 153

Having made this selection, right-click one of the selected nodes and choose Macro > Create Macro
from the contextual menu.

Creating a macro from the selected nodes.

The Macro Editor window appears, filled to the brim with a hierarchical list of every parameter in the
composition you’ve just selected.

The Macro Editor populated with the parameters of all the nodes you selected.

Chapter 6 Node Groups, Macros, and Fusion Templates 154

This list may look intimidating, but closing the disclosure control of the top Text1 node shows us what’s
really going on.

A simple list of all the nodes we’ve selected.

Closing the top node’s parameters reveals a simple list of all the nodes we’ve selected. The Macro
Editor is designed to let you choose which parameters you want to expose as custom editable controls
for that macro. Whichever controls you choose will appear in the Inspector whenever you select that
macro, or the node or clip that macro will become.
So all we have to do now is to turn on the checkboxes of all the parameters we’d like to be able to
customize. For this example, we’ll check the Text3D node’s Styled Text checkbox, the Cloth node’s
Diffuse Color, Green, and Blue checkboxes, and the SpotLight node’s Z Rotation checkbox, so that
only the middle word of the template is editable, but we can also change its color and tilt its lighting
(making a “swing-on” effect possible).

Selecting the checkboxes of parameters we’d like to edit when using this as a template.

Once we’ve turned on all the parameters we’d like to use in the eventual template, we click the Close
button, and a Save Macro As dialog appears.
To have the Title template appear in the Effects Library > Titles category of DaVinci Resolve, save the
macro in the following locations:
– On macOS: Macintosh HD/Users/username/Library/Application Support/Blackmagic Design/
DaVinci Resolve/Fusion/Templates/Edit/Titles
– On Windows: C:\Users\username\AppData\Roaming\Blackmagic Design\DaVinci Resolve\
Support\Fusion\Templates\Edit\Titles
– On Linux: home/username/.local/share/DaVinciResolve/Fusion/Templates/Edit/Titles

Chapter 6 Node Groups, Macros, and Fusion Templates 155

 Choosing where to save a title template for the Edit page in DaVinci Resolve.

Using Your New Title Template

After you’ve saved your macro, you’ll need to quit and reopen DaVinci Resolve. When you open the
Effects Library of the Edit page, you should see your new template inside the Titles category, ready to
go in the Fusion Titles list.

Custom titles appear in the Fusion Titles section of the Effects Library.

Editing this template into the Timeline and opening the Inspector, we can see the parameters we
enabled for editing, and we can use these to customize the template for our own purposes.

Chapter 6 Node Groups, Macros, and Fusion Templates 156

Customizing the template we made.

And that’s it!

Getting Started with a Fusion Transition Template

When creating a Fusion transition template, it’s easiest to start with an existing transition template and
build off that. Three transitions are located in the Fusion Transitions category of the DaVinci Resolve
Effects Library. The simplest transition is the Cross Dissolve, while the most complex example is the
Slice Push.

The Fusion transition templates located in DaVinci Resolve’s Effects Library.

Chapter 6 Node Groups, Macros, and Fusion Templates 157

Creating a Fusion Transition Template
The three Fusion Transitions located in the DaVinci Resolve Effects Library are basically Fusion
compositions that have been turned into macros. Ordinarily, macros are used as building blocks inside
of Fusion so that you can turn frequently-made compositing tricks that you use all the time into your
own nodes. However, you can also use this macro functionality to build transition templates for the
DaVinci Resolve Edit page.
Once you apply a Fusion Transition in the Edit page, you can right-click it and choose Open in
Fusion Page.

Right-clicking over a Fusion Transition in the DaVinci Resolve Edit page.

The Fusion page opens, displaying the node tree used to create the Fusion transition.

The Cross Dissolve node tree in Fusion.

The MediaIn 1 node represents the outgoing clip in the Edit page Timeline. The MediaIn 2 clip
represents the incoming clip. You can modify or completely change the Cross Dissolve effect to
create your own custom transition using any of Fusion’s nodes.

The Fusion Cross Dissolve node tree replaced with Transforms and a Merge node.

TIP: To modify the duration of the Fusion transition from the Edit page Timeline, you must
apply the Resolve Parameter Modifier to any animated parameter. In place of keyframing the
transition, you create the transition using the Scale and Offset parameters of the Resolve
parameter modifier.

Chapter 6 Node Groups, Macros, and Fusion Templates 158

Updating a Fusion Transition
After modifying the transition in Fusion, you can choose to update the transition in the Timeline or
create a new transition, which you save into the Edit page Effects Library. To update the transition, just
return to the Edit page. The transition in the Edit page Timeline reflects the changes you make in the
Fusion page.

Saving a New Fusion Transition

If after modifying the transition in Fusion, you need to save it to the Effects Library to reuse it on other
transitions or other projects, you must make a macro and save it to the Transitions folder.
Start by selecting every node in the Node Editor that you want to include in the transition template,
including the two MediaIn nodes and the MediaOut node.

TIP: Since the transition template must include the MediaIn and MediaOut nodes, the final
steps for saving a transition template must be performed in DaVinci Resolve’s Fusion page
and cannot be performed in Fusion Studio.

Having made this selection, right-click one of the selected nodes and choose Macro > Create Macro
from the contextual menu.

The Macro Editor displaying the parameters of all the nodes you selected.

The Macro Editor window appears, displaying a hierarchical list of every parameter in the composition
you’ve just selected. The order of nodes is based on the order they were selected in the Node Editor
prior to creating the macro.
The Macro Editor is designed to let you choose which parameters you want to display as custom
controls in the Edit page Inspector when the transition is applied.
For transitions, you can choose not to display any controls in the Inspector, allowing only duration
adjustments in the Timeline. However, you can choose a simplified set of parameters for customization
by enabling the checkboxes next to any parameter name.

Chapter 6 Node Groups, Macros, and Fusion Templates 159

Once you enable all the parameters you want to use in the eventual template, click the Close button,
and a Save Macro As dialog appears. Here, you can enter the name of the transition, as it should
appear in the Edit page Effects Library.
To have the transition template appear in the Effects Library > Fusion Transitions category of
DaVinci Resolve, save the macro in the following locations:
– On macOS: Macintosh HD/Users/username/Library/Application Support/Blackmagic Design/
DaVinci Resolve/Fusion/Templates/Edit/Transitions
– On Windows: C:\Users\username\AppData\Roaming\Blackmagic Design\DaVinci Resolve\
Support\Fusion\Templates\Edit\Transitions
– On Linux: home/username/.local/share/DaVinciResolve/Fusion/Templates/Edit/Transitions

Using Your New Transition Template

After you’ve saved your macro, you’ll need to quit and reopen DaVinci Resolve. When you open the
Effects Library on the Edit page, the new transition template is listed in the Video Transitions category,
in the Fusion Transitions list.

A custom Fusion Transition saved in the Edit page Effects Library.

Applying this transition to a cut in the Timeline and opening the Inspector shows the parameters you
enabled for editing, if any.

Getting Started with a Fusion Generator Template

There is one simple Noise Gradient Fusion Generator located in the Effects Library that you can use as
a starting point for creating your own generators.

The Fusion Generator templates located in DaVinci Resolve’s Effects Library.

Chapter 6 Node Groups, Macros, and Fusion Templates 160

To open the Fusion Noise Gradient Generator in the Fusion page, do the following:
1 On the Edit page, drag the Fusion Noise Gradient Generator from the Effects Library to
the Timeline.
2 Right-click over the Noise Gradient Generator and choose Open in Fusion Page from the
pop-up menu.

The Fusion page opens, displaying the node tree that is used to create the Fusion Generator.

Creating a Fusion Generator Template

As easy as it is to begin with the Noise Gradient Generator template, you can just as easily start by
adding a Fusion Composition Effect to a Timeline in the Edit page.

To begin creating a Fusion Generator Template with an empty Fusion composition,

do the following:
1 On the Edit page, drag the Fusion Composition Effect from the Effects Library to the Timeline.
2 Right-click over the Composition Effect and choose Open in Fusion Page from the pop-up menu.

An empty Fusion page with a single MediaOut node opens, ready for you to create a
Fusion Generator.
The Fusion Generator is a solid image generated from any number of tools combined to create a static
or animated background. You can choose to combine gradient colors, masks, paint strokes, or
particles in 2D or 3D to create the background generator you want.

The Fusion Generator node tree creating concentric circles.

Chapter 6 Node Groups, Macros, and Fusion Templates 161

Saving a New Fusion Generator
After creating the generator you want in Fusion, you need to save it to the Effects Library to reuse it in
other projects from the Edit page. To do this, you must create a Macro and save it to the
Generator folder.
Ordinarily, macros are used as building blocks inside of Fusion so that you can turn frequently-made
compositing tricks that you use all the time into your own nodes. However, you also use this macro
functionality to build Generator templates for the DaVinci Resolve Edit page.
Start by selecting every node in the Node Editor that you want to include in the Generator template
including the MediaOut node.
Having made this selection, right-click one of the selected nodes and choose Macro > Create Macro
from the contextual menu.

The Macro Editor displaying the parameters of all the nodes you selected.

The Macro Editor window appears, displaying a hierarchical list of every parameter in the composition
you’ve just selected. The order of nodes is based on the order they were selected in the Node Editor
prior to creating the macro.
The Macro Editor is designed to let you choose which parameters you want to display as custom
controls in the Edit page Inspector when the Generator is applied. You can choose a simplified set of
parameters for customization by enabling the checkboxes next to any parameter name.
Once you enable all the parameters you want to use in the eventual template, click the Close button,
and a Save Macro As dialog appears. Here, you can enter the name of the Transition, as it should
appear in the Edit page Effects Library.
To have the Generator template appear in the Effects Library > Fusion Generators category of
DaVinci Resolve, save the macro in the following locations:
– On macOS: Macintosh HD/Users/username/Library/Application Support/Blackmagic Design/
DaVinci Resolve/Fusion/Templates/Edit/Generators
– On Windows: C:\Users\username\AppData\Roaming\Blackmagic Design\DaVinci Resolve\
Support\Fusion\Templates\Edit\Generators
– On Linux: home/username/.local/share/DaVinciResolve/Fusion/Templates/Edit/Generators

Chapter 6 Node Groups, Macros, and Fusion Templates 162

Using Your New Generator Template
After you’ve saved your macro, you’ll need to quit and reopen DaVinci Resolve. When you open the
Effects Library on the Edit page, the new Generator template is listed in the Generators category, in
the Fusion Generators list.

Custom Fusion Generator saved in the Edit page Effects Library.

Applying this Generator to the Timeline and opening the Inspector shows the parameters you enabled
for editing, if any.

Creating a Fusion Effect Template

You start building a Fusion Effect template by bringing a clip from the Edit page Timeline into Fusion.
This clip is only used for creating the template and will not be saved with the effect.
Once inside Fusion, you use Fusion’s nodes to create the effect you want. You can use a single node
or a hundred, depending on the effect you want to create. For instance, using Fusion’s Color
Correction nodes, you can create a simple color corrector you can use on the Edit page.

To create a simple Color Corrector effect, do the following:

1 Insert the Color Corrector node between the MediaIn and MediaOut nodes.
2 Select the Color Corrector node in the Node Editor, and then press Cmd-A to select the
remaining nodes.
3 Right-click over any of the selected node and choose Macro > Create Macros from the contextual
menu. Enabling the checkboxes in this window determines the parameters that appear in the
Edit page Inspector.
4 The Macro Editor window opens. Here, you can enable the checkboxes for any parameters you
want to be shown in the Edit page Inspector.
5 Enter the name of your effect at the top of the Macro Editor window.
6 To save the Macro, click Close at the bottom of the window, then click Yes in the dialog that
appears asking you to save the changes.

Chapter 6 Node Groups, Macros, and Fusion Templates 163

The Macros must be saved into the correct folder for DaVinci Resolve to recognize the
Macro as an effect.

In the save dialog, save the Macro in the following locations:

– On macOS: Macintosh HD/Users/username/Library/Application Support/Blackmagic Design/
DaVinci Resolve/Fusion/Templates/Edit/Effects
– On Windows: C:\Users\username\AppData\Roaming\Blackmagic Design\DaVinci Resolve\
Support\Fusion\Templates\Edit\Effects
– On Linux: home/username/.local/share/DaVinciResolve/Fusion/Templates/Edit/Effects

To see the effect in the Edit page Effects Library, you’ll need to quit DaVinci Resolve and
relaunch the application.

Creating a Fusion Effect Template for Two or More Layers

If the effect you want to create requires multiple images like a video wall, you start by creating a Fusion
clip on the Edit page Timeline that includes the number of layers you want the effect to have. The clips
are only used to create the number of image inputs for the template and will not be saved with
the effect.
Once inside Fusion, use Fusion’s nodes to create the effect you want.
Save the Macro following the same steps you use for single clip effects. Enable any of the parameters
you want to control in the Edit page. To be able to switch the order of video layers within the effect,
make sure you have the Layer checkbox enabled for all the MediaIn nodes.
Once you’ve saved the Macro and relaunched DaVinci Resolve, to use the effect on multiple timeline
layers, you must create a Fusion clip. The Fusion clip should contain the same number of layers the
effect requires. The order of the Timeline layers, going from the bottom track to the top, matches the
MediaIn numbers. For instance, video track 1 will match the position and appearance of MediaIn1,
video track 2 matches MediaIn2 and so on. If you want to change how tracks map to MediaIn nodes,
you can change the Layer number in the Inspector, assuming you enabled the MediaIn Layer
checkbox when creating the Macro.

Changing Durations of a Template

After you make a template in Fusion, you may want to change its duration in the Edit or Cut page
Timeline. Changing the duration when animation is involved can be complicated, so there are two
Modifiers in Fusion that can help determine how keyframes react when the duration is updated in the
Edit or Cut page Timeline.

Anim Curves Modifier

The Animation Curves Modifier (Anim Curves) is used to dynamically adjust the timing, values, and
acceleration of an animation, even if you decide to change the duration of a Comp. Using this Modifier,
it becomes infinitely easier to stretch or squish animations, create smooth motion, add bouncing
properties, or mirror animations curves without the complexity of manually adjusting splines.
When creating Fusion templates for the Edit or Cut page in DaVinci Resolve, the Anim Curves Modifier
allows the keyframed animation you’ve created in Fusion to stretch and squish appropriately as the
transition, title, or effect’s duration changes on the Edit and Cut page Timelines.

Keyframe Stretcher Modifier

The Keyframe Stretcher modifier is primarily used when creating title templates in Fusion for use in
DaVinci Resolve’s Edit or Cut page. The Keyframe Stretcher modifier is added to any animated
parameter so that the Hold keyframes between the initial animation on screen and the final animation
off-screen stretch when the template is trimmed in the Timeline. This allows the animation to retain its
timing while the static portion of the title stretches to meet the new duration requirements.

Chapter 6 Node Groups, Macros, and Fusion Templates 164

Creating a Custom Template Icon
It is possible to create a custom icon that is embedded with your template thumbnail in the Effects
Library, instead of the default first three letters of the template name.

To create a simple Custom Template Icon:

1 Create a .png file of the icon you want to use for the template. The recommended size is 104 x 58
pixels, but any image will be resized to fit.
2 Name the file exactly the same name as your template, just with a .png extension rather than a
.setting extension.
3 Place the .png image in the same directory as your template.

When you restart DaVinci Resolve, the icon you created will be embedded in the template thumbnail
across all the Effects Libraries in the program.

A Custom Icon added to a fisheye template, before (above) and after (below)

Using Fusion Template Bundles

For ease in distributing templates to other Fusion users, multiple templates can now be bundled
together into a single .drfx file. This file can then be imported back into another Fusion workstation
easily to ensure that all custom templates are the same between computers.
Creating a Fusion Template Bundle requires using a specific directory structure and using your
operating systems file browser and .zip compression utility. The directory structure is listed below, and
you can always find a specific folder from within Fusion by right-clicking on any bin in the Effect Library
and selecting Show Folder.

Folder structure for Templates used in the Edit page:

– Edit
– Effects
– Generators
– Titles
– Transitions

Folder structure for Templates used in the Fusion page:

– Fusion

To create a Fusion Template Bundle:

1 In your OS, create a folder structure above that includes the specific folder for the type of
templates you want to be in the bundle. For example, if you have a transition and two effects
templates, you would create an Edit folder, and two subfolders inside it named Transitions
and Effects.
2 Copy your template (.setting) files into the appropriate directories. You can also include icon files
and any associated assets as well.

Chapter 6 Node Groups, Macros, and Fusion Templates 165

3 Use your OS zip compression utility to create a .zip file of the directory structure.
4 Rename the .zip file in your OS with the “.drfx” extension instead of .zip. The file icon should
change to reflect the new extension.

To import a Fusion Template Bundle:

1 Double-click on a .drfx file in your OS. DaVinci Resolve will launch and a dialogue box will appear
asking if you want to install the template bundle.
2 Drag to the .drfx file from your OS directly into the Fusion page in DaVinci Resolve. A dialogue box
will appear asking if you want to install the template bundle.

To delete a Fusion Template Bundle:

1 Navigate to the appropriate template directory in your file browser.
2 Delete the .drfx file.

IMPORTANT
The Fusion Template Bundle contains all the templates in one file. It does not uncompress
them into separate template files again. Therefore if you delete the .drfx file, all associated
templates inside that bundle will be removed as well.

Chapter 6 Node Groups, Macros, and Fusion Templates 166

Chapter 7

Using Viewers
This chapter covers working with viewers in Fusion, including using onscreen controls
and toolbars, creating groups and subviews, managing viewer Lookup Tables (LUTs),
working with the 3D viewer, and setting up viewer preferences and options.

Contents
Viewer Overview  169
Single vs. Dual Viewers  169
Floating Viewers in Fusion Studio  170
Video Output  170
Clean Feed  170
Loading Nodes into Viewers  170
Clearing Viewers  171
Position and Layout  171
The Viewer Divider  171
Zooming and Panning into Viewers  172
Flipbook Previews  172
Creating Flipbook Previews  172
Playing Flipbook Previews  173
Removing Flipbook Previews  174
Flipbook Preview Render Settings  174
Onscreen Controls  175
Showing and Hiding Onscreen Controls  175
Making Fine Adjustments to Onscreen Controls  176
Toolbars  176
Viewer Toolbar  176
Node Toolbars  176
A/B Buffers  177
Flipping between Buffers  177
Split Wipes between Buffers  177
Moving the Wipe Divider  178

Chapter 7 Using Viewers 167

Subviews  178
Showing and Hiding Subviews  179
Changing the Subview Type  179
Swapping the Subview with the Main View  179
Viewer and Subview Types  180
Viewing Selective Channels  184
Viewing Color Channels  184
Viewing Auxiliary Channels  184
The 3D Viewer  185
Panning, Scaling, and Rotating a 3D Viewer  185
Viewing Objects via Wireframe  185
Changing the POV of a 3D Viewer  185
Copying a Viewer’s POV to a Camera  186
Lighting and Shadows in 3D Viewers  187
Transparency in 3D Viewers  188
Grid  188
Vertex Normals  189
Quad View  189
Quad View Layouts  190
Using Quad Views for 2D Scenes  190
Guides  191
Frame Format Settings  191
Domain of Definition and Region of Interest  192
Domain of Definition (DoD)  192
Region of Interest (RoI)  193
Enabling RoI Controls  193
Managing Viewer Lookup Tables (LUTs)  194
How Lookup Tables Work in Fusion  195
Types of Viewer LUTs  195
Using Viewer LUTs  196
Editing Viewer LUTs  198
LUT Processing Order  199
Applying Multiple LUTs  200
Saving Custom LUTs  200
LUT Files  201
Viewer Preferences and Settings  202
Viewer Settings  202
The Viewer Options Menu  203
Locking the Viewer (Command-L)  203
Additional Viewer Options  204
Status Bar Information  204

Chapter 7 Using Viewers 168

Viewer Overview
Viewers in Fusion display the current frame of the current composition in a variety of ways to help you
see what you’re doing and evaluate the final result of your compositing artistry. Viewers display 2D
images, but they can also display a 3D environment using a 3D View as well as a special Quad viewer
to help you effectively work in three dimensions.

Side-by-side dual viewers: a 3D viewer (left), and a 2D viewer (right)

Additionally, you can expose “subviews” including color inspectors, magnifiers, waveforms,
histograms, and vectorscopes to help you analyze the image as you work.

Viewer with a 3D Histogram subview at the upper left-hand corner

Single vs. Dual Viewers

By default, there are two viewers positioned side by side across the top of the window. However, you
can use the Single/Dual Viewer button to toggle between displaying a single viewer or two viewers
displayed side by side.

The Single/Dual Viewer toggle button

Chapter 7 Using Viewers 169

Floating Viewers in Fusion Studio
In addition to the dual views above the Node Editor, Fusion Studio provides the option to use an
unlimited number of floating viewers. These floating viewers are excellent for taking full advantage of a
dual monitor configuration. Floating viewers can also be set to full-screen mode to make the best use
of screen real estate.
To create a new floating display view, select Window > New View from the menu bar at the top of the
screen. The position and configuration of the new view can be saved in the layout preferences,
if required.

Video Output
When using DaVinci Resolve or Fusion Studio, if Blackmagic video hardware is present in the
computer, then you can select a node to preview directly on that display. While video output can’t be
used for manipulating onscreen controls such as center crosshairs or spline control points, they’re
extremely valuable for evaluating your composition via the output format, and for determining image
accuracy using a properly calibrated display.
The video hardware is configured from the DaVinci Resolve and Fusion Studio preferences.

Clean Feed
When using DaVinci Resolve with dual computer monitors, a full-screen viewer can be displayed on
the secondary monitor from the Fusion page. This displays a third view indicator button under each
node to control what is shown on the second display. To activate this monitor, make sure you do not
have Dual Screen enabled under the Workspace menu and then select Workspace > Video Clean
Feed and select your second computer display from the submenu.

Loading Nodes into Viewers

The Fusion page in DaVinci Resolve and Fusion Studio show two different things when you first open
each application. When you first open the Fusion page, the output of the current empty composition
(the MediaOut1 node) is usually showing in viewer 2. If you’re in dual-viewer mode, viewer 1 remains
empty until you assign a node to one of them. In Fusion Studio, because there are no nodes when you
first begin a comp, nothing is displayed in the viewers.

To load specific nodes into specific viewers, do one of the following:

– Hover the pointer over a node, and click one of two buttons that appear at the bottom-
left of the node.
– Click once to select a node, and press 1 (for the left viewer) or 2 (for the right viewer).
– Right-click a node and choose View On > None/Left View/Right View in the contextual menu.
– Right-click the control header of a node in the Inspector, and choose View On > None/Left View/
Right View from the contextual menu.
– Drag a node and drop it over the viewer you’d like to load it into (this is great for tablet users).

When a node is being viewed, a View Indicator button appears at the bottom left. This is the same
control that appears when you hover the pointer over a node. Not only does this control let you know
which nodes are loaded into which viewer, but they also expose little round buttons for changing
which viewer they appear in.

Chapter 7 Using Viewers 170

 Viewer assignment buttons at the bottom
left of nodes indicate when they’re being
viewed, and which dot is highlighted indicates
which viewer that node is loaded into.

Clearing Viewers
To clear an image from a viewer, click in the viewer to make it active; a light purple outline is displayed
around the active panel. With the viewer active, press the Tilde (~) key. This key is usually found to the
left of the 1 key on U.S. keyboards. The fastest way to remove all images from all viewers is to make
sure none of the viewers is the active panel, and then press the Tilde key.

Position and Layout

When you resize and change the layout of viewers in the composition, that configuration is always
saved with the composition. So each time you open the composition, the size and layout is
remembered. You can prevent this behavior by disabling the Recall Layout checkbox in the Fusion
Global Layout preferences.
If you want all new compositions to open with a certain viewer layout, you can configure the layout of
the two primary viewers, and then use the Grab Document Layout button in the Fusion Global Layout
preferences to remember the layout for any new compositions. To save the position and size of
floating viewers, you use the Grab Program Layout button. Finally, if you want to have the floating
viewers opened automatically when you open Fusion, enable the Create Floating Views checkbox.

The Viewer Divider

You can change the relative sizes of the left and right viewers using the horizontal viewer divider that
runs between them. Drag the viewer divider to increase or decrease the amount of space used by one
viewer. The adjacent viewer will adjust to accommodate the new layout.
The amount of vertical space available for both viewers can be adjusted by dragging the horizontal
scrollbar between the viewers and the work area below them.

The viewer divider bar

Chapter 7 Using Viewers 171

Zooming and Panning into Viewers
There are standardized methods of zooming into and panning around viewers when you need a closer
look at the situation. These methods also work with the Node Editor, Spline Editor, and
Keyframes Editor.

Methods of panning viewers:

– Middle-click and drag to pan around the viewer.
– Hold down Shift and Command and drag the viewer to pan.
– Drag two fingers on a track pad to pan.

Methods of scaling viewers:

– Click a viewer and press the Equals key (=) to zoom in, and the Minus key (-) to zoom out.
– Press the middle and left mouse buttons simultaneously and drag left or right to resize the viewer.
– Hold down the Command key and use your pointer’s scroll control to resize the viewer.
– Hold down the Command key and drag two fingers on a track pad to resize the viewer.
– Hold down the middle mouse button, and then click the left mouse button to zoom in, or click the
right button to zoom out. The scaling uses a fixed amount, centered on the position of the cursor.
– Click a viewer and press Command-1 to resize the image in the viewer to 100 percent.
– Click a viewer and press Command-F or Command-1 to reset the image in the viewer to fit
the viewer.
– Click the Scale Viewer menu and choose Fit or a percentage.
– Right-click on a viewer and choose an option from the Scale submenu of the contextual menu.
This includes a Custom Scale command that lets you type your own scale percentage

Methods of spinning 3D viewers:

– In 3D Perspective view, hold down the Shift key and drag to spin the stage around.

Flipbook Previews
As you build increasingly complex compositions, and you find yourself needing to preview specific
branches of your node tree to get a sense of how various details you’re working on are looking, you
may find it useful to create targeted RAM previews at various levels of quality right in the viewer by
creating a RAM Flipbook. RAM Flipbook Previews are preview renders that exist entirely within RAM
and allow you to render a node’s output at differing levels of quality for quick processing in order to
watch a real-time preview.

Creating Flipbook Previews

Creating a Flipbook Preview is relatively fast, once you know where to look.

To create a Flipbook Preview:

1 Choose the node in your node tree that you want to preview by doing one of the following:
– – Hold down the Option key while dragging a node into the viewer.
– – Right-click a node and choose an option from the Create/Play Preview submenu in
the contextual menu.

Chapter 7 Using Viewers 172

2 When the Preview Render dialog opens, choose the quality, resolution, and motion blur settings
you want to use for the Flipbook Preview.

The Flipbook Preview Render dialog

3 When you’ve chosen the settings you want to use, click Start Render.
The current frame range of the Time Ruler is rendered using the settings you’ve selected, and the
result is viewable in the viewer you selected or dragged into.

Once you’ve created a Flipbook Preview within a particular viewer, right-clicking that viewer presents
Flipbook-specific commands and options to Play, Loop, or Ping-Pong the Flipbook, to open it Full
Screen, to Show Frame Numbers, and to eliminate it.

TIP: If you want to create a Flipbook Preview and bypass the Render Settings dialog by just
using either the default setting or the settings that were chosen last, hold down Shift-Option
while you drag a node into the viewer. The Settings dialog will not appear, and rendering the
preview will start right away.

Playing Flipbook Previews

While the Flipbook Preview is loaded into a viewer, or open in full-screen mode, you can play or scrub
through it using the mouse and the keyboard.

To play back a Flipbook using the mouse, do the following:

– Double-click in the viewer to start playback.

To scrub through a Flipbook using the mouse, do the following:

– Hold down the right mouse button down and drag left or right to scrub through frames.

To play back a Flipbook using the keyboard, do one of the following:

– Press the Spacebar to start or stop playback.
– Hold Shift and press the Spacebar to play in reverse.

To scrub through a Flipbook frame-by-frame using the keyboard, do one of the following:
– Press the Left or Right Arrow keys to move to the previous or next frame.
– Hold Shift and press the Left or Right Arrow keys to jump back or forward 10 frames.
– Press Command-Left Arrow to jump to the first frame.
– Press Command-Right Arrow to jump to the last frame.

Chapter 7 Using Viewers 173

 TIP: The mouse and keyboard shortcuts work in full-screen mode as well.

Removing Flipbook Previews

Once you create a Flipbook Preview, you need to know how to clear it from RAM.

To eliminate a Flipbook you’ve created:

– Right-click within a viewer containing a Flipbook Preview, and choose Remove Preview.

Flipbook Preview Render Settings

This section covers all the settings available for rendering Flipbook Previews to RAM.

Settings
The Settings section of the Preview Render dialog includes three buttons that determine the overall
quality and appearance of your Flipbook Preview. These buttons also have a significant impact on
render times.
– HiQ: When enabled, this setting renders the preview in full image quality. If you need to see
what the final output of a node would look like, then you would enable the HiQ setting. If you are
producing a rough preview to test animation, you can save yourself time by disabling this setting.
– MB: The MB in this setting stands for Motion Blur. When enabled, this setting renders with motion
blur applied if any node is set to produce motion blur. If you are generating a rough preview and
you aren’t concerned with the motion blur for animated elements, then you can save yourself time
by disabling this setting.
– Some: When Some is enabled, only the nodes specifically needed to produce the image of the
node you’re previewing are rendered.

Size
Since RAM Flipbook Previews use RAM, it’s helpful to know how many frames you can render into
RAM before you run out of memory. The Flipbook Preview dialog calculates the currently available
memory and displays how many frames will fit into RAM. If you have a small amount of RAM in your
computer and you cannot render the entire range of frames you want, you can choose to lower the
resolution to a setting that delivers the best quality/duration ratio for your preview.

Network
Network rendering is only available in Fusion Studio. For more information on network rendering,
see Chapter 65, “Rendering Using Saver Nodes” in the DaVinci Resolve Reference Manual or
Chapter 4 in the Fusion Reference Manual.

Shoot On
Sometimes you may not want to render every single frame, but instead every second, third, or fourth
frame to save render time and get faster feedback. You can use the Step parameter to determine the
interval at which frames are rendered.

Frame Range
This field defaults to the current Render Range In/Out set in the Time Ruler to determine the start and
end frames for rendering. You can modify the range to render more or fewer frames.

Configurations
Once you’ve created a useful preview configuration, you can save it for later use by clicking the Add
button, giving it a name, and clicking OK.

Chapter 7 Using Viewers 174

Updating a Preview
This option is designed for the interactive frame-by-frame work of rotoscoping and painting. Right-click
over a preview in the viewer and choose Update from its contextual menu. When active, any frames
that are modified on the previewed node are automatically updated in the preview’s playback. This
lets you reserve the RAM for playback. You can keep it playing on a loop or ping-pong while you work
in another viewer.

Onscreen Controls
When it comes to adjusting images, the Control Panel provides very precise numerical values, but
sometimes visually positioning an element using onscreen controls can get you where you want to go
with less tweaking. The viewers show onscreen controls for manipulating the parameters of the
currently selected node. Common onscreen controls include crosshairs, angle indicators, polylines,
and paint strokes. Each of these controls can be manipulated directly in the viewer using the mouse
or keyboard.

The Angle preview control

The controls shown in viewers are determined by which nodes are selected, not by the node
displayed in the viewer. For example, a downstream blur is easily viewed while manipulating the
controls for a selected polygon mask or merge. If multiple nodes are selected, the controls for every
selected node are shown simultaneously.

Showing and Hiding Onscreen Controls

The onscreen controls for a viewer can be hidden so they don’t interfere with viewing the image.

To toggle the visibility of onscreen controls, do one of the following:

– Click a viewer’s Option menu and choose Show Controls to toggle the controls on or off.
– Right-click in a viewer and choose Options > Show Controls from the contextual menu.
– Select a viewer and press Command-K.

Enabling/Disabling Onscreen Controls in Specific Nodes

Some nodes, like masks, allow disabling of their onscreen controls on a per-node basis, since you
often use multiple Polygon nodes to organize and animate masks.

You can disable some nodes, like the Polygon node, on a per-node basis.

Chapter 7 Using Viewers 175

Making Fine Adjustments to Onscreen Controls
If you want the visual guidance of onscreen controls with the precision of the Inspector, you can use
different keyboard modifiers.
– Up and Down Arrow keys can be used to adjust the vertical position of an onscreen control by
small steps.
– Holding down the Command key while using the Up and Down Arrow keys reduces the scale of
each step by a factor of ten. Holding Shift increases the scale of each step by a factor of ten.

Toolbars
There are two toolbars in the viewer: a viewer toolbar, which always appears at the top of each viewer
and gives you control over what that viewer shows, and an optional node toolbar that appears
underneath that gives you contextual controls based on the node you’ve selected in the Node Editor.

Viewer Toolbar
A viewer toolbar runs across the top of each viewer, providing access to many of the most commonly
used viewer-related settings, as well as an indication of the status of many of the most important
settings. Most of the menus and buttons found on this toolbar are described in detail throughout
this chapter.

The viewer toolbar

Node Toolbars
In addition to the viewer toolbar, a node toolbar is displayed underneath, at the top of the viewer
display area, whenever you select a node that exposes special nodes. Examples of nodes that expose
a toolbar include the text, masks, paths, paint strokes, and the 3D environment.

The node toolbar shown for the Paint node

Customizing the Node Toolbar

If you want to change the size of the buttons that appear in the Node toolbar, or turn on text names for
each node, you can right-click anywhere in the empty area of the toolbar and choose new settings
from the Icon Size and Button Style submenus in the contextual menu.

The contextual menu for the node toolbar

Chapter 7 Using Viewers 176

A/B Buffers
Each viewer has two buffers, each of which can contain images from different nodes, enabling easy
comparison of two different nodes within the same viewer by either toggling between buffers, or via
an adjustable split-wipe. Each buffer can be considered a complete and separate viewer within the
same viewer pane. The A buffer is always shown by default, so when you first load a node into a
viewer, the image loads into the A buffer.

Flipping between Buffers

Switching between buffers is easy, either to view a different image while keeping another image
handy, or to flip between the original image and the affected image for comparison.

To switch between buffers, do one of the following:

– Select a viewer and press comma (,) to select the A buffer or press period (.) to select the B buffer.
– Click the Buffer menu and choose either Switch to A View or Switch to B View.

The Buffer menu lets you switch between buffers

TIP: Each buffer can be set to different display settings—for example, showing different
channels or different viewing LUTs, either applied to different nodes or applied to two
buffered versions of the same node.

Split Wipes between Buffers

You can also wipe between both buffers, providing a more direct means of comparison.

To wipe between buffers, do one of the following:

1 Prepare to wipe between two images by loading different nodes into each buffer, or load the same
node with different viewer options into each buffer.
2 To toggle the split wipe on or off, do one of the following:
a. Click the Switch to Split Wipe View button.
b. Press Forward Slash (/).
3 To adjust the wipe, do one of the following:
a. Move the center of the wipe by dragging the center handle of the wipe divider.
b. Press Command-Option and click anywhere in the viewer to jump the wipe divider to
that location.
c. Change the angle or the wipe by dragging the wipe divider. Dragging the wipe divider while
holding the Shift key snaps it to the nearest 45-degree angle.
d. Panning or zooming the viewer pans and zooms both buffers together.

Chapter 7 Using Viewers 177

4 (Optional) If you want to change the image that’s displayed on that side of the split, you can drag
new nodes onto either half of the viewer.
5 To turn off the wipe, click the Switch to Split Wipe View button again (or press /).

The wipe divider can be adjusted for comparing different areas of the A and B images

Even when you wipe, you can choose different display channels, view LUTs, or other display options
for each buffer individually by clicking on the half of the wipe you want to alter, and then choosing the
options you want that buffer to use. This allows easy comparison of different channels, LUTs, or other
viewer settings while wiping the same image, or different images.

Moving the Wipe Divider

Occasionally, you will have either zoomed in or panned so far from the viewer divider that it’s no
longer visible in the viewer. Holding down Command-Option and clicking anywhere in the image will
cause the viewer divider to jump to the current position of the pointer.

Subviews
A subview is a “mini” viewer that appears within the main viewer. A subview is usually used to show
different information about the image.

The Subview menu with the Histogram subview displayed

Chapter 7 Using Viewers 178

For example, the RGB channels can be viewed in the main viewer, while the alpha channel is displayed
in a subview. For the most part, the subview is a fully functional miniature viewer, with its own
contextual menu and options. It responds to the same keyboard shortcuts and navigation controls as
any other viewer. However, there are several view types designed for use only in the subview,
including the Navigator, Magnifier, Color Inspector, and Image Info.

Showing and Hiding Subviews

Subviews are easily shown and hidden.

To enable the currently selected subview in the Subview menu of a viewer, do one of the following:
– Click the Subview button in the View toolbar.
– Choose Views > Subview > Enabled from the contextual menu.
– Click a viewer, and press the V key.

The Subview button in the viewer toolbar

Changing the Subview Type

The Subview button enables and disables the subview, which usually shows the last subview you
chose. You can change this at any time.

To change which subview type is displayed, do one of the following:

– Click the small arrow to the right of the Subview button to open its menu and choose which
subview you want.
– Right-click within a subview to bring up the subview’s contextual menu.

The Subview drop-down menu and contextual menu show all the available subview types. Once you
choose an option from the list, that view will be displayed in the subview, and the Subview button will
show and hide it as you wish.

Swapping the Subview with the Main View

It’s possible to swap the contents of the main viewer and the subview for select subview types.
However, certain view types, such as the Color Inspector and Magnifier, can only be used in the
subview. In these cases, the swap will not take place.

To swap the contents of the subview with the main view, do one of the following:
– Press Shift-V.
– Right-click in a viewer and choose Views > SubView > Swap from the contextual menu.

Chapter 7 Using Viewers 179

Viewer and Subview Types
Viewers can be changed to show a variety of different information about the image, but not all view
types are available at all times. For example, the 3D Viewer is not available for a 2D node, and some of
the measurement viewers are available only as subviews. Below is detailed information about the
different view types available.

2D Viewer
The 2D Viewer is the default type for showing images. When used as a subview, a different node than
the one used in the main viewer can be displayed by dragging the node into the subview.
This is the only subview type that is not just a different view of the same node in the main viewer.

The subview used as another viewer

3D Image Viewer
The 3D Image Viewer is available when viewing a node from the 3D category.

A 3D Image Viewer as a subview

Histogram
The Histogram Viewer is an analysis node that can be used to identify problems with the contrast and
dynamic range in an image. The graph shows the frequency distribution of colors in the image,
including out-of-range colors in floating-point images. The horizontal axis shows the colors from
shadows to highlights. The vertical axis shows the number of pixels in the image that occur at
each level.
The Histogram Viewer will also display gradient information. You can use the From Image and Perturb
modifiers to output gradients. If you need to see the gradient represented in a histogram, drag the
modifier’s title bar into the viewer.

Chapter 7 Using Viewers 180

The Histogram Viewer type for evaluating
the contrast and color cast in an image

3D Histogram
The more advanced 3D Histogram Viewer shows the color distribution in an image within a 3D cube.
One advantage to a 3D Histogram is that it can accurately represent the out-of-range colors commonly
found in floating-point and high-dynamic-range images. It can also be used to look at vector images
like position, normal, velocity, and so on.

The 3D Histogram Viewer type for

evaluating out-of-range colors

To rotate within a 3D Histogram, do one of the following:

– Hold down the Option key, and drag left or right using the middle mouse button.
– Hold down the middle and right mouse buttons while dragging.

Vectorscope
The Vectorscope Viewer duplicates the behavior of a specific type of video test equipment, displaying
a circular graph that helps to visualize the intensity of chrominance signals.

The Vectorscope Viewer type for evaluating

chrominance in an image

Chapter 7 Using Viewers 181

Waveform
The Waveform Viewer duplicates the behavior of a specific type of video test equipment, displaying a
line or bar graph that helps to visualize the voltage or luminance of a broadcast signal.

The Waveform Viewer type for evaluating

luminance in an image

Navigator
The Navigator can only be used in a subview. It provides a small overview of the entire image, with a
rectangle that indicates the portion of the image that is actually visible in the main viewer. This is useful
when zooming in on an image in the main view.

The Navigator subview for panning the image while zoomed in

Magnifier
The Magnifier can be used only in a subview. It shows a zoomed-in version of the pixels under the
cursor in the main viewer.

The Magnifier subview used to view a zoomed-in version of the image

Chapter 7 Using Viewers 182

Image Info
The Image Info view can only be used in a subview. The Image Info tab shows a horizontal bar across
the top of the image with information about the frame size, pixel aspect, and color depth of the
viewed image.

The Image Info subview for viewing size, pixel

aspect, and color depth information

Color Inspector
The Color Inspector can only be used in a subview. The Color Inspector shows information about the
color channels of the pixel under the cursor. It will show all channels present, even the auxiliary
channels such as Z buffer, XYZ normals, and UV mapping channels.

The Color Inspector subview for

evaluating colors in an image

Metadata
The content of this subview is based entirely on the amount of metadata in your image. Most Loaders
will give the color space and file path for the image. Much more information can be displayed if it exists
in the image.

The Metadata subview for viewing embedded metadata

Chapter 7 Using Viewers 183

Viewing Selective Channels
When compositing, you often deal with individual color components or channels in an image as much
as you deal with the full RGB color of the entire image. The viewers and subviews can display the
isolated color, alpha, depth channels, and even auxiliary channels that make up the image.

Viewing Color Channels

The default view is the full RGB color channel, but to change the channel that is displayed you can use
the Channel toolbar button, the viewer’s contextual menu, or keyboard shortcuts.

The Channel toolbar button for switching

between color, alpha, and depth channels

To toggle between RGB and alpha channels in the active viewer:

– Click the Color button in the viewer toolbar to toggle between full RGB color and that image’s
alpha channel.

To toggle the channel that’s displayed in the active viewer:

– Click the arrow to the right of the Color button to choose a specific channel to view from the list of
available channels in the current image.
– Click the viewer you want to toggle, and press one of the following keyboard shortcuts;
C - Full RGB color display
R - Display red channel
G - Display green channel
B - Display blue channel
A - Display alpha channel
Z - Display Z-buffer channel

Viewing Auxiliary Channels

The viewers support RGBA and Z channels using keyboard shortcuts, but they support other channels
as well. File formats such as OpenEXR often include auxiliary image data that provide more control
and compositing options when working with rendered 3D images. To view auxiliary image data in a
viewer, click the arrow to the right of the RGB button to display the drop-down menu or right-click in
the viewer and choose an option from the Channels submenu of the contextual menu.

Chapter 7 Using Viewers 184

The 3D Viewer
Building a composite in 3D space has different requirements from traditional 2D compositing. When a
node from the 3D category or some particle systems is selected, a 3D Viewer is used to display the
scene. The 3D Viewer shows a representation of a composite in a true GPU-accelerated
3D environment.
For more information on 3D controls, see Chapter 86, “3D Compositing Basics” in the DaVinci Resolve
Reference Manual or Chapter 25 in the Fusion Reference Manual.

Panning, Scaling, and Rotating a 3D Viewer

For the most part, navigation in the 3D Viewer is similar to the navigation in the 2D Viewer. Panning
and zooming work with the same controls even though you’re moving within a 3D space. However,
when viewing a 3D scene, panning changes the point of view and thus the center point for scaling and
rotation, too. A combination of panning and rotation will allow you to move the point of view anywhere
in the scene.
Another small change is that there’s a lower limit to the scale of a 3D scene. Continuing to zoom in past
this limit will instead move (“dolly“) the point of view forward. The mouse wheel will move forward
slowly, and the keyboard will move more quickly.
Critically, the 3D Viewer gives you additional control to rotate the viewer within the three dimensions
of the scene to better see your scene from different angles as you work.

To rotate within a 3D Viewer, do one of the following:

– Hold down the Option key and drag left or right using the middle mouse button.
– Hold down the middle and right mouse buttons while dragging.

The rotation is centered on the middle of the view.

TIP: These rotation controls can be used with the 3D Histogram subview as well.

Viewing Objects via Wireframe

3D composites not only work with 2D images on image planes but can also integrate true geometry,
such as that generated by the Particle system, Text 3D node, imported FBX meshes, and basic
primitives from the 3D toolset. Using a Wireframe view helps to see through a mesh or see the density
of the geometry. It is much easier to see a change in the Subdivision level of an ImagePlane3D in
wireframe than viewing the rendered image.

To display 3D geometry in wireframe, do the following:

– Right-click the 3D Viewer and choose 3D Options > Wireframe from the contextual menu.

Changing the POV of a 3D Viewer

Compositing a 3D scene often requires that you view the scene from different angles to make sure the
position and animation are what you want. While the 3D Viewer uses a perspective camera that allows
you to look at the 3D scene from any angle, you can also switch the 3D Viewer to view from the front,
top, left, or right side of the scene; these are collectively called Orthographic views.
Additionally, if you have a camera or spotlight in your scene, you can switch the viewer to face the
scene from the point of view of those objects.

Chapter 7 Using Viewers 185

To change the 3D viewpoints:
– Right-click the viewer and choose an option from the Camera submenu of the contextual menu.
The choices include Perspective, Front, Top, Left, and Right.

Changing Cameras in a 3D Viewer

If you have one or more camera objects in the 3D scene, they will be listed as options in the contextual
menu. Spotlights and other lights or objects in the scene will be listed in the Other submenu. If you
choose any one of these objects, the 3D Viewer jumps to display the scene from the point of view of
the chosen object. While looking “through“ an object, rotating, panning, or zooming, the viewer will
instead change the rotation, position, and scale of the camera, light, or other object.

Copying a Viewer’s POV to a Camera

There are many times you will have used the default perspective view to set up your scene and then
want that point of view (POV) to become your main camera’s point of view. You can easily position and
orient a camera, light, or other object to match the current point of view shown in the viewer using the
Copy PoV To command.

To copy the point of view in the viewer to a camera, do the following:

1 Set up a 3D Viewer with the point of view you want by zooming, panning, and rotating the viewer.
2 Add a camera to your 3D scene.
3 Right-click anywhere within the 3D Viewer and choose Camera > Copy PoV To >
Camera3DNameOfCamera from the contextual menu.

The Camera3D’s controls will inherit the viewer’s position and angle values.

TIP: The Copy PoV To command uses the object’s own coordinate space; any transformations
performed downstream by another node are not taken into account.

POV Labels
As you switch the POV of the viewer, you can keep track of which POV is currently displayed via a text
label at the bottom-left corner of the viewer. Right-clicking directly on this label, or on the axis control
above it, acts as a shortcut to the Camera submenu, allowing you to easily choose another viewpoint.

The Viewpoint label and Axis control in a 3D Viewer

Chapter 7 Using Viewers 186

Lighting and Shadows in 3D Viewers
Before you add lights to a 3D scene, default lighting is provided. This basic, flat lighting allows you to
see the shading on objects without requiring you to add and set up lights as you work in the 3D
Viewer. Additionally, shadows are hidden by default. Once you start adding lights of your own, you
need to switch modes to see what they affect as you work.

To see the effects of the default light on the scene:

– Right-click within the 3D Viewer and choose 3D Options > Default Lights from the
contextual menu.

When you’re ready to add your own lighting to a scene, you can connect light nodes in various ways to
a Merge 3D node for the scene you’re working on. Once you connect a light to a Merge 3D node, you
need to switch the 3D Viewer over to showing the new, proper lighting.

To toggle lighting rendering within a 3D scene:

– Right-click within the 3D Viewer and choose 3D Options > Lighting from the contextual menu.

A 3D scene using default lights (top), and the same scene with lighting turned on (bottom)

TIP: Attempting to load a Light node into a viewer all by itself will result in an empty
scene, with nothing illuminated. To see the effects of lights, you must view the Merge 3D
node the light is connected to.

Similar to lights, the default 3D Viewer has shadows turned off. To see shadows cast from the
lighting you’ve created, you must turn them on.

Chapter 7 Using Viewers 187

To toggle shadows rendering within a 3D scene:
– Right-click within the 3D Viewer and choose 3D Options > Shadows from the contextual menu.
Enabling shadows will automatically turn on lighting, if it is not already turned on.

A 3D scene with shadows enabled along with the lights

NOTE: The shadows shown in the 3D Viewer are always hard edged. Soft shadows are
available for output to the rest of your composition in the software renderer of the
Renderer3D node.

Transparency in 3D Viewers
Image planes and 3D objects are obscured by other objects in a scene depending on the X, Y, and Z
position coordinates of each object in 3D space. The default method used to determine which
polygons are hidden and which are shown based on these coordinates is called Z-buffering.
Z-buffering is extremely fast but not always accurate when dealing with multiple transparent layers in a
scene. Fortunately, there is another option for more complex 3D scenes with transparency: Sorted.
The Sorted method can be significantly slower in some scenes but will provide more accurate results
no matter how many layers of transparency happen to be in a scene.
The default behavior in the viewer is to use Z-buffering, but if your scene requires the Sorted method,
you can easily change this.

To choose a Sorted method of 3D compositing:

– Right-click anywhere within the 3D Viewer and choose one of the options in the Transparency
submenu of the contextual menu;
– Quick Sort: Reorders the polygons in the scene serially, from back to front, to produce a
reasonably accurate rendering of transparency.
– Full Sort: Renders every polygon in Z order to produce the most accurate rendering of
transparency.

Grid
The 3D Viewer displays a grid that’s used to provide a plane of reference in the 3D scene. By default,
the grid is 24 x 24 units in size, centered on the origin at (0,0,0), and subdivided into large squares of 2
units with small squares of 0.25 units each. These defaults can be altered in the 3D View panel of the
Fusion Settings window, available from the Fusion menu.

Chapter 7 Using Viewers 188

To toggle the grid on and off:
– Right-click anywhere within the 3D Viewer and choose 3D Options > Grid from the
contextual menu.

The default grid of the 3D Viewer grid with its origin at x = 0, y = 0 and z = 0

Vertex Normals
Normals indicate what direction each vertex of 3D geometry is facing, and they are used when
calculating lighting and texturing on an object. When viewing any kind of 3D geometry, including an
image plane or a full FBX mesh, you can display the normals for each object in a scene.

To view the normals in a scene:

– Right-click anywhere within the viewer and choose 3D Options > Vertex Normals from the
contextual menu.

The normals viewed in a 3D scene

Quad View
3D compositing often requires you to view the scene from different points of view to better control
transformations in three dimensions. While you can switch the 3D Viewer to different points of view,
doing so frequently can become cumbersome. Happily, you can instead enable a Quad view, which
divides the viewer into four panes. These panes can then display four different angles of the scene
at one time.

Chapter 7 Using Viewers 189

To toggle the display of the Quad view, do one of the following:
– Right-click anywhere within the viewer and choose Views > Quad View from the contextual menu.
– Press Shift-Q.

A Quad view of a 3D scene

While there are four panes in the Quad view, they all show the same scene. When assigning views
within a Quad view, you can choose between displaying Front, Left, Top, Bottom, and Perspective
orthographic views, or you can choose the view through any camera or spotlight that’s present in
the scene.

To assign different views to panes of a Quad view, do one of the following:

– Right-click directly on the POV label at the bottom left of the pane you want to reassign, and
choose another camera, light, or Point of View from the contextual menu.

Quad View Layouts

There are a variety of Quad view layouts, ranging from four equally sized panels to having three small
panels across the bottom of a larger single panel.

To switch to a different Quad view layout, do the following:

1 Enable the Quad view.
2 Right-click anywhere within the viewer and choose an option from the Views > Quad Layouts
submenu of the contextual menu.

Using Quad Views for 2D Scenes

Quad views aren’t only useful for 3D scenes. They can also be used with 2D scenes, with each pane
showing a different image channel or subview type. For example, one pane can show the image while
the other panes show the alpha channel, a vectorscope, and a histogram.

To assign different channels or subviews to panes of a Quad view for a 2D scene:

1 Right-click in a viewer and choose Views > Quad View.
2 Click once in the pane you want to reassign.
3 Do one of the following:
a. Choose a channel from the Channel Viewer menu.
b. Right-click in the viewer and choose Views, and then choose a Subview from the submenu.

Chapter 7 Using Viewers 190

Guides
Guides are onscreen overlays used to help you compose elements within a boundary or along the
center vertical and horizontal axes. While guides are displayed in the viewer, they’re not rendered into
the scene. There are four commonly used guides that can be displayed, including Monitor Safety, Safe
Title, Center, and Film.

Methods of using guides:

– To display guides in a viewer: Right-click in the viewer and then choose Guides > Show Guides
from the contextual menu, or press Command-G.
– To change the aspect ratio of the displayed guides: Right-click in the viewer and then choose
an option from the Guides > Frame Aspect submenu. The frame aspect is usually set to Default,
which forces the frame aspect to the same resolution as the image that’s displayed in the view.
However, when the frame aspect is set to a specific value, the guides will conform to the exact
boundaries of the specified format and any image area outside of that will be dark gray.
– To show or hide specific guides: Right-click in the viewer and then choose an option from the
Guides submenu. A variety of specific guides are provided, each of which can be individually
enabled and disabled.
– Monitor Safety: Monitor Safety indicates the safe action area viewable on most monitors
and TV screens.
– Safe Title: Safe Title indicates the safe area for titles viewable on all TV and monitor screens.
– Center: Center shows a crosshair for the center point and x- and y-axis of the view.
– Film: Some frame formats include film guides preset for you, whereas some will require
customization. The film guides can be customized in the Preferences > Frame Format window.

The Guides submenu in the viewer’s contextual menu

Frame Format Settings

In the Frame Format panel of the Fusion Settings window (available in the Fusion menu), there are two
film guide settings that you can use to customize these guides.
– Guide 1 contains four fields that specify the offset from the edges of the image for the left,
top, right, and bottom guides, in that order. As with all offsets in Fusion, this is a resolution-
independent number where 1 is the width of the full image and 0.5 is half the width of the image.
– Guide 2’s text box is used to set the aspect ratio of the projection area.

Chapter 7 Using Viewers 191

 The Frame Format Guides settings

Domain of Definition and Region of Interest

As a compositing environment, the Fusion page uses the standard compositing conventions of Region
of Interest (RoI) and Domain of Definition (DoD) to dramatically improve performance.

Domain of Definition (DoD)

In compositing, the Domain of Definition, frequently abbreviated to DoD, refers to a rectangular region
that defines what part of an image actually contains data. DoD makes the concept of an image’s actual
frame somewhat flexible, since rendering is no longer limited to the actual width and height of the
image. This has two effects on the way Fusion renders images.
Firstly, nodes will no longer be required to render portions of the image that will not be affected by the
node. This helps the renderer to optimize its performance. Secondly, Fusion can now keep track of
and apply a node’s effect to pixels that lie outside the visible portion of the image.
For example, consider the output of a Text+ node rendered against a transparent background. The
text occupies only a portion of the pixels in the image. Without Domain of Definition, you would be
required to process every pixel in the image needlessly. With a DoD, you are able to optimize effects
applied to the image, producing faster results and consuming less memory in the process.
The following image shows an image with the DoD outlined.

The DoD is shown as two XY coordinates indicating the corners of an axis-aligned bounding box (in pixels)

For the most part, the DoD is calculated automatically and without the need for manual intervention.
For example, all the nodes in the Generator category automatically generate the correct DoD. For
nodes like Fast Noise, Mandelbrot, and Background, this is usually the full dimensions of the image. In
the case of Text+ and virtually all of the Mask nodes, the DoD will often be much smaller or larger.
The OpenEXR format is capable of storing the data window of the image, and Fusion will apply this as
the DoD when loading such an image through a Loader node and will write out the DoD through the
Saver node.

Chapter 7 Using Viewers 192

When using the Fusion page in DaVinci Resolve, clips from the Edit page timeline or Media Pool will
typically have the DoD default to the full image width of the source media. The exception is media
stored in OpenEXR format.
The DoD is established as soon as the image is created or loaded into the composition. From there, it
passes downstream, where viewers combine it with their Region of Interest in order to determine
exactly what pixels should be affected by the node. As you work, different nodes will automatically
shrink, expand, or move the DoD as they apply their effect to an image, causing the DoD to change
from node to node.

Showing the DoD

If the current DoD for a node is different from the frame size of that image, it’s shown in the tooltip that
appears when the pointer hovers over a node in the Node Editor. The DoD is also visible in the viewer
when you right-click in a viewer and choose Region > Show DoD from the contextual menu.

Setting the DoD Manually in the Node Editor

It is also possible to set the DoD for an image manually using the Tools > Miscellaneous > Auto Domain
node in the Effects Library. This node can be useful when dealing with pre-created media that does
not occupy the full image dimensions. For example, a rendering of a 3D character that walks toward
the camera will frequently occupy only a portion of the image. The Auto Domain node can be used to
animate a DoD that covers the character and ignores the rest of the image, making image processing
more efficient.

Region of Interest (RoI)

The Region of Interest, frequently abbreviated to RoI, is a rectangular region similar to the Domain of
Definition. However, unlike the DoD, which tells the node what pixels are actually present in the image,
the RoI tells the node which pixels actually need to be rendered. When a node renders, it intersects
the current RoI with the current DoD to determine what pixels should be affected.

Enabling RoI Controls

You can turn on the RoI controls to restrict rendering to a small region of the image to significantly
improve performance when you’re only working on a small part of a high-resolution or complex
composition. For example, if you’re using paint to clean up some holes in a matte on the floor of a
composition with many, many high-resolution layers, 3D, and Lighting operations, you can use the RoI
controls to isolate the part of the floor you’re working on, which makes caching that part of the
composition much faster.

To enable the RoI controls, do one of the following:

– Click the RoI button in the 2D Viewer toolbar.
– Right-click in a viewer and choose Region > Show DoD from the contextual menu.
When RoI is enabled and Show is selected from the menu, a rectangular RoI control appears in the
viewer. If this is the first time RoI has been enabled, it will be set to the full width and height of the
image. Otherwise, the last known position of the RoI for that view is used. However, if you want to set
the RoI to a custom area within the frame, you can do one of the following.

To adjust the RoI controls, do one of the following:

– Drag any edge of the RoI rectangle to adjust one side of the RoI.
– Drag a corner to adjust the size of the RoI rectangle from that corner.
– Drag the small circle found at the top left corner of the RoI rectangle to move the RoI without
adjusting its dimensions.
Sometimes, it’s faster to simply draw a rectangle where you want the RoI to be.

Chapter 7 Using Viewers 193

To quickly draw the RoI at the desired size:
1 Choose Set from the viewer menu next to the RoI button, or right-click anywhere within the viewer
and choose Region > Set Region.
2 When the pointer turns into an RoI drawing cursor, drag within the viewer to set a RoI rectangle.
Alternatively, an Auto command sets the RoI to fit whichever pixels are visible at the current zoom/pan
level in the viewer. This lets you quickly limit the RoI to whatever part of the composition you’ve
zoomed into.

To automatically draw the RoI:

– Choose Auto from the viewer menu next to the RoI button.
– Right-click anywhere within the viewer and choose Region > Auto Region.
When you no longer need to use the RoI, you can reset it.

To reset the RoI to the full width and height of the current image, do one of the following:
– Choose Reset from the viewer menu next to the RoI button.
– Right-click anywhere within the viewer and choose Region > Reset Region from the contextual
menu or from the toolbar button menu.
– Disable the ROI control, which will also reset it.

While the RoI Is Active

The RoI is only used for previewing your composition while you work, not for output from Fusion. While
the RoI is active, Fusion will only request rendering of the pixels inside the region when it displays an
image in that viewer. Flipbook Previews that you create in that viewer will also respect the current RoI.
MediaOut and Saver nodes will always use the full image dimensions when writing the image to disk,
ignoring any RoI you’ve set in the viewers.
The RoI improves not only rendering speed and memory use, but it can also reduce file I/O, since
Loaders and MediaIn nodes only load pixels from within the RoI, if one is specified. This does require
that the file format used supports direct pixel access. Cineon, DPX, and many uncompressed file
formats support this feature, as well as OpenEXR and TIFF in limited cases.
Please note that changes to the viewed image size or color depth will cause the pixels outside the RoI
to be reset to the image’s canvas color. This also happens when switching in and out of Proxy mode,
as well as during Proxy mode switching with Auto Proxy enabled. When the image size is maintained,
so are the last rendered pixel values outside the RoI. This can be useful for comparing changes made
within the RoI with a previous node state.

TIP: Right-clicking in a viewer and choosing Options > Show Controls for showing onscreen
controls will override the RoI, forcing renders of pixels for the entire image.

Managing Viewer Lookup Tables (LUTs)

Lookup Tables, or LUTs, can be used to help match the appearance of a viewer to its eventual output
destination. They’re essentially image-processing operations that affect only the image being
previewed in the viewer, not the image data itself. There are two basic ways that LUTs can calculate
color transformations: The first is a simple 1D LUT, and the second is a more sophisticated 3D LUT.

Chapter 7 Using Viewers 194

 – The simplest form of a LUT is a 1D LUT. It accounts for one color channel at a time, so it can make
overall tonality changes but not very specific color changes.
– A 3D LUT looks at each possible color value (red, green, and blue) independently. A 3D LUT
allows for large global changes as well as very specific color changes to be applied to images
very quickly.

How Lookup Tables Work in Fusion

A Lookup Table (LUT) is a table of values used to transform the color and luminance of an image. A 1D
LUT uses a two-column table for input color and output color, while a 3D LUT uses more of a matrix. A
LUT is used primarily to correct for variances in the monitor or the source color space of the image.
You can choose to apply a LUT to all the viewers or apply different LUTs to each viewer.

Image LUTs
Image LUTs can be applied to each viewer. In fact, you can even apply separate Image LUTs for the A
and B buffers of a single viewer. These LUTs can only be applied to 2D images and not to 3D scenes.
Image LUTs are routinely used to get from one scene-referred color space to another. For example, if
you’re working with log-encoded media but want to see how the image will look in the final color
space, you can choose a LUT to make the image transform as a preview.

Buffer LUTs
The Buffer LUT is applied to the viewers regardless of contents, including 3D scenes, 3D materials,
and subview types. Only one Buffer LUT can be applied. If a 2D image is being displayed with an
Image LUT applied, then the Buffer LUT is applied to the result of the image LUT. Buffer LUTs are
typically used to simulate another output color space that’s unique to the display you’re using—for
instance, making a DCI-P3 projector show the image as it would look on an sRGB monitor.

To use a Buffer LUT:

1 Disable the LUT button above the viewer.
2 Right-click in the viewer and choose Global Options > Buffer LUT > Enable.
3 Right-click in the viewer and choose Global Options > Buffer LUT > Type of LUT you want to apply.

When dealing with nonlinear files from many of today’s digital cinema cameras, a modern workflow
would be to convert everything to linear at the beginning of the node tree, then create your composite,
and then apply an Image LUT or Buffer LUT that matches the color space you want it to be in for either
grading in the Color page or for final output.
However, in more elaborate production pipelines, you may need to apply multiple LUTs consecutively.

Types of Viewer LUTs

Aside from the industry standard 1D and 3D LUTs, other types of LUTs are supported, including
script-based Fuse node LUTs and macros assembled from standard nodes. Generally, LUT processing
is performed on the graphics card’s GPU in real time, although the performance of macro-based LUTs
is based on the nodes they contain.

Fusion View LUT

The Fusion View LUT is the default and is a frequently used LUT type. It provides an RGBA curve that
can be used to assign IN/OUT value pairs. This control is identical to that provided by the Color
Curve node.
Since the purpose of the View LUT is to provide an unchanging correction for the monitor or the file’s
color space, however, these splines cannot be animated.

Chapter 7 Using Viewers 195

Log-Lin View LUT
The Log-Lin LUT converts logarithmic data to linear, and vice versa. This can be particularly useful
when used in conjunction with supplied LUT files that expect logarithmic data. It is similar to the
Cineon Log node.

Gamut View LUT

The Gamut LUT converts a source color space to an output color space, with options to deal with
gamma settings, alpha channels, and premultiplication. The Gamut LUT is a frequently used LUT type
to correct the viewer when working with Linear Gamma in the Node editor.

Macro LUTs
Any macro node can also be used as a viewer LUT simply by saving the macro’s .setting file to the
correct Fusion directory.

In DaVinci Resolve, LUTs are saved in the following locations:

– On macOS: Macintosh HD/Users/username/Library/Application Support/Blackmagic Design/
Fusion/LUTs/
– On Windows: C:\Program Files\Blackmagic Design\Fusion\LUTs
– On Linux: home/username/.local/share/DaVinciResolve/Fusion/LUTs

In Fusion Studio, LUTs are saved in the following locations:

– On macOS: Macintosh HD/Users/username/Library/Application Support/Blackmagic Design/
Fusion/LUTs/
– On Windows: C:\Users\username\AppData\Roaming\Blackmagic Design\Fusion\LUTs
– On Linux: home/username/.fusion/BlackmagicDesign/Fusion/LUTs

For this to work, the macro must have one image input and one image output. Any controls exposed
on the macro will be available when the Edit option is selected for the LUT. For more information about
creating macros, see Chapter 67, “Node Groups, Macros, and Fusion Templates” in the
DaVinci Resolve Reference Manual or Chapter 6 in the Fusion Reference Manual.

LUT Presets
All LUTs available to DaVinci Resolve are also accessible to the Fusion page, which includes custom
LUTs you’ve installed, as well as preset LUTs that come installed with DaVinci Resolve, such as the
highly useful VFX IO category that includes a wide variety of miscellaneous to Linear and Linear to
miscellaneous transforms. All of these LUTs appear by category in the viewer LUT menu.

Fuse LUTs
Fuses are scriptable plug-ins that are installed with the application or that you create in Fusion. A fuse
named CT_ViewLUTPlugin can be applied as a LUT to a viewer. You can also script fuses that use
graphics hardware shaders embedded into the LUT for real-time processing. Since fuse LUTs require
shader-capable graphics hardware, they cannot be applied in software. For more information about
Fuses, see the Fusion Scripting Guide located on the Blackmagic Design website.

Using Viewer LUTs

Viewer LUTs can be enabled, edited, and turned off using the viewer LUT button and menu, as well as
by using the viewer contextual menu. This menu shows all LUTs available to Fusion, including custom
LUTs you’ve installed yourself.

Chapter 7 Using Viewers 196

The viewer LUT button and menu from
the Fusion page in DaVinci Resolve

To turn the current viewer LUT on and off:

– Click the LUT button in the viewer toolbar to toggle the viewer LUT on and off.
– The LUT menu can also be found as a submenu in the viewer’s contextual menu.

To choose another viewer LUT:

– Open the menu to the right of the viewer LUT button and choose an option from the
viewer LUT menu.

To apply a Buffer LUT:

1 Right-click anywhere within the viewer and choose Global Options > Buffer LUT > Enable.
2 To choose a specific Buffer LUT, right-click again and choose a LUT from the Global Options >
Buffer LUT submenu.

Buffer LUTs are often useful for applying monitor corrections, which do not usually change
between projects.

To remove a Buffer LUT:

– Right-click anywhere within a viewer and choose Global Options > Buffer LUT >
Enable to uncheck it.

Chapter 7 Using Viewers 197

Editing Viewer LUTs
The viewers are the primary area where composites are assessed, so it’s crucial that they provide an
accurate representation of what the content will look like when it’s played for an audience. The LUT
Editor allows you to customize your viewer’s output to match the gamma and color characteristics of
your eventual playback device, or to test how the current image looks in a completely different color
space, or how it holds up over a range of different color spaces.

To open any editable viewer LUT option’s Editor:

1 Click the LUT button in the viewer toolbar to enable it.
2 Do one of the following:
– Choose Edit from the bottom of the viewer LUT menu.
– Right-click in the viewer and then choose LUT > Edit from the contextual menu.

Editing the Fusion View Lookup Table

Similarly to the Color Curves node, the Fusion View LUT Editor uses spline-based color correction. In
addition to the ability to modify the separate color channels, the LUT has Gain and Gamma sliders. The
Gain slider is helpful for temporarily brightening or darkening the viewed image, allowing easier
examination of shadow or highlight detail. The Color Gamma and Alpha Gamma sliders are used to
duplicate the gamma values of the eventual output device. Video monitors, for example, commonly
have a gamma of 1.7, while computer monitors can range anywhere from 1.6 to 2.2. Alpha Gamma is
applied only when viewing the alpha channel of an image, or when viewing masks.

The LUT Editor for the default Fusion View LUT

Editing the Gamut View LUT

The Gamut View LUT Editor lets you choose a Source and Output color space to guide the
viewer transform.
The Remove and Add Gamma checkboxes let you choose to do the gamut conversion with linear or
nonlinear gamma, or they let you simply remove or add the appropriate gamma values without
changing the color space.
Selecting the Pre-Divide/Post-Multiply checkbox will cause the image’s pixel values to be divided by
the alpha values prior to this conversion, and then re-multiplied by the alpha value after this
conversion. This helps to avoid the creation of illegally additive images, particularly around the edges
of a blue/green key or when working with 3D rendered objects.

Chapter 7 Using Viewers 198

The Gamut View LUT Editor

Editing the Log-Lin LUT

The Log-Lin LUT lets you apply a Log to Lin or Lin to Log operation using the Mode pop-up menu. You
can choose the type of log-encoding to process from the Log Type drop-down, and choose whether
to lock the R, G, and B channels together. A level adjustment lets you redefine the digital range of
values used for the output, while Soft Clip (Knee), Film Stock Gamma, and Conversion Gamma sliders
let you further customize the color transform. Lastly, a Conversion Table field and Browse button let
you add an additional LUT as part of this operation.

The Log-Lin LUT Editor

LUT Processing Order

In elaborate workflows, facilities may apply multiple LUTs in a row before the image is seen. The order
of these is important since each LUT delivers different outputs. For instance, for a Cineon file in Log
color space you may often apply three LUTs. First a Log to Lin conversion, followed by a Fusion View
LUT to apply a color calibration, and a third one to correct it for display on an sRGB monitor, or replace
the last with a 3D DCP LUT if you are viewing on a projector.
When you select a node to be displayed, the image produced is processed before it is shown in the
viewers. The processing order is slightly different for 2D images and 3D scenes.

Chapter 7 Using Viewers 199

2D images first have the image LUT applied, and the result is composited over the checker underlay.
3D scenes are instead rendered with OpenGL.

Image Image LUTs Post Processing Buffer LUT Viewer

Output Macro LUTs Checker underlay (single) Output
Fusion View LUT Dithering

Order of processing

The order of processing for 2D images and 3D scenes

For either 2D or 3D, the result may be drawn to an offscreen buffer where a Buffer LUT can be applied,
along with dithering, a full view checker underlay, and any stereo processing. The final result is then
drawn to the viewer and any onscreen controls are drawn on top.

Applying Multiple LUTs

The viewer contextual menu can be used to apply multiple image LUTs into a processing chain.

To apply an additional LUT, do the following:

1 Right-click anywhere within the viewer.
2 From the viewer’s contextual menu, choose LUT > Add New.
3 From the Add New submenu, choose a LUT to add.

To remove a LUT other than the first LUT, do the following:

1 Right-click anywhere within the viewer.
2 From the viewer’s contextual menu, choose LUT > Delete.
3 From the Delete submenu, choose a LUT to remove.

A complete stacked LUT configuration can be saved to and loaded from a .viewlut file, as
described below.

Saving Custom LUTs

There are a variety of ways to create and use different viewer LUTs in Fusion. You can save LUTs when
you save viewer settings, you can import LUTs that have been exported from Fusion or other
applications, and you can open any one of the various supported LUT file types. In addition, you can
use the standard nodes in Fusion to create macros, which can then be saved and used as a LUT.

LUT Settings
The most straightforward way to save a LUT you have created using the Fusion View LUT Editor is to
use the LUT > Save menu found in the viewer contextual menu. The settings are saved as an ASCII file
with the extension .viewlut in the LUTs folder. Any files with this extension found in that folder will
appear in the Image LUT menus for ease of loading. You can also load the settings that are not found
in the menu by choosing LUT > Load from the viewer’s contextual menu.

Using Viewer Settings

If you’ve modified a LUT, choosing Settings > Save New from the viewer’s contextual menu will save all
the viewer’s settings, including all LUT curves and gain/gamma values. You can save these under
different names, and each settings file can be reloaded at any time by choosing Settings > filename
from the viewer’s contextual menu. Choosing Save Default from the same menu will make these
settings the standard for all new comps.

Chapter 7 Using Viewers 200

Using LUT Curves
The Viewer LUT Edit dialog can be used to import and export LUT curves. You can export the LUT
curves as either ASCII or Saved format. The ASCII (.alut) file format is useful for sharing LUT curves with
other software, whereas the Saved (.lut) file format is preferred for Fusion, as it is more compact,
accurate, and allows further editing.

To export a LUT, do the following:

1 Click the viewer LUT button to enable it.
2 Click the viewer LUT menu, and then choose Edit.
3 Right-click on the LUT Curve Editor, and then choose Export LUT.
4 Select a LUT format at the bottom of the file browser window.
5 Enter a name for the LUT and click Save.

The Import LUT option will load LUT files back into the Curve Editor, or alternatively, if the file has been
saved in Fusion’s LUTs folder, it will appear in the LUT drop-down menu list.

TIP: This is one way to move LUTs between viewers or to and from the Color Curves node or
any other LUT Editor in Fusion.

LUT Files
Any supported LUT files in the LUTs folder can be used by choosing them either from the LUT drop-
down menu or the viewer’s contextual menu. This includes 1D and 3D LUTs such as Fusion’s .lut, .alut
and .alut3 formats, as well as .cube, .shlut, .look, .3dl, and .itx formats. This is a convenient way to
access standard format LUT files for different projects.

Settings and Macros

Since LUTs are a form of color correction, you can also use any node, macro, or group of nodes as a
viewer LUT.

To use a node, group, or macro as a viewer LUT, do the following:

1 Select the node, group, or macro.
2 Right-click over the selected node, and then choose Settings > Save As from the menu.
3 In the file browser, go to the LUTs folder as set in Preferences > Global > Path Map > LUTS.
4 Click Save to save the .settings file.

This allows almost any combination of nodes to be used as a viewer LUT. This is the most flexible
approach but is also potentially the slowest. The LUT nodes must be rendered solely on the CPU,
whereas other methods are GPU-accelerated.

Setting a Default LUT

The default LUT applied when a new composition is created can be assigned in the Viewer panel of
the Fusion Settings window. Clicking the Enable Display LUT checkbox allows you to select a LUT from
the Display LUT plug-ins list.

Chapter 7 Using Viewers 201

The LUT default settings found in the View panel of the Fusion Settings window

Viewer Preferences and Settings

The default settings for each viewer can be changed using the Viewer panel in the Preferences. The
position and size of each floating viewer can also be saved using the Layout menu in the Preferences.

Viewer Settings
It is often preferable to switch between entirely different viewer configurations while working. For
example, while keying, the image may be in the main viewer, and the alpha channel may be in a
subview. Viewer settings toward the end of a project may consist of the histogram, vectorscope, and
waveform, as well as the image in a view set to Quad view.
Fusion provides the ability to quickly load and save viewer settings to help reduce the amount of effort
required to change from one configuration to another.

To save a viewer setting, do the following:

1 Right-click over the viewer you want to save.
2 From the contextual menu, choose Setting > Save New.
3 Enter a name for the settings and click Save.

To load a viewer setting, do the following:

1 Right-click over the viewer you want to load a setting into.
2 From the contextual menu, choose Settings > filename.

Loading and Saving Defaults for a Viewer

The viewer can save new defaults and be returned to its defaults using the Load Defaults and the
Save Defaults options in the Settings portion of the View contextual menu.

Chapter 7 Using Viewers 202

The Viewer Options Menu
The Options menu of the viewer contains several ways you can customize the look and behavior of
the viewer. Many of these options are also in the viewer contextual menu.

Show Controls
When onscreen controls are not necessary or are getting in the way of evaluating the image, you can
temporarily hide them using the Show Controls option. This option is toggled using Command-K.

Checker Underlay
The Checker Underlay shows a checkerboard beneath transparent pixels to make it easier to identify
transparent areas. This is the default option for 2D viewers. Disabling this option replaces the
checkerboard with black.

Pixel Grid
Enabling this option will show a light black grid that outlines the exact boundaries of pixels in the
image when the image is scaled past a certain threshold. The default is Off.

Smooth Resize
The Smooth Resize option uses a smoother bilinear interpolated resizing method when zooming into
an image in the viewer. When Smooth Resize is disabled, scaling uses the nearest neighbor method
and shows noticeable aliasing artifacts but is more useful for seeing the actual pixels of the viewed
image when you zoom all the way down to a pixel level since there is no interpolation. This option is
enabled by default and can be toggled by clicking on the SmR button in the viewer toolbar.

Show Square Pixels

Depending on the frame format preferences and the type of footage loaded, many images may have
pixels that are rectangular instead of square. Both the NTSC and PAL video standards, as well as some
anamorphic film formats, use rectangular pixels. A computer monitor uses perfectly square pixels. To
compensate for this, aspect correction is automatically performed when viewing non-square pixels.
This prevents non-square pixel images from appearing squashed or stretched in the viewer.
You can enable the Show Square Pixels option to override the aspect correction. Show Square Pixels
can also be toggled on and off using the 1:1 button in the viewer toolbar.

Gain/Gamma
Exposes or hides a simple pair of Gain and Gamma sliders that let you adjust the viewed image.
Especially useful for “gamma slamming” a composite to see how well it holds up with a variety of
gamma settings. Defaults to no change.

360º View
Sets the Fusion page viewer to properly display spherical imagery in a variety of formats, selectable
from this submenu. Disable toggles 360 viewing on or off, while Auto, LatLong, Vert Cross, Horiz
Cross, Vert Strip, and Horiz Strip let you properly display different formats of 360º video.

Locking the Viewer (Command-L)

You can lock a viewer to prevent it from updating. The node that’s loaded into that viewer still
processes and the new image is queued for display in the viewer, but until you unlock it, the viewer
does not update. By default, the viewer is unlocked.

Chapter 7 Using Viewers 203

Additional Viewer Options
There are additional commands when you right-click anywhere within a viewer and choose from the
generically named Options submenu.

Alpha Overlay
When you enable the alpha overlay, the viewer will show the alpha channel overlaid on top of the color
channels. This can be helpful when trying to see where one image stops and another begins in a
composite. This option is disabled by default.

Overlay Color
When you turn the alpha overlay on, the default color is to show white for the area the alpha covers.
There are times when white does not show clearly enough, depending on the colors in the image.
You can change the color by choosing a color from the list of Overlay Color options.

Follow Active
Enabling the Follow Active option will cause the viewer to always display the currently active node in
the Node Editor. This option is disabled by default, so you can view a different node than what you
control in the Control Panel.

Show Controls
When onscreen controls are not necessary or are getting in the way of evaluating the image, you can
temporarily hide them using the Show Controls option. This option is toggled using Command-K.

Show Full Color Range

When working with floating-point images, you will occasionally need to visualize the values that fall
outside the normal luminance range. Enabling the Show Full Color Range option using the toolbar
button automatically normalize any image displayed in the viewer. Normalization causes the brightest
pixel in a color channel to be mapped to a value of 1.0 (white) and the darkest pixel to be mapped to a
value of 0.0 (black). Midrange values are scaled appropriately to fit within that range. It is also useful
when viewing Z-buffer or other auxiliary channels, which often use value ranges far different from
those in the color channels.

Show Labels
The Show Labels option lets you toggle the display of the text that sometimes accompanies onscreen
controls in the viewer without disabling the functions that are showing those overlays, and without
hiding the onscreen controls themselves.

Status Bar Information

The status bar at the bottom of the Fusion window provides the exact RGBA and Z values for the pixel
beneath the pointer when it’s hovering within one of the viewers. Additional information about the X
and Y coordinates of the cursor and the exact pixel position are also displayed.

The status bar showing coordinates and color information

Chapter 7 Using Viewers 204

Chapter 8

Editing Parameters
in the Inspector
The Inspector is where you adjust the parameters of each node to do what needs to
be done. This chapter covers the various node parameters and methods for working
with the available controls.

Contents
Overview of the Inspector  206
The Tools and Modifiers Panels  207
Customizing the Inspector  207
Inspector Height  207
Inspector Display Preferences  208
Opening Nodes in the Inspector  208
Pinning Multiple Nodes in the Inspector  209
Hiding Inspector Controls  210
Using the Inspector Header  210
Selecting and Viewing Nodes in the Inspector  211
Using Header Controls  211
Versioning Nodes  212
Parameter Tabs  212
The Settings Tab  212
Inspector Controls Explained  216
Fusion Slider Controls  216
Thumbwheel  217
Range Controls  217
Checkboxes  217

Chapter 8 Editing Parameters in the Inspector 205

Drop-Down Menus  218
Button Arrays  218
Color Chooser and Picker  218
Gradients  220
Modifiers  222
Animating Parameters in the Inspector  222
Removing Animation From a Parameter  223
Attaching a Parameter to an Existing Animation Curve  223
Connecting Parameters  224
Connecting Parameters by Publishing  224
Connecting Parameters by Pick Whipping  224
Contextual Menus  225
Customizing Node Parameters with User Controls  225
An Example of Customizing Directional Blur  226

Overview of the Inspector

While the creation and connection of nodes in the Node Editor determines the tools and order of
operations that make up a composition, the Inspector (previously called the Control Panel) is where
you adjust the various parameters inside each node to do what needs to be done.

Inspector displays the Brightness Contrast controls

Chapter 8 Editing Parameters in the Inspector 206

This chapter covers methods for opening node parameters in the Inspector to edit them in different
ways according to the type of available controls.

To display the Inspector:

– Click the Inspector button on the UI toolbar.

The Tools and Modifiers Panels

The Inspector is divided into two overall panels.
– The Tools panel is where the parameters of selected nodes appear so you can edit them.
– The Modifiers panel is where you edit optional extensions to the tool’s standard toolset as well as
automated expressions that you can attach to individual parameters to create animated effects.
Additionally, certain nodes such as the Paint node generate data such as Strokes, which are saved
in the Modifiers panel.

Modifiers displayed in the Modifiers panel

Customizing the Inspector

You can customize how the Inspector is presented in a variety of ways.

Inspector Height
A small arrow button at the far right of the UI toolbar lets you toggle the Inspector between full-height
and half-height views, depending on how much room you need for editing parameters.

The Maximize button on the left side of the Inspector

In maximized height mode, the Inspector takes up up the entire right side of the UI, letting you see
every control that a node has available, or creating enough room to see the parameters of two or three
pinned nodes all at once. In half-height mode, the top of the Inspector is aligned with the tops of the
viewers, expanding the horizontal space that’s available for the Node Editor.

Chapter 8 Editing Parameters in the Inspector 207

Inspector Display Preferences
By default, you see only selected nodes in the Inspector, and only the Active node is expanded to
show its controls. You can change this behavior by choosing Fusion > Fusion Settings in the Fusion
page or File > Preferences in Fusion Studio and opening the User Interface panel. In the User
Interface, checkboxes manage the display of controls.

Control preferences in the User Interface category

– Auto Control Open: When enabled (the default), whichever node is active automatically opens its
controls in the Inspector. When disabled, selecting an active node opens that node’s Inspector
header in the Inspector, but the parameters remain hidden unless you click the Inspector header.
– Auto Control Hide: When enabled (the default), only selected nodes are visible in the Inspector,
and all deselected nodes are automatically removed from the Inspector to reduce clutter. When
disabled, parameters from selected nodes remain in the Inspector, even when those nodes are
deselected, so that the Inspector accumulates the parameters of every node you select over time.
– Auto Control Close Tools: When enabled (the default), only the parameters for the active node
can be exposed. When disabled, you can open the parameters of multiple nodes in the Inspector
if you want.
– Auto Controls for Selected: When enabled (the default), selecting multiple nodes opens multiple
control headers for those nodes in the Inspector. When disabled, only the active node appears in
the Inspector; multi-selected nodes highlighted in white do not appear.

Opening Nodes in the Inspector

Before you can edit a node’s parameters, you need to open it in the Inspector.

To display a node’s controls in the Inspector:

– Select one or more nodes from the Node Editor, Keyframes Editor, or Spline Editor.

When you select a single node so that it’s highlighted orange in the Node Editor, all of its parameters
appear in the Inspector. If you select multiple nodes at once, Inspector headers appear for each
selected node (highlighted in white in the Node Editor), but the parameters for the active node
(highlighted in orange) are exposed for editing.

Chapter 8 Editing Parameters in the Inspector 208

 Opening multiple nodes in the Inspector

Only one node’s parameters can be edited at a time, so clicking another node’s Inspector header
opens that node’s parameters and closes the parameters of the previous node you were working on.
This also makes the newly opened node the active node, highlighting it orange in the Inspector.

Pinning Multiple Nodes in the Inspector

For instances where you need to work quickly by editing the parameters of multiple nodes at the same
time, you can use the Pin button in the Inspector header of nodes in the Inspector to keep those
parameters exposed in the Inspector, regardless of whether that node is selected and active.

The Pin button of a node’s Inspector

header in the Inspector

While the Pin button is on, that node’s parameters remain open in the Inspector. If you select another
node in the Node Editor, that node’s parameters appear beneath any pinned nodes.

Chapter 8 Editing Parameters in the Inspector 209

 A pinned node on the bottom, with a selected node at the top

You can have as many pinned nodes in the Inspector as you like, but the more you have, the more
likely you’ll need to scroll up or down in the Inspector to get to all the parameters you want to edit.
To remove a pinned node from the Inspector, just turn off its Pin button in the Inspector header.

Hiding Inspector Controls

If you like, Inspector parameters for specific nodes can be hidden so they never appear, even when
that node is selected. This can be useful for preventing accidental changes by you or other
compositors who may be working on a composition in situations where you don’t want to
lock the node.

To Toggle the Inspector controls for a node or or off:

– Right-click on the node in the Node Editor, or on the Inspector header, and choose Modes > Show
Controls from the contextual menu.

Using the Inspector Header

When you select a node, it populates the Inspector with a title bar, or Inspector header, that displays
that node’s name as well as other controls that govern that node. A node’s Inspector header itself has
a variety of controls, but clicking (or double-clicking) on an Inspector header also exposes that node’s
parameters.

A node’s Inspector header

Chapter 8 Editing Parameters in the Inspector 210

When you select multiple nodes at once, you’ll see multiple headers in the Inspector. By default, only
the parameters for the active node (highlighted orange in the Node Editor) can be opened at any
given time, although you can change this behavior in Fusion’s Preferences.

Selecting and Viewing Nodes in the Inspector

Inspector headers are click targets for selecting nodes, opening and closing node parameters, and
other things.

Methods of using headers:

– To select a node using the Inspector header: When multiple nodes are selected, you can make
a node the active node by clicking its Inspector header in the Inspector. As the actively selected
node, the Inspector header and the corresponding node in the Node Editor are highlighted
orange, and its parameters are exposed.
– To load a node into the viewer using the Inspector header: You can view a node by dragging its
header into one of the viewers.
– To view a node’s splines with the control header: If you want to view the animated curves of a
node in the Spline Editor, you can add them by dragging the Inspector header into the Spline
Editor. All animated splines for the parameters of that node will automatically be displayed.

Using Header Controls

The controls found in each node’s Inspector header makes it fast to do simple things.
– To turn nodes off and on: Each Inspector header has a toggle switch to the left of its name, which
can be used to enable or disable that node. Disabled nodes pass image data from the previous
upstream node to the next downstream node without alteration.
– To change the Inspector header name: The name of the node corresponding to that Inspector
header is displayed next. You can change the name by right-clicking the Inspector header to
expose contextual menu commands similar to those found when you right-click a node in the
Node Editor and choosing Rename. Alternatively, you can click an Inspector header and press
F2 to edit its name. A Rename dialog appears, where you can enter a new name and click OK (or
press Return).
– To color-code nodes: A color pop-up menu lets you color code with one of 16 colors. Choose
Clear Color if you want to return that node to the default color.
– To version nodes: Turning on the Versions button displays a Version bar with six buttons.
Versioning is described in the following section.
– To pin Inspector controls: Clicking the Pin button “pins” that node’s parameters in the Inspector so
they remain in place, even if you deselect that node. You can have as many pinned nodes as you
like in the Inspector, but the more you have, the more likely you’ll be scrolling up and down the
Inspector to navigate all the available parameters.
– To lock nodes: Clicking the Lock button locks that node so no changes can be made to it.
– To reset Inspector controls: The rightmost button in the Inspector header is a Reset button that
resets the entire node to the default settings for that node.

Chapter 8 Editing Parameters in the Inspector 211

Versioning Nodes
Each button is capable of containing separate parameter settings for that node, making it easy to save
and compare up to six different versions of settings for each node. All versions are saved along with
the node in the Node Editor for future use.

The Version bar, underneath an Inspector

header, with versions enabled

An orange underline indicates the currently selected version, which is the version that’s currently
being used by your composition. To clear a version you don’t want to use any more, right-click that
version number and choose Clear from the contextual menu.

Parameter Tabs
Underneath the Inspector header is a series of panel tabs, displayed as thematic icons. Clicking one of
these icons opens a separate tab of parameters, which are usually grouped by function. Simple nodes,
such as the Blur node, consist of two tabs where the first contains all of the parameters relating to
blurring the image, and the second is the Settings tab.

The parameter tabs of the Blur node

More complicated nodes have more tabs containing more groups of parameters. For example, the
Delta Keyer has seven tabs: separating Key, Pre-Matte, Matte, Fringe, Tuning, and Mask parameters,
along with the obligatory Settings tab. These tabs keep the Delta Keyer from being a giant scrolling list
of settings and make it easy to keep track of which part of the keying process you’re finessing
as you work.

The parameter tabs of the Delta Keyer node

The Settings Tab

Every node that comes with Fusion has a Settings tab. This tab includes a set of standard controls that
appear for nearly every node, although some nodes have special Settings tab controls that
others lack.

Chapter 8 Editing Parameters in the Inspector 212

 The Settings tab in the Inspector

The following controls are common to most nodes, although some are node-specific. For example,
Motion Blur settings have no purpose in a Color Space node.

Blend
The Blend control is found in all nodes, except the Loader, MediaIn, and Generator nodes. It is used to
blend between the node’s unaltered image input and the node’s final processed output. When the
blend value is 0.0, the outgoing image is identical to the incoming image. Ordinarily, this will cause the
node to skip processing entirely, copying the input straight to the output. The default for this node
is 1.0, meaning the node will output the modified image 100%.

Process When Blend is 0.0

This checkbox forces the node to process even when the input value is zero and the image output is
identical to the image input. This can be useful on certain nodes or third-party plug-ins that store
values from one frame to the next. If this checkbox is disabled on nodes that operate in this manner,
the node will skip being processed when the Blend is set to 0, producing incorrect results on
subsequent frames.

Red/Green/Blue/Alpha Channel Checkboxes

Most nodes have a set of RGBA checkboxes in the Settings tab. These checkboxes let you exclude
any combination of these channels from being affected by that node.

The channel limiting checkboxes in the

Settings panel of a Transform node set so
that only the green channel is affected

Chapter 8 Editing Parameters in the Inspector 213

For example, if you wanted to use the Transform node to affect only the green channel of an image,
you can turn off the Red, Blue, and Alpha checkboxes. As a result, the green channel is processed by
this operation, and the red, blue, and alpha channels are copied straight from the node’s input to the
node’s output, skipping that node’s processing to remain unaffected.

Transforming only the green color channel

of the image with a Transform effect

Skipping Channel Processing

Under the hood, most nodes actually process all channels first, but afterward copy the input image to
the output for channels that have been unchecked. Modern workstations are so fast that this isn’t
usually noticeable, but there are some nodes where deselecting a channel actually causes that node
to skip processing that channel entirely. Nodes that operate this way have a linked set of Red, Green,
Blue, and Alpha checkboxes on another tab in the node.

Channel checkboxes on the Controls tab of the

Blur node indicate that disabled channels won’t
be processed at all, to save rendering time.

In these cases, the Common Control channel checkboxes are instanced to the channel boxes found
elsewhere in the node. Blur, Brightness/Contrast, Erode/Dilate, and Filter are examples of nodes that
all have RGBY checkboxes in the main Controls tab of the Inspector, in addition to the Settings tab.

Apply Mask Inverted

When the Apply Mask Inverted checkbox is enabled, masks attached to the Effect Mask input of that
node are inverted.

TIP: The Apply Mask Inverted checkbox option operates only on effects masks, not on
garbage masks.

Chapter 8 Editing Parameters in the Inspector 214

Multiply By Mask
Selecting this option will cause the RGB values of the masked image to be multiplied by the Mask
channel’s values. This will cause all pixels of the image not included in the mask (i.e., those set to 0) to
become black. This creates a premultiplied image.

Use Object/Use Material (For Masking)

Some 3D animation and rendering software can output to file formats that support auxiliary channels.
Notably, the OpenEXR file format supports Object ID and Material ID channels, either of which can be
used as a mask for an effect. This checkbox determines whether the channels will be used if they are
available. The specific Material ID or Object ID affected is chosen using the next set of controls.

Pick Controls
The Pick Controls are only displayed once the Use Object or Use Material checkbox is enabled. These
controls select which ID is used to create a mask from the Object or Material channels saved in the
image. You use the Pick button to grab IDs from the image in the viewer, the same way you use the
Color Picker to select a color. The image or sequence must have been rendered from a 3D software
package with those channels included.

Correct Edges
The Correct Edges checkbox is only displayed once the Use Object or Use Material checkbox is
enabled. When the Correct Edges checkbox is enabled, the Coverage and Background Color
channels are used to separate and improve the effect around the edge of the object. When disabled
(or no Coverage or Background Color channels are available), aliasing may occur on the edge
of the mask.

Motion Blur
For nodes that are capable of introducing motion, such as Transform nodes, Warp nodes, and so on,
the Motion Blur checkbox toggles the rendering of motion blur on or off for that node. When this
checkbox is enabled, the node’s predicted motion is used to produce the blur caused by a virtual
camera shutter. When the control is disabled, no motion blur is created.
When Motion Blur is disabled, no additional controls are displayed. However, turning on Motion Blur
reveals four additional sliders with which you can customize the look of the motion blur you’re adding
to that node.

Quality
Quality determines the number of samples used to create the blur. The default quality setting of 2 will
create two samples on either side of an object’s actual motion. Larger values produce smoother
results but will increase the render time.

Shutter Angle
Shutter Angle controls the angle of the virtual shutter used to produce the Motion Blur effect. Larger
angles create more blur but increase the render times. A value of 360 is the equivalent of having the
shutter open for one whole frame exposure. Higher values are possible and can be used to create
interesting effects. The default value for this slider is 100.

Center Bias
Center Bias modifies the position of the center of the motion blur. Adjusting the value allows for the
creation of trail-type effects.

Sample Spread
Adjusting Sample Spread modifies the weight given to each sample. This affects the brightness of the
samples set with the Quality slider.

Chapter 8 Editing Parameters in the Inspector 215

Scripting
Scripting fields are present on every node and contain one or more editable text fields that can be
used to add scripts that process when that node is rendering. For more information on the contents of
this tab, please consult the Scripting documentation.

Comments
A Comments field is found on every node and contains a single text field that is used to add comments
and notes to that node. To enter text, simply click within the field to place a cursor, and begin typing.
When a note is added to a node, the comments icon appears in the Control Header and can be seen
in a node’s tooltip when the cursor is placed over the node in the Node Editor. The contents of the
Comments tab can be animated over time, if required.
Additional controls appear under this tab if the node is a Loader. For more information, see Chapter
104, “Generator Nodes.” In the DaVinci Resolve Reference Manual or Chapter 43 in the Fusion
Reference Manual.

Inspector Controls Explained

Although a few nodes use fully customized interface elements that are unique to only that node, the
vast majority of nodes use a mix of sliders, angle wheels, and checkboxes. This section explains how
to use these controls.

Fusion Slider Controls

Slider Controls are used to select a single value from a range of values. You change the value by
dragging the slider or entering a value into the edit box. This is fairly standard behavior for sliders.
However, there is additional functionality that can increase your productivity when making changes
with sliders.
Clicking on the gutter to the left or right of the handle will increase or decrease the value. Holding
Command while clicking on the gutter will adjust the values in smaller increments. Holding Shift while
clicking will adjust the value in larger increments.

Hold Command while clicking in the gutter

to move in smaller increments

Once you click directly on a slider handle, you can make changes to its value using the Left and Right
Arrow keys. The Command and Shift keys can again be used to modify the value in larger or smaller
increments.
While slider controls use a minimum and maximum value range, entering a value in the edit box
outside that range will often expand the range of the slider to accommodate the new value. For
example, it is possible to enter 500 in a Blur Size control, even though the Blur Size sliders default
maximum value is 100. The slider will automatically adjust its maximum displayed value to allow entry
of these larger values.
If the slider has been altered from its default value, a small circular indicator will appear below the
gutter. Clicking on this circle will reset the slider to its default.

Chapter 8 Editing Parameters in the Inspector 216

Thumbwheel
A Thumbwheel control is identical to a slider except it does not have a maximum or minimum value. To
make an adjustment, you drag the center portion left or right or enter a value directly into the edit box.
Thumbwheel controls are typically used on angle parameters, although they do have other
uses as well.

Thumbwheel controls for X,Y, and Z rotation with

arrows on either end for fine-tuning adjustments

You can use the arrowheads at either end of the control to fine tune your adjustments. Once the
thumbwheel has been selected either by dragging or using the arrow keys, you can use the Left and
Right Arrows on your keyboard to further adjust the values. As with the slider control, the Command
and Shift keys can be used to increase or decrease the change in value in smaller or larger
increments.
If the thumbwheel has been altered from its default value, a small circular indicator will appear below
above the thumbwheel. Clicking on this circle will reset the thumbwheel to its default.

Range Controls
The Range controls are actually two separate controls, one for setting the Low Range value and one
for the High Range value. To adjust the values, drag the handles on either end of the Range bar. To
slide the high and low values of the range simultaneously, drag from the center of the Range bar. You
can also expand or contract the range symmetrically by holding Command and dragging either end of
the Range bar. You find Range controls on parameters that require a high and low threshold, like the
Matte Control, Chroma Keyer, and Ultra Keyer nodes.

A Matte Threshold Range control

TIP: You can enter floating-point values in the Range controls by typing the values in using
the Low and High numeric entry boxes.

Checkboxes
Checkboxes are controls that have either an On or Off value. Clicking on the checkbox control will
toggle the state between selected and not selected. Checkboxes can be animated, with a value of 0
for Off and a value of 1.0 or greater for On.

Checkboxes used to select options for tracking

Chapter 8 Editing Parameters in the Inspector 217

Drop-Down Menus
Drop-down menus are used to select one option from a menu. Once the menu is open, choosing one
of the items will select that entry. When the menu is closed, the selection is displayed in the Inspector.

Drop-down menu in the TimeSpeed node

Drop-down menu selections can be animated, with a value of 0 representing the first item in the list, 1
representing the second, and so forth.

Button Arrays
Button arrays are groups of buttons that allow you to select from a range of options. They are almost
identical in function to drop-down menu controls, except that in the case of a button array it is possible
to see all of the available options at a glance. Often button arrays use icons to make the options more
immediately comprehensible.

The Lens Type button array in the Defocus node

Color Chooser and Picker

The Color panel is displayed wherever a parameter requires a color as its value, such as the Fill or
Outline color in the Text+ node. The selected color is shown in a swatch with an Eyedropper to its
right, and below the swatch is the Color Chooser.

The Color panel with transparency preview

The Color panel is extremely flexible and has four different techniques for selecting and
displaying colors.

TIP: Color can be represented by 0–1, 0.255, or 0–65000 by setting the range you want in
the Preferences > General panel.

macOS and Windows Color Nodes

Clicking on the color swatch will display the operating system’s standard Color Selection node.

Chapter 8 Editing Parameters in the Inspector 218

macOS Colors panel Windows Color dialog

Each operating system has a slightly different layout, but the general idea is the same. You can choose
a color from the swatches provided—the color wheel on macOS, or the color palette on Windows.
However you choose your color, you must click OK for the selection to be applied.

The Color Chooser

You also have access to the built-in color chooser, which includes sections for choosing grayscale
values, as well as the currently chosen hue with different ranges of saturation and value. A hue bar and
alpha bar (depending on the node) let you choose different values.

The color chooser in the Background node

Picking Colors from an Image

If you are trying to match the color from an image in the viewer, you can hold down the cursor over the
Eyedropper, and then drag the pointer into the viewer. The pointer will change to an Eyedropper, and
a pop-up swatch will appear above the cursor with the color you are hovering over and its values.
When you are over the color you want, release the mouse button to set the color.

The Eyedropper with color swatch

Chapter 8 Editing Parameters in the Inspector 219

The Color Picker normally selects from a single pixel in the image, but you can adjust the size of the
selection by dragging into the viewer with the Eyedropper, and then holding Command and dragging
out a rectangle for the sample size you want. The size change applies to all Color Pickers until the size
is changed again.

Gradients
The Gradient Control bar is used to create a gradual blend between colors. The Gradient bar displays
a preview of the colors used from start to end. By default, there are two triangular color stops: one on
the left that determines the start color, and one on the right that determines the end color.

The default Gradient controls

Gradient Type
The Gradient Type button array is used to select the form used to draw the gradient. Linear draws the
gradient along a straight line from the starting color stop to the ending color stop.

Linear gradient

Reflect draws the gradient by mirroring the linear gradient on either side of the starting point.

Reflect gradient

Square draws the gradient by using a square pattern when the starting point is at the center of
the image.

Square gradient

Chapter 8 Editing Parameters in the Inspector 220

Cross draws the gradient using a cross pattern when the starting point is at the center of the image.

Cross gradient

Radial draws the gradient in a circular pattern when the starting point is at the center of the image.

Radial gradient

Angle draws the gradient in a counter-clockwise sweep when the starting point is at the center of
the image.

Angle gradient

Start and End Position

The Start and End Position controls have a set of X and Y edit boxes that are useful for fine-tuning the
start and end position of the gradient. The position settings are also represented by two crosshair
onscreen controls in the viewer, which may be more practical for initial positioning.

Gradient Colors Bar

The Gradient Colors bar is used to select the blending colors for the gradient. The default two color
stops set the start and end colors. You can change the colors used in the gradient by selecting the
color stop, and then using the Eyedropper or color wheel to set the new color.
You can add, move, copy, and delete colors from the gradient using the Colors bar.

To add a color stop to the Gradient Colors bar:

1 Click anywhere along the bottom of the Gradient Colors bar.
2 Use the Eyedropper or color wheel to set the color for the color stop.

To move a color stop on the Colors bar:

– Drag a color stop left or right along the Gradient Color bar.

To copy a color stop on the Colors bar:

– Hold Command while you drag a color stop.

Chapter 8 Editing Parameters in the Inspector 221

To delete a color stop from the Colors bar, do one of the following:
– Drag the color stop up past the Gradient Colors bar.
– Select the color stop, then click the red X button to delete it.

Interpolation Space
The Gradient Interpolation Method pop-up menu lets you select what color space is used to calculate
the colors between color stops.

Offset
When you adjust the Offset control, the position of the gradient is moved relative to the start and end
markers. This control is most useful when used in conjunction with the repeat and ping-pong modes
described below.

Once/Repeat/Ping-Pong
These three buttons are used to set the behavior of the gradient when the Offset control scrolls the
gradient past its start and end positions. The Once button is the default behavior, which keeps the
color continuous for offset. Repeat loops around to the start color when the offset goes beyond the
end color. Ping-pong repeats the color pattern in reverse.

1x1, 2x2, 3x3, 4x4, 5x5

These buttons control the amount of sub-pixel precision used when the edges of the gradient become
visible in Repeat mode, or when the gradient is animated. Higher settings will take significantly longer
to render but will be more precise.

Gradient Contextual Menu

Gradients have their own contextual menu that you can bring up by right-clicking on the Gradient bar.
In the Gradient contextual menu are options for animating, publishing, and connecting one gradient to
another. There is also a gradient-specific modifier that builds a custom gradient by sampling colors
from the output of a node in the Node Editor.

Modifiers
Modifiers are expressions, calculations, trackers, paths, and other mathematical components that you
attach to a parameter to extend its functionality. When a modifier is attached to a parameter, its
controls will appear separately in the Inspector Modifiers tab.

To attach a modifier:
1 Right-click over the parameter to which you want to attach a modifier.
2 Make a selection from the Modifier submenu in the contextual menu.

Animating Parameters in the Inspector

Fusion can keyframe most parameters in most nodes, in order to create animated effects such as
animated transforms, rotoscoping with splines, dynamically altering warping behaviors, and so on; the
list is endless. For convenience, a set of keyframing controls are available within the Inspector next to
each keyframable parameter. These controls are:
– A gray Keyframe button to the right each keyframable parameter. Clicking this gray button creates
a keyframe at the current position of the playhead, and turns the button orange.
– When you add a keyframe to a parameter, moving to a new frame and changing the parameter will
automatically add a keyframe at the current position.

Chapter 8 Editing Parameters in the Inspector 222

 – Whenever the playhead is sitting right on top of a keyframe, this button turns orange. Clicking an
orange Keyframe button deletes the keyframe at that frame and turns the button gray again.
– Small navigation arrows appear to the right and left if there are more keyframes in those
directions. Clicking on navigation arrows to the right and left of keyframes jumps the playhead to
those keyframes.

Orange Keyframe buttons in the Inspector show there’s a keyframe at that frame

Once you’ve keyframed one or more parameters, the node containing the parameters you
keyframed displays a Keyframe badge, to show that node has been animated.

A keyframed node displays a Keyframe

badge in the Node Editor

Once you’ve started keyframing node parameters, you can edit their timing in the Keyframes Editor
and/or Spline Editor. For more information about keyframing in Fusion, see Chapter 60, “Animating in
Fusion’s Keyframe Editor” in the DaVinci Resolve Reference Manual or Chapter 9 in the Fusion
Reference Manual.

Removing Animation From a Parameter

To remove all keyframes from a parameter:
1 Right-click over the name of the keyframed parameter in the Inspector.
2 Choose Remove “node name:parameter name” from the contextual menu.

TIP: If you change the default spline type from Bézier, the contextual menu will display the
name of the current spline type.

Attaching a Parameter to an Existing Animation Curve

Multiple parameters can be connected to the same animation curve. This can be an invaluable
timesaver if you are identically animating different parameters in a node.
To connect a second parameter to the same animation curve:
1 Right-click on the second parameter you want to attach.
2 In the contextual menu, hover over the Connect To submenu.
3 In the Connect To submenu, choose the name of the animated parameter.

Chapter 8 Editing Parameters in the Inspector 223

Connecting Parameters
It is often useful to connect two parameters together even without an animation curve. There are two
methods you can use.

Connecting Parameters by Publishing

If you want to tie two parameters together so adjusting one adjusts the other, you must connect them
together using the Publish menu command on the first parameter and the Connect menu command on
the second parameter.

To Publish and Connect parameters:

1 Right-click the name of the parameter you want to publish, and choose Publish from the
contextual menu.
2 Right-click on the second parameter you want to attach, and choose the name of the parameter
you just published from the Connect To submenu.

The Publish contextual menu

Connecting Parameters by Pick Whipping

You can also use simple expressions to link two parameters together. By using simple expressions via
pick whipping, values can be connected and combined visually without the need to publish a value
first. The pick whip is a temporary line drawn from one parameter to another in order to create a link
between the two.

To link two parameters using a pick whip:

1 Double-click the field of a parameter you want to pick whip to another parameter, type =, and then
press the Return key.
2 When Pick Whip controls appear underneath the parameter, drag a “whip” from the Add button to
the target parameter.
Now, adjusting the target parameter automatically adjusts the original parameter.

Pick whipping one parameter to another

TIP: Disabling the Auto Control Close node’s General preference, and then selecting two
nodes in the Node Editor will allow you to pick whip two parameters from different nodes.

Chapter 8 Editing Parameters in the Inspector 224

The Expression field can further be used to add mathematical formulas to the value received from the
target parameter.
For more information on Pick Whipping and Expressions, see Chapter 73, “Using Modifiers,
Expressions, and Custom Controls” in the DaVinci Resolve Reference Manual or Chapter 12 in the
Fusion Reference Manual.

Contextual Menus
There are two types of contextual menus you can invoke within the Inspector.

Node Contextual Menus

To display the Node Context menu from the Inspector, right-click on the Inspector header. The node’s
contextual menu includes the same menu options that are accessed by right-clicking on a node in the
Node Editor. See Chapter 66, “Working in the Fusion Node Editor,” in the DaVinci Resolve Reference
Manual or Chapter 5 in the Fusion Reference Manual for more information on these options.

Parameter Contextual Menus

The contextual menu for individual parameters is accessed by right-clicking over the parameter’s
name, slider, thumbwheel, range control, button array, or other control type. For example, right-clicking
on a slider will provide the slider’s contextual menu, with options to animate the control or add
additional modifiers. Many of these options were described in this chapter.

Customizing Node Parameters

with User Controls
The user interface for each node in Fusion is designed to provide access to the parameters in a logical
manner. Sometimes, though, you may want to add, hide, or change the controls. This is commonly
done for simple expressions and macros, but it can be done for usability and aesthetic reasons for
favorites and presets.
User custom controls can be added or edited via the Edit Control dialog. Right-click the name of a
node in the Inspector (in the header bar) and choose Edit Control from the contextual menu. A new
window will appear, titled Edit Control.

The Edit Control window

In the Input attributes, you can select an existing control or create a new one, name it, define the type,
and assign it to a tab. In the Type attributes, you define the input controls, the defaults and ranges, and
whether it has an onscreen preview control. The Input Ctrl attributes box contains settings specific to
the selected node control, and the View Ctrl attributes box contains settings for the preview
control, if any.

Chapter 8 Editing Parameters in the Inspector 225

All changes made using UserControls are stored in the node instance itself, so they can be copy/
pasted, saved to a setting, added to the Bins, or added to your favorites.

An Example of Customizing Directional Blur

In the following example, let’s suppose we wanted to create a more intuitive way of controlling a linear
blur than using the Length and Angle sliders independently.

Default Directional Blur controls in the Inspector

We could use the Center input control, along with its preview control, to set an angle and distance
from directly within the viewer using expressions.
1 Right-click the label for the Length parameter, choose Expression from the contextual menu, and
then paste the following expression into the Expression field that appears:
-sqrt(((Center.X-.5)*(Input.XScale))^2+((Center.Y-.5)*(Input.YScale)*(Input.
Height/Input. Width))^2)

2 Next, right-click the label for the Angle parameter, choose Expression from the contextual menu,
and then paste the following expression into the Expression field that appears:
atan2(Center.Y-.5)/(Input.OriginalWidth/Input.X , .5-Center.X) * 180 / pi

Directional Blur controlled by the Center’s position

Chapter 8 Editing Parameters in the Inspector 226

This functions fine, but the controls are confusing. The Center control doesn’t work as the center
anymore, and it should be named “Blur Vector” instead. The controls for the Length and Angle aren’t
meant to be edited, so they should be hidden away, and we’re only doing a linear blur, so we don’t
need the buttons for Radial or Zoom. We just need to choose between Linear and Centered.

Adding Another Control

For the first task, let’s rename the Center. From the Add Control window, select Center from the ID list.
A dialog will appear asking if you would like to Replace, Hide, or Change ID. We’ll choose Replace.
Now we are editing the Center input. We’ll change the Name to Blur Vector, set the Type to Point, and
the Page to Controls, which is the first tab where the controls are normally. Press OK, and our new
input will appear on our node in the Node Editor. The ID of the control is still Center, so our
SimpleExpressions did not change.
To hide the Length and Angle, we’ll run the UserControls script again. This time when we select the
Length and Angle IDs, we’ll choose Hide in the dialog. Press OK for each.
Finally, to change the options available in the Type, we have two options. We can hide the buttons and
use a checkbox instead, or we can change the MultiButton from four entries to two. Let’s try both.
To add the checkbox, run UserControls again, but this time instead of selecting an existing ID, we’ll
type Centered into the Name. This will set the name and the ID of our input to Centered. The Type
is set to Number, and the Page is set to Controls. Now in the Type Attributes, set the Input Ctrl to
be CheckboxControl. Press OK, and now we have our checkbox. To make the new control affect
the Type, add a SimpleExpression to the Type:
iif(Centered==1, 2, 0).

Once that’s done, we can use the UserControls to hide the Type control.
To make a new MultiButton, run the UserControl script, and add a new control ID, TypeNew. You
can set the Name to be Type, as the Names do not need to be unique, just the IDs. Set the Type to
Number, the Page to Controls, and the Input Ctrl to MultiButtonControl. In the Input Ctrl attributes,
we can enter the names of our buttons. Let’s do Linear and Centered. Type them in and hit Add for
each. Press OK, and we have our new buttons with the unneeded options removed. To make this
new control affect the original Type, add a SimpleExpression to the Type:
iif(TypeNew==0, 0, 2).

Once that’s done, we can use the UserControls to hide the original Type control.

Directional Blurs with UserControls applied

Chapter 8 Editing Parameters in the Inspector 227

Chapter 9

Animating in Fusion’s
Keyframes Editor
This chapter covers how you can keyframe effects in the Inspector, and how you can
edit clips, effects, and keyframes in the Keyframes Editor.

Contents
Keyframing in the Inspector  229
Removing Animation in the Inspector  230
Attaching a Parameter to an Existing Animation Curve  230
Keyframes Editor Overview  230
Keyframes Editor Tracks  231
The Timeline Header  231
The Playhead  232
Spreadsheet  232
Scaling and Panning the Timeline  233
Working with Segments in the Timeline  233
Moving Segments in the Timeline  233
Trimming Segments  234
Holding the First or Last Frame  234
Working with Keyframes in the Timeline  234
Drag and Drop Keyframe Editing  235
Keyframe Editing Using the Time Editor  235
The Keyframe Spreadsheet  235
Duplicating Spline Keyframes  236
Time Stretching Keyframes  236
Showing Keyframe Values  236
Timeline Filters  236
Selected Filtering  238

Chapter 9 Animating in Fusion’s Keyframes Editor 228

Sorting in the Timeline  238
The Tree Item Order Menu  238
The Sort Menu  239
Markers  239
Jumping to Markers  240
Renaming Markers  240
Show Marker List  240
Deleting Markers  241
Autosnap  241
Autosnap Points  241
Autosnap Markers  241
The Spreadsheet Editor  242
Selecting a Node to Edit  242
Inserting Keyframes  242
Selecting Multiple Nodes to Edit  242
Customizing the Keyframes Editor  243
Line Size  243
Display Point Values  243

Keyframing in the Inspector

Most parameters in most effects nodes can be keyframed in order to create animated effects such as
animated transforms, rotoscoping with splines, dynamically altering warping behaviors, and more; the
list is endless.
For convenience, a set of keyframing controls is available within the Inspector next to each
keyframable parameter. These controls are:
– A gray Keyframe button to the right each keyframable parameter. Clicking this gray button creates
a keyframe at the current position of the playhead, and turns the button orange.
– Whenever the playhead is sitting right on top of a keyframe, this button turns orange. Clicking an
orange Keyframe button deletes the keyframe at that frame and turns the button gray again.
– Small navigation arrows appear to the right and left if there are more keyframes in those
directions. Clicking on navigation arrows to the right and left of keyframes jumps the playhead to
those keyframes.

Orange Keyframe buttons in the Inspector

show there’s a keyframe at that frame

Chapter 9 Animating in Fusion’s Keyframes Editor 229

 Once you’ve keyframed one or more parameters, the node containing the parameters you
keyframed displays a Keyframe badge to show that node has been animated.

A keyframed node displays a Keyframe

badge in the Node Editor

Once you’ve started keyframing node parameters, you can edit their timing in the Keyframes
Editor and/or Spline Editor.

Removing Animation in the Inspector

To remove a keyframed spline from a parameter:
1 Right-click the keyframe control of the parameter you want to remove animation from.
2 Choose Remove [Name of parameter] from the contextual menu.

Attaching a Parameter to an Existing Animation Curve

Multiple parameters can be connected to the same animation curve. This can be an invaluable
timesaver if you are identically animating different parameters in a node.

To connect a second parameter to the same animation curve:

1 Right-click on the second parameter you want to attach.
2 In the contextual menu, hover over the Connect To submenu.
3 In the Connect To submenu, choose the name of the animated parameter.

Keyframes Editor Overview

The Keyframes Editor is essentially a timeline view of your composition, within which each clip and
effect node in your composition is represented by a track. These tracks have the same color coding as
the nodes they represent and are labeled where appropriate. A Time Ruler at the top indicates the
timing of your composition, while numerous controls let you control the contents of the
Keyframes Editor.
The Keyframes Editor can be used for one of two things:
– To adjust the timing of elements in a project, whether they’re clips or effects. You can trim, slide,
and extend clips, adjust the timing of an animation spline, or trim the duration of an effects node.
You can freely rearrange the order of nodes in the Timeline without affecting the layering order of
your composition. All compositing operations are handled in the Node Editor, while the Keyframes
Editor manages the timing of your composition.
– To create and/or edit keyframes that you’ve applied to effects in a track-based manner, you can
retime keyframes, add and delete keyframes, and even edit keyframe values

Chapter 9 Animating in Fusion’s Keyframes Editor 230

 The Keyframes Editor

To show the Keyframes Editor, do one of the following:

– Click the Keyframes Editor button in the UI toolbar to toggle visibility of the Keyframes Editor
on and off.
– Press F7 on the keyboard.

Keyframes Editor Tracks

While each clip and effect node in your composition is represented by a track, keyframed parameters
are exposed either as keyframes superimposed upon the track to which they’re applied (as seen on
the MOVEMENT track), or they can be opened up onto their own tracks for more precise editing, one
keyframe track per keyframed parameter, by clicking a disclosure control to the left of that track’s
name in the Timeline header (as seen under the “Drip1” track).

The Timeline tracks

The Timeline Header

The Timeline header area on the left side of the Timeline is a hierarchical list of all tracks in a
composition. Each track displays the name of its corresponding node, a lock button, and a disclosure
control for revealing keyframe tracks for each keyframe animation, modifier, and mask that’s
attached to it.

The Timeline header area

Chapter 9 Animating in Fusion’s Keyframes Editor 231

Collapse/Open All
A quick way to open or close all available keyframe tracks at once is to use the Expand/Collapse Tool
Controls commands in the Keyframe Timeline Option menu.

The Playhead
As elsewhere in Fusion, the playhead is a red vertical bar that runs through the Timeline view to
indicate the position of the current frame or time. The Keyframes Editor playhead is locked to the
viewer playhead, so the image you’re viewing is in sync.
You must click on the playhead directly to drag it, even within the Timeline ruler (clicking and dragging
anywhere else in the Timeline ruler scales the Timeline). Additionally, you can jump the playhead to a
new location by holding down the Command-Option keys and clicking in the track area (not the
Timeline ruler).

The playhead about to be

dragged by the pointer

Spreadsheet
If you turn on the Spreadsheet and then click on the name of a layer in the keyframe track, the numeric
time position and value (or values if it’s a multi-dimensional parameter) of each keyframe appear as
entries in the cells of the Spreadsheet. Each column represents one keyframe, while each row
represents a single aspect of each keyframe.

Editing keyframes in the Spreadsheet

For example, if you’re animating a blur, then the Key Frame row shows the frame each keyframe is
positioned at, and the Blur1BlurSize row shows the blur size at each keyframe. If you change the Key
Frame value of any keyframe, you’ll move that keyframe to a new frame of the Timeline.

Chapter 9 Animating in Fusion’s Keyframes Editor 232

Scaling and Panning the Timeline
At the top, a series of zoom and framing controls let you adjust the work area containing the layers.
– A Horizontal zoom control lets you scale the size of the editor.
– A Zoom to Fit button fits the width of all tracks to the current width of the Keyframes Editor.
– A Zoom to Rect tool lets you draw a rectangle to define an area of the Keyframes Editor to zoom into.
– A Sort pop-up menu lets you sort or filter the tracks in various ways.
– An Option menu provides access to many other ways of filtering tracks and controlling
visible options.

Working with Segments in the Timeline

Most of the work in the Timeline involves trimming and aligning clip segments.

To select a single segment in the Timeline, do one of the following:

– Click the node’s name in the header.
– Click the node’s segment in the Timeline.

To add another segment to the selection, do one of the following:

– Hold Command and click additional segments to select discontiguous selections.
– Select a segment, and then hold Shift and click another segment to make a contiguous selection
of all segments in between.

To remove a segment from the selection, do the following:

– Hold Command and click a selected segment to deselect it.

TIP: Selecting a node’s name from the Timeline header also selects the node’s tile in the
Node Editor, with its controls displayed in the Inspector.

Moving Segments in the Timeline

To move the position of a segment, drag on the node’s segment in the Keyframes Editor. The cursor
will resemble a bar with two arrows pointing in either direction. Moving a segment changes where that
clip begins and ends in the composition.

The Move cursor

Chapter 9 Animating in Fusion’s Keyframes Editor 233

Trimming Segments
Trimming segments has different effects on Loaders, MediaIn and Effect nodes:
– Trimming a Loader or MediaIn node is similar to trimming clips in an editing application, in
that you’re changing the in and out points of the range of media that clip makes available to
your composition.
– Trimming the segments of effect nodes instead modifies the range of that node’s effect in the
composition. Outside of the trimmed region, that effect node will behave as if it were disabled.

TIP: Shortening the duration of effects nodes can optimize processing. Imagine a Loader or
MediaIn node that represents a clip that’s 100 frames long and is connected to a Defocus
node that’s animated from frames 80–100. There is little to no point in processing the defocus
node between frames 0–79, so trimming the defocus segment to start at frame 80 in the
Timeline will effectively prevent it from rendering and consuming either memory or processor
time until needed.

To trim a segment in the Timeline, do the following:

– Drag on either end of the node’s segment in the Timeline.
The cursor changes to a vertical bar with a single arrow when the cursor is in the right location to trim.

The Trim cursor

Holding the First or Last Frame

If you want to hold a Loader’s first or last frame of a clip for a certain number of frames, also called a
freeze frame, you can hold Command while you drag beyond the first or last of the segment in
the Timeline.

Working with Keyframes in the Timeline

Keyframes can be drawn in one of two ways. When keyframe tracks are closed, they’re drawn over the
node’s segment. Clicking on the disclosure icon to the left of the node’s name in the track header
expands the display so each keyframed parameter has its own track in the Timeline, enabling
precise editing.
Furthermore, each keyframe track, whether open or closed, exposes a miniature curve overlay that
provides a visual representation of the rise and fall of keyframed values. This little overlay isn’t
directly editable.

The Drip1 segment has its keyframe tracks exposed, while the Text1 segment has
its keyframe tracks collapsed so they’re displayed within the segment.

Chapter 9 Animating in Fusion’s Keyframes Editor 234

Drag and Drop Keyframe Editing
Here are pointer-based keyframe editing methods that will get you started.

Methods of selecting keyframes:

– Click a single keyframe to select it.
– Drag a bounding box over a series of keyframes to select them all.
– Command-click to select discontiguous keyframes.
– Shift-click the first and last of a range of keyframes to select a contiguous range.

Methods of adjusting keyframes:

– You can drag keyframes left and right to reposition them in time.
– You can right-click one or more selected keyframes and use contextual menu commands to
change keyframe interpolation, copy/paste keyframes, or even create new keyframes.

Keyframe Editing Using the Time Editor

A drop-down and editing field at the bottom right of the Keyframes Editor lets you numerically edit the
timing, in frames, of any selected keyframe, making it easy to make precise adjustments.

To change the position of a keyframe using the toolbar, do one of the following:
– Select a keyframe, and then enter a new frame number in the Time Edit box.
– Choose T Offset from the Time Editor drop-down, select one or more keyframes, and enter a
frame offset.
– Choose T Scale from the Time Editor drop-down, select one or more keyframes,
and enter a frame offset.

The Time button can switch to Time Offset or Time Scale for moving keyframes.

The Keyframe Spreadsheet

If you turn on the Spreadsheet and then click on the name of a layer in the keyframe track, the numeric
time position and value (or values if it’s a multi-dimensional parameter) of each keyframe appear as
entries in the cells of the Spreadsheet. Each column represents one keyframe, while each row
represents a single aspect of each keyframe.

Editing keyframes in the Spreadsheet

For example, if you’re animating a blur, then the Key Frame row shows the frame each keyframe is
positioned at, and the Blur1BlurSize row shows the blur size at each keyframe. If you change the Key
Frame value of any keyframe, you’ll move that keyframe to a new frame of the Timeline.

Chapter 9 Animating in Fusion’s Keyframes Editor 235

Duplicating Spline Keyframes
Keyframes can be duplicated, either onto the same keyframe track or onto different tracks. This can
save you time if you need to repeat a keyframe sequence at another time on the same segment, or
even just create identically-timed keyframes on two different segments.

To duplicate keyframes, do the following:

1 Select one or more keyframes you want to duplicate.
2 Hold Command and drag one of the selected keyframes to a new position.

Time Stretching Keyframes

If you select a range of keyframes in a keyframe track, you can turn on the Time Stretch tool to show a
box you can use to squeeze and stretch the entire range of keyframes relative to one another, to
change the overall timing of a sequence of keyframes without losing the relative timing from one
keyframe to the next. Alternatively, you can turn on Time Stretch and draw a bounding box around the
keyframes you want to adjust to create a time-stretching boundary that way. Click the Time Stretch tool
again to turn it off.

Time stretching keyframes

Showing Keyframe Values

When a node and its accompanying segment have animated parameters, keyframes appear as
colored tick marks in keyframe tracks to indicate when animated changes occur. If the tracks and
splines are open on a parameter, choosing Show Values from the Keyframes Editor Option menu
shows editable fields beneath each keyframe. These fields show each keyframe’s current value and
allow you to edit them simply by entering a new number.

Show values from the Keyframes Editor Option menu

Timeline Filters
When a composition grows to include hundreds of nodes, locating specific node layers can quickly
become difficult. Timeline filters can be created and applied to sift out nodes that are not necessary to
the current operation. The Global Timeline preferences include a number of pre-made filters that you
can enable, or you can create new ones as needed.

To use a Timeline filter:

Open the Keyframes Editor Option menu and choose an item from the top of the menu. Default
Timeline filters include;
– Show All, which shows all node layers in the current composition.
– Show None, which hides all layers.

Chapter 9 Animating in Fusion’s Keyframes Editor 236

 – Show Tools at Current Time, which only displays node layers under the playhead.
– If you’ve created custom filters, they appear here as well, in alphabetical order.

To go back to showing everything:

– Choose Show All from the Keyframes Editor Option menu. All layers will reappear.

Choosing a Timeline filter

To create a Timeline filter:

1 Choose Create/Edit Filters from the Keyframes Editor Option menu to open the Timeline panel of
the Fusion Settings window. This is where you can create new Timeline filters.

The Global Timeline preferences for enabling filters

2 Click the New button, enter a name for your new filter setting, and click OK. The filter you created
is now selected in the Filter pop-up menu at the top.
3 Use the “Settings for filters” list to turn on the checkboxes of nodes you want to be seen and turn
off the checkboxes of nodes you want to filter out. Each category of node can be turned on and
off, or you can open up a category’s disclosure control to turn individual nodes on and off. Clicking
Invert All immediately turns off all node categories.
4 When you’re finished creating filters, click the Save button to hide the Fusion Settings window.

Filters that you’ve created in the Timeline panel of the Fusion Settings window appear in the
Keyframes Editor Option menu.

Chapter 9 Animating in Fusion’s Keyframes Editor 237

To delete a filter:
1 Choose Create/Edit Filters from the Keyframes Editor Option menu to open the Timeline panel of
the Fusion Settings window. This is where you can delete Timeline filters.
2 Choose the filter you want to delete from the Filter pop-up menu.
3 Click the Delete button, and when a dialog asks if you really want to do that, click OK.

Selected Filtering
Choosing “Show only selected tools” from the Keyframes Editor Option menu filters out all segments
except for layers corresponding to selected nodes. This option can be turned on or off.

TIP: When “Show only selected tools” is enabled, you can continue to select nodes in the
Node Editor to update what’s displayed in the Keyframes Editor.

Sorting in the Timeline

You can change the order in which the nodes are displayed from top to bottom in the Timeline.
– You can use the Tree Item Order Selection menu to sort the tracks by an assigned number.
– You can use the Sort pop-up menu.

The Tree Item Order Menu

Right-clicking over any track on the Keyframes Editor will display a contextual menu that contains the
Tree Item Order Selection submenu. Choosing Start from the submenu allows you to start numbering
each item in the track header by clicking on them. The first item you click will be #1, the second item
#2, the third #3, and so on. Once you have selected all the items in the order you want them to be
organized in the Keyframes Editor, right-click over a track, and from the Tree Item Order Selection
submenu, choose End. The items will be ordered using the assigned numbers, with #1 appearing
above #2, which appears above #3, and so on. The first items in the Keyframe track list will always be
the nodes that are the root of the node tree. The numbered nodes will appear in order after the root
nodes. For example, if the node tree starts with a background node, and then connects to a Fast
Noise, Blur, and Color Corrector, the background node will always appear at the top of the Keyframes
Editor track list because it is the root node.

The Keyframes Editor Tree Item Order Selection menu

If you begin numbering nodes in the track header and change your mind or decide on a different
order, you can choose Restart to begin numbering again or choose Cancel to keep the current order.

Chapter 9 Animating in Fusion’s Keyframes Editor 238

The Sort Menu
The Sort menu reorders how the layers of each node appear in the Keyframes Editor. Setting the
menu back to All Tools will display them in a linear order, scanning the Node Editor from left to right
and top to bottom. This is the default setting.

The Timeline Sort Order menu

– All Tools: Forces all tools currently in the Node Editor to be displayed in the Keyframes Editor.
– Hierarchy: Sorts with the most background layers at the top of the header, through to the most
foreground layers at the bottom, following the connections of the nodes in the Node Editor.
– Reverse: The opposite of Hierarchy, working backward from the last node in the Node Editor
toward the most background source node.
– Names: Sorts by the alphabetical order of the nodes, starting at the top with the beginning
of the alphabet.
– Start: Orders layers based on their starting point in the composition. Nodes that start earlier
in the Global project time are listed at the top of the header, while nodes that start later are
at the bottom.
– Animated: Restricts the Timeline to showing animated layers only. This is an excellent mode to
use when adjusting the timing of animations on several nodes at once.

Markers
Markers help identify important frames in a project that might affect how you keyframe animation.
They may indicate the frame where a dragon breathes fire at a protagonist, the moment that someone
passes through a portal, or any other important frame in a composition that you need to keep track of.
Markers added to the Timeline in the Cut, Edit, Fairlight, or Color page will appear in the Keyframes
Editor and Spline Editor of the Fusion page. They can also be added from the Keyframes Editor or the
Spline Editor while working in Fusion Studio or the Fusion page. Markers in Fusion appear as a small
handle with a line extending vertically through the graph view when selected.

A marker being moved in the Keyframed Editor

Chapter 9 Animating in Fusion’s Keyframes Editor 239

To create a marker, do the following:
– Right-click at a frame in the Timeline Ruler of the Keyframes Editor and choose Add Marker from
the contextual menu.

The most important attribute of a marker is its position. For it to add value, a marker must be placed on
the frame you intended it to be on. Hovering the cursor over a marker displays a tooltip with its current
frame position. If it is on the wrong frame, you can drag it along the Time Ruler to reposition it.
Markers added to the Time Ruler are editable in the Fusion page, and the changes appear back in the
other DaVinci Resolve pages. Time Ruler markers can be added, moved, deleted, renamed, and given
descriptive notes from within Fusion’s Keyframes or Spline Editor.

NOTE: Markers attached to clips in the Edit page Timeline are visible on MediaIn nodes in
Fusion’s Keyframes Editor but not editable. They are not visible in the Spline Editor.

Jumping to Markers
Double-clicking a marker jumps the playhead to that marker’s position.

Renaming Markers
By default, a marker uses the frame number in its name, but you can give it a more descriptive name to
go along with the frame number, making it easier to identify. To rename a marker in Fusion, right-click
over the marker and choose Rename Guide from the contextual menu. Enter a name in the dialog
and click OK.

The Marker contextual menu is accessed by right-clicking

over a marker or Keyframe Editor Time Ruler.

Show Marker List

Markers can be used to jump to specific locations in a composition using the Marker List. If you
right-click over a marker or within the Keyframe Editor Time Ruler to bring up the contextual menu, you
can choose Show Marker List, or press Shift-G, to display the Marker List dialog. The Marker List is a
floating dialog that will remain on top of the main window until closed.
The Marker List shows all the current markers in the composition, listed according to their position in
time along with any custom name you’ve given them. If you double-click a marker’s name from the list,
the playhead jumps to the marker’s location.

Chapter 9 Animating in Fusion’s Keyframes Editor 240

 The Marker List dialog allows you to navigate
through a composition using markers.

There is a pair of checkboxes beside the names of each marker. One is for the Spline Editor, and one
is for the Keyframes Editor. By default, markers are shown in both the Spline Editor and Keyframes
Editor, but you can deselect the appropriate checkbox to hide the markers in that view.

Deleting Markers
You can delete a marker by dragging it up beyond the Time Ruler and releasing the mouse. You can
also use the marker’s contextual menu to choose Delete Marker.

Autosnap
To help with precisely positioning keyframes and the start and end of segments as you drag in the
Timeline, you can have them snap to a field, a frame, or to markers. The Autosnap option is accessed
through the Keyframes Editor’s contextual menu. There are two submenu options for autosnapping.
One option controls the snapping behavior when you drag keyframes, control points, or the starting
and ending edges of segments. The other option controls the snapping behavior of markers.

Autosnap Points
When you drag keyframes or the edges of segments, often you want them to fall on a specific frame.
Autosnap restricts the placement of keyframes and segment edges to frame boundaries by default,
but you have other options found in the contextual menu. To configure autosnapping on keyframes
and segment edges, right-click anywhere within the Keyframes Editor and choose Options > Autosnap
Points from the contextual menu. This will display the Autosnap Points submenu with options for the
snapping behavior. The options are:
– None: None allows free positioning of keyframes and segment edges with subframe accuracy.
– Frame: Frame forces keyframes and segment edges to snap to the nearest frame.
– Field: Field forces keyframes and segment edges to snap to the nearest field,
which is 0.5 of a frame.
– Guides: When enabled, the keyframes and segment edges snap to markers.

Autosnap Markers
When you click to create a new marker, the default behavior is that it will snap to the closest frame. If
you reposition the marker, it also snaps to the nearest frame as you drag. This behavior can be
changed in the Keyframes Editor’s contextual menu by choosing from the Options > Autosnap Markers
submenu. The options are:
– None: Markers can be placed anywhere with subframe accuracy.
– Frame: Frame forces all markers to snap to the nearest frame.
– Field: Field forces all markers to snap to the nearest field.

Chapter 9 Animating in Fusion’s Keyframes Editor 241

The Spreadsheet Editor
The Spreadsheet Editor is a separate panel that can be displayed beneath the Keyframes Editor. It is
used to compactly show the numeric values of the keyframes for selected parameters in the
Keyframes Editor’s header, via a table with rows and columns, showing time and value.

The Spreadsheet Editor showing editable data for six keyframes

To reveal the Spreadsheet Editor, click on the Spreadsheet button in the toolbar. The Spreadsheet will
split the Work Area panel and appear below the Keyframes Editor’s interface.

Selecting a Node to Edit

To display a node’s timing in the Spreadsheet, select the node’s name in the Keyframes Editor header.
The Start and End points of the selected node will appear in the keyframe’s line of the Spreadsheet.
To edit an animation parameter in the Spreadsheet Editor, select the parameter in the Keyframes
Editor header. The keyframe row includes a box for each frame number that contains a keyframe.
The value of the keyframe is displayed in the cell below the frame number. Clicking on a cell allows
you to change the frame number the keyframe is on or the parameter’s value for that keyframe.

Clicking on the parameter’s keyframe

value allows you to change it

TIP: Entering a frame number using a decimal point (e.g., 10.25 or 15.75) allows you to set
keyframes on a subframe level to create more natural animations.

Inserting Keyframes
You can also add new keyframes to an animation by clicking in an empty keyframe cell and entering
the desired time for the new keyframe. Using the cell under the new keyframe, you can enter a value
for the parameter.

Selecting Multiple Nodes to Edit

Multiple splines and nodes can be edited together in the Spreadsheet. By default, selecting a new
parameter in the Timeline header will replace the parameter and keyframes currently listed in the
Spreadsheet Editor. Holding down Command, you can click on additional parameters on different
nodes to add them to the Spreadsheet.

TIP: You can use the Tab and Shift-Tab key shortcuts to move the selection right or left in the
Spreadsheet Editor.

Chapter 9 Animating in Fusion’s Keyframes Editor 242

Customizing the Keyframes Editor
There are a few ways you can change the appearance of the Keyframes Editor to better fit your needs.
All these options are found by right-clicking anywhere within the Keyframes Editor and choosing an
option from the contextual menu that appears.

Line Size
The Line Size option controls the height of each Timeline segment individually. It is often useful to
increase the height of a Timeline bar, especially when editing or manipulating complex splines.

Methods of increasing or decreasing the height of segments:

– To change the hight of just one segment: Right-click anywhere within the Keyframes Editor
and choose a size from the Line Size submenu. The options are Minimum, Small, Medium,
Large, and Huge.
– To change the height of all segments: Right-click anywhere within the Keyframes Editor
and choose a size from the All Line Size submenu. The options are Minimum, Small, Medium,
Large, and Huge.

Display Point Values

A more traditional view of keyframes is to view them as control points instead of vertical bars, making
them easier to select for some people. From the Timeline contextual menu, you can right-click
anywhere within the Keyframes Editor and choose Options > Display Point Values to change how
keyframes look.

The Options submenu for changing Display Point Values

Here are the two options, compared.

Keyframes displayed as bars (left), and keyframes displayed as Point Values (right).

Chapter 9 Animating in Fusion’s Keyframes Editor 243

Displaying Audio Waveforms
You can display a MediaIn node’s audio waveform in the Keyframes Editor and use it as a guide as you
add and move keyframes.

Waveforms are displayed in the Keyframes Editor for all MediaIn nodes

To display the audio waveform in the Keyframes Editor:

1 Open the Keyframes Editor.
2 Click the disclosure arrow next to the MediaIn node to view the audio waveform for that clip.

To change the size of an audio waveform display:

1 Open the Keyframes Editor.
2 In the Keyframes Editor, select the audio track of the waveform you want to modify.
3 Right-click over the audio waveform and choose Line Size > Minimum/Small/Medium/Large/Huge.
When using Fusion Studio, you can view the audio waveforms in the Keyframes Editor by
displaying the Saver node.

To view the Audio Waveform in Fusion Studio, do the following:

1 Open the Keyframes Editor.
2 Expand the Saver track to view the audio waveform.
When you want to find the precise location of an audio beat, transient, or cue, you can slowly drag
over the audio waveform to hear the audio. If you need to see more resolution in the waveform
display, you can increase the size.

To change the size of an audio waveform display:

1 Open the Keyframes Editor.
2 Right-click over the audio waveform and choose Line Size > Minimum/Small/Medium/Large/Huge.

TIP: Right-clicking a track in the Keyframes Editor and choosing All Line Size > Minimum/
Small/Medium/Large/Huge changes all the tracks and audio waveforms in the
Keyframes Editor.

Chapter 9 Animating in Fusion’s Keyframes Editor 244

Chapter 10

Animating in Fusion’s
Spline Editor
This chapter covers how you can keyframe effects and control animations in
Fusion’s Spline Editor.

Contents
Spline Editor Overview  246
Spline Editor Interface  247
The Graph, Header, and Toolbar  247
Renaming Splines  248
Changing Spline Colors  249
Navigating Around the Spline Editor  249
Markers  250
Autosnap  252
Creating Animation Splines  253
Animating with Different Spline Types  253
Working with Keyframes and Splines  255
Adding Keyframes  255
Locked and Unlocked Controls Points  256
Selecting, Moving, and Deleting Keyframes  257
Showing Key Markers  258
Copying and Pasting Keyframes  258
Time and Value Editors  260
Modifying Spline Handles  261
Reducing Points  262

Chapter 10 Animating in Fusion’s Spline Editor 245

Filtering the Spline Editor  262
Working with Filters  263
Changing a Spline’s Status  264
Selection States  265
Reshaping Splines Using the Toolbar  266
Interpolation  266
Reversing Splines  268
Looping Splines  268
Time Stretching  270
Shape Box  271
Ease In/Out  272
Importing and Exporting Splines  273

Spline Editor Overview

The Spline Editor is the main area where animation is manipulated and refined. You primarily use the
Spline Editor to show the changing values of parameters over time in the form of splines. Whereas
Keyframes explicitly set the value of a parameter on a given frame, Splines are lines or curves that
interpolate the values between keyframes. Once you set a keyframe on a parameter, a spline is
created and can be displayed in the Spline Editor so you can make further refinements to the
animation. However, the Spline Editor is more advanced than a standard curve editor since it can also
display functions, which may not be splines, like changing characters within a text string or
mathematical expressions that drive your animations.

The Spline Editor with three animated parameters

What Are Splines?

All animation uses splines that describe the value of a parameter at any given point in time. The Spline
Editor graph shows the time of the comp along the horizontal axis and the value of the parameter
along the vertical axis.

Chapter 10 Animating in Fusion’s Spline Editor 246

The advantage of using splines to represent animation instead of keyframes as in the Keyframes Editor
is that splines allow you to manipulate the interpolation between keyframes. For example, if a
keyframe is set at a value of 1.0 on a parameter for frame 1, followed by a keyframe value of 10.0 for
frame 10, the values between keyframes are smoothly interpolated or calculated based on the shape
of the spline. Using the functions and controls in the Spline Editor, you have a fantastic amount of
control over that interpolation.

Spline Editor Interface

The Spline Editor is not visible by default, but you can show it at any time by clicking the Spline button
in the user interface toolbar. You can also display the Spline Editor by right-clicking on a node in
the Node Editor or a segment in the Keyframes Editor and choosing Edit Splines from the
drop‑down menu.
The Spline Editor can be open alongside the Node Editor or Keyframes Editor, or displayed separately
in order to take up the entire work area.

The Graph, Header, and Toolbar

The Spline Editor has three main working areas: the graph, header, and toolbar. On the left side of the
Spline Editor is the header, which shows a list of animated parameters. The majority of the panel is
taken up by the splines displayed in the graph area, and a toolbar runs along the bottom, providing a
variety of ways to manipulate the splines.

Graph
The graph is the largest area of the interface. It is here that you see and edit the animation splines.
There are two axes in the graph. The horizontal axis represents time, and the vertical axis represents
the spline’s value. A thin bar, called the playhead, runs vertically through the graph to represent the
current time as it does in the Timeline Editor. You can drag the playhead to change the current time,
updating the frame displayed in the viewers.

Removing a spline from a parameter removes the animation

Spline Editor Header

The header provides a mechanism for determining what splines are visible in the graph. It shows the
name of each spline in the project beneath the tool that contains that parameter. The checkbox beside
each name shows whether that spline is currently displayed in the graph and whether the spline can
be edited.

Chapter 10 Animating in Fusion’s Spline Editor 247

The Spline header

Spline Editor Toolbar

The toolbar across the bottom of the Spline Editor represents the most common operations applied to
an animation spline. The various operations represented in the toolbar are all accessible from the
graph’s context menu as well, but the following buttons provide a faster shortcut.

Spline toolbar buttons

Playhead
The playhead is the thin red vertical bar that runs vertically through the Spline Editor graph and
represents the current time of the comp. You can drag the playhead to change the current time.

Status Bar
The status bar in the lower-right corner of the Fusion window regularly displays information about the
position of the pointer, along with the time and value axes.

Contextual Menus
There are two contextual menus accessible from the Spline Editor. The Spline contextual menu is
displayed by right-clicking over the graph, while the Guide contextual menu is displayed by right-
clicking on the Time Ruler above the graph.

Right-click on the horizontal axis Time Ruler for the Guide menu

Renaming Splines
The name of a spline in the header is based on the parameter it animates. You can change the name
of a spline by right-clicking on it in the header and choosing Rename Spline from the contextual menu.

Chapter 10 Animating in Fusion’s Spline Editor 248

Changing Spline Colors
Each spline in the graph is assigned a different color, making individual splines easier to identify when
multiple splines are visible at once. When the spline is active, a round color swatch is displayed next to
the spline’s name in the header.

To change the color of a spline:

1 Right-click on the name of the spline in the header and choose Change Color from the
contextual menu.
2 Select the new color from the dialog box that appears and click OK.

Navigating Around the Spline Editor

It is often necessary to magnify and pan around the graph area to ensure that the splines you want to
work on are visible. In general, scaling and panning the Spline Editor works the same as in all
navigable parts of the Fusion interface. However, there are several unique functions to the Spline
Editor for controlling your view based on the height, width, and selection of multiple animation splines.
The most obvious navigation methods to use are the scale sliders and buttons located in the upper
left of the Spline Editor panel.

The Zoom Height and Zoom Width sliders, Fit button, and Zoom to
Rectangle button can be used to navigate around the graph

To scale and pan using the sliders and buttons:

– Zoom Height and Zoom Width sliders let you change the height and width of the graph area.
– The Fit button attempts to rescale the view so that all currently active splines fit within the graph.
– The Zoom to Rectangle button (Command-R) allows you to draw a bounding box around the area
of the graph you want centered and scaled.

To scale using the axis labels:

– Place the mouse pointer over the rulers for the horizontal or vertical axis and drag to resize
the graph on that axis only. The view is scaled centered on the original position of the
pointer on the ruler.

Dragging in the Time Ruler scales

the graph horizontally

To scale and pan with the mouse and/or keyboard:

– With the Spline Editor active, press the + and - keys on your keyboard to zoom in and
out of the graph.
– You can also zoom to a specific control point by holding down the Command key and scrolling the
middle mouse wheel. The mouse pointer location determines the area that gets magnified.
– Position the mouse pointer over the graph, and then hold down the middle mouse button. With the
middle mouse button pressed down, click once on the left mouse button to zoom in and once on
the right button to zoom out.

Chapter 10 Animating in Fusion’s Spline Editor 249

To pan the graph area:
– Drag left or right using the middle mouse button or use the scroll bar along the bottom and right
side of the graph.

Using the Graph contextual menu for navigation:

There are several ways to navigate the graph area using the Spline Editor contextual menu as well.
– Choose Scale > Scale to Fit (Command-F) to fit all active splines into the graph area.
– Choose Scale > Scale to Rectangle (Command-R) to draw a bounding box around the area
of the graph you want centered and scaled. This has the same effect as clicking the Zoom to
Rectangle button.
– Choose Scale > Default to reset the scaling of the graph area to default values.
– Choose Scale > Zoom In/Zoom Out to scale the graph area. This performs the same functions as
pressing the + and - keys on the keyboard.
– Choose Scale > Auto Fit to scale the graph to fit all splines dynamically as you make splines visible
and hidden. If the scaling is changed with Auto-Fit enabled, the graph area will scroll as you play
the comp to view all the keyframes.
– Choose Scale > Auto Scroll to scroll the graph area if the splines fall outside the graph horizontally
as you play.
– Choose Scale > Manual to disable all automatic attempts at showing splines in the graph.
– Choose Options > Fit Times to automatically scale along the X-axis to fit the selected spline.
All visible splines are taken into account, not just the newly selected spline. With this option off,
activating a new spline will not change the horizontal scale.
– Choose Options > Fit Values to automatically scale along the Y-axis to fit the selected spline.
All visible splines are taken into account, not just the newly selected spline. With this option off,
activating a new spline will not change the vertical scale.

Markers
Markers help identify important frames in a project. They may indicate a frame where a ray gun shoots
a beam in the scene, the moment that someone passes through a portal in the image, or any other
important event in the composite.
Markers added to the Timeline in the Cut, Edit, Fairlight, or Color page will appear in the Keyframes
Editor and Spline Editor of the Fusion page. They can also be added from the Keyframes Editor or the
Spline Editor while working in Fusion Studio or the Fusion page. Markers appear along the top of the
horizontal axis Spline Editor’s Time Ruler. They are displayed as small blue shapes, and when selected,
a line extends from each guide down vertically through the graph.

NOTE: Markers attached to clips in the Cut, Edit, Color, or Fairlight pages Timeline are not
visible in Fusion’s Spline Editor.

Unselected markers appear as blue shapes along the top, while selected
markers display a vertical line running through the graph

Chapter 10 Animating in Fusion’s Spline Editor 250

Working with Markers
Markers call attention to a particular frame within a comp. They can be named, displayed in a list, and
edited. After you add markers, you can easily jump the playhead between them, change their position,
or delete them altogether.

Markers can be added by right-

clicking in the horizontal time axis

To create a marker:
– Right-click in the horizontal axis Time Ruler and choose Add Guide.

To delete a marker, do one of the following:

– Drag the marker up outside the Spline Editor panel.
– Right-click on the marker and then choose Delete Guide from the menu.
– Select the marker and then press Delete or Backspace on the keyboard.
– From the Marker List, select a guide in the list and click the Del button.

To move a marker to a new frame, do one of the following:

– Drag the marker handle along the time axis.
– Right-click in the marker area and choose Options > Enable Guide Grabs, and then drag the
marker’s vertical line to move the guide.

To move the Playhead to a marker:

– Right-click on a marker and choose Set Current Time To [Frame number].

Using the Marker List

The Marker List is a list of all markers in the current comp. It can display the markers from either the
Keyframes Editor, the Spline Editor, or both panels simultaneously. Clicking on the frame number or
name of a guide causes the current time to change to that marker’s frame. Since the Marker List is a
floating window, it can remain open, allowing you to quickly jump to different markers while you work
in the Spline Editor.

To show the Marker List:

– Right-click in the horizontal axis and choose Show Marker List, or press Shift-G.

Chapter 10 Animating in Fusion’s Spline Editor 251

 The Marker List above shows markers in the current comp.

If markers currently exist in the comp, they are automatically displayed in the Marker List, regardless of
whether they were added in the Keyframes Editor or the Spline Editor or any other page in
DaVinci Resolve. You can also add markers directly from the Marker List, which can be helpful if you
have multiple markers you need to add, and you know the rough timing.

To add a guide from the Marker List:

1 Click the Add button in the Marker List window.
2 Enter a frame number in the Time field.
3 Press Tab or click the Close button to close the Marker List.

To name a marker, do one of the following:

– In the Marker List, double-click in the Name column, to the right of the frame number, and enter a
name for the marker.
– Right-click over a marker in the horizontal axis and choose Rename Marker. In the dialog that
opens, enter a name for the marker.

Displaying Marker with the Timeline

The Marker List window includes checkboxes next to each marker that determines whether a marker
displays in the Spline Editor, the Keyframes Editor, both, or neither. By default, when you create
markers, they are active in both panels. To hide a marker from appearing in either panel, deselect the
appropriate checkbox.

Autosnap
To assist in precisely positioning keyframe control points along the horizontal (time) axis, you can
enable the Spline Editor’s Autosnap function. Right-clicking over a spline and choosing Options >
Autosnap provides a submenu with four options.
– None: Allows free, sub-frame positioning of the keyframes.
– Frame: Keyframes snap to the nearest frame.
– Fields: Keyframes snap to the nearest field.
– Guides: Keyframes snap to the nearest marker.

Autosnapping and Markers

By default, a newly-created marker snaps to the closest frame. Moving markers with the mouse also
snaps them to the current frame. You can change this behavior by selecting Options > Autosnap
Markers > None or by selecting Options > Autosnap Markers > Field from the contextual menu.

Chapter 10 Animating in Fusion’s Spline Editor 252

Creating Animation Splines
Animation splines are created automatically when you keyframe a parameter in the Inspector or the
Keyframes Editor. However, you can create an animation spline without first having to add a keyframe.

To create a spline:
– Right-click on the parameter to be animated in the Inspector, and choose Animate from the
contextual menu.

Choosing Animation from the menu

displays a spline in the Spline Editor

Selecting Animate from the contextual menu connects the parameter to the default spline type. This is
usually a Bézier Spline unless you change the default spline in the Defaults panel of the Fusion
Preferences.

Deleting Animation Splines

To remove an animation spline from a parameter, right-click on the control and select Remove [tool
parameter’s name] from the contextual menu. Removing a spline from a parameter only deletes the
spline if no other tool in the composition is connected to the same spline at that time.

Removing a spline from a parameter

removes the animation

Animating with Different Spline Types

A Bézier spline is the default spline unless changed in the Preferences. However, if you want to use a
spline type other than Bézier for the animation curve, you can choose the spline type from the Modify
With contextual submenu before creating any keyframes.

Chapter 10 Animating in Fusion’s Spline Editor 253

 Three spline types in the Modify With menu

– Bézier Spline: Bézier splines are the default curve type. Three points for each keyframe on the
spline determine the smoothness of the curve. The first point is the actual keyframe, representing
the value at a given time. The other two points represent handles that determine how smoothly
the curve for the segments leading in and out of the keyframe are drawn. Bézier is the most used
spline type because Bézier splines allow you to create combinations of curves and straight lines.

Bézier spline

– Modify with > B-Spline: B-splines use a single point to determine the smoothness of the curve.
Instead of using handles, a single control point determines the value as well as the smoothness
of the curve. Holding down the W key while dragging left or right on the control point adjusts the
tension of the curve.

B-spline

– Modify with > Cubic Spline: Cubic splines are similar to Bézier splines, in that the spline
passes through the control point. However, Cubic splines do not display handles and always
make the smoothest possible curve. In this way, they are similar to B-splines. This spline type
is almost never used.

Chapter 10 Animating in Fusion’s Spline Editor 254

 Cubic spline

– Modify with > Natural Cubic Spline: Natural Cubic splines are similar to Cubic splines, except that
they change in a more localized area. Changing one control point does not affect other tangents
beyond the next or previous control points.

Natural Cubic spline

Working with Keyframes and Splines

Once you animate a parameter and display the Spline Editor, you can manipulate the spline’s
keyframes (and thus the animation) in a variety of ways. By selecting Keyframe control points, you can
move, copy, and change the interpolation of your animation.

Adding Keyframes
Once you create one keyframe, additional keyframes are automatically added to a spline whenever
you move the playhead and change the value of that spline’s parameter. For example, if you change
the strength of an animated glow at frame 15, a keyframe with the new value occurs on frame 15.
In the Spline Editor, control points can also be added directly to a spline by clicking on the spline
where you want to add the new keyframe.

Adding Keyframes at the Playhead

If you want to add a new keyframe at the current playhead location, pressing Command-K on the
keyboard or right-clicking in the graph and choosing Set Key adds a keyframe under the playhead.

Adding Equal Keyframes

If you want to hold a value over several frames, right-clicking in the graph area and choosing Set Key
Equal To displays a submenu to add a new keyframe with a value equal to the next or the
previous keyframe.

Chapter 10 Animating in Fusion’s Spline Editor 255

 To hold a value between two keyframes choose from the Set Key Equal To submenu

Locked and Unlocked Controls Points

When animating the Center X/Y or Pivot X/Y parameters on any tool, you create a displacement spline
in the Spline Editor. The displacement spline represents the relative offset position of the animated
object along its path. Since the displacement spline is relative, keyframes use a value between 0.0
and 1.0. A displacement value of 0.0 in the Spline Editor indicates that the object is at the very
beginning of a path. A value of 1.0 indicates that the object is positioned at the end of the path.

The displacement spline represents the relative position along a motion path

Displacement paths are composed of locked and unlocked points. Whether a point is locked is
determined by how you added it to the polyline. Locked points on the spline have an associated point
in the viewer’s motion path; unlocked points do not have a corresponding point in the viewer’s motion
path. Each has a distinct behavior, as described below.

TIP: You can convert displacement splines to X and Y coordinates by right-clicking over the
motion path in the viewer and choosing Path#: Polyline > Convert to X/Y Path.

Chapter 10 Animating in Fusion’s Spline Editor 256

Locked Points
Locked points are the motion path equivalents of keyframes. They are created by moving the
playhead position and changing the parameter value. These points indicate that the animated object
must be in a specified position on a specified frame. Since these keyframes are only related to
position along the path, they can only be moved horizontally along the spline’s time axis.
The locked points appear as larger-sized lock icons in the Spline Editor. Each locked key has an
associated point on the motion path in the viewer.
Deleting a locked point changes the overall timing of the motion.

Unlocked Points
Unlocked points are created by clicking directly on the spline in the Spline Editor. These points give
additional control over the acceleration along the motion path without adjusting the path itself.
Conversely, you can add unlocked points in the viewer to control the shape of the motion path without
changing the timing.
You can change an unlocked point into a locked point, and vice versa, by selecting the point(s),
right-clicking, and choosing Lock Point from the contextual menu.
For more information on motion paths and locked keyframes, see Chapters 60 and 62 in the
DaVinci Resolve manual or Chapters 9 and 11 in the Fusion Studio manual.

Selecting, Moving, and Deleting Keyframes

The placement of keyframes greatly affects the style of the animation. Using the graph, you can select
keyframes and move them up or down to change their value or move them left and right to change the
timing. Keyframes can be copied and pasted between splines and parameters.

Methods of selecting keyframes:

– Click directly on a keyframe on the spline, or drag a bounding box around the keyframe.
– Drag a bounding box that encompasses multiple keyframes to select more than one.
– To add or remove a keyframe from the current selection, hold down the Command key while
selecting the keyframes. This will remove currently selected keyframes and add currently
unselected keyframes.
– Press Command-A or right-click in the graph area and choose Select Points > Select All from the
contextual menu to select all keyframes from the active splines.

Moving Keyframes
You can freely move keyframes with the mouse, keyboard, or the edit point controls. Keyframes can
even pass over existing points as you move them. For instance, if a keyframe exists on frame 5 and
frame 10, the keyframe at frame 5 can be repositioned to frame 15.

To move keyframes with the mouse:

– Drag the selected keyframe to its new position in the graph. If more than one keyframe is
selected, all selected keyframes will be moved simultaneously.
– Hold down the Option key before dragging the keyframe to constrain its motion to a single axis.

To move keyframes with the keyboard:

– The Up and Down Arrow keys will adjust the value of the keyframes by a small amount.
– The current scale of the graph determines the degree of vertical movement applied to the value
with each key press. The closer the zoom in the spline, the finer the adjustment.
– Hold down the Shift key while pressing the Up or Down Arrow keys to increase the value
adjustment in larger increments.

Chapter 10 Animating in Fusion’s Spline Editor 257

To move keyframes using edit fields:
– The Value and Time Editors are found on the far right of the toolbar. These number fields allow
explicit values to be entered for selected keyframes. These controls are explained in more depth
later in this chapter.

To delete one or more keyframes:

– Select one or more keyframes and press the Delete or Backspace key on the keyboard. This
only removes the keyframes; it does not remove the spline, even if there are no keyframes on the
spline. To remove the spline, right-click over the parameter in the Inspector and choose Remove
[parameter name].

Showing Key Markers

You can adjust the position of the keyframes in time, without worrying about manipulating splines, by
using the key markers. The horizontal time axis can show markers that indicate the position of each
keyframe. The display of these markers is enabled by right-clicking in the graph and choosing Show >
Key Markers from the contextual menu, or by clicking on the Show Key Markers button in the toolbar.

The key markers show keyframes in the horizontal axis using the same color as the splines

Copying and Pasting Keyframes

To precisely match animation, Keyframes can be copied to a new location on the same spline or onto
completely different splines and different tools.
There are two options in the graph’s contextual menu for copying keyframes. Choosing Copy Points
(Command-C) copies all selected points. Choosing Copy Value copies a single point identified by the
pointer from multiple selected points. This does not deselect your selection set, and you can pick out
numbers as needed.

To copy and paste points to a new location on the same spline:

1 Select the desired keyframes on the spline.
2 Right-click over the spline and choose Copy Points from the contextual menu or press
Command-C.
3 Click in an empty area of the graph to deselect all the copied points.
4 Move the playhead to the area of the spline where you want the points pasted and press
Command-V.
Or, move the pointer over the spline where you want the points pasted, and when the spline
highlights, right-click and choose Paste Points/Value.

Chapter 10 Animating in Fusion’s Spline Editor 258

Alternatively, you can copy and paste keyframes by dragging them with the mouse. After you select
the points, hold down the Command key and drag the points along the spline to where you want
them pasted.

To copy and paste keyframes from one spline to another:

1 Make one spline the active visible spline and select the desired keyframes on the spline.
2 Right-click over the spline and choose Copy Points from the contextual menu or press
Command-C.
3 Set the spline to viewed or disabled using the status checkbox next to the spline’s name in
the header.
4 Make the destination spline the active visible spline and select the keyframe on the spline where
the new keyframes should be pasted.
5 Right-click and choose Paste Points/Value or press Command-V.

You can copy a single point’s value from a group of selected points. Since this process does not
deselect the selected set, you can continue picking out values as needed without having to
reselect points.

To copy and paste a keyframe value:

1 Make one spline the active visible spline and select all the keyframes on the spline.
2 Right-click over a single point and choose Copy Value from the contextual menu.
3 Set the spline to viewed or disabled using the status checkbox next to the spline’s name in
the header.
4 Make the destination spline the active visible spline and select the keyframe on the spline where
the new keyframe should be pasted.
5 Right-click and choose Paste Points/Value or press Command-V.

Keyframes can also be pasted with an offset, allowing you to duplicate a spline shape but increase the
values or shift the timing using an offset to X or Y.

To paste keyframes points and values with an X or Y offset:

1 Make one spline the active visible spline and select the desired keyframes on the spline.
2 Right-click over the spline and choose Copy Points from the contextual menu or press
Command-C.
3 Set the spline to viewed or disabled using the status checkbox next to the spline’s name in
the header.
4 Make the destination spline the active visible spline and select the keyframe on the spline where
the new keyframes should be pasted.
5 Right-click and choose Paste with Offset. In the Offset dialog, enter the Y value, which will be
added to the values of the pasted keyframes.

TIP: You cannot copy and paste between different spline types. For instance, you cannot
copy from a Bézier spline and paste into a B-spline.

Chapter 10 Animating in Fusion’s Spline Editor 259

Time and Value Editors
The Time and Value Editors in the lower-right corner of the Spline Editor are used to change the
position and parameter value of a keyframe by entering a number into the number field for each
button. Each field can switch between three modes that help modify the time and value of a keyframe
in three precise but distinct ways. The default mode for each field takes the explicit frame number or
parameter value at which you want the keyframe set. The other modes offset and scale the keyframe’s
position or value.

Use the number fields to enter in a value or a

specific time to change the selected keyframe

Time Editor
The Time Editor is used to modify the current time of the selected keyframe. You can change the Time
mode to enter a specific frame number, an offset from the current frame, or spread the keyframes
based on the distance (scale) from the playhead. You can select one of the three modes from the Time
mode drop-down menu.

Three time editing modes are selectable

from the Time mode drop-down menu

Time
The number field shows the current frame number of the selected control point. Entering a new frame
number into the field moves the selected control point to the specified frame. If no keyframes are
selected or if multiple keyframes are selected, the field is empty, and you cannot enter a time.

Time Offset
Selecting T Offset from the drop-down menu changes the mode of the number field to Time Offset. In
this mode, the number field offsets the selected keyframes positively or negatively in time. An offset of
either positive or negative values can be entered. For example, entering an offset of 2 moves a
selected keyframe from frame 10 to 12. If multiple keyframes were selected in the previous example, all
the keyframes would move two frames forward from their current positions.

Time Scale
Selecting T Scale from the drop-down menu changes the mode of the number field to Time Scale. In
this mode, the selected keyframes’ positions are scaled based on the position of the playhead. For
example, if a keyframe is on frame 10 and the playhead is on frame 5, entering a scale of 2 moves the
keyframe 10 frames forward from the playhead’s position, to frame 15. Keyframes on the left side of the
playhead would be scaled using negative values.

Value Editor
The Value Editor is used to modify the selected keyframe’s parameter value using one of three Value
modes. You can change the Value mode to enter a specific value for a parameter, an offset from the
value, or to spread the values. The mode is chosen from the Value mode drop-down menu.

Chapter 10 Animating in Fusion’s Spline Editor 260

 Three value editing modes are selectable
from the Value mode drop-down menu

Value
The number field shows the value of the currently selected keyframes. Entering a new number into the
field changes the value of the selected keyframe. If more than one keyframe is selected, the displayed
value is an average of the keyframes, but entering a new value will cause all keyframes to adopt
that value.

Value Offset
Choosing Offset from the drop-down menu sets the Value Editor to the Offset mode. In this mode, the
value for the selected keyframes are offset positively or negatively. An offset of either positive or
negative values can be entered. For example, entering a value of -2 changes a value from 10 to 8. If
multiple keyframes are selected, all the keyframes have their values modified by -2.

Value Scale
Choosing Offset from the drop-down menu sets the Value Editor to the Scale mode. Entering a new
value causes the selected keyframes’ values to be scaled or multiplied by the specified amount. For
example, entering a value of 0.5 changes a keyframe’s value from 10 to 5.

Modifying Spline Handles

All Bézier spline keyframes have a pair of control handles to shape the spline as it passes through the
key point. These handles are only displayed when the keyframe is selected. Initially, these handles are
set to linear, creating straight line changes between keyframes. However, any control point can be
made smooth by right-clicking over it and choosing Smooth or pressing Shift-S.

Bézier splines can mix linear and smooth curves

Dragging on a keyframe’s handles adjusts the slope of the segments passing through the spline. By
default, the two control handles on a control point are locked together so that if one moves, the one
on the other side moves with it. This maintains a constant tension through the keyframe. There are
situations, however, when it is desirable to modify these control handles separately for a more
pronounced curve or effect.

To temporarily break the control handles on a Bézier spline, moving one

independently of the other:
1 Select the control point to be modified.
2 Hold down the Command key and drag one of the control handles. They will now move
independently of each other, as long as the Command key is held down.

Chapter 10 Animating in Fusion’s Spline Editor 261

To treat all Bézier handles as independent in the Spline Editor:
– Right-click in the graph and choose Independent Handles from the Options contextual menu.
Enabling this option causes all the Bézier handles to be independent. This is the same as using the
Command key when moving a handle, except it is applied to all control points until it is disabled.

Reducing Points
When there are too many control points too close together on a spline, you can choose Reduce Points
to decrease their number, making it easier to modify the remaining points. The overall shape of the
spline is maintained as closely as possible while eliminating redundant points from the path.

To reduce the number of points on a spline:

1 Select the range of keyframes you want to thin out.
2 Right-click in the graph area and choose Reduce Points from the contextual menu.
3 When the Reduce Points dialog appears, drag the slider to a lower value.

You can set the slider value as low as possible as long as the spline still closely resembles the shape
of your original spline.

TIP: When the value is 100, no points will be removed from the spline. Use smaller values to
eliminate more points.

Filtering the Spline Editor

The animation splines for multiple parameters can be displayed within the Spline Editor simultaneously,
and Fusion offers several ways for you to choose which spline to view and which to edit.
A complex composition can easily contain dozens, if not hundreds, of animation curves. As a
composition grows, locating a specific spline can become more difficult. There are two ways to filter
the splines shown in the Spline Editor: display selected tools only or create a filter to show only
certain tools.
The Spline Editor includes different ways to control which splines are displayed. The majority of these
options are available in the Options menu, located in the upper-right corner of the Spline Editor panel.

The Options menu is used to control which

splines are displayed in the Spline Editor

Chapter 10 Animating in Fusion’s Spline Editor 262

 – Show Only Selected Tools: You can choose to limit the splines displayed in the Spline Editor by
showing only the splines from selected tools. Choosing this option at the top of the Options menu
displays only the splines for tools currently selected in the Node Editor.
– Show All/None: The default behavior of the Spline Editor displays all the splines for all the nodes
with animated parameters. You can override this by enabling Show Only Selected Tools in the
Options menu. You can also disable the Show All setting by choosing Show None, in which case
the Spline Editor remains empty.
– Expose All Controls: The Expose All Controls option is a way of not filtering the parameters.
Choosing this option displays all parameters in the Spline Editor header for all nodes in the
Node Editor. It can be a fast way of activating one of the parameters and automatically adding an
animation spline for it if one does not exist.
With a large number of nodes displayed, which themselves might have a large number of
parameters, this might lead to cluttering and slowing down the interface. This option is most
effective when used in conjunction with the Show Only Selected Tool option to limit the number of
nodes and parameters displayed and yield optimum performance.
– Follow Active: The Follow Active option is located by right-clicking in the graph and choosing
Options > Follow Active. This option provides a way to filter the splines in the graph while not
filtering the header list of tools. Where the Show Only Selected Tool option hides other tools in
the header, the Follow Active option leaves the header displaying all the tools but automatically
enables only the splines of the Active tool.

Working with Filters

Filters allow you to select the specific types and classes of tools shown in the Spline Editor and
Keyframes Editor. For example, you can make a filter that shows only particle nodes or one that only
shows color correction and brightness/contrast tools.

To create a filter:
1 From the Options menu, choose Create/Edit Filters.
2 Click the New button to create a new filter and name the new filter in the dialog box.
3 Enable a checkbox next to the entire category or the individual tools in each category to
determine the tools included in the filter.

Enable each tool you want to keep in the Spline Editor when the filter is selected

The Invert All and Set/Reset All buttons can apply global changes to all the checkboxes, toggling the
selected states as described.

Chapter 10 Animating in Fusion’s Spline Editor 263

To switch the selection state of the categories when creating a filter list:
1 Click the Invert All button.
2 After configuring the custom filter, click the Save button to close the Settings dialog and save
the filter.

To enable all checkboxes or disable all checkboxes:

1 Click the Set/Reset All button as many times as needed until all categories are either checked or
unchecked.
2 After configuring the custom filter, click the Save button to close the Settings dialog and save
the filter.

To apply a filter to the Spline Editor:

– Choose the desired filter by name from the Options menu. The filter applies to both the Spline
Editor and the Timeline.

Each filter you create is listed in the Options menu

To disable a filter and show all tools in the Spline Editor again:
– Choose Show All from the Options menu.

Changing a Spline’s Status

The Spline header is a hierarchical list of animated parameters and their parent nodes. Clicking the
disclosure arrow next to a tool’s name reveals all the names of the animated parameters on that tool.
Clicking directly on the parameter name in the Spline header activates that spline for
display and editing.

The Spline header with the Brightness Contrast

tool and its animated Gain and Gamma parameters

Tool Status Checkbox

Next to the name of each spline is a checkbox that indicates the spline’s status. When you select a
parameter name, the checkbox becomes active, allowing you to see and edit the spline in the graph.
There are three selection modes for each checkbox: active, viewed, and disabled. Clicking directly on

Chapter 10 Animating in Fusion’s Spline Editor 264

the checkbox will toggle it between these three states. Changing the state of the parent node
checkbox sets the state for all splines for that node.
– Active: When the checkbox is enabled with a check mark, the spline is displayed in the graph and
can be edited.
– Viewed: When the checkbox is enabled with a solid gray box, the spline is visible in the graph but
cannot be edited. It is read-only.
– Disabled: When the checkbox is clear, the spline is not visible in the graph and cannot be edited.

Checkboxes determine which splines

are visible and editable

Selection States
There are three selection options, labeled Select All Tools, Deselect All Tools, and Select One Tool,
that determine how the items in the Spline Editor header behave when a checkbox or label is selected
to activate a spline. These states are located in the Options menu in the upper-right corner of the
Spline Editor.

The Options menu selection states make it easier to

select and deselect all parameters in the header

– Select All Tools: Choosing this option activates all splines for editing.
– Deselect All Tools: Choosing this option sets all spline checkboxes to disabled.
– Select One Tool: This option is a toggle. When Select One Tool is chosen from the menu, only
one spline in the header is active and visible at a time. Clicking on any spline’s checkbox will set
it to active, and all other splines will be cleared. When disabled, multiple splines can be active
in the header.

Chapter 10 Animating in Fusion’s Spline Editor 265

Selection Groups
It is possible to save the current selection state of the splines in the header, making selection groups
that can easily be reapplied when needed. To create a selection group, right-click over any parameter
in the header or in an empty area of the graph and choose Save Current Selection from the contextual
menu. A dialog will appear to name the new selection.
To reapply the selection group, choose the selection group by name from the Set Selection menu in
the same contextual menu. Other context menu options allow selection groups to be renamed
or deleted.

Reshaping Splines Using the Toolbar

There are several ways to manipulate the shape of a spline, thereby altering the animation that spline
generates. Other than manually adjusting Bézier handles, you can quickly squish, stretch, loop, and
reverse a spline. You can also quickly change the interpolation between keyframes from the default
linear motion to more natural smooth motion. All these options are provided at the bottom of the
Spline Editor in the toolbar. The toolbar is divided into different groups for setting interpolation,
reversing splines, looping splines, time stretching, and reshaping splines.

Interpolation
Keyframes are specific frames in an animation where control points are set to exact values on a given
parameter. Interpolation is the method used to fill in the unknown values between two keyframes.
Fusion automatically interpolates between two keyframes. However, you may want to modify the
interpolation to achieve a specific style of animation. The Spline Editor includes several interpolation
methods you can choose from using the toolbar.

Interpolation buttons in the toolbar: Smooth,

Linear, Invert, Step In, and Step Out

Smooth
A smoothed segment provides a gentle keyframe transition in and out of the keyframe by slightly
extending the direction handles on the curve. This slows down the animation as you pass through the
keyframe. To smooth the selected keyframe(s), press Shift-S or click the toolbar’s Smooth button.

Smooth interpolation between keyframes

Chapter 10 Animating in Fusion’s Spline Editor 266

Linear
A linear segment effectively takes the shortest route between two control points, which is a straight
line. To make the selected keyframe(s) linear, press the L key or click the Linear button in the toolbar.

Linear interpolation between keyframes

TIP: Invert is used only for non-animated LUT splines, which are currently only available in the
LUT Editor window.

Step In/Step Out

On occasion, it is not desirable to have any interpolation between two keyframes. Instead, the value of
one keyframe may hold its value until another keyframe changes it. For these cases, use the Step In or
Step Out mode.
Step In causes the value of the previous keyframe to hold, then jump straight to the value of
the next keyframe.

Step In holds a value until the next

keyframe is reached in the comp

Step Out causes the value of the selected keyframe to hold right up to the next keyframe.

Step Out switches immediately to the

next keyframe value in a comp

Chapter 10 Animating in Fusion’s Spline Editor 267

Step In and Step Out modes can be set for selected keyframes by clicking on the toolbar buttons for
each mode, or by right-clicking and choosing the appropriate option from the contextual menu. The
keyboard shortcuts I and O can also be used to enable Step In and Step Out on selected keyframes.

Reversing Splines
Reverse inverts the horizontal direction of a segment of an animation spline. To apply reverse, choose
a group of points in a spline and click the Reverse button, or right-click and choose Reverse from the
contextual menu, or press the V key. The group of points is immediately mirrored horizontally in the
graph. Points surrounding the reversed selection may also be affected.

The Reverse button in the toolbar

Looping Splines
It is often useful to repeat an animated section, either infinitely or for a specified number of times, such
as is required to create a strobing light or a spinning wheel. Fusion offers a variety of ways to repeat a
selected segment.

The various Loop buttons in the toolbar

Set Loop
To repeat or loop a selected spline segment, select the keyframes to be looped. Select Set Loop from
the contextual menu or click on the Set Loop button in the toolbar. The selected section of the spline
repeats forward in time until the end of the global range, or until another keyframe ends the
repeating segment.

A looped section in the graph

Changing and Removing the Loop

You can change the looped segment by modifying any of the keyframes or control points originally
used to create the loop. Simply select one of the originating key points, make any necessary
modifications, and the looped segment updates. To remove the loop, select the keyframes you used
to create the loop, and then click the Loop button in the toolbar.

Chapter 10 Animating in Fusion’s Spline Editor 268

Ping-Pong
The Ping-Pong Loop mode repeats the selected segment, reverses each successive loop, and then
repeats. Ping-pong looping can be enabled on the selected segments from the context menu or
the toolbar.

A ping-pong section in the graph

Relative Loop
The Relative Loop mode repeats the segment like the Loop, but each repetition adds upon the last
point of the previous loop so that the values increase steadily over time.

A Relative Loop section in the graph

Looping Backward
You can choose Set Pre-Loop by right-clicking in the graph area and choosing it from the contextual
menu. This option contains the same options for looping as the Loop option buttons in the toolbar,
except that the selected segment is repeated backward in time rather than forward.

Repeating Splines X Number of Times

You can duplicate splines and repeat them a set number of times by right-clicking in the graph area
and choosing Duplicate from the contextual menu. Duplicated splines are like looped splines, except
that the selected segment repeats only a specified number of times, and each repetition is a copy
rather than an instance. Adjustments to the original segment do not alter the shape of its repetitions.
The Duplicate modes are only accessed from the Duplicate contextual menu, which reveals a
submenu with all the looping modes described above. Selecting any of these modes opens a dialog in
which the number of repetitions can be entered.

Chapter 10 Animating in Fusion’s Spline Editor 269

The Duplicate contextual submenu

Gradient Extrapolation
You can choose Gradient Extrapolation by right-clicking in the graph area and choosing it from the
contextual menu. This option continues the trajectory of the last two keyframes.

The Gradient Extrapolation applied to the spline

Time Stretching
Time Stretching allows for a selected group of keyframes to be proportionally stretched or squashed.
This allows you to change the duration of the animation while keeping the relative distance between
each keyframe. To enable spline stretching, select the group of keyframes that you want to time
stretch, and then choose Modes > Time Stretching from the graph’s contextual menu or click the Time
Stretch button in the toolbar.

The Time Stretch button in the toolbar

Chapter 10 Animating in Fusion’s Spline Editor 270

When you have more than one keyframe selected on the spline, enabling Time Stretch surrounds the
outer keyframes with two vertical white bars. Drag on the white vertical bars to stretch or shrink the
timing of the spline segments within the bars. Drag these bars back and forth to stretch or squash the
spline segment.

The Time Stretch bars in the graph

TIP: If no keyframes are selected when you enable Time Stretch, drag a rectangle to set the
boundaries of the Time Stretch.

To disable the Time Stretching mode:

– Click on the Time Stretch button in the toolbar again or reselect Modes > Time Stretching from the
contextual menu.

Shape Box
The Shape Box transform mode is similar to Time Stretching; however, it can adjust the vertical scaling
of keyframe values as well as time.

To enable the Shape Box, do one of the following:

– Select a group of keyframes, and then choose Modes > Shape Box from the contextual menu.
– Select a group of keyframes, and then click the Shape Box button in the toolbar.
– Select a group of keyframes, and then press Shift-B to enable or disable the Shape Box mode.

The Shape Box button in the toolbar

Chapter 10 Animating in Fusion’s Spline Editor 271

A white rectangle outlines the selected points when the mode is enabled. To scale, skew, or stretch
the spline, drag on any of the control points located around the box. To move all the keyframes, drag
on the box edges.

The Shape Box in the graph

TIP: If no points are selected, or if you want to select a new group of keyframes, you can drag
out a new rectangle at any time.

Ease In/Out
For a more precise way to adjust the length of Bézier direction handles attached to selected
keyframes, you can use the Spline Ease dialog. To show the dialog, select a keyframe in the graph,
and then choose Edit > Ease In/Out from the graph’s contextual menu or press T on the keyboard.
The Ease In/Out controls appear above the graph area. You can drag over the number fields to adjust
the length of the direction handles or enter a value in the fields.

The Ease In/Out controls above the graph area

Clicking the Lock In/Out button will collapse the two sliders into one, so any adjustments apply to both
direction handles.

Chapter 10 Animating in Fusion’s Spline Editor 272

Importing and Exporting Splines
Spline shapes can be imported and exported from or to an ASCII text file. This makes it easier to save
complex spline curves for later reuse, or to transfer tracking, path, and animation data from one
application to another. Exported splines are assigned the file extension .spl for easy identification.

To export a spline:
1 Select the active spline in the Spline Editor.
2 Right-click on the spline in the graph area to display the contextual menu.
3 Choose from three format options in the submenu.
4 Enter a name and location in the file browser dialog, and then click Save.

Exporting a spline gives you three options. You can export the Samples, Key Points, or All Points.
Samples adds a control point at every frame to create an accurate representation of the spline.
Key Points replicates the control point positions and values on the spline using linear interpolation.
All Points exports the spline as you see it in the Spline Editor, using the same position, value, and
interpolation.

To import a spline:
1 Add an animation spline for the parameter.
2 In the Spline Editor, right-click on the animation spline and select Import Spline from the
contextual menu.
3 In the File Browser dialog, select the spline curve .spl file, and then click Open.

Importing a new curve will replace any existing animation on the selected spline.

Chapter 10 Animating in Fusion’s Spline Editor 273

Chapter 11

Animating with
Motion Paths
Layers and 3D objects can move along a designated spline shape to create motion
path animations. This chapter discusses how you can create, edit, and use motion
paths in Fusion.

Contents
Animating Using Motion Paths  275
Types of Motion Paths  276
Polyline Path  276
Path Modifier  279
Controlling Speed and Orientation along a Path  280
XY Path  282
Types of Control Points  284
Locked Points  284
Unlocked Points  287
Locking and Unlocking Points  288
Tips for Manipulating Motion Paths  288
Compound Motion Paths Using Path Centers  288
Copying and Pasting Motion Paths  289
Removing Motion Paths  289
Recording Motion Paths  290
Importing and Exporting Polylines  290
Native Format  290

Chapter 11 Animating with Motion Paths 274

Animating Using Motion Paths
Motion paths are created from splines (polylines) including paint strokes and masks that have a path
modifier applied to them. The path modifier defines the movement that transforms ordinary spline
shapes into motion paths. You apply the path modifier either explicitly to the Center X/Y or Pivot X/Y
parameters or implicitly by keyframing the Center or Pivot parameters. In either case, the spline path is
displayed and then visually adjusted in the viewers. For all motion paths, the coordinate control in the
viewer represents the position of an object or effect, such as a Transform or Rays node’s center that
moves along the path. Coordinate controls are represented onscreen with vertical and horizontal
positioning arrows or an X.

A Center Offset onscreen control

for an object’s Transform node

The following nodes have parameters that can be animated using path modifiers to move an image
around the composition. These include, but are not limited to:
– Transform: Center X/Y can be animated to move an image around.
– DVE: Center X/Y can be animated to move an image around.
– Merge: Center X/Y can be animated to move the Foreground connected image around.
– Paint: Stroke Controls > Center X/Y can be animated to move a stroke around.
– Camera 3D node: Translation X/Y/Z
– Shape 3D node: Translation X/Y/Z

The following nodes have parameters that can be animated using paths to alter the direction of a
visual effect. These include, but are not limited to:
– Directional Blur: Center X/Y can be animated to change the direction of the blur.
– Hot Spot: Primary Center X/Y can be animated to move the hot spot around.
– Rays: Center X/Y can be animated to change the angle at which rays are emitted.
– Polygon/BSpline/Ellipse/Rectangle/Triangle mask: Center X/Y can be animated to
move the mask.
– Corner Positioner: Top Left/Top Right/Bottom Left/Bottom Right X/Y can be animated to move
each corner of the corner-pinned effect.
– Vortex: Center X/Y can be animated to move the center of the warping effect.

NOTE: It’s not possible to add a motion path to a one-dimensional value, such as blur
strength or merge angle. However, you can use the Spline Editor to edit these kinds of values
in a visual way.

Chapter 11 Animating with Motion Paths 275

Types of Motion Paths
There are three types of motion paths: Polyline paths, XY paths, and 3D motion paths for 3D scenes.
– A Polyline path is generated by applying the path modifier. It uses two splines to define the path;
one for the shape of the path displayed in the viewer, and a Displacement spline for the speed of
the object along the path, displayed in the Spline Editor. The Polyline path is the default type of
path modifier, and most documentation in this chapter assumes that this type is used.
– The XY path modifier employs a spline for the X position of the point and another
for the Y position. The XY path modifier is explained in detail later in this chapter.
– 3D motion paths pertain only to positional controls within 3D scenes.

Polyline Path
Polyline paths are the easiest motion paths to work with. You can use the spline shape in the viewer to
control the shape of the path, while a single Displacement curve in the Spline Editor is used to control
the acceleration along the path. The most obvious way to create a Polyline motion path is by
keyframing the Center X/Y parameter of a Transform node in the Inspector.

To create a Polyline motion path using the Center X/Y parameter in the Inspector:
1 Position the playhead on the frame where the motion will begin.
2 In the Inspector, click the gray Keyframe button to the right of the Center X and Y parameters.
This action applies the path modifier in the Modifiers tab in the Inspector.
3 Adjust the Center X and Y for the first keyframe position.
4 Position the playhead on the frame where the motion should change or stop.
5 In the Inspector, change the Center X and Y parameters to set a keyframe for the new location
automatically.
6 In the viewer, modify and refine the motion path by selecting a control point and using any of the
spline controls in the viewer toolbar.
7 Open the Spline Editor and adjust the Displacement spline to control the speed of the object
along the path.

Keyframing Center X/Y is not the only way to apply the path modifier. An alternative method is to apply
the path modifier to the Center X/Y parameter either in the Inspector or using the coordinate control in
the viewer.

To apply a Polyline path to the Center X/Y parameter:

1 Position the playhead on the frame where the motion will begin.
2 Position the center coordinate control for the object, or effect, at its starting position.
3 Right-click the onscreen center coordinate control in the viewer (or over the Center X and Y
parameter in the Inspector), and choose Path from the contextual menu for that control.

Chapter 11 Animating with Motion Paths 276

 Applying a path modifier to the center of a Transform node

The object now has a path modifier applied, so without setting a keyframe you can drag the object
to begin creating a motion path in the viewer.
4 Move the playhead to a new frame.
5 Drag the onscreen coordinate control or adjust the Offset or Center values in the Inspector. A
keyframe is automatically created on the motion path, and a polyline is drawn from the original
keyframe to the new one.

Two keyframes spaced several frames apart display a

motion path showing the direction of animation.

6 The position of the center control is interpolated between the two keyframes. You can continue
adding points by moving the playhead and adjusting the object’s position until the entire motion
path you need is created. For motion paths, there’s no need to close the spline path; you can
leave it open.
7 Upon completion, set the polyline to Insert and Modify mode by selecting a point on the path
and pressing Command-I or clicking the Insert and Modify button on the toolbar. Don’t worry too
much about the overall shape of the motion path at this point. The shape can be refined further by
adding additional points to the polyline and modifying the existing points.

Chapter 11 Animating with Motion Paths 277

A final alternative method for creating a motion path is to draw a spline shape first and then connect a
path modifier to the spline. Using any of Fusion’s spline tools, you can draw the shape of the path and
then connect the path modifier to the published spline. Once the path modifier and the published
spline are connected, you can keyframe the Displace parameter to move an image along the path.
This method is useful when you want to use a paint stroke or mask shape as a motion path.

To use a mask shape or paint stroke as a path:

1 Draw a polygon mask or paint stroke in the shape of the path you want to use.
You do not have to close the mask shape.

An open Polyline mask shape

2 When done drawing the shape, click the Insert and Modify button in the viewer toolbar to leave
the mask shape as an open spline.
3 At this point you can select any of the control points along the spline and press Shift-S to make
them smooth or Shift-L to make them linear.
All mask polylines have animation enabled by default, but that is usually not desirable for a motion
path. You will need to remove this keyframe animation if you are using a mask shape.
4 At the bottom of the Inspector, right-click on the “Right-click here for shape animation” label and
choose Remove Polygon1Polyline.

Right-clicking at the bottom of the Inspector to

remove auto-keyframing from a mask shape

5 Right-click at the bottom of the Inspector again and select Publish to give other nodes access to
this spline shape. (For a paint stroke, you will need to make the Stroke editable first by clicking the
Make Editable button in the Stroke Controls.)

Chapter 11 Animating with Motion Paths 278

 This enables the Modifiers tab with the Published Polyline modifier. This published spline can be
used to define the shape of splines in other nodes.
6 Connect a Transform node to the image you want to have follow the path.
7 Right-click over the Center X/Y parameter in the Inspector and choose Path.
This adds a path modifier into the Modifiers tab.
8 In the Inspector, click the Modifiers tab and double-click the Path1 heading to open its parameters.
The Displacement parameter already has a keyframe on it automatically. You’ll want to remove that
so you can set your own.
9 Click the red keyframe button to the right of the Displacement parameter to remove it.

Remove the automatic displacement keyframe

prior to creating your own keyframes.

10 At the bottom of the Modifiers tab, right-click on “Right-click here for shape animation” and choose
Connect To > Polygon1Polyline.
11 To quickly see where your object has gone, drag the Displacement slider back and forth.
12 You may want to use the Size parameter to adjust the size of the overall path.
The Displacement slider is meant to be keyframed for animating the object along the path.

Path Modifier
In terms of functionality, it makes no difference which method you use to generate the path modifier.
All the above methods are just different ways to get to the same point. Whichever way you decide to
add the path modifier, the Modifiers tab contains controls for the path.

Creating a path adds controls to the

Modifiers tab in the Inspector.

Chapter 11 Animating with Motion Paths 279

You can use the path modifier controls in the Inspector to change the position, size, and rotation of the
entire path shape. The Displacement parameter is represented as a spline in the Spline Editor, which
determines the object’s position along the path and the Heading Offset is used for the orientation
along the path.

Controlling Speed and Orientation along a Path

Every Polyline path has an associated Displacement spline in the Spline Editor. The Displacement
spline represents acceleration, or the position of the animated object along its path, represented as a
value between 0.0 and 1.0.

The Displacement curve of a Poly path represents the acceleration of an object on a path.

Smaller values are closer to the beginning of a path, while larger values are increasingly closer to the
end of the path.
For instance, let’s say you have a bumblebee that bobs up and down as it moves across the screen. To
have the bee accelerate as it moves up and down but slow down as it reaches its peaks and valleys
you use the Displacement curve.

A curvy path defined by a spline shape

The curved shape path does not define how fast the bee moves. The speed of the bee at any point
along the path is a function of the Displacement parameter. You can control the Displacement
parameter either in the Modifiers tab or in the Spline Editor.

Chapter 11 Animating with Motion Paths 280

To use the Displacement parameter to control the speed of an object on a path, do the following.
1 Position the playhead at the start of the animation.
2 In the Modifiers tab, drag the Displacement parameter to 0.0. This positions the object at the start
of the path.
3 Click the Keyframe button to the right of the Displacement parameter.
4 Position the playhead somewhere further into the comp and drag the Displacement parameter
until the object is where you want it to be based on the current frame.
5 Continue updating the playhead and the Displacement parameters at key points in the comp until
you have reached the end of the path.

After the initial animation is set, you can use the Displacement curve in the Spline Editor to adjust
the timing.

To adjust the timing of an object along a path:

1 Open the Spline Editor and enable the Displacement spline in the header.
2 Move control points horizontally closer together to increase the speed between the two points
while maintaining the location of the object along the path.
A longer, flatter curve between two points indicates a slower rate of change.
3 Drag a control point up or down to change its location on the path while maintaining the timing
between two points.

A Displacement curve in the Spline Editor

TIP: Holding down the Option key while clicking on the spline path in the viewer will add a
new point to the spline path without adding a Displacement keyframe in the Spline Editor.
This allows you to refine the shape of the path without changing the speed along the path.

Using a Path Modifier to Adjust Orientation

The Heading parameter is used to adjust the orientation of the object along the path. For instance, if
you want the bee in our animation to auto-orient based on the direction of the path, you can connect
the bee’s angle to the Heading parameter.

To connect the Heading to an object’s angle:

1 Right-click over an object’s angle parameter in a Transform node.
2 Choose Connect To > Path > Header.

Chapter 11 Animating with Motion Paths 281

In our example comp, the bee now auto-orients as the path descends and rises.

The Transform’s angle parameter connected to the path modifier’s Heading parameter

XY Path
Unlike a Polyline path, the XY path modifier uses separate splines in the Spline Editor to calculate
position along the X-axis and along the Y-axis.

To animate a coordinate control using an XY path modifier:

– Right-click on the center coordinate control in the viewer or the Center X/ Y parameter in the
Inspector, and then choose Modify With > XY Path from the contextual menu.

Adding the XY Path modifier to a Center parameter in the Inspector

At first glance, XY paths work like Polyline paths. To create the path once the modifier is applied,
position the playhead and drag the onscreen control where you want it. Position the playhead again
and move the onscreen control to its new position. The difference is that the control points are only
there for spatial positioning. There is no Displacement parameter for controlling temporal positioning.

Chapter 11 Animating with Motion Paths 282

 XY path modifier controls in the
Modifiers tab of the Inspector

Instead of dragging in the viewer, you can use the controls in the Modifiers tab to create a motion path,
while using the object’s original Inspector controls as an offset to this motion path. You can use the
XYZ parameters to position the object, the Center X/Y parameters to position the entire path, the Size
and Angle to scale and rotate the path, and the Heading Offset control to adjust the orientation.

Using an XY path modifier to animate a piece of text

Using the XY Paths in the Spline Editor

The Spline Editor for the XY path displays the X and Y channel splines. Changes to the path in the
viewer or the Inspector will be displayed as keyframes on these splines in the Spline Editor. Unlike a
Polyline path, XY paths do not include a Displacement curve. The speed of the object along the path
is tied to the path itself and cannot be separated from the timing of the keyframes that define that path.

TIP: XY path and Poly path can be converted between each other from the contextual menu.
This gives you the ability to change methods to suit your current needs without having to
redo animation.

The advantage of the XY path modifier is that you can explicitly set an XY coordinate at a specific time
for more control.

Chapter 11 Animating with Motion Paths 283

XY Path displays X and Y curves in the Spline Editor but does not include a Displacement control.

Switching Default Paths

If you want to change the default path type to XY path, you can choose Fusion > Preferences > Globals
in Fusion Studio or Fusion > Fusion Settings in DaVinci Resolve. In the Settings window, select the
Defaults category and choose XY Path for the Points With drop-down menu. The next time you
keyframe the Center X/Y parameter or choose Animate from CenterX/Y’s contextual menu, an XY path
modifier will be used instead of a Polyline path.

Types of Control Points

The control points along an XY path in the viewer are locked to the control points on the X and Y
curve in the Spline Editor. The number of points are identical, and adding a control point in one place
adds it to the other. That is not the case with a Polyline path. Polyline paths are composed of locked
and unlocked points. Whether a point is locked is determined by how it was added to the Polyline.
Locked points on the motion path in the viewer will have an associated point on the Displacement
spline in the Spline Editor; unlocked points will not have corresponding points. Each has a distinct
behavior, as described below.

Locked Points
Locked points are the motion path equivalents of keyframes. They are created by changing the
playhead position and moving the animated control. These points indicate that the animated control
must be in the specified position at the specified frame.
The locked points are displayed as larger-sized hollow squares in the viewer. Each locked key has an
associated point on the path’s Displacement curve in the Spline Editor.

Locked points on a path in the viewer

Deleting a locked point from the motion path will change the overall timing of the motion.

Chapter 11 Animating with Motion Paths 284

To change the duration of a path using locked points:
1 Connect an object to a Transform node.
2 Position the where you want to start the motion path.

A graphic placed on the right side of the frame

3 Set the playhead at frame 0.

4 Select the Transform node and in the Inspector, click the Keyframe button
to the right of the Center X/Y parameter.
This adds the path modifier and creates the first locked point of the path.
5 Position the playhead at frame 45.
6 Move the object’s center to the lower center of the screen.

Moving the playhead and repositioning the bee adds a second locked point

This sets the second locked point.

7 View the Spline Editor and display Path1’s Displacement spline.

Chapter 11 Animating with Motion Paths 285

 The path’s Displacement spline with locked keyframes

At a value of 0.0, the control will be located at the beginning of the path. When the value of the
Displacement spline is 1.0, the control is located at the end of the path.
8 Select the keyframe at frame 45 in the Displacement spline and drag it to frame 50.
The motion path is now 50 frames long, without making any changes to the motion path’s shape.
If you try to change this point’s value from 1.0 to 0.75, it cannot be done because the point is the
last in the animation, so the value must be 1.0 in the Displacement spline.
9 Position the playhead on frame 100 and move the bee center to the upper-left corner of
the screen.

Moving locked points changes the duration of a motion path without changing its shape

This will create an additional locked point and set a new ending for the path.

Chapter 11 Animating with Motion Paths 286

Unlocked Points
Unlocked points are created when additional points are added to the motion path while in Insert and
Modify modes. These points are used to adjust the overall shape of the motion path, without directly
affecting the timing of the motion. This means you can add whatever unlocked points you want to
reshape a motion path, without that having any effect on the timing of the animation on that path. This
makes it vastly easier to fine-tune a motion path spatially after you’ve perfected the timing of the
animation temporally.
Unlocked points do not have corresponding points on the path’s Displacement spline. They appear in
the viewer as smaller, solid square points.

To add unlocked points to a motion path, do the following:

1 Select a motion path spline by using the Tab key to cycle controls until the path is selected.
2 To insert points along a path, click the Insert and Modify button in the toolbar.
3 Click on the path and create two new points: one half-way between the first and the second
points, and the other half-way between the second and the third points.
The two points just added are not present in the motion path’s Displacement spline. These are
unlocked points, used to shape the motion but unrelated to the timing of the path.

Unlocked points added to the motion path are not displayed on the Displacement spline

You can add unlocked points to the Displacement spline as well. Additional unlocked points in the
Spline Editor can be used to make the object’s motion pause briefly.

To pause motion along a motion path using an unlocked point:

– Select a locked point on the Displacement spline and then hold down the Command key while
dragging the point horizontally to another frame. The point is copied as an unlocked point to the
new frame.

Chapter 11 Animating with Motion Paths 287

 Create a pause in the motion by copying locked points

Knowing the difference between locked and unlocked points gives you independent control over the
spatial and temporal aspects of motion paths.

Locking and Unlocking Points

You can change an unlocked point into a locked point, and vice versa, by selecting the point(s) and
choosing the Lock Point option from the contextual menu.

Tips for Manipulating Motion Paths

There are a variety of ways you can create and edit motion paths in the viewer.

Compound Motion Paths Using Path Centers

Every motion path has a defined center represented by a crosshair. Path centers allow paths to be
connected to other controls and behave in a hierarchical manner, which is an exceptionally powerful
way of creating complex motion by combining relatively simple paths.
A useful example of this technique would be animating the path of a bee in flight. A bee often flies in a
constant figure eight pattern while simultaneously moving forward. The easy way of making this
happen involves two paths working together.
In the following example, the bee would be connected to a first path in a Transform node, which would
be a figure eight of the bee moving in place. This first path’s center would then be connected to
another path defining the forward motion of the bee through the scene via a second Transform node.

To create a compound motion path:

1 Create a figure-eight motion path by keyframing an object or drawing a polyline mask.
(If using a polygon mask, you’ll need to remove the auto-keyframing and publish the mask prior to
the next step.)
2 Add a polyline mask and create a smooth curve spline that travels across the screen.
3 At the bottom of the Inspector, right-click over the “Right-click here for shape animation” label and
choose Remove Polygon Polyline to remove the auto-animation behavior.
4 Right-click over the label again and choose Publish.
5 Select the object’s Transform node and click the Modifiers tab.
6 Right-click over the Path1 Center X/Y parameter and choose Path.
7 At the bottom of Path2, choose Connect To > Polygon: Polyline.
8 Keyframe the Path2 Displacement to move the object along the second path.

Chapter 11 Animating with Motion Paths 288

 Two motion paths working together

Copying and Pasting Motion Paths

It is possible to copy an entire motion path to the clipboard and then paste it onto another node or
path or composition.

Methods of copying and pasting motion paths:

– To copy a motion path: In the Inspector’s Modifiers tab, right-click on the path’s control header
and choose Copy from the contextual menu.
– To cut a motion path out of a node: In the Inspector, right-click on the path’s control header and
choose Cut from the contextual menu.
– To paste the copied path over another path: In the Inspector, right-click on the path’s control
header and choose Paste from the contextual menu.

When pasting a path, the old motion path will be overwritten with the one from the clipboard.

Removing Motion Paths

There are multiple ways to remove or delete a path, and they all involve a right-click contextual menu.
Removing a path modifier does not remove the object or the spline shape; it only removes the
animation from the object and the modifier from the Modifiers tab in the Inspector.

To remove a path, do one of the following:

– In the Modifiers tab, right-click over the Path1 header and choose Delete from the menu.
– In the Inspector, right-click over the Center XY parameter and choose Remove Path1
from the menu.
– Right-click the center coordinate control in the viewer for the object you’re animating, and choose
Remove Path1 from the submenu of the NameOfObject; Center submenu.

Removing an entire motion path at once

Chapter 11 Animating with Motion Paths 289

Recording Motion Paths
You can animate both of the control’s spatial and temporal information at the same time using the
Record mode. This is useful when both position and speed are crucial to achieve the desired result.
Right-click on the desired path and select Record from the contextual menu. This displays a submenu
of available data that may be recorded.
Use the Record Time option in conjunction with the Draw Append mode to create complex motion
paths that will recreate the motion precisely as the path is drawn.
The time used to record the animation may not suit the needs of a project precisely. Adjust the path’s
Displacement spline in the Spline Editor to more correctly match the required motion.

Importing and Exporting Polylines

You can import and export polyline shapes into a common editable ASCII text file or its native format.
These methods are used to save a particularly useful or generic mask or path for future use or for use
in another application, such as Maya or LightWave. You can also import FXF, SSF, or Nuke shape files.

Native Format
To save a polyline shape in Fusion’s native ASCII format, right-click on the header of the Mask node in
the Inspector and select Settings > Save As from the contextual menu. Provide a name and path for
the saved file and select OK to write a file with the .setting extension. This file will save the shape of a
mask or path, as well as any animation splines applied to its points or controls.
To load the saved setting back into Fusion, you first create a new polyline of the same type, and then
select Settings > Load from the mask’s context menu or drag the .setting file directly into the
Node Editor.
If you want to move a polyline from one composition to another, you can also copy the
node to the clipboard, open your second composition, and paste it from the clipboard into the new
composition.

Chapter 11 Animating with Motion Paths 290

Chapter 12

Using Modifiers,
Expressions, and
Custom Controls
Some of the most powerful aspects of Fusion are the different ways it allows you to
go beyond the standard tools delivered with the application. This chapter provides an
introduction to a variety of advanced features, including Modifiers, Expressions, and
Scripting, which can help you extend the functionality and better integrate Fusion into
your studio.

Contents
The Contextual Menu for Parameters in the Inspector  292
Using Modifiers  292
Adding the Right Modifier for the Job  292
Adding Modifiers to Individual Parameters  292
Combining Modifiers and Keyframes  293
Publishing a Parameter  293
Connecting Multiple Parameters to One Modifier  294
Adding and Inserting Multiple Modifiers  294
Performing Calculations in Parameter Fields  295
Using SimpleExpressions  295
Pick Whipping to Create an Expression  298
Removing SimpleExpressions  298
Customizing User Controls  298
FusionScript  302

Chapter 12 Using Modifiers, Expressions, and Custom Controls 291

The Contextual Menu for
Parameters in the Inspector
Most of the features in this chapter are accessed via a contextual menu that appears when you
right-click most parameters in the Inspector. Different contextual menus are available based on where
in the Inspector you right-click. Specifically, right-clicking over parameter names or sliders displays a
feature-rich contextual menu that can add animation, expression fields, modifiers to extend
functionality, as well as publishing and linking capabilities, allowing you to adjust multiple controls
simultaneously.

Using Modifiers
Parameters can be controlled with modifiers, which are extensions to a node’s toolset. Many modifiers
can automatically create animation that would be difficult to achieve manually. Modifiers can be as
simple as keyframe animation or linking the parameters to other nodes, or modifiers can be complex
expressions, procedural functions, external data, third-party plug-ins, or fuses.

Adding the Right Modifier for the Job

Which modifiers are available depends on the type of parameter you’re right-clicking over. Numeric
values, text, polylines, gradients, and points each have different sets of modifiers that will work with
them, so the Modify With menu is filtered based on each parameter.

Adding Modifiers to Individual Parameters

You add modifiers to a parameter using the Inspector’s contextual menu, or by right-clicking the
onscreen control for a parameter in the viewer. Either way, a dynamic list of modifiers that are
appropriate for the selected parameters is displayed. For instance, a Perturb modifier can be added to
any slider to auto-animate the parameter by randomly wiggling the value. Once the modifier is added,
controls are displayed in the Modifiers tab to adjust the speed and amplitude of the random animation.

The Inspector contextual menu is used to add a Perturb

modifier to the Center X and Y parameters

A modifier’s controls are displayed in the Modifiers tab of the Inspector. When a selected node has a
modifier applied, the Modifiers tab will become highlighted as an indication. The tab remains grayed
out if no modifier is applied.

Chapter 12 Using Modifiers, Expressions, and Custom Controls 292

 The Modifiers tab with two modifiers applied

Modifiers appear with header bars and header controls just like the tools for nodes. A modifier’s title
bar can also be dragged into a viewer to see its output.

Combining Modifiers and Keyframes

Modifiers that auto-animate parameters like Perturb and Shake can be combined with keyframes to
create more natural, organic looking animations. For instance, you can create the general motion path
for an element by keyframing the Center X and Y parameters, and then apply a modifier to the same
parameters to create a secondary wiggling motion.

To combine a keyframed motion path with a Perturb modifier:

1 Add a Transform to an image like a butterfly or spaceship.
2 Select the Transform node in the Node Editor.
3 In the Inspector, click the Keyframe button to the right of the Center X and Y parameters.
4 Position the butterfly or spaceship image where you want the start of the animation to begin.
5 Continue to move the playhead in the render range and reposition the image until you create a
figure-8 motion path.
6 Right-click over the Center X label in the Inspector and choose Modify With > Perturb.
7 Click the Modifiers tab at the top of the Inspector and adjust random, wiggling motion by setting
the Strength, Wobble, and Speed parameters while the animation plays.

Publishing a Parameter
The Publish modifier makes the value of a parameter available, so that other parameters can connect
to it. This allows you to simultaneously use one slider to adjust other parameters on the same or
different nodes. For instance, publishing a motion path allows you to connect multiple objects to the
same path.

Publish a parameter in order to

link other parameters to it

Chapter 12 Using Modifiers, Expressions, and Custom Controls 293

Once a parameter is published, you can right-click another parameter and choose Connect To >
[published parameter name] from the contextual menu. The two values are linked, and changing the
parameter value of one in the Modifiers tab changes the other.
Using the pick whip between two parameters provides similar linking behavior with more
flexibility. Pick whipping between parameters is covered later in this chapter.

Connecting Multiple Parameters to One Modifier

A single modifier or published parameter can be applied to multiple parameters so they all act as one.
This is handled through the Connect To submenu. As with modifier assignment, the list is filtered to
show only options that are suitable for the parameter you’ve right-clicked on. When you do this, the
connection is bidirectional; editing either parameter will cause the other parameter to change.

The Connect To menu connects

modifiers to multiple parameters

Adding and Inserting Multiple Modifiers

Modifiers can be connected to each other and branched, just like nodes in the Node Editor. For
example, the Calculation modifier outputs a number, but has two Number parameters, both of which
can have modifiers added to them. If you want to insert a modifier between the existing modifier and
the modified parameter, use the Insert submenu of the parameter’s contextual menu.

Available Modifiers in Fusion:

– Bézier Spline: Adds a Bézier spline to the Spline Editor for animating the selected parameter.
– B-Spline Spline: Adds a B-Spline spline to the Spline Editor for animating the selected parameter.
– Calculation: Creates an indirect link that includes a mathematical expression
between two parameters.
– CoordTransform Position: Calculates the current 3D position of a given object even after multiple
3D transforms have repositioned the object through the node tree hierarchy.
– Cubic Spline: Adds a Cubic spline to the Spline Editor for animating the selected parameter.
– Expression: Allows you to add a variable or a mathematical calculation to a parameter, rather than
a straight numeric value. The Expression modifier provides controls in the Modifiers tab, giving
you more room and parameters than the SimpleExpression
– From Image: This modifier takes color samples of an image along a user-definable line and
creates a gradient from those samples.
– Gradient Color Modifier: Creates a customized gradient and maps it into a specified time range to
animate a value.
– KeyStretcher: Used to stretch keyframes in a Fusion Title template when trimming the template in
the Edit page or Cut page timeline.
– MIDI Extractor: Modifies the value of a parameter using the values stored in a MIDI file.
– Natural Cubic Spline: Adds a Natural Cubic spline to the Spline Editor for animating the
selected parameter.

Chapter 12 Using Modifiers, Expressions, and Custom Controls 294

 – Offset (Angle, Distance, Position): The three Offset modifiers are used to create variances,
or offsets, between two positional values. For instance, when this modifier is added to a size
parameter, you can change the size of an object using the distance between two onscreen
controls (position and offset).
– Path: Produces two splines to control the animation of of an object: An onscreen motion path
(spacial) and a Time spline visible in the Spline Editor (temporal).
– Perturb: Generates smoothly varying random animation for a given parameter.
– Probe: Auto-animates a parameter by sampling the color or luminosity of a specific pixel or
rectangular region of an image.
– Publish: The first step in linking two non-animated parameters is to use the Publish modifier to
publish a parameter. That allows other parameters to use the Connect To submenu and link to the
published parameter.
– Shake: Similar to Perturb, Shake generates smoothly varying random animation
for a given parameter.
– Track: Attaches a single point tracker to the selected parameter. The tracker can then track an
object onscreen to animate the parameter. This is quicker and more direct than using the normal
Tracker node; however, it offers less flexibility since the resulting tracker is only a single point and
can only be used for the selected parameter.
– Vector Result: Similar to the Offset modifier, Vector Result is used to offset position parameters
using origin, distance, and angle controls to create a vector. This vector can then be used to adjust
any other parameter.
– XY Path: Produces an X and Y spline in the Spline Editor to animate the position of an object.

For more information on all modifiers available in Fusion, see Chapter 122, “Modifiers.” In the
DaVinci Resolve Reference Manual and Chapter 61 in the Fusion Reference Manual.

Performing Calculations in Parameter Fields

You can enter simple mathematical equations directly in a number field to calculate a desired value.
For example, typing 2.0 + 4.0 in most number fields will result in a value of 6.0. This can be helpful
when you want a parameter to be the sum of two other parameters or a fraction of the
screen resolution.

Using SimpleExpressions
Simple Expressions are a special type of script that can be placed alongside the parameter it is
controlling. These are useful for setting simple calculations, building unidirectional parameter
connections, or a combination of both. You add a SimpleExpression by entering an equals sign directly
in the number field of the parameter and then pressing Return.

Entering an equals sign opens the

SimpleExpression field with Pick Whip control

Chapter 12 Using Modifiers, Expressions, and Custom Controls 295

An empty field will appear below the parameter, and a yellow indicator will appear to the left. The
current value of the parameter will be entered into the number field. Using Simple Expressions, you
can enter a mathematical formula that drives the value of a parameter or even links two different
parameters. This helps when you want to create an animation that is too difficult or impossible to set
up with keyframing. For instance, to create a pulsating object, you can use the sine and time functions
on a Size parameter. Dividing the time function can slow down the pulsing while multiplying it can
increase the rate.

A SimpleExpression using the sine and time functions

Inside the SimpleExpression text box, you can enter one-line scripts in Lua with some Fusion-specific
shorthand. Some examples of Simple Expressions and their syntax include:

Expression Description

time This returns the current frame number.

Merge1.Blend This returns the value of another input, Blend, from

another node, Merge1.

Merge1:GetValue(“Blend”, time-5) This returns the value from another input, but sampled
at a different frame, in this case five frames before the
current one.

sin(time/20)/2+.5 This returns a sine wave between 0 and 1.

iif(Merge1.Blend == 0, 0, 1) This returns 0 if the Blend value is 0, and returns 1 if

it is not. The iff() function is a shorthand conditional
statement, if-then-else.

iif(Input.Metadata.ColorSpaceID == This returns 0 if the image connected to the current

“sRGB”, 0, 1) node’s Input is tagged with the sRGB colorspace.
When no other node name is supplied, the expression
assumes the Input is coming from the current node. It
is equivalent to self.Input. The Input in most, but not
all, Fusion nodes is the main image input shown in
the Node Editor as an orange triangle. Images have
members that you can read, such as Depth, Width,
Metadata, and so on.

Point(Text1.Center.X, Text1. Unlike the previous examples, this returns a Point, not
Center.Y-.1) a Number. Point inputs use two members, X and Y. In
this example, the Point returned is 1/10 of the image
height below the Text1’s Center. This can be useful for
making unidirectional parameter links, like offsetting
one Text from another.

Text1.Center - Point(0,.1) This is similar to the previous expression.

This SimpleExpression returns Text instead of a
Number or Point.

Chapter 12 Using Modifiers, Expressions, and Custom Controls 296

 Expression Description

Text(“Colorspace: “..(Merge1. The string inside the quotes is concatenated with the
Background.Metadata.ColorSpaceID) metadata string, perhaps returning:
Colorspace: sRGB

Text(“Rendered “..os.date(“%b %d, This returns a much larger Text, perhaps

%Y”).. “ at “..os.date(“%H:%M”)..”\n something like:
on the computer “..os.
Rendered Nov 12, 2019 at 15:43 on
getenv(“COMPUTERNAME”).. “ running
the computer Rn309 running Windows_
“..os. getenv(“OS”)..”\n from the
NT from the comp \\SRVR\Proj\Am109\
comp “..ToUNC(comp.Filename))
SlateGenerator_A01.comp

os.date(“%H:%M”) The OS library can pull various information about the

computer. In the previous example, os.date gets the
date and time in hours:minutes.

“..os.getenv(“COMPUTERNAME”).. “ Any environment variable can be read by os.getenv,

running “..os. in this case the computer name and the operating
system.

”\n from the comp “..ToUNC(comp. To get a new line in the Text, \n is used. Various
Filename) attributes from the comp can be accessed with the
comp variable, like the filename, expressed as a
UNC path.

TIP: When working with long SimpleExpressions, it may be helpful to drag the Inspector panel
out to make it wider or to copy/paste from a text editor or the Console.

After setting an expression that generates animation, you can open the Spline Editor to view the
values plotted out over time. This is a good way to check how your SimpleExpression evaluates
over time.

A sine wave in the Spline Editor, generated by the expression used for Text1: Size

For more information about writing Simple Expressions, see the Fusion Studio Scripting Guide, and the
official Lua documentation.

Chapter 12 Using Modifiers, Expressions, and Custom Controls 297

Pick Whipping to Create an Expression
With a SimpleExpression field open, a + button is displayed on the left side. Dragging the + button
onto another control, or “pick whipping,” links the two parameters, similar to the Connect To menu.
However, unlike a Connect To parameter link, the pick whip allows you to modify the connection by
further editing the expression.

Pick whipping to connect one

parameter to another quickly

SimpleExpressions can also be created and edited within the Spline Editor. Right-click on the
parameter in the Spline Editor and select Set Expression from the contextual menu. The
SimpleExpression will be plotted in the Spline Editor, allowing you to see the result over time.

Removing SimpleExpressions
To remove a SimpleExpression, right-click the name of the parameter, and choose Remove Expression
from the contextual menu.

Customizing User Controls

Each tool’s parameters are organized in a logical order in the Inspector. The most used controls are
closer to the top, and the more subtle refinement controls are lower in the list. Sometimes, though, you
may want to add, hide, or change the controls. You often need to do this for SimpleExpressions and
macros, but you may also do this for usability and aesthetic reasons for favorites and presets.
Custom controls can be added or edited via the Edit Control dialog, which you access by right-clicking
over the node’s name in the Inspector and choosing Edit Controls from the menu.

Choose Edit Controls to create

and customize parameters

Chapter 12 Using Modifiers, Expressions, and Custom Controls 298

In the Edit Control dialog, you use the ID menu to select an existing parameter or create a new one.
You can name the control and define whether it is a text field, number field, or a point using the Type
attributes list.

Edit Control dialog

You use the page list to assign the new control to one of the tabs in the Inspector. There are also
settings to determine the defaults and ranges, and whether it has an onscreen preview control. The
Input Ctrl box contains settings specific to the selected Type, and the View Ctrl attributes box contains
a list of onscreen preview controls to be displayed, if any.
All changes made using the Edit Controls dialog get stored in the current tool instance, so they can be
copied/ pasted to other nodes in the comp. However, to keep these changes for other comps, you
must save the node settings, and add them to the Bins in Fusion Studio or to your favorites.
As an example, we’ll customize the controls for a DirectionalBlur:

The Inspector for a default direction blur

Let’s say we wanted a more interactive way of controlling a linear blur in the viewer, rather than using
the Length and Angle sliders in the Inspector. Using a SimpleExpression, we’ll control the length and
angle parameters with the Center parameter’s onscreen control in the viewer. The SimpleExpression
would look something like this:

Chapter 12 Using Modifiers, Expressions, and Custom Controls 299

For Length:
sqrt(((Center.X-.5)*(self.Input.XScale))^2+((Center.Y-.5)*(self.Input.YScale)*(self.
Input.Height/self.Input.Width))^2)

For Angle:
atan2(.5-Center.Y , .5-Center.X) * 180 / pi

DirectionalBlur controlled by the Center’s position.

This admittedly somewhat advanced function does the job fine. Dragging the onscreen control adjusts
the angle and length for the directional blur. However, now the names of the parameters are
confusing. The Center parameter doesn’t function as the center anymore. It is the direction and length
of the blur. It should be named “Blur Vector” instead. You no longer need to edit the Length and Angle
controls, so they should be hidden away, and since this is only for a linear blur, we don’t need the Type
menu to include Radial or Zoom. We only need to choose between Linear and Centered. These
changes can easily be made in the Edit Controls dialog.

To change the Inspector parameters for the example above:

1 In the Edit Control dialog, select the Center from the ID list.
2 A dialog will appear asking if you would like to Replace, Hide, or Change ID. We’ll choose Replace.
3 Change the Name to Blur Vector.
4 Set the Type to Point.
5 Select Controls in the Page list. (Controls is the first tab in the Inspector, normally.)
6 Click OK.

DirectionalBlur Center parameter name changed to Blur Vector.

Chapter 12 Using Modifiers, Expressions, and Custom Controls 300

The new Blur Vector parameter now appears in the Inspector. The internal ID of the control is still
Center, so our SimpleExpressions did not change.

To hide the Length and Angle parameters in the Inspector:

1 In the Edit Control dialog, select the Length from the ID list.
2 Select Controls from the Page list.
3 In the input Ctrl list, select Node.
4 Click OK.
5 In the Edit Control dialog, select the Angle from the ID list.
6 In the input Ctrl list, select Node.
7 Click OK.

Finally, to remove Radial and Zoom options from the Type menu:
1 In the Edit Control dialog, select the Type from the ID list.
2 Select Controls from the Page list.
3 Select Radial from the Items list and click Del to remove it.
4 Select Zoom from the Items list and click Del to remove it.
5 Click OK.

The Type menu now includes only two options.

If you want to replace the Type menu with a new checkbox control, you can do that by creating a new
control and a very short expression.

To create a new control:

1 In the Edit Control dialog, enter Center Blur in the Name field.
2 Select the New Control from the ID list.
3 Set the Type to Number, and set the Page to Controls.
4 Set the Input Ctrl to CheckboxControl.
5 Click OK.

To make this new checkbox affect the original Type menu, you’ll need to add a SimpleExpression to
the Type:
iif(TypeNew==0, 0, 2)

The “iif” operator is known as a conditional expression in Lua script. It evaluates something based on a
condition being true or false.

Chapter 12 Using Modifiers, Expressions, and Custom Controls 301

FusionScript
Scripting is an essential means of increasing productivity. Scripts can create new capabilities or
automate repetitive tasks, especially those specific to your projects and workflows. Inside Fusion,
scripts can rearrange nodes in a comp, manage caches, and generate multiple output files for delivery.
They can connect Fusion with other apps to log artist time, send emails, or update webpages and
databases.
FusionScript is the blanket term for the scripting environment in Fusion. It includes support for Lua as
well as Python 2 and 3 for some contexts. FusionScript also includes libraries to make certain common
tasks easier to do with Lua and Python within Fusion.
You can run interactive scripts in various situations. Common scripts include:
– Utility Scripts, using the Fusion application context, are found under the File > Scripts menu.
– Comp Scripts, using the composition context, are found under the Script menu or entered into the
Console.
– Tool Scripts, using the tool context, are found in the Tool’s context menu > Scripts.

Other script types are available as well, such as Startup Scripts, Scriptlibs, Bin Scripts, Event Suites,
Hotkey Scripts, Intool Scripts, and SimpleExpressions. Fusion Studio allows external and command-
line scripting as well and network rendering Job and Render node scripting.
FusionScript also forms the basis for Fuses and ViewShaders, which are special scripting-based
plug-ins for tools and viewers that can be used in both Fusion and Fusion Studio.
For more information about scripting, see the Fusion Scripting Documentation, accessible from the
Documentation submenu of the Help menu.

Chapter 12 Using Modifiers, Expressions, and Custom Controls 302

Chapter 13

Bins
This chapter covers the bin system in Fusion Studio. Bins allow for the storage and
organization of clips, compositions, tool settings, and macros, similar to the Media
Pool and Effects Library in DaVinci Resolve. It includes a built-in Studio Player for
creating a playlist of multiple shots and their versions. Bins can be used in a server
configuration for organizing shots and collaborating with other team members across
the studio.

Contents
Bins Overview  304
Bins Interface  304
Viewing and Sorting Bins  305
Organizing Bins  306
Adding and Using Content  308
File Type Details  309
Using Content from Bins  309
Jog and Shuttle  310
Stamp Files  310
Using the Studio Player  311
Playing a Single Clip  311
Creating a Reel  312
Connecting Bins Over a Network  319
Adding a Remote Bin Entry  320
Accessing Remote Bins  321
Permissions  321
Studio Player and Bin Server  321

Chapter 13 Bins 303

Bins Overview
Bins are folders that provide an easy way of accessing commonly used tools, settings, macros,
compositions, and footage. They can keep all your custom content and resources close at hand, so
you can use them without searching through your hard drives. Bins can also be shared over a network
to improve a collaborative workflow with other Fusion artists.
Bins are only available in Fusion Studio.

The Bins window

Bins Interface
The Bins window is separated into two panels. The sidebar on the left is a list of the bins, while the
Content panel on the right displays the selected bin’s contents.

To open the Bins window, do one of the following:

– Choose File > Bins from the menu bar.
– Press Command-B.

The Bins sidebar

Chapter 13 Bins 304

 TIP: When opening the Bins window on macOS, the window may open behind the current
Fusion Studio window. Check in the dock for the Bins window icon or move the Fusion Studio
window to locate the Bins window if you are working on a single monitor.

The sidebar organizes content into bins, or folders, using a hierarchical list view. These folders can be
organized however they suit your workflow, but standard folders are provided for Clips, Compositions,
Favorites, Projects, Reels, Settings, Templates, and Tools. The Tools category is a duplicate of all the
tools found in the Effects library. The Tools bin is a parent folder, and parent folders contain subfolders
that hold the content. For instance, Blurs is a subfolder of the Tools parent folder. Parent folders can be
identified by the disclosure arrow to the left of their name.
When you select a folder from the sidebar, the contents of the folder are displayed in the Contents
panel as thumbnail icons.

The Bins icon view

Viewing and Sorting Bins

A contextual menu is used to access most of a bin’s features. You show the contextual menu by
right-clicking in an empty area in the Contents panel. Right-clicking on an item will show the same
menu with additional options for renaming, playing, or deleting the item.

Icon or List View

One use of the contextual menu is to switch between viewing the contents as thumbnail icons or
as a list.

To view a bin’s contents in List view:

1 Right-click in an empty area of the Contents panel.
2 From the contextual menu, choose View > Details.

Each bin in the sidebar can be set to List view or Icon view independently of each other. So while you
may have some bins you want to see as a list, others may be easier to view as icons.

Chapter 13 Bins 305

The Bins List view

Sort Order in List View

Clicking on the heading of a column in List view will sort the list in ascending order, and clicking it
again will reverse the sort order.

Icon Size in Icon View

The icons can be adjusted to small, medium, large, or huge by clicking the Size button in the bottom
toolbar or right-clicking in an empty area of the Contents panel to bring up the contextual menu and
choosing a size from the Icon Size submenu.

Use the Size button to select

the icon size in the bin

Organizing Bins
Once you begin adding your own categories and content, you can have hundreds of items that need
to be organized. To keep all of your elements accessible, you’ll want to use some basic organization,
just like keeping files and documents organized on your computer.

To create a new folder in the sidebar:

1 In the sidebar, select the parent folder under which the new folder will be listed.
2 Right-click in an empty area of the Contents panel.
3 From the contextual menu, choose New > New Folder.
4 Enter a name for the new folder, and then click OK on the dialog.

Chapter 13 Bins 306

 The New Folder menu

You can also click the New Folder icon in the toolbar.

To rename a bin folder:

1 Right-click on the folder icon in the Contents panel.
2 Choose Rename from the contextual menu or press F2 on the keyboard.

To move a folder into or out of a parent folder:

1 Select the parent folder that contains the folder you want to move.
2 In the Contents panel, drag the folder into the sidebar where you want it moved.

Drag a folder from the Contents panel to move it in the sidebar.

When you drag a folder onto another folder in the sidebar, you create a hierarchical subfolder.
Dragging it to the Library parent folder at the top of the sidebar will add it to the top level of the
Bins window.

To remove a folder from the bins:

1 Select the folder in the Contents panel.
2 Press Command-Delete (Backspace on Windows and Linux).

TIP: You cannot undo removing a folder from the Bins window.

Chapter 13 Bins 307

Adding and Using Content
You can add and use different types of content with bins. Fusion Studio compositions, tools, saved tool
settings, macros, tool groups, and any file format that is supported in Fusion can be added to bins and
then used in compositions at a later time.

To add an item to a bin:

1 Select a bin in the sidebar where you want to add the content.
2 Right-click in the Contents panel.
3 Choose New > … from the contextual menu.
4 Select the media, settings, macros, or comps in the file browser, and then click Open.

Add Item in the contextual menu

To remove an item from the bins:

1 Select the folder in the Contents panel.
2 Right-click on the item and choose Delete from the contextual menu or press Command-Delete.

The Delete function in the contextual menu

TIP: Unsupported files like PDFs can also be saved in the bin, and the application that
supports it will launch when you double-click the file. It’s a useful location for scripts
and notes.

If you have an operating system file browser window open, you can drag files directly into a bin as
well. When adding an item to the bins, the files are not copied. A link is created between the content
and the bins, but the files remain in their original location.

Chapter 13 Bins 308

File Type Details
Some types of content have additional methods of being added to bins. Some additional methods are
because of the file type and some are because of where they are located.

Projects and Media

In addition to using the New… contextual menu as explained earlier, Fusion Studio project files with the
extension “.comp” and media can also be added to bins by dragging them to the Contents panel from
a file browser.

Tool Settings
If you want to add a node with custom settings to a bin, first save the node as a setting by right-clicking
over it in the Node Editor and choosing Settings > Save As. Once the setting is saved, you can add it
to a bin by dragging it from a File Browser into the Bins window.

Image Sequences and Stills

Image sequences are automatically identified on disk and loaded as clips rather than stills, so it is not
necessary to select more than one frame from an image sequence when dragging it into a bin.
To ignore the image sequence and import only a single frame, hold Shift when you drag the frame into
a bin. This can be useful when trying to import a single still image from a series of still shots with a
DSLR. The numbers maybe sequential, but you just need one still image from the series.

Using Content from Bins

Once you have content in your bins, you’ll want to add them to a composition. In general, you can
either drag the content directly into the Node Editor or double-click it to add it; however, each type of
content behaves a little differently when added to the Node Editor.

Media
Dragging media into the Node Editor from a bin window creates a new Loader that points to the media
on disk. Still files or photos are automatically set to loop.

Compositions
To add a composition, you must right-click and choose Open. Dragging a comp item onto an open
composition will have no effect. When a composition is added, it is opened in a new window. It is not
added to the existing composition.

Tools, Tool Settings, and Macros

When you add tools from the Tools categories to a composition, the methods you use and results you
get are similar to adding tools using the toolbar buttons or the Effects Library. Dragging a tool allows
you to place it anywhere in the Node Editor, unconnected, or, if you drag it over a connection line,
inserted between two existing tools. Double-clicking a tool in the bin will insert it after the Active tool
in the Node Editor. Dragging a tool from a bin into a viewer will insert that tool after the currently
viewed tool.
Settings and macros work a bit differently than tools. They can only be added to the Node Editor by
dragging and droping. Dragging a setting or macro allows you to place it in the Node Editor,
unconnected, or, if you drag it over a connection line, inserted between two existing tools.

Chapter 13 Bins 309

Jog and Shuttle
You can scrub clips in Icon view in the bin using one of two modes. Jog mode is the default mode. It
moves the clip forward and backward as long as you are dragging the mouse. Once the mouse stops,
the clip pauses.
You can choose Shuttle mode by right-clicking over the clip’s thumbnail in the bin and choosing Scrub
Mode > Shuttle.

Jog and Shuttle contextual menu

Shuttle mode begins playing the clip forward or backward once you press the right mouse button and
drag either left or right. The clip continues to play until the mouse button is released or you reach the
end of the clip.

Stamp Files
Stamp files are low-resolution, local proxies of clips, which are used for playback of clips stored on a
network server, or for very large clips.

To create a stamp file for a clip:

1 Right-click on the clip in the bin.
2 Choose Create Stamp from the contextual menu.

The Status bar at the top of the Bins window shows the progress of the stamp creation. Since the
stamp creation is a background process, you can queue other stamps and can continue working with
the composition.

Chapter 13 Bins 310

Using the Studio Player
The Studio Player is a timeline-based playback interface built into the Bins window. It allows you to
play and organize versions of compositions, make notes, and collaborate on shots and projects. The
resolution-independent player uses any format that Fusion Studio can ingest, like EXR, ProRes, BMD
RAW, QuickTime, and others. Clips use a RAM cache for playback, so even large formats can loop
playback as long as there is available memory. Clips can contain audio and are output to a video
monitor using Blackmagic Design Decklink or UltraStudio devices, allowing you to screen dailies and
review shots. You can add annotation notes to the project, and setting up a Bin Server allows multiple
artists to access the Studio Player.

Here are some Studio Player highlights:

– Clips can be played as single events, looped, or using a ping-pong playback, with a definable
loop range.
– Clip metadata can be viewed, with live update during scrub/play.
– The Timeline interface can create a playlist for reviewing multiple clips.
– Per-shot color adjustment controls allow for consistent display of clips from different formats.
– Annotation notes can be typed on each clip and version, as well as the entire project.
– Audio Scratch track can be enabled for each clip during playback.
– Clip versions are stored in the same project to allow for quick access to previous work and for
comparison of progress.
– Guide overlays can be customized to show monitor/title safety and show crops to
various output formats.
– Blackmagic Design UltraStudio and DeckLink playback devices are supported for reviewing clips
on video monitors and projectors.
– The fully collaborative workflow automatically synchronizes reel changes, annotations, and color
adjustments across multiple workstations, allowing multiple artists or supervisors to access the
same projects simultaneously.
– Remote sync allows multiple Studio Players to follow the master. Actions performed on the master,
such as playback and scrubbing, will also be executed on the slaves, allowing the reel to be
reviewed across multiple workstations or sites.
– Studios can automate tasks using the Fusion scripting engine to control the Studio Player.

Playing a Single Clip

Clips created from image sequences, MOV files, and some RAW formats can be previewed using the
Studio Player without having to first add the clip to a node tree.

To play a clip in the Studio Player, do one of the following:

– Double-click a clip in a bin to open the Studio Player.
– Select the clip and click the Play button at the bottom of the bin window.

Chapter 13 Bins 311

The Studio Player includes a large viewer, a Timeline, and a toolbar along the bottom.

Once you have the clip open in the Studio Player, you can click the Play button in the toolbar at the
bottom of the window.

Scrubbing the Timeline

You can quickly preview the clip by scrubbing through it rather than playing it.

To scrub a clip in Studio Player:

1 In the Timeline, drag the playhead to the area you want to scrub over.
2 Use the Left and Right Arrow keys to move one frame forward or backward.

Closing the Studio Player

After you have finished previewing in the Studio Player, you can return to the Bins to switch to a new
clip or continue working in Fusion.

To close the current clip in the Studio Player and return to the bins:
– Click the three-dot Options menu in the lower-left corner and choose Close.

Creating a Reel
A reel is a playlist or clip list that is viewed either as storyboard thumbnails or a timeline. In the bin, you
create a new reel item to hold and play back multiple clips. The thumbnail for a reel appears with a
multi-image border around it, making it easer to identify in the Bin window.

Chapter 13 Bins 312

To create a reel in the current bin:
– Right-click in an empty area of the bin and choose New > Reel
– Click the Reel button along the bottom toolbar.

Use the New Reel button to create

a reel in the current bin.

Once created and named, the reel appears in the current bin.
Double-clicking the reel opens the Studio Player interface along the bottom of the Bin window. An
empty Studio Player appears in the top half of the window.

To add a clip to a reel:

1 Click the three-dot Options menu in the lower-left corner and choose Show Bins.
2 Drag a clip from one of the bins to the empty reel in the lower half of the window.

The toolbar across the bottom of the interface has various controls for setting a loop, showing and
adjusting color, playback transport controls, collaboration sync, guide overlays, and frame number and
playback speed in fps.

The toolbar along the bottom of the Studio Player includes controls to customize playback

Toolbar buttons
– Set Loop In/Out: Sets the start and end frame for playing a section of the Timeline in a loop.
– Shot: Sets the loop for the entire clip.
– Reset: Disables the loop mode.
– M: Shows Metadata of the image.
– RGB and alpha: Toggles between displaying the color and alpha of the clip.
– Brightness Gamma: Adjusts the brightness and gamma of the viewer and is applied to all clips.
Individual clip color controls can also be applied using another menu.
– Video: Outputs the image to Blackmagic Design DeckLink and UltraStudio devices.
– Transport controls: Used to play forward, backward, fast forward, and fast backward, as well as go
to the start and end of a clip.
– Sync: A three-way toggle allowing the Studio Player to be controlled or to control another player
over a network. The Off setting disables this functionality.
– Guide buttons: These three buttons control the visibility of three customizable guide settings.

Chapter 13 Bins 313

Inserting Shots
Clips or comps from the bin can be dragged to the storyboard area of the reel to add and
organize a playlist.

You can insert a shot between existing clips by positioning

the new clip in between existing items in the reel.

To play the clips in a reel:

1 Click the three-dot Options menu in the lower-left corner and choose Show Player.
2 Position the playhead where you want to begin playing and click the Play button.

Creating Versions
Alternatively, you can add a version to an existing clip by dragging the new item on top.

Chapter 13 Bins 314

Versions of a shot will appear as stacked icons in the storyboard reel. The number of stacks in the icon
indicate the number of versions included with that clip. In the example below, the first shot has two
versions, the second shot has four versions, and the last clip has only one version.

Version Menu
You can choose which version to view by right-clicking over the clip in the storyboard and selecting it
from the Version > Select menu.

The Version menu also includes options to move or rearrange the order of the clip versions as well as
remove a version, thereby deleting it from the stack.

Shot Menu
The per-clip Shot menu includes functions to Rename the shot, Remove the shot, Trim the clip’s In and
Out points, add notes, adjust the color, and add an audio soundtrack.

Chapter 13 Bins 315

 – Rename: Allows you to change the name of the shot.
– Remove: Deletes the entire shot and all the versions from the project reel.
– Trim: Opens the Trim dialog to adjust the clip In point and Out point on the Timeline.
– Notes: Opens the Notes window to the right of the interface, allowing you to add a note to the
specific shot.
– Color: Opens the Color Adjustments panel to perform Lift, Gamma, Gain, Brightness, and Contrast
adjustments to clips.
– Audio: Allows you to attach an audio file to a clip in the Reel.

When notes are added, they are time and date stamped as well as name stamped.
The naming is from the bin login name and computer name.

Selecting Color from the Shot menu allows you to make tonal adjustments per clip using Color
Decision List (CDL) style controls.

The Audio menu option can import an audio .wav file

that will play back along with the selected clip.

Options Menu
The three-dot Options menu in the lower left of the interface displays the menu that can be used to
switch between viewer and the bin in the top half of the window. It is also used to clear the memory
used in playback by selecting Purge Cache.

Chapter 13 Bins 316

 The Options menu includes options
to switch the top half of the window
between the viewer or bin contents.

Selecting Reel > Notes opens the Notes dialog to add annotations text to the entire reel project.
The Reel > Export option saves the reel to disk as an ASCII readable format so it can be used
elsewhere or archived.

The Reel submenu opens an area for

production notes on the entire reel.

The View menu is used when you want to switch between the reel storyboard layout and a
Timeline layout.

Guides
You can assign customizable guide overlays to three Guide buttons along the bottom of the Studio
Player. Fusion includes four guides to choose from, but you can add your own using the XML Guide
format and style information provided at the end of this chapter. You assign a customizable guide to
one of the three Guide buttons by right-clicking over a button and selecting a guide from the list. To
display the guide, click the assigned button.

Three toolbar Guide buttons enable

guide overlays in the viewer.

Guides can be customized and placed in the the Guides folder:

– On macOS: Macintosh HD/Users/username/Library/Application Support/Blackmagic Design/
Fusion/Guides/
– On Windows: C:\Users\username\AppData\Roaming\Blackmagic Design\Fusion\Guides.
– On Linux: home/username/.fusion/BlackmagicDesign/Fusion/Guides.

Chapter 13 Bins 317

Guides are a simple XML formatted text document saved with the .guide extension, as defined below.
This makes it easy to create and share guides.

Custom Guide Format

The guides are files that have drawing instructions a bit like code, like this:
Guide
{
     Name = “10 Pixels”,
     Elements =
        {
            HLine { Y1=”10T” },
            HLine { Y1=”10B” },
            VLine { X1=”10L” },
            VLine { X1=”10R” },
        },
}

Or an example of safe area:

Guide
{
     Name = “Safe Frame”,

     Elements =
        {
            HLine { Y1=”10%”, Pattern = 0xF0F0 },
            HLine { Y1=”90%”, Pattern = 0xF0F0 },
            HLine { Y1=”95%” },
            HLine { Y1=”5%” },
            VLine { X1=”10%”, Pattern = 0xF0F0 },
            VLine { X1=”90%”, Pattern = 0xF0F0 },
            VLine { X1=”95%” },
            VLine { X1=”5%” },
            HLine { Y1=”50%”, Pattern = 0xF0F0, Color = { R =
1.0, G = 0.75, B = 0.05, A=1.0 } },
            VLine { X1=”50%”, Pattern = 0xF0F0, Color = { R =
1.0, G = 0.75, B = 0.05, A=1.0 } },
        },
}

Chapter 13 Bins 318

Guide Styles
The style of a guide is defined by a set of properties that appear in the format shown below:
<HLine Y1=“33%” Pattern=“C0C0” Color=“FFFFFFFF”/>

– HLine: Draws a horizontal line and requires a Y-value, which is measured from the top of the
screen. The Y-value can be given either in percent (%) or in absolute pixels (px).
– Vline: Draws a vertical line and requires an X-value, which is measured from the left of the screen.
The X-value can be given either in percent (%) or in absolute pixels (px).
– Pattern: The Pattern value is made up of four hex values and determines the visual appearance of
the line.
Examples for such patterns include:
>>FFFF draws a solid line ________________
>>EEEE a dashed line -------------------
>>ECEC dash-dot line -.-.-.-.-.-.-.-.-.-.-.
>>ECCC dash-dot-dot -..-..-..-..-..-..-..
>>AAAA dotted line ………………

– Color: The Color value is composed of four groups of two hex values each. The first three groups
define the RGB colors; the last group defines the transparency. For instance, the hex value for
pure red would be #FF000000, and pure lime green would be #00FF0000
– Rectangle: Draws a rectangle, which can be empty or filled, and supports the same pattern and
color settings described above.
It requires two X- and two Y-values to define the extent <Rectangle Pattern=“F0F0” X1=“10%”
Y1=“10%” X2=“90%” Y2=“90%”>.

– FillMode: Applies to rectangles only and defines whether the inside or the outside of the
rectangle should be filled with a color. Leave this value out to have just a bounding rectangle
without any fill.
>>FillMode = (“None”|”Inside”|”Outside”)

– FillColor: Applies to rectangles only and defines the color of the filled area specified by FillMode.

>>FillColor=“FF000020”

Connecting Bins Over a Network

You can share bins among computers running Fusion on the network so that multiple visual effects
artists can share assets, presets, and even entire compositions. These shared bins are called remote
bins, and everyone can share one or more remote bin in a studio.

To connect to a remote system and display its bins:

1 Choose Fusion Studio > Preferences.
2 In the Preferences dialog, select Global > Bins > Servers in the list.

Chapter 13 Bins 319

 The bin servers Preferences panel

This panel shows a list of the available bin servers, with buttons below for entries to be added to or
deleted from the list.

Adding a Remote Bin Entry

If you want to add another Remote bin to the list of available Remote bins, you can click the Add button
in the bin servers Preferences panel. The text controls below the button will become enabled for
editing. In the Server field, type the system name or IP address where the bin is hosted.

Add the IP address where the bin server is hosted.

Then add a User name and Password if one is needed to access the server.
The Library field lets you name the bins. So if you want to create a bin for individual projects, you
would name it in the Library field, and each project would gets its own bin.
The Application field allows larger studios to specify some other program to serve out the
bin requests.

Chapter 13 Bins 320

Once you’ve finished setting up the bin server information and clicked Save in the Preferences
window, you can open the Bins window to test your bin server. Opening the Bins window is the first
time your connection to the server will be tested. If it cannot connect, the bin server will still be listed,
with access denied or unavailable marked next to the name on the bins sidebar.
There is no practical limit to the number of bins that can be accessed.

Accessing Remote Bins

Bin servers behave just like a local bins. Any bin added in the preferences show in the Bins sidebar as
another top-level item. The available bins are shown by name with a status and, if required, a
password. Bins unavailable to you are marked as (unavailable).

Permissions
Unlike other directories on a server, your access to bins on a network is stored in the bin document.
The bins themselves contain all the users and passwords in plain text, making it easy for someone to
administer the bins.

Studio Player and Bin Server

Reel projects can be shared by multiple artists across the studio via the bin server system, reviewing
and adding versions and notes, all independently at the same time. With the Sync function, multiple
people can collaborate together with synced playback and scrubbing.

The Sync button is a three-way toggle

button: Off, Slave, and Master

When the Sync function is On, the transport controls can be set to control the playback or follow the
master controller.

Master enables controlling other

slave players on the network

The local transport controls can be disabled,

causing playback to follow the master

Chapter 13 Bins 321

Chapter 14

Fusion Connect
This chapter goes into detail on how to use the Fusion Connect AVX2 plug-in
with an Avid Media Composer editing system. The Fusion Connect AVX plug-in
is only available with Fusion Studio.

Contents
Fusion Connect Overview  323
System Requirements  323
The Effects Palette  323
The Layer Input Dialog  324
Applying Fusion Connect to a Transition Point  324
Export Clips  324
Edit Effect  325
Browse for Location  325
Auto Render in Fusion  326
Red on Missing Frames  326
Compress Exported Frames  326
Edit Effect Also Launches Fusion  326
Versioning  326
Create New Version  326
Version  326
About RAW Images  326
About Color Depth  327
Manual vs. Auto-Render  327
Fusion/Avid Project Relationship  329
Rendering with Fusion  329
Directory Structure of Fusion Connect Media  330
Advanced Project Paths  331
Configuring Paths on macOS  331
Configuring Paths on Windows  332
Fields and Variables  332
Environment Variables  333

Chapter 14 Fusion Connect 322

Fusion Connect Overview
Fusion Connect is an AVX2 plug-in for Avid Media Composer. It allows editors to create a conduit
between the Timeline in Avid editing products and Fusion Studio. Fusion Connect exports clips from
the Avid Timeline as image sequences and assembles Fusion compositions that allow you to work
your magic on the content.
Fusion can be started automatically by the plug-in if Fusion is installed on the same system, or it can
be used on remote computers to modify the composition.

System Requirements
Fusion Connect has the following requirements:
– Supported Avid products: Media Composer 8.x
– Supported product: Fusion Studio 8.1 or later
– Installation: Two files will be installed in your Media Composer:
– Fusion Connect.avx
– BlackmagicFusionConnect.lua

– Avid’s default directory: \Avid\AVX2_Plug-ins

The Effects Palette

After launching Media Composer, Fusion Connect is located in the Blackmagic Effects Palette
category. As with any segment or transition effect in the Effects Palette, you can apply the Fusion
Connect AVX2 plug-in to any clip. This includes filler*, edit transition point, or video track layers on the
Avid Timeline.

Avid Media Composer Effects palette showing the

Blackmagic category and Fusion Connect AVX plug-in.

Chapter 14 Fusion Connect 323

The Layer Input Dialog
When you apply the Fusion Connect AVX plug-in to a clip or a layer in the Timeline, you are presented
with an AVX Optional Inputs dialog box.

The Fusion Connect AVX plug-in can access multiple

layers from the Media Composer Timeline.

Once the layer count is selected, Fusion Connect will be applied to the Timeline.
– Select the layer count equal to the number of video track layers you want to ingest into Fusion.
– Filler can be used as a layer.
– Fusion Connect will allow a maximum of eight layers.

Applying Fusion Connect

to a Transition Point
If you apply Fusion Connect to a transition point, no dialog box will display, and the
AVX plug-in will simply be applied to the Timeline transition point.

A Fusion Connect AVX plug-in applied

as a transition to a cut in the Timeline

You can use the Avid dialog boxes or smart tools to adjust the length and offset of
the transition to Start, Center, End, or Custom.

Export Clips
By pressing the Export Clips button in the Effects Editor, Fusion Connect exports all the associated
clips as image sequences, to provide access to them in Fusion. Any previously exported existing
images are overwritten, ensuring that all media needed by Fusion is accessible. Performing an export
is desired if you want to use Fusion on a different computer than the one with Media Composer
installed.
When using Fusion on the same computer as Media Composer, there is no need to export the clips
explicitly by checking the Export Clips checkbox. Without this option enabled, Fusion Connect saves

Chapter 14 Fusion Connect 324

the source frames each time images are displayed, scrubbed, or played back from the Timeline.
Depending on your Media Composer Timeline settings, these interactively exported images might be
half-resolution based on Avid proxy settings. When scrubbing around the Timeline, only a few
frames—namely, those that are fully displayed during scrubbing—might be written to disk.

TIP: Set your Timeline Video Quality button to Full Quality (green) and 10-bit color bit depth. If
the Timeline resolution is set to Draft Quality (green/yellow) or Best Performance (yellow),
Fusion receives subsampled, lower-resolution images.

Fusion Connect AVX plug-in controls in the Effects Editor

Edit Effect
After exporting the clips, the Edit Effect button performs three subsequent functions:
– Creates a Fusion composition, with Loaders, Savers, and Merges (for layers), or a Dissolve
(for transitions). This function is only performed the first time a comp is created when the
Fusion Connect AVX2 plug-in is applied.
– Launches Fusion (if installed on the machine), if it is not already launched.
– Opens the Fusion comp associated with created effects.

Browse for Location

The Fusion Connect media folders are created on the drive where the associated Avid media resides,
defaulting to the root level of that drive. However, you can choose a new location for the Fusion
Connect media.

To choose another location to store and access the media:

1 In the Media Composer Effects Editor, click the Browse for Location button.
2 In the File Browser, change the location and create additional folders if desired.

The path settings field in the Effects Editor updates to show the current location. If you apply Fusion
Connect to another clip in the Timeline, the last location is remembered.

Chapter 14 Fusion Connect 325

Auto Render in Fusion
The Auto Render button is a toggle that allows you to automatically render your Fusion comp from
within Avid. Please note that this method of rendering has limitations and can be significantly slower
than rendering directly in Fusion. It is mostly used for batch rendering on the Avid Timeline. Auto Render
also exports the necessary media without having to manually execute the Export Clips function first.

Red on Missing Frames

The Red on Missing Frames button is a toggle used to display red images within the Avid Timeline
viewer (Timeline monitor) if no rendered frames from Fusion Studio can be found, or if the rendered
frames are not of a high enough resolution. When disabled, the original untouched frames will be
shown instead of the red frame.

Compress Exported Frames

When enabled, this button creates Fusion RAW files with compression for, both exported and
rendered frames. This creates smaller file sizes, saving disk space. As with any other compression
algorithm, the compression adds time to the write process of the file sequence.

Edit Effect Also Launches Fusion

When enabled, this button opens Fusion Studio when the Edit Effect button is clicked, allowing for a
more direct Avid/Fusion workflow. When disabled, this button will not launch Fusion Studio when the
Edit Effect button is clicked, but a Fusion .comp file is created. This is useful when working with Fusion
Studio on a separate computer than the one running the Media Composer software.

Versioning
Creating visual effects is almost always an iterative process. You’ll often need to create revisions after
your first pass at the effect. Built into Fusion Connect is a versioning feature that lets you create
multiple revisions of an effect and switch between them from within Media Composer.

Create New Version

This checkbox creates a copy of the current comp without affecting the original.
Any changes that are rendered in the copy will be written to a new folder and become another version
of the rendered result played on the Avid Timeline. Previous versions of the comp and their rendered
results are accessible using the Version slider.

Version
This slider selects which version of the comp is used in the Media Composer Timeline. It can be used
to interactively switch from one version to the other in order to compare the results.

About RAW Images

Fusion Connect creates a Fusion RAW file image sequence for intermediate folders between Avid
Media Composer and Fusion Studio, in order to preserve all the image information. This allows the
images to reside on disk and not take up space in RAM. The benefits of using Fusion RAW include:
– The ability to continue the editing process while an effect is rendering
– The ability to take advantage of network rendering
– The ability to retime footage
– The ability to run Fusion Studio remotely

Chapter 14 Fusion Connect 326

About Color Depth
Fusion Connect derives its images directly from the RGB data within Avid Media files. This allows the
images to be codec agnostic. All RAW files from Avid that begin as 8- or 10-bit images are remapped
to 16-bit float in Fusion. Rendered results from Fusion are processed in 16-bit float to maintain the full
color fidelity supported by Media Composer.

Manual vs. Auto-Render

While Auto-Render is the easier workflow, the manual approach offers faster renderings in Fusion and
more control over the performance and memory usage on your system.
– In the manual workflow, Fusion Studio is not required to be installed on the Avid system itself but
can reside on any other computer.
– For Auto-Render, Fusion Studio must be installed on the local computer.
The following diagram shows typical workflows for manual and automatic renders.

Manual Workflow Auto-Render Workflow

1 Add Connection effect 1 Add Connection effect

Ensure thatAuto-Render is ticked.

2 Click Export Clip

Creates Fusion RAW files at the given location on the

media drive. These files are not deleted automatically,
and manual cleanup of old files will be rquired to
avoid running out of disk space.

3 Click Edit Effect 2 Click Edit Effect

Create a comp file when first clicked, but will not Creates Fusion RAW files as Export Clip would do.
overwrite this file when clicked again. Attempts to Also creates a comp file when first clicked, but will
launch Fusion and load the comp. If Fusion is not not overwrite this file when clicked again. Launches
installed locally, the comp can be accessed manually Fusion and loads the comp.
from a different machine via the network.

4 Edit comp in Fusion 3 Edit comp in Fusion

Work your magic here. Work your magic here.

5 Save comp in Fusion 4 Save comp in Fusion

This step is important! This step is important!

Do not forget to save the comp. Do not forget to save the comp.

6 Render comp in Fusion

You can continue working in Avid while Fusion renders.

Your Timeline will update automatically with any frame
that is succesfully rendered.

5 Render clip in Avid

In this case, both the Fusion comp and the Avid clip
will be rendered simultaneously. If a full-size rendered
7 Render clip in Avid
frame is not found, the full size/depth source frames
are exported automatically for that time. Fusion is then
Optional step, but recommended.
instructed to start rendering from that point.
The resulting frame is loaded and returned to MC,
and process is repeated for each frame thereafter.

Avid/Fusion layer to comp relationship for auto and manual renders

Chapter 14 Fusion Connect 327

Once the initial trip from Avid to Fusion is complete, depending on the type of clip to which you
assigned the effect, in Fusion Studio you will be presented with one of the following:
– A Loader node representing a single clip.
– Two or more Loaders connected to Merge nodes representing layers.
– Two Loaders connected to a Dissolve node representing a transition.

In all three node tree layouts outlined above, there will also be a Saver node. The Saver node is
automatically set to render to the directory that is connected back to the Media Composer Timeline
with the correct format. If for some reason the file format or file path are changed in the Saver node,
the Fusion Connect process will not render correctly.

TIP: Due to the design of the AVX2 standard, hidden embedded handles are not supported.
To add handles, prior to exporting to Fusion, increase the length of your clip in the Media
Composer Timeline to include the handle length.

Fusion Node Editor representations of a single clip segment effect in the Media Composer
Timeline (left), a multi-layered composite (center), and the transition (right)

TIP: If segments in the Avid Timeline have different durations, move the longest clip to the top
layer and apply Fusion Connect to that clip. This will ensure all clips are brought into Fusion
without being trimmed to the shortest clip.

Chapter 14 Fusion Connect 328

Fusion/Avid Project Relationship
The frame rate and image size preferences created in Media Composer are adopted within Fusion’s
frame rate preferences. This allows for consistency in formats for the roundtrip process from Avid to
Fusion and back to Avid. The format settings do not prevent you from using or mixing any other sized
imaging within the composition as Fusion is resolution independent.

Fusion Connect AVX uses frame rate and resolution from the Media Composer Timeline.

Rendering with Fusion

When you perform a render of your comp inside Fusion Studio, the results are rendered to the output
folder created by Fusion Connect during the initial application of the plug-in to the Timeline. Upon
rendering, you immediately see the results of the rendered Fusion comp in the Avid Timeline. Even
while Fusion is rendering, you can continue with the edit process on any clip except for the associated
clip being rendered at the time.

Chapter 14 Fusion Connect 329

Directory Structure of
Fusion Connect Media
Fusion Connect creates a logical folder structure that is not affiliated with the Avid Media Files folder
but rather the Fusion Connect AVX2 plug-in. Based on data gathered during the AVX application to
the Timeline, a logical folder hierarchy is automatically created based on Avid projects, sequences,
and clips. This structure allows for multiple instances of Fusion Studio to access the media and multiple
instances of the AVX to relate to a single Fusion comp. In a re-edit situation, media is never duplicated
but is overwritten to avoid multiple copies of identical media.

Avid ProjectName

Master directory for the Avid project.

This can be set within user defined directories.

Avid SequenceName

This directory contains the Fusion comp-files

and sub-directories, named after the Avid clip.

Bob_v01.comp
Charlie_v01.comp

Fusion composition(s) named

after the Avid clip(s).

Bob

Directory named after the Avid clip.

Charlie

Directory named after the Avid clip.

Avid

Directory named ‘Avid’, containing

the exported source clip(s).

Charlie_0000.raw
Charlie
Charlie_0000.raw
Charlie_0000.raw
Directory named after the
exported clip. Image sequence named
after the exported clip.

Dave Dave_0000.raw
Dave_0000.raw
Optional directory named after Dave_0000.raw
the exported clip.
Optiona second source clip
for Charlie_v01.comp.

xyz_name
xyz_0000.raw
Optional directory named after xyz_0000.raw
the exported clip. xyz_0000.raw

Any additional source clip

for Charlie_v01 comp.

Fusion

Directory named ‘Fusion’’ containing

the rendered clip.

Charlie_0000.raw
Render_v01
Charlie_0000.raw
Charlie_0000.raw
Directory named ‘Render_’ plus the
version number of the Fusio comp. Rendered images sequence
named after the Avid clip.

Render_v02

Directory named ‘Render_’ plus the

version number of the Fusion comp.

If you apply the effect to a transition, the naming behavior might be somewhat different.

Chapter 14 Fusion Connect 330

By default, Media Composer refers to the two clips of a transition as “Clip_001” and “Clip_002”. Based
on the naming convention, Fusion Connect will create folders with matching names. If such folders
already exist, because another transition has already used up “Clip_001” and “Clip_002”, the numbers
will be incremented automatically.
Likewise, “_001” will be added incrementally to the group folder name, if a folder of that name already
exists. The corresponding comp file will be named accordingly.

Fusion Connect AVX creates folder structures in the OS to save media and
Fusion compositions. Those names are reflected in the Timeline.

You will notice that the Fusion Connect icon is a green dot (real-time) effect. If your hardware is fast
enough, the results that populate the plug-in will play in real time. However, it’s recommended that you
render the green dot effect, which will force an MXF precompute to be created to guarantee real-
time playback.

Advanced Project Paths

The Fusion Connect AVX2 plug-in controls the pathing of Fusion Connect’s .raw media and Fusion’s .
comp files as well as showing and hiding project and sequence level directories. This is achieved
through environment variables, which are set in the operating system. This gives you the most flexible
control over pathing your media, and as the name depicts, you can change variables (controls) in
various application environments. This is useful for network storage and when Fusion Studio is running
on other systems.

Configuring Paths on macOS

When using Fusion Connect on macOS, the Configure Path Defaults dialog looks like this:

Environment variable path maps on macOS.

Default paths can be configured using variables similarly as on Windows, but for added convenience it
is possible to enter any desired path defaults directly into fields in the dialog, without the need for
using environment variables.

Chapter 14 Fusion Connect 331

Configuring Paths on Windows
When using Fusion Connect on Windows, the Configure Path Defaults dialog looks like this:

Environment variable path maps on Windows.

Fusion Connect can define the user variables directly in the Fusion Connect plug-in. Click the
Configure Path Defaults button to launch the path defaults dialog editor. In the Options section of the
Fusion Connect AVX2 plug-in, click the triangle to reveal the path details.

Fields and Variables

Variables are the control title defined specifically by the application that is being controlled, while
values are the instructions that tell a variable what to do. The fields and variables that can be used on
macOS and Windows are described in the following table:

Field Variable Environment Variable Description

$PROJECT CONNECT_PROJECT Overrides the current Avid Project name.

Project Name $DRIVE CONNECT_DRIVE Drive or folder for all Connect projects

$SEQUENCE – Name of Avid sequence

$SEQPATH CONNECT_SEQUENCE_PATH Folder for all Connect files in this sequence

Sequence Path $GROUP – Unique name of this Connect instance

$CLIP – Name of exported clip

Clip Path – CONNECT_CLIP_PATH Folder for exported clips from Avid

Out Path – CONNECT_OUT_PATH Folder for rendered results from Fusion

Comp Path – CONNECT_COMP_PATH Location and name of Fusion comp file

Chapter 14 Fusion Connect 332

Environment Variables
The pathing can be set in the environment variables of the system, so that IT management of the
project paths can be achieved.

Accessing the Environment Variables on Windows

The quickest way to access environment variables is through your Windows control panel by searching
the word “env” without the quotes. You have a choice of editing at a user level or system level.

User Variables
Click on the link that says “Edit environment variables for your account.”

System Variables
Click on the link that says “Edit the system environment variables.”

Accessing the Environment Variables on macOS

To set an environment variable on macOS, you must use the Terminal window. Environment variables
on macOS are added to the .bash_profile directory for the current user.

User Variables
For system-wide operations, place the environment variable in ~/.bash_profile

TIP: System variables control the environment throughout the operating system, no matter
which user is logged in.
User variables always overrule any system variable; therefore, the user variable always wins
if control for a specific function is duplicated in the user and system variable.

System Variables
For system-wide operations, place the environment variable in /etc/profile

Environment Variables and What They Mean

The Fusion Connect AVX plug-in can use environment variables to set different locations for certain
folders or files. Each computer OS has a unique way of entering environment variables, but in every
OS, the variable must be typed exactly as shown.

TIP: If you type directly in Fusion Connect’s Path Editor, you do not have to type the variable,
just the value. You also can make modifications without having to restart the
Media Composer! The only caveat is that in order to remove a variable, you must exit
Media Composer and clear the environment variable in the Windows interface or
macOS Terminal and restart the Media Composer.

Chapter 14 Fusion Connect 333

Other values you can control derived from your Avid Bin include:

Values Description

$DRIVE This will force the directory to the drive where the Avid media is stored.

This will force a directory based on the Avid project name for which the media was
$PROJECT
digitized/imported or AMA linked.

This will force a directory based on the Avid SEQUENCE name for which the media
$SEQUENCE
was digitized/imported or AMA linked.

Here is an example of how a variable can be set up to support project and sequence names within
your directory.

Environment variable project and sequence name examples on Windows

Here is the same example in the Windows environment variable editor.

Windows environment user variable editor

Chapter 14 Fusion Connect 334

Chapter 15

Preferences
This chapter covers the various options that are available from the
Fusion Preferences Window.

Contents
Preferences Overview  336
Categories of Preferences  337
Preferences In Depth  340
AVI  341
Defaults  341
Flow  343
Frame Format  345
General  346
GPU  348
Layout  350
Loader  351
Memory  353
Network  354
Path Maps  356
Preview  359
QuickTime  360
Script  361
Spline Editor  362
Splines  364
Timeline  365
Tweaks  366
User Interface  368
Video Monitoring  370
View  371
VR Headsets  372

Chapter 15 Preferences 335

Bins/Security  374
Bins/Server  375
Bins/Settings  376
EDL Import  377
Customization  378
Shortcuts Customization  378
Customizing Preferences  379

Preferences Overview
The Preferences window provides a wide variety of optional settings available for you to configure
Fusion’s behavior to better suit your working environment. These settings are accessed via the
Preferences window. The Preferences window can be opened from a menu at the top of the interface.

In DaVinci Resolve, to open the Fusion Preferences window, do one of the following:
– On macOS, switch to the Fusion page and choose Fusion > Fusion Settings.
– On Windows, switch to the Fusion page and choose Fusion > Fusion Settings.
– On Linux, switch to the Fusion page and choose Fusion > Fusion Settings.

In Fusion Studio, to open the Fusion Preferences window, do one of the following:
– On macOS, choose Fusion Studio > Preferences.
– On Windows, choose File > Preferences.
– On Linux, choose File > Preferences.

The Preferences window includes composition settings

Chapter 15 Preferences 336

Global and Composition Preferences
The Preferences window is divided into a category sidebar on the left and the settings panel on the
right. In Fusion Studio, there are two levels of preferences: Global and Composition. The Fusion page
in DaVinci Resolve uses just a single Global preference that affects every project, new and existing.
The Global preferences are used to set options specific to Fusion’s overall behavior as well as defaults
for each new composition. The Composition preferences in Fusion Studio can further modify the
currently open composition without affecting the Global preferences or any other composition that is
open but not displayed.

Categories of Preferences
The first entry in the Preferences sidebar is assigned to the Global preferences. Clicking the Global
and Default Settings disclosure arrow reveals the following sections.

Fusion Studio Global and Default settings

3D View
The 3D View preferences offer control over various parameters of the 3D Viewers, including grids,
default ambient light setup, and Stereoscopic views.

AVI (Windows Fusion Studio Only)

The AVI preferences configure the default compression options when creating AVI files from a Saver
node. These settings can be overridden on a case-by-case basis using the Format tab in the Saver’s
Inspector.

Defaults
The Defaults preferences are used to select default behavior for a variety of options, such as
animation, global range, timecode display, and automatic tool merging.

Chapter 15 Preferences 337

Flow
You use the Flow preferences to set many of the same options found in the Node Editor’s contextual
menu, like settings for Tile picture, the Navigator, and pipe style.

Frame Format
The Frame Format preferences are used to create new frame formats as well as select the default
image height and width when adding new creator tools like Background and Text+. You also set the
frame rate for playback.

General
The General preferences contain options for the general operation, such as auto save, and gamma
settings for color controls.

GPU (Fusion Studio Only)

The GPU preferences include options for selecting specific GPU acceleration methods based on your
computer platform and hardware capabilities. It is also used for enabling caching, and debugging GPU
devices and tools.

Layout (Fusion Studio Only)

You can use the Layout preferences to save the exact layout of Fusion’s windows.

Loader (Fusion Studio Only)

Using the Loader preferences, you can set options for the default Loader’s depth and aspect ratio as
well as define the local and network LoaderCache settings.

Memory (Fusion Studio Only)

Memory management for multi-frame and simultaneous branch rendering is configured in the Memory
preferences.

Network (Fusion Studio Only)

The Network rendering preferences are used to configure options such as selecting a render master,
email notification, and whether the machine can be used as a render slave.

Path Map
Path Map preferences are used to configure virtual file path names used by Loaders and Savers as
well as the folders used by Fusion to locate comps, macros, scripts, tool settings, disk
caches, and more.

Preview (Fusion Studio Only)

The Preview preferences is where you configure the Preview creation and playback options.

QuickTime (macOS Fusion Studio Only)

This section lets you preconfigure the QuickTime codec used for rendering.

Script
The Script preferences include a field for passwords used to execute scripts externally, programs to
use for editing scripts, and the default Python version to use.

Chapter 15 Preferences 338

Spline Editor
The Spline Editor preferences allow you to set various spline options for Autosnap behavior, handles,
markers, and more.

Splines
Options for the handling and smoothing of animation splines, Tracker path defaults, onion-skinning,
roto assist, and more are found in the Splines preference.

Timeline
The Timeline preferences is where you create and edit Timeline/Spline filters and set default options
for the Keyframes Editor.

Tweaks
The Tweaks preferences handle miscellaneous settings for modifying the behavior when loading
frames over the network and queue/network rendering.

User Interface
These preferences set the appearance of the user interface window and how the Inspector is
displayed.

Video Monitoring (Fusion Studio Only)

The Video Monitoring preferences is where you can configure your Blackmagic video display
hardware for monitoring on an HD, Ultra HD, or DCI 4K display.

View
The View preferences are used to manage settings for viewers, including default control colors,
Z-depth channel viewing ranges, default LUTs, padding for fit, zoom, and more.

VR Headsets
The VR Headsets preferences allow configuration of any connected Virtual Reality headsets, including
how stereo and 3D scenes are viewed.

Bins (Fusion Studio Only)

There are three panels as part of the Bins preferences: a Security panel where you set users and
passwords for serving the local bins; a Servers panel used to select which remote Bin servers are
connected; and a Settings panel for stamp rendering.

Import
The Import settings contain options for EDL Import that affect how flows are built using the data
from an EDL.

Chapter 15 Preferences 339

Preferences In Depth
Within each category is a deep set of controls for configuring Fusion so that it better fits your working
environment. The preferences contain both software and hardware options that affect all newly
created comps. The following section explains every option located in the preferences categories.

3D View
The 3D View preferences contain settings for various defaults in the 3D Viewers, including grids,
default ambient light setup, and Stereoscopic views.

The 3D View preferences

Grid
The Grid section of the 3D View preferences configures how the grid in 3D Viewers are drawn.
– Grid Antialiasing: Some graphics hardware and drivers do not support antialiased grid
lines, causing them to sort incorrectly in the 3D Viewer. Disabling this checkbox will disable
antialiasing of the grid lines. To turn off the grid completely, right-click in a 3D Viewer and choose
3D Options > Grid.
– Size: Increasing the Size value will increase the number of grid lines drawn. The units used for the
spacing between grid lines are not defined in Fusion. A “unit” is whatever you want it to be.
– Scale: Adjusting the overall scaling factor for the grid is useful, for example, if the area of the grid
appears too small compared to the size of your geometry.

Chapter 15 Preferences 340

Perspective Views
The Perspective Views section handles the appearance of the perspective view in both a normal and
stereoscopic project.
– Near Plane/Far Plane: These values set the nearest and furthest point any object can get to or
from the camera before it is clipped. The minimum setting is 0.05. Setting Near Plane too low and
Far Plane too far results in loss of depth precision in the viewer.
– Eye Separation/Convergence/Stereo Mode: This group of settings defines the defaults when
stereo is turned on in the 3D Viewer.

Orthographic Views
Similar to the Perspective Views, the Orthographic Views (front, top, right, and left views) section sets
the nearest and furthest point any object can get to or from the viewer before clipping occurs.

Fit to View
The Fit to View section has two value fields that manage how much empty space is left around objects
in the viewer when the F key is pressed.
– Fit Selection: Fit Selection determines the empty space when one or more objects are selected
and the F key is pressed.
– Fit All: Fit All determines the empty space when you press F with no objects selected.

Default Lights
These three settings control the default light setup in the 3D Viewer.
The default ambient light is used when lighting is turned on and you have not added a light to the
scene. The directional light moves with the camera, so if the directional light is set to “upper left,” the
light appears to come from the upper-left side of the image/camera.

AVI
The AVI preference is only available in Fusion Studio on Windows. It configures the default AVI codec
settings when you select AVI as the rendering file format in the Saver node.
– Compressor: This drop-down menu displays the AVI codecs available from your computer. Fusion
tests each codec when the application opens; therefore, some codecs may not be available if the
tests indicate that they are unsuitable for use within Fusion.
– Quality: This slider determines the amount of compression to be used by the codec. Higher values
produce clearer images but larger files. Not all codecs support the Quality setting.
– Key Frame Every X Frames: When checked, the codec creates keyframes at specified intervals.
Keyframes are not compressed in conjunction with previous frames and are, therefore, quicker to
seek within the resulting movie. Not all codecs support the keyframe setting.
– Limit Data Rate To X KB/Second: When checked, the data rates of the rendered file are limited to
the amount specified. Not all codecs support this option. Enter the data rate used to limit the AVI
in kilobytes (kB) per second, if applicable. This control does not affect the file unless the Limit Data
Rate To option is selected.

Defaults
The choices made here are used to determine Fusion’s behavior when new tools are added to the
Node Editor and when parameters are animated.

Chapter 15 Preferences 341

 The Defaults preferences

Default Animate
The Default Animate section is used to change the type of modifier attached to a parameter when the
Animate option is selected from its contextual menu. The default option is Nothing, which uses a
Bézier spline to animate numeric parameters and a path modifier for positional controls.
– Number With and Point With: Drop-down lists are used to select a different modifier for the new
default. For example, change the default type used to animate position by setting the Point with
the drop-down menu to XY Path.
Choices shown in this menu come from installed modifiers that are valid for that type of parameter.
These include third-party plug-in modifiers, as well as native modifiers installed with Fusion.

Auto Tools
The Auto Tools section determines which tools are added automatically for the most common
operations of the Background tools and Merge operations.
– Background: When set to None, a standard Background tool is used; however, the drop-down
menu allows you to choose from a variety of tools including 2D and 3D tools to customize the
operation to your workflow.
– Merge: When set to None, nothing happens. When set to Merge, connecting the outputs of two
tools or dragging multiple clips on the Node Editor uses a standard Merge. Other valid options for
this are Anaglyph, Channel Booleans, and Dissolve.
– Use Merge Only When Connecting Outputs Directly: When this option is active, Merges are not
automatically added when you drag multiple clips from the Finder or Windows Explorer onto the
Flow area.

Global Range
Using the Start and End fields, you can define the Global Start and End frames used when creating
new compositions.

Chapter 15 Preferences 342

Time Code
You use this option to determine whether new compositions will default to showing SMPTE Time Code
or frames (Feet + Frames) to represent time.

Flow
Many of the same options found in the Node Editor’s contextual menu, like settings for Tile Picture, the
Navigator, and Pipe Style, are found in this category.

The Flow preferences

Force
The Force section can set the default to display pictures in certain tool tiles in the Node Editor rather
than showing plane tiles. The Active checkbox sets pictures for the actively selected tool, the All
checkbox enables pictures for all tiles, and the Source and Mask checkbox enables tile pictures for
just Source and Mask tools.
When All is enabled, the picture shown will either be a thumbnail of the image rendered by the tool if
the tool has rendered, or if the Show Thumbnails option is disabled, the tool’s default icon is used.
Concatenated transforms will also show a default icon.
– Show Modes/Options: Enabling this option will display icons in the tool tile depicting various
states, like Disk Caching or Locked.
– Show Thumbnails: When this checkbox is selected, tool tiles set to show tile pictures will
display the rendered output of the tool. When the checkbox is cleared, the default icon for the tool
is used instead.

Chapter 15 Preferences 343

Options
The Options section includes several settings that control or aid in the layout and alignment of tools in
the Node Editor.
– Arrange to Grid: This enables a new node tree’s Snap to Grid option to force the tool layout to
align with the grid marks in the flow.
– Arrange to Connected: Tools snap to the vertical or horizontal positions of other tools they are
connected to.
– Auto Arrange: This option enables the Node Editor to shift the position of tools as needed to
make space when inserting new tools or auto-merging layers.
– Show Grid: This enables or disables the display of the Node Editor’s background grid.
– Auto Remove Routers: Pipe Routers or “elbow nodes” in the Node Editor are considered
“orphaned” if the tools connected to either the input or output are deleted. When this option is
enabled, Orphaned Routers are automatically deleted.
– Pipes Always Visible: When enabled, the connection lines between tools are drawn over the top
of the tool tiles.
– Keep Tile Picture Aspect: Enabling this option forces tool tile thumbnail pictures to preserve the
aspect of the original image in the thumbnail.
– Full Tile Render Indicators: Enabling this checkbox causes the entire tile to change color when it
is processing. This can make it easier to identify which tools are processing in a large composition.
The coloring itself will form a progress bar to alert you to how close slower tools are to finishing
their process.
– Show Instance Links: This option is used to select whether Instance tools will show links,
displayed as green lines, between Instance tools.
– Navigator: The Navigator is a small square overview of the entire composition. It is used to quickly
navigate to different parts of a node tree while you are zoomed in. The checkboxes in this section
determine when the Navigator is displayed, if at all.
– On: The Navigator will always be visible.
– Off: The Navigator will always be hidden.
– Auto: The Navigator will only be visible when the Node Editor’s contents exceed the currently
visible Work area.

– Pipe Style: This drop-down menu selects which method is used to draw connections between
tools. The Direct method uses a straight line between tools, and Orthogonal uses horizontal and
vertical lines.
– Build Direction: When auto-building or laying out a node tree, Build Direction controls whether
tools are organized horizontally or vertically.
– Scale: The Scale menu allows you to select the default zoom level of the Node Editor when a new
composition is created.

Pipe Grab Distance

The Pipe Grab Distance slider allows you to choose how close the pointer must be (in pixels) to a
connection line in the node tree when selecting them.

Link Grab Distance

The Link Grab Distance slider allows you to choose how close the pointer must be (in pixels) to a knot
on a node before a connection is made or removed. 0 pixels means you must be directly on the knot,
while a maximum setting of 20 means you can be 20 pixels from a knot and still connect to it.

Chapter 15 Preferences 344

Group Opacity
This slider controls the opacity of an expanded group’s background in the Node Editor.

Frame Format
Frame Format preferences allow you to select the resolution and frame rate for the nodes that
generate images like Background, fast noise, and Text+. It also sets the color bit depth for final
renders, previews, and interactive updates in the viewer. The color bit depth settings only apply to
Fusion Studio. Rendering in DaVinci Resolve always use 32-bit float.

The Frame Format preferences

Default Format
This drop-down menu is used to select the default resolution for Generator tools from a list of presets.
This is only a default setting; these settings can be overridden using the Resolution settings in a
node’s Inspector.
Use the Edit boxes to change any of the default settings. When creating a new setting, press the New
button and enter a name for the setting in the dialog box that appears and enter the parameters.

Chapter 15 Preferences 345

Settings
The Settings section defines the format that is selected in the Default Format menu. You can modify an
existing format or create a new one.
– Width/Height: When creating a new format for the menu or modifying an existing menu item, you
specify the Width or Height in pixels of the format using these fields.
– Frame Rate: Enter or view the frames per second played by the format. This sets the default
Frame Rate for previews and final renders from the Saver tool. It also sets the playback for the
comp itself, as well as the frame to time code conversion for tools with temporal inputs.
– Has Fields: When this checkbox is enabled, any Creator or Loader tool added to the Node Editor
will be in Fields process mode.
– Film Size: This field is used to define how many frames are found in one foot of film. The value is
used to calculate the display of time code in Feet + Frames mode.
– Aspect Ratio: These two fields set the pixel aspect ratio of the chosen frame format.
– Guide 1: The four fields for Guide 1 define the left, top, right, and bottom guide positions for
the custom guides in the viewer. To change the position of a guide, enter a value from 0 to 1.
The bottom-left corner is always 0/0, the top-right corner is always 1/1. If the entered value’s
aspect does not conform to the frame format as defined by the Width and Height parameters, an
additional guide is displayed onscreen. The dotted line represents the image aspect centered
about Guide 1’s Center values.
– Guide 2: This setting determines the image aspect ratio in respect to the entire frame format width
and height. Values higher than 1 cause the height to decrease relative to the width. Values smaller
than 1 cause height to increase relative to width.
– New: You use the New button to create a new default setting in the drop-down menu. Once you
click the button, you can name the setting in the dialog box that appears.
– Copy: The Copy button copies the current setting to create a new one for customization.
– Delete: The Delete button will remove the current setting from the default drop-down list.

Color Depth
The three menus in the Color Depth section are used to select the color mode for processing preview
renders, interactive renders, and full (final) renders. Processing images at 8-bit is the lowest color
depth and is rarely sufficient for final work these days but is acceptable for fast previews. 16-bit color
has much higher color fidelity but uses more system resources. 16-bit and 32-bit float per channel uses
even more system resources and is best for digital film and HDR rendered images.
Generally, these options are ignored by the composition unless a Loader or Creator tool’s Color Depth
control is set to Default.

General
The sections contained in the General preferences affect the behavior of the Inspector as well as
some other user interface elements.

Chapter 15 Preferences 346

 The General preferences

Usability
Usability has a number of project, Node Editor, and user interface settings that can make the
application easier to work with, depending on your workflow.
– Auto Clip Browse: When this checkbox is enabled, the File Browser is automatically displayed
when a new Loader or Saver is added to the Node Editor.
– New Comp on Startup: When checked, a new, empty project is created each time Fusion Studio is
launched. This has no effect in DaVinci Resolve’s Fusion page.
– Summarize Load Errors: When loading node trees or “comps” that contain unknown tools (e.g.,
comps that have been created on other computers with plug-ins not installed on the current
machine), the missing tools are summarized in the console rather than a dialog being presented
for every missing tool.
– Save Compressed Comps: This option enables the saving of compressed node trees, rather than
ASCII based text files. Compressed node trees take up less space on disk, although they may take
a moment longer to load. Node trees containing complex spline animation and many paint strokes
can grow into tens of megabytes when this option is disabled. However, compressed comps
cannot be edited with a text editor unless saved again as uncompressed.
– Show Video I/O Splash: This toggles whether the Splash image will be displayed over the video
display hardware. This is only applies to Fusion Studio.
– Use Simplified Copy Names: This option reduces the occurrence of underscores in tool names
when copying.
– Show Render Settings: When this checkbox is selected, the Fusion Render Settings dialog will be
displayed every time a render is started in Fusion Studio. Holding Shift while starting a render will
prevent the display of the dialog for that session, using whatever settings were applied during the
last render. Disabling this option reverses this behavior.

Chapter 15 Preferences 347

 – Mouse Wheel Affects the Window Under the Pointer: Normally the mouse wheel or trackpad
swiping works in the currently active window. With this option enabled, it will work in the window
underneath the cursor, so you don’t have to click into a window first to make it active.
– Frames Start From: This designates the starting frame number for clip times in the Loader and its
Clip list.
– Show Color As: This setting determines the numeric scale used to represent colors. The available
options are Normalized (0 to 1), 8-bit (0 to 255), and 16-bit (0 to 65,535). This does not affect the
actual processing or quality of the image, but it can make the mental math sometimes used to
figure out adjustments a bit easier.

Auto Save
The Auto Save settings only apply to Fusion Studio. To set auto backups for the Fusion page in
DaVinci Resolve, use the DaVinci Resolve Project Load and Save Preferences.
When Auto Save is enabled in Fusion Studio, comps are automatically saved to a backup file at regular
intervals defined by the Delay setting. If a backup file is found when attempting to open the comp, you
are presented with the choice of loading either the backup or the original.
If the backup comp is opened from the location set in the Path Map preference, saving the backup will
overwrite the original file. If the backup file is closed without saving, it is deleted without affecting the
original file.
– Save Before Render: When enabled, the comp is automatically saved before a preview or final
render is started.
– Delay: This preference is used to set the interval between Auto Saves. The interval is set using
mm:ss notation, so entering 10 causes an Auto Save to occur every 10 seconds, whereas entering
10:00 causes an Auto Save every 10 minutes.

Proxy
– Update All, Selective, No Update: The Update mode button is located above the toolbar. You
can use this preference to determine the default mode for all new comps. Selective is the usual
default. It renders only the tools needed to display the images in the Display view. All will render
all tools in the composition, whereas None prevents all rendering.
– Standard and Auto: These sliders designate the default ratio used to create proxies when the
Proxy and Auto Proxy modes are turned on. These settings do not affect the final render quality.

Even though the images are being processed smaller than their original size, the image viewing scales
in the viewers still refer to original resolutions. Additionally, image processing performed in Proxy
Scale mode may differ slightly from full-resolution rendering.
The Proxy and Auto Proxy size ratios may be changed from within the interface itself by right-clicking
on the Prx and APrx buttons above the toolbar and selecting the desired value from the
contextual menu.

GPU
The GPU preference is only available in Fusion Studio. In DaVinci Resolve, you can configure the GPU
processing in Resolve’s Memory and GPU preferences.
In Fusion Studio, the GPU preference is used to specify the GPU acceleration method used for
processing, based on your computer platform and hardware capabilities. It is also used for enabling
caching and debugging GPU devices and tools.

Chapter 15 Preferences 348

 The GPU preferences

Options
The GPU options include radio buttons to select whether the GPU is used when processing and, if so,
which computer framework is used for communicating with the GPU.
– GPU Tools: This preference has three settings: Auto, Disable, and Enable. When set to Disable, no
GPU acceleration is used for tools or third-party plug-ins. Fuses may still require GPU acceleration.
If Enable is selected, GPU acceleration is available for tools and plug-ins, if appropriate
drivers are installed.
– API: The API setting selects the GPU processing method to use.
– Device: The Device setting determines which GPU hardware to use in the case of multiple GPUs.
The Auto setting gives priority to GPU processing; however, if it is unavailable, Fusion uses the
platform default. Currently, both the AMD and CPU options require either the AMD Catalyst 10.10
Accelerated Parallel Processing (APP) technology Edition driver or the ATI Stream SDK 2.1 or later
to be installed. The Select setting allows you to choose the device explicitly.

Debugging
The more advanced preferences located in this section are designed for diagnostics and analyzing
GPU operations.
– Verbose Console Messages: Enabling this option causes information to be shown in the Console.
For example, Startup Logs, Compiler Warnings, and Messages.
– OpenGL Sharing: Enabling this option shares system RAM with onboard GPU RAM to create a
larger, but slower, OpenGL memory pool.
– Clear Cache Files: This option will clear already compiled GPU code and then
recompile the kernels.

Chapter 15 Preferences 349

Layout
The Layout preferences are only available in Fusion Studio. To save a Layout in DaVinci Resolve’s
Fusion page, use the Workspace > Layout Presets menu. The Layout options are used to control the
layout, size, and position of various windows in Fusion’s interface at startup or when a comp is created.

The Layout preferences

There are a lot of options, but in practice, you simply organize the interface the way you prefer it on
startup and when a new composition is created, then open this Preferences panel and click on the
three buttons to grab the Program Layout, the Document Layout and the Window Settings.

Program Layout
The Program Layout is used to save the overall Fusion interface window and any open floating
windows. Each new composition you open within the lager overall Fusion interface window will adhere
to these preferences.
– Grab Program Layout: Pressing this button stores the application’s overall current position
and size.
– Run Mode: This menu is used to select the application’s default mode at startup.
You choose between a Maximized application window, a Minimized application, or a Normal
application display.
– Use the Following Position and Size: When checked, the values stored when Grab Program
Layout was selected will be used when starting Fusion Studio.
– Create Floating Views: When checked, the position and size of the floating viewers will be saved
when the Grab Program Layout button is used.

Chapter 15 Preferences 350

Document Layout
The Document Layout is used to save the layout of panels and windows for the current Fusion comp.
– Recall Layout Saved In Composition: When checked, all Document Layout settings in the controls
below will be recalled when a saved composition is loaded.
– Grab Document Layout: Pressing this button stores the entire interface setup, including all the
internal positions and sizes of panels and work areas.
– Window: When multiple windows on the same composition are used, this menu is used to select
the window to which the Window Settings will apply.

Window Settings
Rather than saving entire comp layouts, you can save position and size for individual floating windows
and panels within a comp using the Window Settings.
– Automatically Open This Window: When checked, the selected window will automatically be
opened for new flows.
– Grab Window Layout: Pressing this button stores the size and position of the selected window.
– Run Mode: Select the default run mode for the selected window. You can choose between a
Maximized window, a Minimized window, or a Normal window display.
– Use Grabbed Position and Size: When checked, the selected window will be created using the
stored position and size.

Loader
The Loader preferences are only available in Fusion Studio. Using the Loader preferences, you can set
options for the default Loader’s color depth and aspect ratio as well as define the local and network
cache settings.

The Loader preferences

Chapter 15 Preferences 351

Defaults
The Defaults section includes two settings to determine how color depth and aspect ratio are handled
for Loaders.
– Loader Depth: The Loader Depth defines how color bit depth is handled when adding a Loader.
Choosing Format means that the correct bit depth is automatically selected, depending on the
file format and the information in the file’s header. Choosing Default sets the bit depth to the value
specified in the Frame Format preferences.

Cache
The Cache preferences allow you to control how disk caching operates in Fusion. You can set how and
where the cache is generated, when the cache is removed, how the cache reacts when source files
are not available, as well as many other cache related options. This is not to be confused with RAM
cache, which is controlled in the Memory preferences.
– Disable All Local Caching: This setting disables local caching.
– Cache Files from Network DiskCaches: If a tool has disk caching enabled, and the disk cache
files are stored remotely on the network, then enabling this option will use a local copy of those
cache files, similarly to the local cache on a networked Loader.
– Enable Local Caching of Loaders: Files will be copied into the LoaderCache path set below or in
the Path Maps preferences.
– Cache Multi-Frame Files: Files like AVI or QuickTime will be copied into the LoaderCache path.
This may take some time if the file is large.

– Don’t Cache Files from Local Disks: Files that do not sit on a network drive will not be copied into
the LoaderCache path. You can disable this option if you have, for example, a fast SSD cache drive
and want to use it for local files as well to speed up file access while working interactively.
– Only Files Smaller Than xxx MB.: Files larger than the value set here will not be copied into the
LoaderCache path.
– Cache Path Separator Character: When Enable Local Caching of Loaders is enabled, you can use
this setting to rebuild the path of the original files in LoaderCache.
For instance, given the default “!” character, the original path X\Project\MyShots\ Shot0815\ will be
translated into X!Project!MyShots!Shot0815! in the LoaderCache path. Other separator characters
may be used, including the “\” character, which will use subdirectories in LoaderCache: X\Project\
MyShots\Shot0815\.
– If Original File Is Missing: This setting provides three options to determine the caching behavior
when the original files can’t be found. The Fail option behaves exactly as the Default Loader in
Fusion. The Loader will not process, which may cause the render to halt. The Load Cache option
loads the cache even though no original file is present.The Delete Cache option clears missing
files from the cache.
– Cache Location: For convenience, this is a copy of the LoaderCache path set in the Path Maps
preferences.
– Explore: This button opens the LoaderCache path in the macOS X Finder window
or a Windows Explorer window.
– Clear All Cache Files: This button deletes all cached files present in the LoaderCache path.

Chapter 15 Preferences 352

Memory
The Memory preferences are only available in Fusion Studio. To control Fusion’s memory when using
the Fusion page in DaVinci Resolve, open DaVinci Resolve’s Memory and GPU preferences.
Occasionally, it will be necessary to adjust the Memory preferences in order to make the best use of
available memory on the computer. For example, some people prefer a higher cache memory for
faster interactive work, but for final renders the cache memory is often reduced, so there’s more
memory available for simultaneous processing of tools or multiple frames being rendered at once.

The Memory preferences

Caching Limits
The Caching Limits include options for Fusion’s RAM cache operation. Here, you can determine how
much RAM is allocated to the RAM cache for playing back comps in the viewer.
– Limit Caching To: This slider is used to set the percentage of available memory used for the
interactive tool cache. Available memory refers to the amount of memory installed in the computer.
When the interactive cache reaches the limit defined in this setting, it starts to remove lower
priority frames in the cache to clear space for new frames.
– Automatically Adjust In Low Memory Situations: This checkbox will set the caching to adjust
when memory is low. The console will display any cache purges.
– Leave At Least X MBytes: This setting is used to set the hard limit for memory usage. No matter
what the setting of the Cache Limit, this setting determines the amount of physical memory
available for use by other applications. Normally, this value should not be smaller than 25 MBytes.

Chapter 15 Preferences 353

Interactive Render
The Interactive Render option allows you to optimize Fusion’s processing based on the amount of
RAM you have installed in your system.
– Simultaneous Branching: When checked, more than one tool will be processed at the same time.
Disable this checkbox if you are running out of memory frequently.

Final Render
These settings apply to memory usage during a rendering session, either preview or final, with no
effect during an interactive session.
– Render Slider: This slider adjusts the number of frames that are rendered at the same time.
– Simultaneous Branching: When checked, more than one branch of a node tree will be
rendered at the same time. If you are running low on memory, turn this off to increase
rendering performance.

Network
The Network preferences are only available in Fusion Studio. These preferences are used to set up
and control network rendering in Fusion Studio. The majority of settings are found in the Render
Manager dialog.

The Network preferences

Chapter 15 Preferences 354

Submit Network Render Compositions
In these fields, you enter the Master Name and IP address of the computer that will manage all
network renders sent from this machine. If a standalone render master is in use on the network, these
fields may be pre-filled and may not be editable. This is done to prevent multiple unauthorized render
masters from being created by each person in a facility.
To re-enable editing of the master name and IP, create the environment variable FUSION_NO_
MANAGER and set the value to True. Check your operating system user guide for how to create
environment variables.

General
The General preferences are designed with the most used options at the top in the General section.
These options determine in what capacity the system is used during network rendering.
– Make This Machine a Render Master: When enabled, Fusion will accept network render
compositions from other computers and manage the render. It does not necessarily mean that this
computer will be directly involved in the render, but it will submit the job to the render nodes listed
in the Render Manager dialog.
– Allows This Machine to Be Used as a Network Slave: When enabled, this computer can be used
as a Render node and will accept compositions for network rendering. Deselect it to prevent other
people from submitting compositions to render on this computer.
– Render on All Available Machines: Enable this checkbox to ignore groups and priorities
configured in the Render Manager. Compositions submitted from this computer for network
rendering will always be assigned to every available slave.

Email Notification
You can use the Email Notification section to set up who gets notified with status updates regarding
the render jobs and the network.
– Notify Options: These checkboxes cause emails to be sent when certain render events take
place. The available events are Queue Completion, Job Done, and Job Failure.
– Send Email to: Enter the address or addresses to which notifications should be sent. You separate
multiple addresses with a semicolon.
– Override Sender Address: Enter an email address that will be used as the sender address. If this
option is not selected, no sender address is used, which may cause some spam filters to prevent
the message from being delivered to the recipient.

Server Settings
This section covers Clustering and Network Rendering. For more information on these settings and
clustering, see Chapter 65, “Rendering Using Saver Nodes” in the DaVinci Resolve Reference Manual
or Chapter 4 in the Fusion Reference Manual.

Chapter 15 Preferences 355

Path Maps
Path Maps are virtual paths used to replace segments of file paths with variables. For example, define
the path ‘movie_x’ as actually being in X\Shows\Movie_X. Using this example, Fusion would
understand the path ‘movie_x\scene_5\ scan.000.cin’ as actually being X:\Shows\ Movie_X\scene_5\
scan.000.cin.

The Path Map preferences

For Fusion Studio, there are two main advantages to virtual path maps instead of actual file paths. One
is that you can easily change the path to media connected to Loaders (for example, when moving a
comp from one drive to another), without needing to make any changes in the composition. The other
advantage is when network rendering, you can bypass the different OS filename conventions.
– Enable Reverse Mapping of Paths Preferences: This checkbox is at the top of the Path Map
settings. When enabled, Fusion uses the built-in path maps for entries in the path’s settings
when applying mapping to existing filenames. The main benefit is for Fusion Studio. Enabling
this checkbox causes Loaders to automatically use paths relative to the location of the saved
composition when they are added to the Node Editor. For more information on using relative paths
for Loaders, see Chapter 105, “IO Nodes,” in the DaVinci Resolve Reference Manual or Chapter 44
in the Fusion Reference Manual.

As with other preferences in Fusion Studio, paths maps are available in both Global and Composition
preferences. Global preferences are applied to all new compositions, while Composition path maps
are only saved with the current composition. Composition path maps will override Global path maps
with the same name.

Chapter 15 Preferences 356

The Global paths maps are divided into three sections:
– System Path Maps: The operating system determines system path maps, and they define Fusion’s
global locations. You can override specific System path maps using the Defaults or current
Composition Path Map settings. If you change your mind at a later time, you are always able to
return to Fusion’s “factory” defaults using the System path maps. There are several top-level path
maps established in the System Path Map settings.
– AllData: The folder where Fusion saves all shared application data.
– AllDocs: The folder where Fusion saves the public/shared document folder.
– AllLUTs: The nested LUTs path in the Defaults section, where Fusion saves LUTs.
– Fusion: The folder where Fusion Studio app is installed. For example, if you open Fusion from
C:\Program Files\Fusion, then the path Fusion:\Help refers to C:\Program Files\Fusion\Help. If
you instead used a copy of Fusion found in \\post-server\fusion\16, then Fusion:\Help would
expand to \\post-server\fusion\16\Help.
– FusionLibs: The Fusion libraries used for the application.
– Profile: The folder where default Fusion preferences file is saved.
– Profiles: The folder where Fusion individual user preferences are saved.
– Programs: The location of Fusion Studio or DaVinci Resolve.
– SystemFonts: The folder where the OS saves fonts that appear for Text+ and Text 3D nodes.
– Temp: The system’s temporary folder.
– UserData: The folder where Fusion saves all user-specific miscellaneous roaming data. The
individual elements included in the roaming data are listed in the Default Path Maps section.
For Fusion Studio on Windows, this is “C:\Users\username\AppData\Roaming\”. On Linux, this
will be “$HOME/Blackmagic/Fusion”. On macOS, this is “Users/UserName/Library/Application
Support/Blackmagic Design/Fusion”.
– UserDocs: The folder where Fusion saves the user’s document folders.

– Default Path Maps: The Defaults are user-editable path maps. They can reference the System
paths, as part of their paths. For instance. the Temp folder is defined in the System path and used
by the Default DiskCache path map to refine the nested location (Temp:DiskCache). Default path
maps can also redirect paths without using the Global System path maps. After you change a
Default, the updated setting can be selected in the Preferences window, and a Reset button at the
bottom of the Preferences window will return the modified setting to the System default.
– AutoSaves: This setting determines the Fusion Comp AutoSave document’s location, set in the
Fusion General preferences.
– Bins: Sets the location of Fusion Studio bins. Since the bins use pointers to the content, the
content is not saved with the bin. Only the metadata and pointers are saved in the bins.
– Brushes: Points Fusion to the folder that contains custom paintbrushes.
– Comps: The folder where Fusion Studio compositions are saved. On macOS or Windows, the
default location is in Users/YourUserName/Documents/Blackmagic Design/Fusion.
– Config: Stores Configuration files used by Fusion Studio during its operation.
– Defaults: Identifies the location of node default settings so they can be restored if overwritten.
– DiskCache: Sets the location for files written to disk when using the Cache to Disk feature. This
location can be overridden in the Cache to Disk window.
– Edit templates: The location where Fusion macros are saved in order to appear as templates in
the DaVinci Resolve Effects Library.
– Filters: Points to a folder containing Convolution filters like sharpen, which can be used for the
Custom Filter node.

Chapter 15 Preferences 357

 – Fonts: The default path map for Fonts points to the operating system fonts folders. Changing
this will change the fonts that are available in the Text+ or Text 3D nodes as well as any Fusion
Title Template. In DaVinci Resolve. This path map does not affect the five additional Edit page
titles (L Lower 3rd, R Lower 3rd, M Lower 3rd, Scroll, and Text.)
– Fuses: Points to a folder containing Fusion Fuses plug-ins.
– FusionTemplates: Location where Fusion macros are saved in order to appear as templates in
Fusion’s Effects Library.
– Guides: Location where custom viewer guide overlays are stored.
– Help: Identifies where Fusion Studio PDF files are located.
– Layouts: Location where Fusion Studio custom window layouts are saved.
– Libraries: Points to a support folder where custom Effects Library items can be stored.
– LoaderCache: The Fusion Studio Loader preferences allow the Loader to cache when
reading from a slow network. This path map point to the local drive location for that cache.
– LuaModules: Location for Lua Scripting modules.
– LUTs:
– Macros: Points to the location for user created macros. The macros saved to this location
appear in the macros category of the Effects Library and in the right-click Edit Macro
contextual menu.
– Plugins: This refers to user specific OpenFX plug-ins that you do not want loaded for all users.
– Previews: Path map used for the older style, file sequence flipbook previews.
– Queues: Location of the Render manager list.
– Scripts: Location of Lua and Python scripts. This path can be further refined into specific scripts
for tools (nodes), comps, and other specific script types.
– Settings: Location where custom Node settings are saved.
– Stamps: Location for preview movies generated in a Fusion Studio bin. This is an outdated path
map since bins now include the Studio Player.
– Templates: Location of the Templates folder. Saving Macros to the Template folder will cause
them to appear in the Effects Library in a Templates category. In Fusion Studio, the Templates
category does not appear until a Macros is saved into the folder.
– Thumbs: Location for clip thumbnails generated in a Fusion Studio bin. This is an outdated path
map since bins include now include the Studio Player.
– UserPaths: Used for locations of studio- or facility-specific tools, like custom plug-ins and
scripts located on a central server.

– User Path Maps: User paths are new paths that you have defined that do not currently exist in the
Defaults settings.
– Comp refers to the folder where the current composition is saved. For instance, saving media
folders within the same folder as your Fusion Studio comp file is a way to use relative file paths
for Loaders instead of actual file paths.

Modifying a System Path Map

To modify an existing System path map, select the path map in the System section. Click the folder
icon at the bottom of the Preferences window, and enter the name of the path map in the From field
below. Enter the value of the path map in the To: field.

Chapter 15 Preferences 358

Modifying a Default Path Map
To modify an existing Default path map, select the path map in the Default section. Click the folder
icon at the bottom of the Preferences window, and enter the name of the path map in the From field
below. Enter the value of the path map in the To: field.

Creating a User Path Map

To create a path map, click on the New button and enter the name of the path map in the From field
below. Enter the value of the path map in the To: field.

Deleting a Path Map

To delete a user-created path map, select it from the list and click the Delete button. System and
Default path maps cannot be deleted; only user created path maps can be removed from the
Path Maps list.

Nesting Path Maps

When defining your own path map, you can use an existing path map in the new definition.
For example, define a path map called ‘Episode’ that maps to MyDrive\ Projects\Episode1. Then create
new path maps called Renders and Stills that map to Episode\ Renders_v1 and Episode\Stills_v1.

Preview
Preview is only available in Fusion Studio. Previews in DaVinci Resolve use the Scratch Disk setting in
the Media Storage preferences.
In the Preview preferences, you configure the creation and playback options for preview renders.

The Preview preferences

Chapter 15 Preferences 359

Options
– Render Previews Using Proxy Scaling: When checked, this option scales down the images to the
preview size for the Loader and Creator tools. This causes much faster rendering. If this option is
disabled, frames will be rendered at full size and are then scaled down.
– Skip Frames to Maintain Apparent Framerate: When checked, frames are skipped during
playback of Flipbooks and file sequences to maintain the frame rate setting.
– Show Previews for Active Loaders: This setting determines whether the preview playback
controls are shown below the Inspector when a Loader with a valid file is activated.
– Show Previews for Active Savers: This setting determines whether the preview playback controls
below the Inspector are shown when a Saver with a valid file is activated.
– Display File Sequences On: This setting determines which viewer or external monitor is used for
the interactive and file sequence playbacks as well as for the scrubbing function in the bins.

QuickTime
The QuickTime preferences are only available in Fusion Studio on macOS. These settings configure
the default QuickTime codec settings when you select QuickTime as the rendering file format in the
Saver node.

The QuickTime preferences

– Compressor: This drop-down menu displays the QuickTime codecs available from your computer.
Fusion tests each codec when the program is started; therefore, some codecs may not be
available if the tests indicate that they are unsuitable for use within Fusion.
– Quality: This slider is used to determine the amount of compression to be used by the codec.
Higher values produce clearer images but larger files. Not all codecs support the Quality setting.

Chapter 15 Preferences 360

 – Key Frame Every X Frames: When checked, the codec will create key frames at specified
intervals. Key frames are not compressed in conjunction with previous frames and are, therefore,
quicker to seek within the resulting movie. Not all codecs support the key frame setting.
– Limit Data Rate To X KB/Second: When checked, the data rates of the rendered file will be limited
to the amount specified. Not all codecs support this option. Enter the data rate used to limit the
QuickTime in kilobytes (kB) per second, if applicable. This control will have no effect if the Limit
Data Rate To option is not selected.

Script
The preferences for Scripting include a field for passwords used to execute scripts from the command
line and applications for use when editing scripts.

The Script preferences

Login
There are three login options for running scripts outside of the Fusion application.
– No Login Required to Execute Script: When enabled, scripts executed from the command line, or
scripts that attempt to control remote copies of Fusion, do not need to log in to the workstation in
order to run.
– Specify Custom Login: If a username and password are assigned, Fusion will refuse to process
incoming external script commands (from FusionScript, for example), unless the Script first logs in
to the workstation. This only affects scripts that are executed from the command line, or scripts
that attempt to control remote copies of Fusion. Scripts executed from within the interface do not
need to log in regardless of this setting. For more information, see the Scripting documentation.

Chapter 15 Preferences 361

 – Use Windows Login Validation: When using Fusion on Windows, enabling this option verifies
the user name and password (also known as credentials) with the operating system before running
the script.

Options
– Script Editor: Use this preference to select an external editor for scripts. This preference is used
when selecting Scripts > Edit.

Python Version
– Two options are presented here for selecting the version of Python that you plan on using
for your scripts.

Spline Editor
The Spline Editor preferences allow you to set various spline options for Autosnap behavior, handles,
markers, and more. This only affects splines displayed in the Spline Editor, not splines created in the
viewer using the polygon tool or paths.

The Spine Editor preferences

Chapter 15 Preferences 362

Spline Editor Options
These settings control the spline behavior in the Spline Editor, as well as the appearance of the
graph area.
– Independent Handles: Enabling this option allows the In or Out direction handle on newly created
key frames to be moved independently without affecting the other. This option is also available via
the Options submenu when right-clicking in the Spline Editor graph.
– Follow Active: The Spline Editor focuses on the currently active tool. This option is also available
via the Options submenu when right-clicking in the Spline Editor graph.
– Show Key Markers: Small colored triangles will be displayed at the top of the Spline Editor
Time Ruler to indicate key frames on active splines. The colors of the triangles match the colors of
the splines. This option is also available via the Show submenu when right-clicking in the Spline
Editor graph.
– Show Tips: Toggles if tooltips are displayed or not. This option is also available via the Show
submenu when right-clicking in the Spline Editor graph.
– Autosnap Points: When moving points in the Spline Editor, these will snap to the fields or frames
or can be moved freely. This option is also available via the Options submenu when right-clicking
in the Spline Editor graph.
– Guides: When moving points in the Spline Editor, these will snap to guides as well. This option is
also available via the Options submenu when right-clicking in the Spline Editor graph.
– Autosnap Guides: When moving or creating guides, these will snap to the fields or frames or can
be moved freely. This option is also available via the Options submenu when right-clicking in the
Spline Editor graph.
– Autoscale: Keeps the Spline Editor scales intact on changing the editable spline content of the
graph. This scale is also available via the Options submenu when right-clicking in the Spline Editor
graph.
– Scroll: Scrolls horizontally and vertically to show all or most of the spline points. This option is also
available via the Scale submenu when right-clicking in the Spline Editor graph.
– Fit: Zooms to fit all points within the spline graph, if necessary. This option is also available via the
Scale submenu when right-clicking in the Spline Editor graph.

LUT View Options

LUT splines are not currently available in the Spline Editor, and therefore these options are not
operational in the current version.
– Independent Handles: Enabling this option allows the In or Out direction handle on newly created
key frames to be moved independently without affecting the other.
– Show Key Markers: Small colored triangles will be displayed at the top of the Spline Editor Time
Ruler to indicate key frames on active splines. The colors of the triangles match the colors of
the splines.
– Show Tips: Toggles whether tooltips are displayed.

Chapter 15 Preferences 363

Splines
Options for the handling and smoothing of animation splines, tracker path defaults, and rotoscoping
are found in the Splines preferences.

The Splines preferences

– Autosmooth: Automatically smooths out any newly created points or key frames on the splines
selected in this section. You can choose to automatically smooth animation splines, B-Splines,
polyline matte shapes, LUTs, paths, and meshes.
– B-Spline Modifier Degree: This setting determines the degree to which the line segments
influence the resulting curvature when B-Splines are used in animation. Cubic B-Splines determine
a segment through two control points between the anchor points, and Quadratic B-Splines
determine a segment through one control point between the anchor points.
– B-Spline Polyline Degree: This setting is like the one above but applies to B-Splines
used for masks.
– Tracker Path Points Visibility: This setting determines the visibility of the control points on tracker
paths. You can show them, hide them, or show them when your cursor hovers over the path, which
is the default behavior.
– Tracker Path: The default tracker creates Bézier-style spline paths. Two other options in this
setting allow you to choose B-Spline or XY Spline paths.
– Polyline Edit Mode on Done: This setting determines the state of the Polyline tool after you
complete the drawing of a polyline. It can either be set to modify the existing control points on the
spline or modify and add new control points to the spline.
– Onion Skinning: The Onion Skinning settings determine the number of frames displayed while
rotoscoping, allowing you to preview and compare a range of frames. You can also adjust if the
preview frames only from the frame prior to the current frame, after the current frames, or split
between the two.

Chapter 15 Preferences 364

Timeline
The Timeline preferences is where you create and edit Keyframes Editor/Spline Editor filters and set
default options for the Keyframes Editor.

The Timeline preferences

Filter/Filter to Use
The Filter menu populates the hierarchy area below the menu with that setting. It lets you edit the
filters. The Filter to Use menu selects the default filter setting located in the Keyframes Editor
Options menu.

Settings for Filters

This area is used to create a new filter and define its settings. You start by first clicking the New button
and entering the name of the new Filter. You then select any of the tools that you want the filter to
contain. Only tools that are checked will appear in the Keyframes Editor or Spline Editor when the filter
is selected. You can also create a copy of the filter using the Copy button or remove a filter from the
list by clicking the Delete button.

Timeline Options
The Timeline Options configure which options in the Keyframe Editor are enabled by default. A series
of checkboxes correspond to buttons located in the Timeline, allowing you to determine the states of
those buttons at the time a new comp is created. For more information on the Keyframes Editor
functions, see Chapter 70, “Animating in Fusion’s Keyframes Editor” in the DaVinci Resolve Reference
Manual or Chapter 9 in the Fusion Reference Manual.
– Autosnap Points: When moving points in the Keyframes Editor, the points will snap to the fields or
to the frames, or they can be moved freely.

Chapter 15 Preferences 365

 – Guides: When moving points in the Keyframes Editor, the point will snap to the guides that are
placed in the Timeline graph.
– Autosnap Guides: When moving or creating guides, the guides will snap to the fields or to the
frames, or they can be moved freely.
– Autoscale: Keeps the Timeline scales intact while changing the editable spline content in the
graph. When set to scroll, the Timeline scrolls horizontally and vertically to show all or most of the
spline points when changing the editable spline content in the graph. When set to Fit, the Timeline
zooms to fit all points within the graph, if necessary.
– Tools Display Mode: This menu controls the default sort order of the tools displayed in the
Keyframes Editor. The default can be changed using the Sort order menu in the upper right of the
Keyframes Editor.

Tweaks
The Tweaks preferences handle a collection of settings for fine-tuning Network rendering in Fusion
Studio and graphics hardware behavior.

The Tweaks preferences

Network
The Network section is used to control and monitor the health of communication packets over TCP/IP
when rendering over a network in Fusion Studio.
– Maximum Missed Heartbeats: This setting determines the maximum number of times the network
is checked before terminating the communication with a Render node.
– Heartbeat Interval: This sets the time between network checks.

Chapter 15 Preferences 366

 – Load Composition Timeout: This timeout option determines how long the Render Manger will wait
for a composition to load before moving on to another task.
– Last Slave Restart Timeout: This timeout option determines how long the Render Manager will
wait for a render salve to respond before using another render slave.

File I/O
The File I/O options are used to control the performance when reading frames or large media files
from both direct and networked attached storage.
– I/O Canceling: This option enables a feature of the operating system that allows queued
operations to be canceled when the function that requested them is stopped. This can improve
the responsiveness, particularly when loading large images over a network.
Enabling this option will specifically affect performance while loading and accessing formats that
perform a large amount of seeking, such as the TIFF format.
This option has not been tested with every hardware and OS configuration, so it is recommended
to enable it only after you have thoroughly tested your hardware and OS configuration using drive
loads from both local disks and network shares.
– Enable Direct Reads: Enabling this checkbox uses a more efficient method when loading a large
chunk of contiguous data into memory by reducing I/O operations. Not every operating system
employs this ability, so it may produce unknown behavior.
– Read Ahead Buffers: This slider determines the number of 64K buffers that are use to read ahead
in a file I/O operation. The more buffers, the more efficient loading frames from disk will be, but the
less responsive it will be to changes that require disk access interactively.

Area Sampling
The Area Sampling options allow you to fine-tune the RAM usage on Render nodes by trading off
speed for lower RAM requirements.
– Automatic Memory Usage: This checkbox determines how area sampling uses available memory.
Area sampling is used for Merges and Transforms. When the checkbox is enabled (default), Fusion
will detect available RAM when processing the tool and determine the appropriate trade-off
between speed and memory.
If less RAM is available, Fusion will use a higher proxy level internally and take longer to render.
The quality of the image is not compromised in any way, just the amount of time it takes to render.
In node trees that deal with images larger than 4K, it may be desirable to override the automatic
scaling and fix the proxy scale manually. This can preserve RAM for future operations.
– Pre-Calc Proxy Level: Deselecting the Automatic Memory will enable the Pre-Calc Proxy Scale
slider. Higher values will use less RAM but take much longer to render.

Open GL
This section controls how Fusion makes use of your graphics card when compositing in 3D with the
Renderer 3D node. Most settings may be left as they are, but since OpenGL hardware varies widely in
capabilities and different driver revisions can sometimes introduce bugs, these tweaks can be useful if
you are experiencing unwanted behavior.
– Disable View LUT Shaders: OpenGL shaders can often dramatically accelerate View LUTs, but
this can occasionally involve small trade-offs in accuracy. This setting will force Fusion to process
LUTs at full accuracy using the CPU instead. Try activating this if View LUTs do not seem to be
giving the desired result.

Chapter 15 Preferences 367

 – Use Float16 Textures: If your graphics hardware supports 16-bit floating-point textures, activating
this option will force int16 and float32 images to be uploaded to the viewer as float16 instead,
which may improve playback performance.
– Texture Depth: Defines in what depth images are uploaded to the viewer.
– Auto: The Auto option (recommended) lets Fusion choose the best balance of performance
and capability.
– int8: Similar to the Use Float16 Textures switch, this option can be used to force images to be
uploaded to the Display View as int8, which can be faster but gives less range for View LUT
correction.
– Native: The Native option uploads images at their native depth, so no conversion is done.

– Image Overlay: The Image Overlay is a viewer control used with Merge and Transform tools
to display a translucent overlay of the transformed image. This can be helpful in visualizing the
transformation when it is outside the image bounds but may reduce performance when selecting
the tool if cache memory is low. There are three settings to choose from: None, Outside, and All.
– None: This setting never displays the translucent overlay or controls, which can reduce the
need for background renders, in some cases resulting in a speed up of the display.
– Outside: This will display only those areas of the control that are outside the bounds of the
image, which can reduce visual confusion.
– All: Displays all overlays of all selected tools.

– Smooth Resize: This setting can disable the viewer’s Smooth Resize behavior when displaying
floating-point images. Some older graphics cards are not capable of filtering floating-point
textures or may be very slow. If Smooth Resize does not work well with float images, try setting
this to flt16 or int.
– Auto Detect Graphics Memory (MB): Having Fusion open alongside other OpenGL programs
like 3D animation software can lead to a shortage of graphics memory. In those cases, you can
manually reduce the amount of memory Fusion is allowed to use on the card. Setting this too low
or too high may cause performance or data loss.
– Use 10-10-10-2 Framebuffer: If your graphics hardware and monitor support 30-bit color (Nvidia
Quadro/AMD Radeon Pro, and some Nvidia GeForce/AMD Radeon), this setting will render
viewers with 10 bits per primary accuracy, instead of 8 bits. Banding is greatly reduced when
displaying 3D renders or images deeper than 8-bit.

User Interface
The User Interface preferences set the appearance of the user interface window and how the
Inspector is displayed.

Chapter 15 Preferences 368

 The User Interface preferences

Appearance
When enabled, the Use Gray Background Interface checkbox will change the color of the background
in Fusion’s panels to a lighter, more neutral shade of gray.

Controls
This group of checkboxes manages how the controls in the Inspector are displayed.
– Auto Control Open: When disabled, only the header of the selected node is displayed in the
Inspector. You must double-click the header to display the parameters. When enabled, the
parameters are automatically displayed when the node is selected.
– Auto Control Hide: When enabled, only the parameters for the currently active tool (red outline)
will be made visible. Otherwise, all tool headers will be visible and displayed based on the Auto
Control Open setting.
– Auto Control Close Tools: When enabled, only the active (red outlined) tool in the Node Editor will
have controls displayed. Any previous active node’s tools will be closed in the Inspector. When
disabled, any number of tools may be opened to display parameters at the same time. This setting
has no effect if the Auto Control Hide checkbox is enabled.
– Auto Control Close Modifiers: When enabled, only one modifier’s parameters will be displayed for
the active node. Any additional modifiers for the active node will show only their header.
– Auto Control Advance: If the Auto Control Advanced checkbox is enabled, the Tab key and
Return/Enter key will cause the keyboard focus to advance to the next edit box within the
Inspector. When disabled, Return/Enter will cause the value entered to be accepted, but the
keyboard focus will remain in the same edit box of the control. The Tab key can still be used to
advance the keyboard focus.
– Show Controls for Selected: When this option is disabled, only the active tool’s parameters are
shown in the Inspector. By default, it is enabled, showing controls for the active tool as well as all
selected tools.

Chapter 15 Preferences 369

 – Combined Color Wheel: When the Color Corrector tool is displayed in the Inspector, enabling
this checkbox will show one color wheel with buttons to switch between the master, shadow,
midtones, and highlight channels. Otherwise, four color wheels are displayed in the Inspector.
– Gamma Aware Color Controls: This setting adjusts color correction nodes when working
with Rec. 709 images in a non-color managed project. Rec. 709 images appear correct on the
computer monitor because monitors have a gamma adjustment built in. When working in the Rec.
709 color space without color management, enabling Gamma Aware color removes the gamma,
applies the color correction as if it where linear, and then reapplies the gamma. For Rec. 709
images, enable the Gamma Aware setting and enter a Gamma value of 2.4. In a color managed
linear project, this should be set to Off or a value of 1.0. When dealing with mixed color spaces,
Fusion reads the metadata from the image and sets the Aware gamma value based on the
metadata available.
– Grab Distance: This slider ranges from 1 to 10 and defaults to 5. It designates the active area
around the mouse pointer and can be modified if you have difficulties in selecting points for
modification in paths and spline curves. Smaller values will require a more accurate selection with
the mouse pointer.

Touch Scrolling and Mouse Wheel

This group of settings allows you to configure which, if any, keyboard modifiers are needed to pan or
zoom a panel when using a trackpad or middle mouse wheel.

Video Monitoring
This setting is only available in Fusion Studio. Control over video hardware for the Fusion Page is done
in the DaVinci Resolve preferences. The Video Monitoring preferences are used to configure the
settings of Blackmagic Design capture and playback products such as DeckLink PCIe cards and
UltraStudio i/O units.

The Video Monitoring preferences

Chapter 15 Preferences 370

Video Output
This group of drop-down menus allows you to select the type of video I/O device you have installed,
the output resolution, and the pixel format. These settings have nothing to do with your rendered
output; it is only for your display hardware.
The Output HDR over HDMI settings are used to output the necessary metadata when sending high
dynamic range signals over HMDI 2.0a and have it correctly decided by an HDR capable video display.
The Auto setting detects the image’s values and outputs HDR. This will not affect non HDR images.
The Always setting forces HDR on all the time. This can can be useful when checking non HDR and
HDR grades.
When Auto or Always is selected, you can then set the “nit” level (slang for cd/m2) to whatever peak
luminance level your HDMI connected HDR display is capable of.

Stereo Mode
This group of settings configures the output hardware for displaying stereo 3D content.
– Mono will output a single non stereo eye.
– Auto will detect which method with which the stereo images are stacked.
– Use the Vstack option if the stereo images are stacked vertically as left on top and
right at the bottom.
– Use the Hstack option if the stereo images are stacked horizontally as left and right.
The Swap eyes checkbox will swap the eyes if stereo is reversed.

View
The View preferences are used to manage settings and default controls for viewers.

The View preferences

Chapter 15 Preferences 371

Saved View Settings
The area at the top of the view preferences lists the currently saved settings that you create from the
viewer’s contextual menu. You can use the Rename and Delete buttons to manage the selected
entries in the list. For more information on the viewer and its contextual menu, see Chapter 68 “Using
Viewers” in the DaVinci Resolve Reference Manual or Chapter 7 in the Fusion Reference Manual.

Settings for View

Each viewer has its own preferences. The Settings for View drop-down menu is used to select the
viewer you want to configure.

Control Colors
The Control Colors setting allows you to determine the color of the active/inactive onscreen controls.

Color Picking Area Size

You can use these width/height controls to set the number of pixels sampled when using the Color
Picker in the viewers.

Displayed Depth Range

The Displayed Depth Range setting controls the view normalization of the Z-Channel.

Fit Margin
The Fit Margin setting determines how much padding is left around the frame when the Fit button is
pressed or Fit is selected from the viewer’s contextual menu.

Display LUT Plug-Ins

This list shows the available display LUTs and activates the selected one as default.

VR Headsets
The VR Headsets preferences allow configuration of any connected Virtual Reality headsets, including
how stereo and 3D scenes are viewed.

Chapter 15 Preferences 372

Headset Options
The Headset options are used to select the type of VR headset you are using to view the composite
as well as the video layout of the 360° view.

API
– Disabled: Disabled turns off and hides all usage of headsets.
– Auto: Auto will detect which headset is plugged in.
– Occulus: Occulus will set the VR output to the Oculus headset.
– OpenVR: OpenVR will support a number of VR headsets like the HTC Vive.

360° Video Format

– Auto: Auto will detect the incoming image layout from the metadata and image frame aspect.
– VCross and HCross: VCross and HCross are the six square faces of a cube laid out in a cross,
vertical or horizontal, with the forward view in the center of the cross, in a 3:4 or 4:3 image.
– VStrip and HStrip: VStrip and HStrip are the six square faces of a cube laid vertically or
horizontally in a line, ordered as Left, Right, Up, Down, Back, Front (+X, -X, +Y, -Y, +Z, -Z), in
a 1:6 or 6:1 image.
– LatLong: LatLong is a single 2:1 image in equirectangular mapping.
– Enable Mirror Window: Enable Mirror Window will show a window displaying the
headset user’s live view.

Stereo
Similar to normal viewer options for stereo 3D comps, these preferences control how a stereo
3D comp is displayed in a VR headset.

Mode
– Mono: Mono will output a single non stereo eye.
– Auto: Auto will detect the method with which the stereo images are stacked.
– Vstack: Vstack stereo images are stacked vertically as left on top and right at the bottom.
– Hstack: Hstack stereo images are stacked horizontally as left and right.
– Swap Eyes: Swap eyes will swap the eyes if stereo is reversed.

3D
Similar to normal viewer options for 3D comps, these preferences control how a 3D comp is displayed
in a VR headset.

Lighting
– Disabled lighting is off.
– Auto will detect if lighting is on in the view.
– On will force lighting on in the VR view.

Sort Method
– Z buffer sorting is the fast OpenGL method of sorting polygons.
– Quick Sort will sort the depth of polygons to get better transparency rendering.
– Full Sort will use a robust sort and render method to render transparency .
– Shadows can be on or off.
– Show Matte Objects will make matte objects visible in view or invisible.

Chapter 15 Preferences 373

Bins/Security
Bins preferences are only available in Fusion Studio. These preferences are used to manage the Bin
users and their permissions.

The Bins Security preferences

Users List
The Users List is a list of the users and their permissions. You can select one of the entries to edit their
settings using the User and Password edit boxes.
– Add: The Add button is used to add a new user to the list by entering a username and password.
– Remove: Click this button to remove the selected entry.

User
This editable field shows the username for the selected Bin Server item. If the username is unknown,
try “Guest” with no password.

Password
Use this field to enter the password for the Bin user entered in the Users list.

Permissions
The administrator can set up different permission types for users.
– Read: This will allow the user to have read-only permission for the bins.
– Create: This will allow the user to create new bins.
– Admin: This gives the user full control over the bins system.
– Modify: This allows the user to modify existing bins.
– Delete: This allows the user to remove bins.

Chapter 15 Preferences 374

Bins/Server
These preferences are used to add Bin Servers to the list of bins Fusion will display in the Bins dialog.

The Bin Servers preferences

Servers
This dialog lists the servers that are currently in the connection list. You can select one of the entries
to edit its settings.
– Add: Use this button to add a new server to the list.
– Remove: Click this button to remove the selected entry.

Server
This editable field shows the name or IP address of the server for the selected entry in the list.

User
This editable dialog shows the username for the selected Bin Server item.

Password
Use this field to enter the password for the server entered in the Server list.

Library
The Library field lets you name the bins. If you wanted to create a bin for individual projects, you would
name it in the Library field and each project would gets its own bin.

Application
The Application field allows larger studios to specify some other program to serve out the
Bin requests.

Chapter 15 Preferences 375

Bins/Settings
These preferences are used to control the default behavior of bins.

The Bins Settings preferences

Stamp Quality
The Stamp Quality is a percentage slider that determines the compression ratio used for Stamp
thumbnail creation. Higher values offer better quality but take up more space.

Stamp Format
This drop-down list determines whether the Stamp thumbnails will be saved as compressed or
uncompressed.

Options
– Open Bins on Startup: When Open Bins on Startup is checked, the bins will open automatically
when Fusion is launched.
– Checker Underlay: When the Checker Underlay is enabled, a checkerboard background is used
for clips with alpha channels. When disabled, a gray background matching the Bin window is used
as the clip’s background.

Chapter 15 Preferences 376

EDL Import
The EDL Import options are used to determine how compositions are created from imported CMX-
formatted EDL files.

The EDL Import preferences

Flow Format
This drop-down menu provides three options that determine how the node tree is constructed for the
imported EDL file.
– Loader Per Clip: A Loader will be created for each clip in the EDL file.
– A-B Roll: A node tree with a Dissolve tool will be created automatically.
– Loader Per Transition: A Loader with a Clip list will be created, representing the imported EDL list.

Use Shot Names

When checked, shot names stored in the EDL file are used to locate the footage.

Chapter 15 Preferences 377

Customization
The following section covers the customization of preferences that are not technically part of the
Preferences window. Using Fusion Studio’s Hotkey Manager window, you can customize the keyboard
shortcuts, making the entire process of working in Fusion not only faster but potentially more familiar if
you are migrating from another software application. You can also customize Fusion with environment
variables to switch between different preferences files, allowing different working setups based on
different users or job types. Both of these customization options are only available in Fusion Studio.

Shortcuts Customization
Keyboard shortcuts can be customized in Fusion Studio. You can access the Hotkey Manager by
choosing Customize HotKeys from the View menu.

Fusion has active windows to focus attention on those areas of the interface, like the Node Editor,
the viewers, and the Inspector. When selected, a gray border line will outline that section. The
shortcuts for those sections will work only if the region is active. For example, Command-F in the
View will scale the image to fit the view area; in the Flow view, Command-F will open the Find tool
dialog; and in the Spline editor, it will fit the splines to the window.
On the right is a hierarchy tree of each section of Fusion and a list of currently set hotkeys. By
choosing New or Edit, another dialog will appear, which will give specific control over that hotkey.

Chapter 15 Preferences 378

Creating a new keyframe will give you the key combo to press, and this Edit Hotkey dialog will appear
where the Action can be defined at top right: pressed, repeated, or released. The Name and
abbreviated Short Name can be set, as can the Arguments of the action.

Customizing Preferences
Fusion Studio’s preferences configure Fusion’s overall application default settings and settings for
each new composition. Although you access and set these preferences through the Preferences
window, Fusion saves them in a simple text format called Fusion.prefs.
These default preferences are located in a \Profiles\Default folder and shared by all Fusion users on
the computer. However, you may want to allow each user to have separate preferences and settings,
and this requires saving the preferences to different locations based on a user login.
To change the saved location of the preferences file requires the use of environment variables.

Setting the Preferences Location

When you fist open Fusion, the environment variable FUSION_PROFILE_DIR defines the folder that
contains the Profiles folder. If this variable defines a valid path, then the preferences are saved to this
folder. If the FUSION_PROFILE_DIR does not exist, then Fusion attempts to create it. If it cannot create
the path, then the preferences are stored in the default Path Map location: AllData:\Profiles.
Typically, all users share the same preferences.If you want each user to save separate preferences
within their home folder, you must create another environment variable with the name FUSION_
PROFILE (e.g., FUSION_PROFILE=jane). Using this second environment variable, Fusion will look for
the preferences in the PROFILE_DIR of the user profile. Using a login script, you can make sure the
FUSION_PROFILE is set to the name of the logged in user.

Creating a Master Preferences File

When working with multiple Fusion users in a studio, you may want to standardize on a few settings.
Using the FUSION_MasterPrefs environment variable, you can create one or more site-wide
preferences in addition to your local personal preferences.
FUSION_MasterPrefs must contain the full path to at least one preferences file. If you have multiple
preferences paths, separate them using semicolons. Fusion does not write to these prefs files, and
they may contain a subset of all available settings. You may change settings in these files and use
them only where local prefs do not already exist unless you set the Locked flag.

Locking Preferences
If the line “Locked = true,” appears in the main table of a master file, all settings in that file are locked
and override any other preferences. Locked preferences cannot be altered by the user.

Chapter 15 Preferences 379

 PART 2

2D Compositing
Chapter 16

Controlling Image
Processing and
Resolution
This chapter covers the overall image-processing pipeline. It discusses color bit-depth
and how to control the output resolution in a resolution-independent environment.

Contents
Fusion’s Place in the DaVinci Resolve Image-Processing Pipeline  382
Source Media into the Fusion Page  382
Forcing Effects into the Fusion Page  382
Output from the Fusion Page to the Color Page  383
What Viewers Show in Different DaVinci Resolve Pages  383
Managing Resolution In Fusion  383
Changing the Resolution of a Clip  384
Compositing with Different-Resolution Clips  384
Sizing Between DaVinci Resolve Pages  385
Color Bit Depths  385
Understanding Integer vs. Float  385
Setting Color Depth in Fusion Studio  386
Combining Images with Different Color Depths  387
Advantages of Floating-Point Processing  388

Chapter 16 Controlling Image Processing and Resolution 381

Fusion’s Place in the DaVinci Resolve
Image-Processing Pipeline
When working in a single unified environment like DaVinci Resolve, it is important to understand the
order of operations among the pages. DaVinci Resolve exposes some of this via the order of the page
buttons at the bottom of the screen, with the Media, Cut, and Edit page at the beginning of the chain
and the Color, Fairlight, and Deliver page at the end. However, this isn’t the whole story, especially
when it comes to the Fusion page. The following sections describe where the Fusion page fits in the
image-processing chain of DaVinci Resolve.

Source Media into the Fusion Page

For ordinary, single clips coming in from the Edit or Cut page, the MediaIn node in the Fusion page
represents the source media, as modified by the Clip Attributes window. Although you select the clip
from the Edit or Cut page Timeline, in the Fusion page, the clip is accessed from the Media Pool.

TIP: The decoding or debayering of RAW files occurs prior to all other operations, and as
such, any RAW adjustments will be displayed correctly in the Fusion page.

This means you have access to the entire source clip in the Fusion page, but the render range is set to
match the duration of the clip in the Timeline. You also use the full resolution of the source clip, even if
the Timeline is set to a lower resolution. However, none of the Edit or Cut page Inspector adjustments
carry over into the Fusion page, with the exception of the Lens Correction adjustment.
When you make Zoom, Position, Crop, or Stabilization changes in the Edit or Cut page, they are not
visible in the Fusion page. The same applies to any Resolve FX or OpenFX third-party plug-ins. If you
add these items to a clip in the Edit or Cut page, and then you open the Fusion page, you won’t see
them taking effect. All Edit and Cut page timeline effects and Inspector adjustments, with the
exception of the Lens Correction adjustment, are computed after the Fusion page but before the Color
page. If you open the Color page, you’ll see the Edit and Cut page transforms and plug-ins applied to
that clip, effectively as an operation before the grading adjustments and effects you apply in the Color
page Node Editor.
With this in mind, the order of effects processing in the different pages of DaVinci Resolve can be
described as follows:

Source RAW Clip Fusion Edit/Cut Page Edit/Cut Color

Media Debayering Attributes Effects Inspector Plug-ins Effects
Adjustments Resolve FX

TIP: Retiming applied to the clip in the Edit page Timeline is also not carried over into the
Fusion page.

Forcing Effects into the Fusion Page

There is a way you can force clips with Edit page Inspector adjustments, plug-ins, retiming, and Color
page grades into the Fusion page, and that is to turn that clip into a compound clip. When Edit page
effects and Color page grading are embedded within compound clips, MediaIn nodes corresponding
to compound clips route the effected clip into the Fusion page. However, bringing a compound clip
into the Fusion page does change the resolution of the source clip to match the Timeline resolution.
For more information, see the section “Sizing Between DaVinci Resolve Pages” in this chapter.

Chapter 16 Controlling Image Processing and Resolution 382

Output from the Fusion Page to the Color Page
The composition output from the Fusion page’s MediaOut node are passed on via the Color page’s
source input, with the sole exception that if you’ve added plug-ins to that clip in the Edit or Cut page,
then the handoff from the Fusion page to the Color page is as follows:

Fusion Edit/Cut Page Edit Page Color

Effects Inspector Plug-Ins Effects
Adjustments

What Viewers Show in Different DaVinci

Resolve Pages
Owing to the different needs of compositing artists, editors, and colorists, the viewers show different
states of the clip.
– The Edit page source viewer: Always shows the source media, unless you’re opening a
compound clip that’s been saved in the Media Pool. If Resolve Color Management is enabled, then
the Edit page source viewer shows the source media at the Timeline color space and gamma.
– The Edit page Timeline viewer: Shows clips with all Edit page effects, Color page grades,
and Fusion page effects applied, so editors see the program within the context of all effects
and grading.
– The Fusion page viewer: Shows Media Pool source clips at the Timeline color space and gamma,
but no Edit page Inspector adjustments or Resolve FX effects and no Color page grades.
– The Color page viewer: Shows clips with all Edit page effects, Color page grades, and
Fusion page effects applied.

Managing Resolution In Fusion

There is no formal resolution to a comp in Fusion. Even though opening Fusion > Fusion Settings in the
Fusion page or Preferences in Fusion Studio allows you to set the Width and Height in the Frame
Format panel, those settings only affect the size of Fusion-generated images, like the Background
tool, Fast Noise, and Text+ tool.  The actual resolution of your composition is initially determined by the
source resolution of the input image. However, it can be modified at any time using a variety of
operations and nodes. For example, if you read in a full HD 1920 x 1080 resolution image, your comp
starts at full HD 1920 x 1080 resolution. This is regardless of the Timeline resolution when you are
using the Fusion page in DaVinci Resolve. The initial resolution of the Fusion comp is the size of the
source media. Depending on how you combine images and the nodes you use, the output comp
resolution can be maintained or modified.

TIP: The output of the Fusion page is placed back into the Edit page Timeline based on
DaVinci Resolve’s Image Sizing setting. By default, DaVinci Resolve uses an image sizing
setting called Scale to Fit. This means that even if the Fusion page outputs a 4K composition,
it conforms to 1920 x 1080 if that is what the project or a particular Timeline is set to. Changing
the image sizing setting in DaVinci Resolve’s Project Settings affects how Fusion
compositions are integrated into the Edit page Timeline.

Chapter 16 Controlling Image Processing and Resolution 383

Changing the Resolution of a Clip
If your comp uses a single image, you can change the pixel output resolution in several ways. Three
common tools that change the pixel resolution of a clip are the Resize, Scale, and Crop nodes. A fourth
node, Letterbox, is less commonly used but also changes the pixel resolution of a clip.
These four nodes are located in the Transform category of the Effects library. Resize is also located in
the toolbar.
– Crop: Sets the output resolution of the node using a combination of X and Y size along with X and
Y offset to cut the frame down to the size you want. Crop removes pixels from the image, so if you
later use a Transform node and try to move the image, those pixels are not available.
– Letterbox: Sets the output resolution of the node by adding horizontal or vertical black edges
where necessary to format the frame size and aspect ratio.
– Resize: Sets the output resolution of the node using absolute pixels.
– Scale: Sets the output resolution of the node using a relative percentage of the current input
image size.

TIP: To change resolution and reposition a frame without changing the pixel resolution of a
clip, use the Transform node.

Compositing with Different-Resolution Clips

When you composite images with different resolutions using the Merge node, the image that’s
connected to the orange background input determines the output resolution of the Merge node.
Often, it’s easiest to control the comp resolution right at the start by connecting a node with the
desired output resolution you want to the orange background input on the Merge node. A Background
node is often used in this situation because it consumes meager system resources.

A Background node determines the output resolution of the merge

The Background node sets the output size, and the foreground image is cropped if it is larger.

A Background node created at 1280 x 720 crops the larger foreground.

However, all the pixels of the larger foreground are available for repositioning.

Chapter 16 Controlling Image Processing and Resolution 384

Sizing Between DaVinci Resolve Pages
The order of sizing operations between DaVinci Resolve pages is a bit more nuanced. It’s important to
understand which sizing operations happen in the Fusion page, and which happen after, so you know
which effects alter the image that’s input to the Fusion page, and which effects alter the page’s output.
For example, lens correction, while not strictly sizing, is nonetheless an effect that changes how the
image begins in your Fusion composition. However, the Edit or Cut page stabilization function is an
effect that comes after the Fusion page, so it does not appear in the composition you’re creating.
The order of sizing effects in the different pages of DaVinci Resolve can be described as follows:

Super Edit/Cut Page Fusion Edit/Cut Page Input Output

Scale Lens Transforms Transforms Sizing Sizing
Correction

Sizing with Compound and Fusion Clips

Another way to modify the resolution before clips get handed off from the Edit page to the Fusion
page is to create a compound clip or a Fusion clip. Both compound clips and Fusion clips change the
working resolution of the individual clips to match the Timeline resolution. For instance, if two 4K clips
are stacked one on top of the other in an HD timeline, creating a compound or Fusion clip resizes the
clips to HD. The full resolution of the individual 4K clips is not available in Fusion and is therefore
handed off to the Color page at the rescaled size. To maintain the full resolution of source clips, bring
only one clip into the Fusion page from the Edit or Cut page Timeline, and then bring other clips into
the Fusion composition using the Media Pool. Of course, if your clips are full HD and your timeline is
full HD, then creating a Fusion clip or compound clip does not affect the resolution.

Color Bit Depths

The term bit depth describes how many colors are available in the color palette used to make up an
image. The higher the bit depth, the greater the precision of color in the image, and therefore the
greater the color reproduction. The higher precision is most apparent in gradients with subtle changes.
Lower bit-depth gradients have noticeable banding artifacts, whereas higher bit-depth images can
reproduce more colors, so fewer, if any, banding artifacts occur. The Fusion page within
DaVinci Resolve always uses 32-bit float bits per channel precision to process images. However, in
Fusion Studio you can choose to process images with 8-bit integer, 16-bit integer, 16-bit float, and
32-bit float bits per channel. Although always working at 16-bit float or 32-bit float will produce the best
quality, it may be more efficient to use a lower bit depth if your images are 8-bit or 16-bit integer
formats to begin with.

Understanding Integer vs. Float

Generally, 8-bit integer color processing is the lowest bit depth you’ll come across for video formats.
8-bit images come from older or consumer-grade video equipment like mobile phones and
camcorders. If you try to perform any significant gamma or color correction on 8-bit images, you can
often see more visible banding.
16-bit integer color depth doubles the amount of precision, eliminating problems with banding.
Although you can select 16-bit integer processing for an 8-bit clip, it does not reduce banding that
already exists in the original file. Still, it can help when adding additional effects to the clip. This sounds
like the best solution until you realize that many digital cameras like Blackmagic Design URSA Mini Pro
and others record in formats that can capture over-range values with shadow areas below 0.0 and
super highlights above 1.0, which are truncated in 16-bit integer.

Chapter 16 Controlling Image Processing and Resolution 385

The 16-bit float color depth sacrifices a small amount of the precision from standard 16-bit integer color
depth to allow storage of color values less than 0 and greater than 1.0. 16-bit float, sometimes called
half-float, is most often found in the OpenEXR format and contains more than enough dynamic range
for most film and HDR television purposes yet requires significantly less memory and processing time
than is required for full float, 32-bit images.

Preserving over-range values allows you to

change exposure while maintaining highlights

Processing at 32-bit float can work with shadow areas below 0.0 and highlights above 1.0, similar to
16-bit float, except with a much greater range of precision but also much greater memory and
processing requirements.

Setting Color Depth in Fusion Studio

As we said earlier, DaVinci Resolve always processes at 32-bit float bits per channel; however, you can
use less memory and still achieve more-than-acceptable results using the Performance Mode setting
located in the User > Playback Preferences panel.
Fusion Studio automatically uses the color depth that makes the most sense for each file format.
For example, if you read in a JPEG file from disk, then the color depth for the Loader is set to 8 bits per
channel. Since the JPEG format is an 8-bit format, loading the image at a greater color depth would
generally be wasteful. If a 16-bit TIFF is loaded, the color depth is set to 16 bits. Loading a DPX file
defaults to 32-bit float, whereas OpenEXR generally defaults to 16-bit float. However, you can override
the automatic format color depth using the settings found in the Import tab of the Loader node’s
Inspector. The Loader’s Inspector, as well as the Inspector for images generated in Fusion (i.e., text,
gradients, fast noise, and others), has a Depth menu for 8-bit, 16-bit integer, 16-bit float, and
32‑bit float.

The Loader’s Inspector Color Bit Depth settings

Configuring Default Color Depth Preferences

The default color depth setting forces the tool to process based on the settings configured in the
Node Editor’s Frame Format preferences. These are used to set a default value for color depth,
applied when a Generator tool is added to the Node Editor. There are three drop-down menus to

Chapter 16 Controlling Image Processing and Resolution 386

configure color depth in the preferences. They specify the different color depths for the interactive
session, final renders, and preview renders.
To improve performance as you work on your comp, you can set the Interactive and Preview depth to
8-bits per channel, while final renders can be set to 16-bit integer. However, if your final render output
is 16-bit float or 32-bit float, you should not use the integer options for the interactive setting. The final
results may look significantly different from interactive previews set to integer options.

The Frame Format Color Depth settings

If you aren’t sure what the color depth process is for a tool, you can position the pointer over the
node’s tile in the Node Editor, and a tooltip listing the color depth for that node will appear on the
Status bar.

Hover over a node to view its Color Bit Depth setting.

TIP: When working with images that use 10-bit or 12-bit dynamic range or greater, like
Blackmagic RAW or Cinema DNG files, set the Depth menu in the Inspector to 16-bit float or
32-bit float. This preserves highlight detail as you composite.

Combining Images with Different Color Depths

You can combine images with different color depths in a single composition. When images of different
color depths are combined, the image from the background input of the node determines the bit depth
output, and the foreground image is adjusted to match.

Chapter 16 Controlling Image Processing and Resolution 387

Advantages of Floating-Point Processing
There are two major advantages to floating-point processing that make the additional RAM
requirements and longer render times worth your while. The first benefit is that floating-point values
are more accurate than integer values. The second benefit is the preservation of shadow and highlight
values that go beyond the normal tonal range.

Greater Accuracy
Using 16- or 32-bit floating-point processing prevents the loss of accuracy that can occur when using
8- or 16-bit integer processing. The main difference is that integer values cannot store fractional or
decimal values, so rounding occurs in all image processing. Floating-point processing allows decimal
or fractional values for each pixel, so it is not required to round off the values of the pixel to the closest
integer. As a result, color precision remains virtually perfect, regardless of how many operations are
applied to an image.
If you have an 8-bit pixel with a red value of 75 (dark red) and that pixel is halved using a Color
Correction tool, the pixel’s red value is now 37.5. Since you cannot store decimal or fractional values in
integers, that value is rounded off to 37. Doubling the brightness of the pixel with another Color
Correction tool should bring back the original pixel value of 75 but because of rounding 37 x 2 is 74.
The red value lost a full point of precision due to integer rounding on a very simple example. This is a
problem that can result in visible banding over several color corrections. Similar problems arise when
merging images or transforming them. The more operations that are applied to an image, the more
color precision is lost to rounding when using 8- or 16-bit integer processing.

Accessing Extended Highlights and Shadows

Increasingly more productions are capturing out-of-range images thanks to digital cinema cameras like
the Blackmagic URSA Mini Pro and even the Pocket Cinema 6K camera. These cameras capture very
high dynamic range RAW images and maintain color detail even in heavily over or underexposed
frames. The extended white color detail can also give very nice, natural results when blurred, glowed,
color corrected, or even just when faded or dissolved. While it is possible to work with these RAW
images using integer data, doing so results in the loss of the extended range values, losing all detail in
the highlights and shadows. Float processing makes working with logarithmic RAW images
considerably easier by preserving highlight and shadow detail.
If you have an 8-bit pixel that has a red value of 200 (bright red) and a Color Gain tool is used to
double the brightness of the red channel, the result is 200 x 2, or 400. However, 8-bit color values are
limited to a range of 0 through 255. So the pixel‘s value is clipped to 255, or pure red. If now the
brightness is halved, the result is half of 255, or 127 (rounded), instead of the original value of 200.
When processing floating-point colors, pixel values brighter than white or darker than black are
maintained. There is no value clipping. The pixel is still shown in the viewer as pure red, but if float
processing is used instead of 8-bit, the second operation where the gain was halved would have
restored the pixel to its original value of 200.

Using Float with 8-Bit HD Video

There is also some value to using float color depths with an 8-bit HD video when the images require a
lot of color correction. Using float helps maintain precision by avoiding the rounding errors common to
8-bit processing, as described above.

Detecting Extended Highlight and Shadow Values

Although floating-point processing preserves extended values below 0.0 and greater than 1.0, also
called “out-of-range values,” the viewer still displays them as black or white. This can make it difficult
for you to determine the overall dynamic range of an image.

Chapter 16 Controlling Image Processing and Resolution 388

To discover whether there are out-of-range values in a viewed image:
– Right-click in the viewer and choose Options > Show Full Color Range.

Use the Show Full Color Range pop-up menu to detect out-of-range images.

Enabling this display mode rescales the color values in the image so that the brightest color in the
image is remapped to a value of 1.0 (white), and the darkest is remapped to 0.0 (black).
The 3D Histogram subview can also help visualize out-of-range colors in an image. For more
information, see Chapter 68, “Using Viewers” in the DaVinci Resolve Reference Manual, or Chapter 7
in the Fusion Reference Manual.

Clipping Out-of-Range Values

When processing in floating point, there may be situations where the out-of-range values in an image
need to be clipped. The Brightness/Contrast tool provides checkboxes that can be used to clip
out-of-range values to 0 or 1.
For example, there may be files that contain out-of-range alpha values. Since the alpha channel
represents the opacity of a pixel, it makes little sense to be more than completely transparent or more
than fully opaque, and compositing such an image may lead to unexpected results. To easily clip alpha
values below 0 and above 1, add a Brightness/Contrast toolset to Clip Black and Clip White, with only
the Alpha checkbox selected.

Clip White and Clip Black settings in Brightness/

Contrast can be used to clip mattes.

Alternatively, you can clip the range by adding a Change Depth node and switching to 8-bit or 16-bit
integer color depths.

Chapter 16 Controlling Image Processing and Resolution 389

Chapter 17

Managing Color
for Visual Effects
This chapter discusses LUTs, color space conversions, and the value of
compositing with linear gamma while previewing the image in the viewer using the
gamma of your choice.

Contents
Color Management  391
All Compositing Is Math  391
Introducing Color Management in Fusion  392
Converting to Linear Gamma  392
Applying LUTs to a Viewer  395
Using Resolve Color Management  397
Using ACES Color Management in Resolve  398
Using OCIO for ACES Color Management in Fusion  399
Applying OCIO LUTs in the Viewer  400

Chapter 17 Managing Color for Visual Effects 390

Color Management
The simplified goal of color management is to make sure that the image you see on your computer
screen is what your audience sees when they view it on a television, cinema screen, or mobile device.
For such a simple goal, problems arise that you, as the compositor, must manage. These problems
begin with the fact that our eyes see luminance one way, and a computer display represents
luminance differently.
Each capture device records images using a nonlinear tonal curve or gamma curve to compensate for
this difference. Specifically, Rec. 709 HD gamma curves are designed so that when shown on
HD displays, the images have built-in compensation for the display. The result is that HD images on
HD displays appear normal to us.
Digital cinema cameras have taken the concept of gamma curves further. They use gamma curves as a
way to maximize the bit depth of an image and store a wider dynamic range. Digital cinema cameras’
gamma curve (often collectively referred to as log gamma), give more attention to the darker mid-
tones where the human eye is most sensitive. This allows them to save images with brighter highlights
and more detail in shadows.

A Rec. 709 HD gamma curve (left) and a nonlinear, or log gamma, curve (right)

The problem is that these images do not look normal on any monitor. Clips recorded with a log gamma
curve typically have a low contrast, low saturated appearance when viewed on an sRGB computer
display or Rec. 709 HD video monitor. This problem is easy to fix using a LookUp Table, or LUT. A LUT
is a form of gamma and color correction applied to the viewer to normalize how the image is displayed
on your screen.

A clip displayed with a nonlinear, log gamma curve (left) and corrected in the viewer using a LUT (right)

All Compositing Is Math

The LUT applied to the viewer only solves the problem in the viewer. Now we come to the larger
problem. The image data is still using a log gamma curve. Fusion, and every other image-processing
application, operates with the assumption that the image data has linear gamma. The image-
processing filters you apply to images use standard math functions, like 1 + 1 = 2. Common operations
such as those that add pixels like Brightness, or divide pixels (a.k.a. “unpremultiply”), or composite

Chapter 17 Managing Color for Visual Effects 391

modes that include multiplication such as “screen,” and many other compositing tasks assume that 1 + 1
always equals 2. In other words, if you perform an operation that doubles the amount of brightness,
then every pixel should be twice as bright. However, if you are starting with a nonlinear gamma curve,
pixels are not being adjusted linearly, so some pixels might end up 1.2 x as bright, 1.7 x as bright, or 2.4
x as bright. Now the math is 1 + 1 = 3. The further your images are from linear gamma, the more
pronounced the math error. A Rec. 709 HD clip shows less error than a log gamma clip from a digital
cinema camera. However, an error is still an error, and the more compositing operations you perform
on the image, the more the error is compounded.
You can see a more practical example when you apply filtering effects, such as a blur, to an image with
any gamma setting. The image probably looks fine. However, if you convert the image to a linear
gamma first and then apply the blur, the images (especially those with extremely bright areas) are
processed with greater accuracy, and you should notice a different and superior result.
The answer to these problems is to manage your color before compositing.

Introducing Color Management in Fusion

Images loaded into Fusion by default are not color managed. The image is displayed directly from the
file to the viewer without any interpretation or conversion. However, Fusion includes nodes that
convert the output of each image to linear gamma at the beginning of your composite. The same
nodes can convert from linear back to your desired output gamma at the end of your composite, just
prior to the Saver or MediaOut node.

A log clip converted to linear and then converted for output

To manually set up a linear gamma workflow in Fusion:

1 Use a Gamut or CineonLog node after all MediaIn or Loader nodes to convert them to linear.
2 Apply a GAMUT View LUT to the viewers to correct the display of a linear image to sRGB
or Rec. 709.
3 Before a Saver or MediaOut node, insert a Gamut or CineonLog node to convert from linear to
your target output format.

Converting to Linear Gamma

Whether an image comes from the Edit page in DaVinci Resolve, or from a Loader in Fusion Studio, the
color and gamma are read directly into Fusion, with no modification. For some simple operations on
sRGB or Rec 709 clips, this may be fine, but it’s not always the ideal way to work, especially for
log-encoded media. The ideal way to work with log-encoded media is to convert images to linear
gamma, since the majority of image-processing operations in Fusion expect gamma to be linear and
will produce superior results.

TIP: 3D rendered CGI images are often generated as EXR files with linear gamma, and
converting them is not necessary. However, you should check your specific files to make sure
they are using linear gamma.

Chapter 17 Managing Color for Visual Effects 392

Fusion includes several kinds of nodes you can use to convert the image out of each MediaIn or
Loader node to linear gamma at the beginning of your composite, and then convert from linear back to
your desired output gamma at the end of your composite. These include:
– CineonLog node: The CineonLog node, found in the Film category of the Effects Library,
performs a conversion from any of the formats in the Log Type menu to linear, and also reverses
the process, adding log gamma back to a clip. This is most often used for images coming from
common digital cinema cameras like BlackMagic Design, Arri, or Red. The CineonLog node is
added directly after a MediaIn or Loader node. The Mode menu chooses the direction of the
conversion to or from linear. 

Add a CineonLog tool to convert log gamma

curves to linear and vice versa.

– Gamut node: The Gamut node, found in the Color category of the Effects Library, lets you perform
linear conversions based on color space. This node converts to linear or from linear and is often
inserted after a MediaIn or Loader node or just before a MediaOut or Saver node. Depending
on where you insert the node, you either choose from the Source Space controls or the Output
Space controls.

Add a Gamut tool to convert gamma curves

to linear based on color space.

When converting media to linear gamma, set the Source Space menu to the color space of your
source material. For instance, if your media is full 1080 HD ProRes, then choose ITU-R BT.709
(scene) for gamma of 2.4. Then, enable the Remove Gamma checkbox if it isn’t already enabled, to
use linear gamma.

Chapter 17 Managing Color for Visual Effects 393

 Source Space is used to convert to linear gamma.

When converting from linear gamma for output, you insert the Gamut node before your output
node, which is a Saver in Fusion Studio or a MediaOut node in DaVinci Resolve’s Fusion page.
Make sure the Source Space menu is set to No Change, and set the Output Space to your output
color space. For instance, if your desired output is full 1080 HD, then choose either sRGB or ITU-R
BT.709 (scene) for gamma of 2.4. Then, enable the Add Gamma checkbox if it isn’t already
enabled, to format the output of the Gamut node for your final output.

Output Space is used to convert from linear gamma.

– MediaIn and Loader nodes: MediaIn and Loader nodes have Source Gamma Space controls in
the Inspector that let you identify and remove the gamma curve without the need to add another
node. If your files include gamma curve metadata like RAW files, the Auto setting for the Curve
Type drop-down menu reads the metadata and uses it when removing the gamma curve. When
using intermediate files or files that do not include gamma curve metadata, you can choose either
a log gamma curve by choosing Log from the Curve Type menu or a specific color space using
the Space option from the menu. Clicking the Remove Curve checkbox then removes the gamma
curve, converting the image to linear gamma.

Chapter 17 Managing Color for Visual Effects 394

 MediaIn and Loader nodes include a Remove
Curve checkbox in the Inspector.

– FileLUT node: The FileLUT node, found in the LUT category of the Effects Library, lets you do a
conversion using any LUT you want, giving you the option to manually load LUTs in the ALUT3,
ITX, 3DL, or CUBE format to perform a gamma and gamut conversion. Although LUTs are very
commonly placed at the end of a node tree for final rendering, you’ll get more accurate gamma
and color space conversions using the Gamut and CineonLog nodes to transform your MediaIn
and Loader nodes into linear.

Applying LUTs to a Viewer

Images converted to a linear gamma don’t look correct. They usually look very dark, with extremely
bright highlights and oversaturated colors. Happily, even though the image may appear to be
incorrect, the fact that Fusion can work entirely with floating-point color data means that you’re not
actually clipping or losing any image data. It just looks completely wrong when viewing the linear state
of your image data directly.

A clip displayed with a nonlinear, log gamma curve (left) and the clip transformed to linear gamma (right)

It would be impossible to work if you couldn’t view the image as it’s supposed to appear within the
final gamut and gamma you’ll be outputting. For this reason, each viewer has a LUT menu that lets you
enable a “preview” color space and/or gamma conversion, while the node tree is processing correctly
in linear gamma.

Chapter 17 Managing Color for Visual Effects 395

 Apply a Gamut View LUT to preview the image
in your intended output color space.

To preview the images in the viewer using sRGB or Rec. 709 color space:
1 Enable the LUT button above the viewer.
2 From the Viewer LUT drop-down menu, choose either a Gamut View LUT, or a LUT from the VFX
IO category that transforms linear to Rec. 709 or sRGB.
3 If you choose the Gamut View LUT, then choose Edit from the bottom of the LUT menu to
configure the LUT.
4 In the LUT Editor, set the Output Space to the target color space you want.
5 Enable the Add Gamma checkbox to apply the gamma curve based on the selected color space.

Gamut View LUT Output Space set to Rec. 709

for HD with the Add Gamma checkbox enabled

TIP: If your monitor is calibrated differently, you need to select a LUT

that matches your calibration.

Chapter 17 Managing Color for Visual Effects 396

Whether you use the Gamut View LUT or a LUT for your specific monitor calibration, you can save the
viewer setup as the default.

To Save the Gamut LUT setup as the default viewer setup:

– Right-click in the viewer, and then choose Settings > Save Defaults.
For every comp, the viewer will now be preconfigured based on the saved defaults.
For more information on Viewer LUTs, see Chapter 68 “Using Viewers” in the DaVinci Resolve
Reference Manual or Chapter 7 in the Fusion Reference Manual.

Using Resolve Color Management

If you’re using the Fusion page within Davinci Resolve, you have the option of enabling
DaVinci Resolve’s scene-referred color management, instead of inserting Gamut and CineonLog
nodes. When DaVinci YRGB Color Managed (RCM) is enabled, the color of MediaIn nodes in the
Fusion page is handled differently. The RCM Timeline color space setting is used for all MediaIn nodes
in the Fusion page, which then automatically get converted to linear gamma. The MediaOut node then
gets converted back into the Timeline color space to get graded in the Color page or further edited in
the Edit page Timeline.
Despite the seeming complexity of color management, using RCM is actually simple. In essence, all
you have to do is (A) turn on RCM, (B) assign the appropriate Input Color Space to each clip in the
Media Pool based on its recording device, and (C) choose the Timeline and Output Color Space
combination you want to use.

To enable Resolve Color Management:

1 Open the Color Management panel of the Project Settings.
2 Choose DaVinci YRGB Color Managed from the Color Science drop-down menu.
3 Choose the Input Color Space for the clips in the Media Pool.
4 Set the desired options for the Timeline Color Space and Output Color Space RCM settings.

When DaVinci YRGB Color Managed is enabled, Timeline color

space is used for all MediaIn nodes in the Fusion page.

To override the input color space for differently recorded clips in the Media Pool:
1 Enable DaVinci YRGB Color Management as explained above.
2 Save and close the Settings dialog.
3 In the Media Pool, select the clip or clips you want to assign a new Input Color space.
4 Right-click one of the selected clips.
5 Choose the Input Color Space that corresponds to those clips from the contextual menu.

Chapter 17 Managing Color for Visual Effects 397

Using RCM eliminates a few steps, since the input color space math used to transform the source
preserves all wide-latitude image data, making highlights easily retrievable without any extra steps.
With RCM enabled, there is no need to insert CineonLog or Gamut nodes while in the Fusion page.
The transforms from and to linear are done automatically based on the RCM settings. Switching to the
Fusion page converts the images to linear and enables the LUT button in the viewers with the
Managed LUT selected. The Managed LUT uses the RCM settings to take a linear image and display it
based on the RCM output color space.
For more information on Resolve Color Management, see Chapter 8, “Data Levels, Color Management,
and ACES” in the DaVinci Resolve Reference Manual.

Using ACES Color Management in Resolve

The ACES (Academy Color Encoding Specification) color space is another standard for managing color
throughout an entire production. It’s designed to make start-to-finish, scene-referred color
management a reality for digital cinema workflows. Just like DaVinci’s RCM, ACES makes it easier to
extract high-precision, wide-latitude image data from raw camera formats, to preserve high-quality
image data from acquisition through the color grading process and to output high-quality data for
broadcast viewing, film printing, or digital cinema encoding.
ACES works by assigning an IDT (Input Device Transform) to every camera and acquisition device. The
IDT specifies how media from that device is converted into the ACES color space. At the end of the
pipeline, an ODT (Output Device Transform) is applied to convert the image data from ACES color
space into the gamut of your final output.
Similar to setting up RCM, DaVinci Resolve’s color management project settings can be configured for
ACES, which carries through the Edit, Fusion, and Color pages.

NOTE: When using Fusion Studio, the OpenColorIO (OCIO) framework is used for ACES color
management.

The Color Science drop-down menu in the Color Management panel of the Project Settings is used to
set up the ACES color management in DaVinci Resolve.

When ACES is enabled, IDT and ODT are used to identify input and output devices.

Chapter 17 Managing Color for Visual Effects 398

 – Color Science: Using this drop-down menu, you can choose either ACEScct or ACEScc color
science. This is primarily a personal preference since they are mostly identical, but the shadows
respond differently to grading operations. In the Fusion page, images are automatically converted
to linear, so whoever does the grading has more of a reason to choose one or the other.
– ACEScc: Choose ACEScc color science to apply a standard Cineon style log encoding to the
ACES data before it is processed by DaVinci Resolve.
– ACEScct: This variation of ACEScc adds a roll-off at the toe of the image to make color
correction lift operations “feel” more like they do with film scans and LogC encoded images.

– ACES Version: When you’ve chosen one of the ACES color science options, this menu becomes
available to let you choose which version of ACES you want to use. As of DaVinci Resolve 16, you
can choose either ACES 1.0.3 or ACES 1.1 (the latest version).
– ACES Input Device Transform: This menu lets you choose which IDT (Input Device Transform) to
use for the dominant media format in use.
– ACES Output Device Transform: This menu lets you choose an ODT (Output Device Transform)
with which to transform the image data to match your required deliverable.
– Process Node LUTs In: This menu lets you choose how you want to process LUTs in the Color
page and does not affect the Fusion page.

For more information on ACES within Davinci Resolve, see Chapter 8, “Data Levels, Color Management
and ACES” in the DaVinci Resolve Reference Manual.

Using OCIO for ACES

Color Management in Fusion
When using Fusion Studio or not using color management in DaVinci Resolve, you have the option to
use OpenColorIO nodes in Fusion to composite in an ACES color space.
OpenColorIO (OCIO) is an open-source color management framework for visual effects and computer
animation. OCIO is compatible with the Academy Color Encoding Specification (ACES). Three OCIO
nodes located in the Color category of the Effects Library allow you to use OCIO color space
transforms in Fusion.
– OCIO CDL Transform node allows you to create, save, load, and apply a Color Decision
List (CDL) grade.
– OCIO Color Space allows sophisticated color space conversions based on an OCIO config file.
– OCIO File Transform allows you to load and apply a variety of LookUp Tables (LUTs).

Using OCIO for converting MediaIn or Loader nodes to linear gamma is based on the OCIO Color
Space node. Placing the OCIO Color Space node directly after a Loader (or MediaIn in
DaVinci Resolve) displays the OCIO Source and Output controls in the Inspector.

OCIO Color Space nodes can be used to work

in an ACES color managed environment.

Chapter 17 Managing Color for Visual Effects 399

Within the Inspector for OCIO Color Space node, Fusion includes default Source and Output
transforms for standard color spaces. However, to use the full OCIO standard, you’ll need to download
and install the OCIO config file. You can download the config file from the OCIO website. https://
opencolorio.org
Clicking the Browse button in the Inspector will allow you to navigate to the downloaded config file.
From the download, locate the ACES 1.0.3 or later folder and select the file config.ocio.

Source and Output menus are populated based

on the config.ocio file that you download.

The Source menu is used to choose the color profile for your Loader or MediaIn node. The default raw
setting shows an unaltered image, essentially applying no color management to the clip. The selection
you make from the menu is based on the recording profile of your media.
The Output menu is set based on your deliverables. When working in Fusion Studio, typically the
Output selected is ACEScg, to work in a scene linear space.

Applying OCIO LUTs in the Viewer

The viewer also includes OCIO View LUTs to calibrate the viewers. Once the OCIO Color Space View
LUT is selected from the LUT menu above the viewer, choosing Edit for the same menu opens a dialog
where you can load the OCIO config file.

The OCIO Color Space View LUT is located

in the Viewer LUT menu in Fusion Studio.

By default, the same standard options are available in the View LUT. However, clicking the Browse
button allows you to load the same config file you loaded into the OCIO Color Space node. Once
loaded, all the expanded OCIO options are available. If you selected the OCIO Color Space node to
output ACEScg, you use the OCIO View LUT to go from a source setting of linear sRGB to an output
setting of sRGB or Rec. 709 in most cases.

Chapter 17 Managing Color for Visual Effects 400

 The OCIO Color Space View LUT dialog is
used to configure the viewer when using the
OCIO Color Space node in the Node Editor.

TIP: If your monitor is calibrated differently, you will need to select a LUT that matches your
calibration.

To set the LUT using OCIO:

1 Click the LUT menu and choose the OCIO Color Space View LUT.
2 From the same menu, select Edit.
3 In the View LUT editor that opens, set the source’s color space to lin sRGB
4 Set the output space to sRGB or REC 709, assuming you are viewing on a standard computer
monitor. You now see a normalized image in the viewer, but all color operations will be on
linear images.

Whether you use the OCIO Color Space LUT or a LUT for your specific monitor calibration, you can
save the viewer setup as the default.

To save the OCIO ColorSpace LUT setup as the default viewer setup:
– Right-click in the viewer, and then choose Settings > Save Defaults. Now, for every comp, the
viewer is preconfigured based on the saved defaults.

Chapter 17 Managing Color for Visual Effects 401

Chapter 18

Understanding
Image Channels
This chapter seeks to demystify how Fusion handles image channels and, in the
process, show you how different nodes need to be connected to get the results you
expect. It also explains the mysteries of premultiplication, and presents a full
explanation of how Fusion is capable of using and even generating auxiliary data.

Contents
Channels in Fusion  403
Types of Channels Supported by Fusion  403
Fusion Node Connections Carry Multiple Channels  404
Node Inputs and Outputs  405
Node Colors Tell You Which Nodes Go Together  408
Using Channels in a Composition  409
Channel Limiting  410
Adding Alpha Channels  411
How Channels Propagate During Compositing  412
Rearranging or Combining Channels  413
Understanding Premultiplication  413
The Rules of Premultiplication  415
Alpha Channel Status in MediaIn and Loader Nodes  417
Controlling Premultiplication in Color Correction Nodes  417
Controlling Premultiplication With Alpha Divide and Alpha Multiply  418
Multi Channel Compositing  418
Compositing with Beauty Passes  419
Working with Auxiliary Channels  424

Chapter 18 Understanding Image Channels 402

Auxiliary Channels Explained  426
Propagating Auxiliary Channels  433
Nodes That Use Auxiliary Channels  434
Image Formats That Support Aux Channels  435
Creating Auxiliary Channels in Fusion  435

Channels in Fusion
Fusion introduces some innovative ways of working with the many different channels of image data
that modern compositing workflows encompass. This chapter’s introduction to color and data
channels and how they’re affected by different nodes and operations is a valuable way to begin the
process of learning to do paint, compositing, and effects in Fusion.
If you’re new to compositing, or you’re new to the Fusion workflow, you ignore this chapter at your
peril, as it provides a solid foundation to understanding how to predictably control image data as you
work in this powerful environment.

Types of Channels Supported by Fusion

Digital images can be divided into separate streams of image data called Channels, each of which
carries a specific kind of image data. Nodes that perform different image processing operations
typically expect specific channels to provide predictable results. You are probably familiar with the
three standard color channels of red, green, and blue, but there are many others. This section
describes the different kinds of channels that Fusion supports.

RGB Color channels

The red, green, and blue channels of any still image or movie clip combine additively to represent
everything we can see via visible light. Each of these three channels is a grayscale image when seen
by itself. When combined additively, these channels represent a full-color image.

Alpha Channels
An alpha channel is an embedded fourth channel that defines different levels of transparency in an
RGB image. Alpha channels are typically embedded in RGB images that are generated from computer
graphics applications. In Fusion, white denotes solid areas, while black denotes transparent areas.
Grayscale values range from more opaque (lighter) to more transparent (darker).
If you’re working with an imported alpha channel from another application for which these conventions
are reversed, never fear. Every node capable of using an alpha channel is also capable of inverting it.

Single-Channel Masks
While similar to alpha channels, mask channels are single channel images, external to any RGB image
and typically created by Fusion within one of the available Mask nodes. Mask nodes are unique in that
they propagate single-channel image data that defines which areas of an image should be solid and
which should be transparent. However, masks can also define which parts of an image should be
affected by a particular operation, and which should not. Mask channels are designed to be
connected to specific mask inputs of nodes including Effect Mask, Garbage Mask, and Solid
Mask inputs.

Chapter 18 Understanding Image Channels 403

Auxiliary Channels
Auxiliary channels (covered in more detail later in this chapter), describe a family of special-purpose
image data that typically expose 3D data in a way that can be used in 2D composites. For example,
Z-Depth channels describe the depth of each pixel in an image along a Z axis, while an XYZ Normals
channel describes the orientation (facing up, facing down, or facing to the left or right) of each pixel in
an image. Auxiliary channel data is generated by rendering 3D images, so it usually accompanies or is
embedded with RGB images generated by 3D modeling and animation applications. These channels
can also be generated from within Fusion via the Renderer 3D node, which outputs a 3D scene that
you’ve assembled and lit as 2D RGBA channels, with optionally accompanying auxiliary channels.
The reason to use auxiliary data is that 3D rendering is computationally expensive and time-
consuming, so outputting descriptive information about a 3D image that’s been rendered empowers
compositing artists to make sophisticated alterations in 2D. You can add motion blur, perform
relighting, and composite with depth information faster than re-rendering the 3D source material
over and over.

TIP: You can view any of a node’s channels in isolation using the Color drop-down menu in
the viewer. Clicking the Color drop-down menu reveals a list of all channels within the
currently selected node, including red, green, blue, or auxiliary channels.

Fusion Node Connections Carry Multiple Channels

The connections that pass image data from one node to the next in Fusion’s Node Editor are capable
of carrying multiple channels of image data. That means that a single connection may route RGB, or
RGBA, or RGBAZ-Depth, or even just Z-Depth, depending on the Input you connect to and the
node’s function.
In the following example, the two MediaIn nodes output RGB data. However, the Delta Keyer creates
an alpha channel and combined it with MediaIn2’s RGB image. The RGB-A of the Delta Keyer becomes
the foreground image that the Merge node can use to create a two-layer composite.

The alpha channel generated by the Delta key is used for

compositing by the foreground input of the Merge node.

NOTE: Node trees shown in this chapter may display MediaIn nodes found in
DaVinci Resolve’s Fusion page; however, Fusion Studio Loader nodes are interchangeable
unless otherwise noted.

Running multiple channels through single connection lines makes Fusion node trees simple to read,
but it also means you need to keep track of which nodes process which channels to make sure that
you’re directing the intended image data to the correct operations.

Chapter 18 Understanding Image Channels 404

Node Inputs and Outputs
By default, MediaIn nodes in the Fusion page and Loader nodes in Fusion Studio output RGBA
channels. When you connect one node’s output to another node’s input, the active channels are
passed from the upstream node to the downstream node, which then processes the image according
to that node’s function. Only one node output can be connected to a node input at a time. In this
simple example, a MediaIn node’s output is connected to the input of a Highlight node to create a
sparkly highlight effect.

MediaIn node connected to a Highlight node connected to MediaOut node in the Fusion page.

When connecting nodes, a node’s output carries the same channels no matter how many times the
output is “branched.” You cannot send one channel out on one branch and a different channel out on
another branch.

The MediaIn node’s output is branched but carries the same RGB channels to both inputs.

Using Multiple Inputs

Most nodes have two inputs, one for RGBA and another for an effect mask that can be optionally used
to limit the effect of that node to a particular part of the image. However, some nodes have three or
even more inputs, and it’s important to make sure you connect the correct image data to the
appropriate input to obtain the desired result. If you connect one node’s output to another node’s input
and nothing happens, chances are you’ve connected to the wrong input.
For example, the MatteControl node has a background input and a foreground input, both of which
accept RGBA channels. However, it also has SolidMatte, GarbageMatte, and EffectsMask inputs that
accept alpha or mask channels to modify the transparency of the Node’s output. If you want to perform
the extremely common operation of using a MatteControl node to create an alpha channel using a
Polygon node for rotoscoping an image, you need to make sure that you connect the Polygon node to
the GarbageMatte input to obtain the correct result. The GarbageMatte input is automatically set to
alter the alpha channel of the foreground image. If you connect to any other input, your Polygon mask
may not produce expected results. 

Chapter 18 Understanding Image Channels 405

Polygon node connected to the GargabeMatte input of a MatteControl node for rotoscoping

In another example, the DeltaKeyer node has a primary input (labeled “Input”) that accepts RGBA
channels, but it also has three Matte inputs. These SolidMatte, GarbageMatte, and EffectsMask inputs
on the Delta Keyer accept alpha or mask channels to modify the matte being extracted from the image
in different ways.

The DeltaKeyer combines the multiple mask nodes in different ways..

If you position your pointer over any node’s input or output, a tooltip appears in the Tooltip bar at the
bottom of the Fusion window, letting you know what that input or output is for, to help guide you to
using the right input for the job. If you pause for a moment longer, another tooltip appears in the Node
Editor itself.

(Left) The node input’s tooltip in the Tooltip bar, (Right) The node tooltip in the Node Editor

Connecting to the Correct Input

When you’re connecting nodes, pulling a connection line from the output of one node and dropping it
right on top of the body of another node makes a connection to the default input for that node, which
is typically the “Input” or “Background” input.

Side by side, dropping a connection on a node’s body to connect to that node’s primary input

Chapter 18 Understanding Image Channels 406

However, if you drop a connection line right on top of a specific input, then you’ll connect to that input,
so it’s important to be mindful of where you drop connection lines as you wire up different node
trees together.

Side by side, dropping a connection on a specific node input, note how the inputs
rearrange themselves afterwards to keep the node tree tidy-looking

TIP: If you hold the Option key down while you drag a connection line from one node onto
another, and you keep the Option key held down while you release the pointer’s button to
drop the connection, a menu appears that lets you choose which specific input you want to
connect to, by name.

Inputs Require Specific Channels

Usually, you’re prevented from connecting a node’s output to another node or node input that’s not
compatible with it. For example, if you try to connect a Text3D node’s output directly to the input of a
regular Merge node, it won’t work; 3D nodes do not produce RGB images, they produce 3D geometry
data, so you must first connect to a Renderer3D node that creates the RGB output appropriate for 2D
compositing operations.
In other cases, connecting the wrong image data to the wrong node input won’t give you any error, it
simply fails to produce the result you were expecting, necessitating you to troubleshoot the
composition. If this happens to you, check the Fusion Effects section of this manual to see if the node
you’re trying to connect to has any limitations as to how it must be attached.

TIP: This chapter tries to cover many of the easy-to-miss exceptions to node connection that
are important for you to know, so don’t skim too fast.

Always Connect the Background Input First

Many nodes combine images in different ways, using “background” and “foreground” inputs. This
includes the Merge node, the Matte Control node, and the Channel Booleans node, to cite common
examples. The color of inputs on a node can help you make the right corrections. For instance,
background inputs are always orange, and foreground inputs are always green.
When you first connect any node’s output to a multi-input node, you usually want to connect the
background input first. This is handled for you automatically when you first drop a connection line onto
the body of a new multi-input node. The orange-colored background input is almost always connected
first (the exception is Mask nodes, which always connect to the first available Mask input). This is good
because you want to get into the habit of always connecting the background input first.
If you connect to only one input of a multi-input node and you don’t connect to the background input,
you may find that you don’t get the results you wanted. This is because each multi-input node expects
that the background is connected before anything else so that the internal connections and math used
by that node can be predictable.

Chapter 18 Understanding Image Channels 407

 TIP: The only node to which you can safely connect the foreground input prior to the
background input is the Dissolve node, which is a special node that can be used to either
dissolve between two inputs, or automatically switch between two inputs of unequal duration.

Node Colors Tell You Which Nodes Go Together

Each node in Fusion accomplishes a single type of effect or operation. These single-purpose nodes
make it easier to decipher a complex composition when examining its node tree. Single-purpose
nodes also make it easier to focus on fine-tuning specific adjustments, one at a time, when assembling
an ever-growing tree.
Because each Fusion node has a specific function, they’re categorized by type to make it easier to
keep track of which nodes require what types of image channels as input, and what image data you
can expect each node to output. These general types are described here.

A node tree showing the main categories of node colors

Blue MediaIn and Loader Nodes, and Green Generator Nodes

Blue MediaIn nodes and blue Loader nodes add clips to a composite, and green Generator nodes
create images. Both types of nodes output RGBA channels (depending on the source and generator),
and may optionally output auxiliary channels for doing advanced compositing operations.
Because these are sources of images, both kinds of nodes can be attached to a wide variety of other
nodes for effects creation besides just 2D nodes. For example, you can also connect MediaIn nodes
to Image Plane 3D nodes for 3D compositing, or to pEmitter nodes set to “Bitmap” for creating
different particle systems. Green Generator nodes can be similarly attached to many different kinds of
nodes, for example, attaching a FastNoise node to a Displace 3D node to impose undulating effects to
3D shapes.
Shape nodes are also green, although they must be attached to a specialized set of gray modifier and
render nodes (all of which begin with the letter “s” and appear in the Shape category of the
Effects Library.

Chapter 18 Understanding Image Channels 408

2D Processing Nodes, Color Coded by Type
These encompass most 2D processing and compositing operations in Fusion, all of which process
RGBA channels and pass along auxiliary channels. These include:
– Orange Blur nodes
– Olive Color Adjustment nodes (color adjustment nodes additionally concatenate with one another)
– Pink Paint nodes
– Dark orange Tracking nodes
– Tan Transform node (transform nodes additionally concatenate with one another)
– Teal VR nodes
– Dark brown Warp nodes
– Gray, which includes Compositing nodes as well as many other types.

Additionally, some 2D nodes such as Fog and Depth Blur (in the Deep Pixel category) accept and use
auxiliary channels such as Z-Depth to create different perspective effects in 2D.

TIP: Two 2D nodes that specifically don’t process alpha channel data are the Color Corrector
node and the Gamut node. The Color Correction node lets you color correct a foreground
layer to match a background layer without affecting an alpha channel. The Gamut node lets
you perform color space conversions to RGB data from one gamut to another without
affecting the alpha channel.

Purple Particle System Nodes

These are nodes that connect to create different particle systems, and they’re incompatible with other
kinds of nodes until you add a pRender node that outputs 2D RGBA and auxiliary data that can be
composited with other 2D nodes and operations.

Dark Blue 3D Nodes

These are 3D operations, which generate and manipulate 3D data (including auxiliary channels) that
are incompatible with other kinds of nodes until processed via a Renderer 3D node, which then
outputs RGBA and auxiliary data.

Brown Mask Nodes

Masks output single-channel images that can only be connected to one another (to combine masks) or
to specified Mask inputs. Masks are useful for defining transparency (Alpha masks), defining which
parts of an image should be cropped out (Garbage masks), or defining which parts of an image should
be affected by a particular node operation (Effects masks).

Using Channels in a Composition

When you connect one node’s Output to another node’s Input, you feed all of the channels that are
output from the upstream node to the downstream node. 2D nodes, which constitute most simple
image processing operations in Fusion, propagate all channel data from node to node, including RGB,
alpha, and auxiliary channels, regardless of whether or not that node actually uses or affects a
particular channel.

Chapter 18 Understanding Image Channels 409

2D nodes also typically operate upon all channel data routed through that node. For example, if you
connect a node’s output with RGBA and XYZ Normals channels to the input of a Vortex node, all
channels are equally transformed by the Size, Center, and Angle parameters of this operation,
including the alpha and XYZ normals channels, as seen in the following screenshot.

(Left) The Normal Z channel output by a rendered torus, (Right) The Normal Z channel after the output is
connected to a Vortex node; note how this auxiliary channel warps along with the RGB and A channels

This is appropriate because in most cases, you want to make sure that all channels are transformed,
warped, or adjusted together. You wouldn’t want to shrink the image without also shrinking the alpha
channel along with it, and the same is true for most other operations.
On the other hand, some nodes deliberately ignore specific channels when it makes sense. For
example, the Color Corrector and Gamut nodes, both of which are designed to alter RGB data
specifically, do not affect auxiliary channels. This makes them convenient for color-matching
foreground and background layers you’re compositing, without worrying that you’re altering the depth
information accompanying that layer.

TIP: If you’re doing something exotic and you actually want to operate on a channel that’s
usually unaffected by a particular node, you can always use the Channel Booleans node to
reassign the channel. When doing this to a single image, it’s important to connect that image
to the background input of the Channel Booleans node, so the alpha and auxiliary channels
are properly handled.

Channel Limiting
Most nodes have a set of Red, Green, Blue, and Alpha buttons in the Settings tab of that node’s
Inspector. These buttons let you exclude any combination of these channels from being affected by
that node.

The channel limiting buttons in the Settings panel of a

Transform node, so only the Green channel is affected

For example, if you wanted to use the Transform node to affect only the green channel of an image,
you can turn off the Green, Blue, and Alpha buttons. As a result, the green channel is processed by
this operation, and the red, blue, and alpha channels are copied straight from the node’s input to the
node’s output, skipping that node’s processing to remain unaffected.

Chapter 18 Understanding Image Channels 410

 Transforming only the green color channel of the image with a Transform effect

Skipping Channel Processing

Under the hood, most nodes process all channels first, but afterward copy the input image to the
output for channels that have been enabled. Modern workstations are so fast that this isn’t usually
noticeable, but there are some nodes where deselecting a channel actually causes that node to skip
processing that channel entirely. Nodes that operate this way have a linked set of Red, Green, Blue,
and Alpha buttons on another tab in the node. In these cases, the Common Control channel buttons
are instanced to the channel buttons found elsewhere in the node.
Blur, Brightness/Contrast, Erode/Dilate, and Filter are examples of nodes that all have RGBA buttons in
the main Controls tab of the Inspector, in addition to the Settings tab.

Adding Alpha Channels

Much of visual effects compositing has to do with placing a foreground subject over a background.
Possibly the most fundamental method is through the use of an alpha or matte channel. If an alpha
channel is not contained within the clip, you add one via keying or rotoscoping. While more specific
methods are covered in detail in later chapters, here’s an example of how this is handled within Fusion.
In the case of extracting an alpha matte from a green screen image, you typically connect the image’s
RGB output to the “Input” input of a Keyer node such as the Delta Keyer, and you then use the keyer’s
controls to extract the matte. The Keyer node automatically inserts the alpha channel that’s generated
alongside the RGB channels, so the output is automatically RGBA. Then, when you connect the keyer’s
output to a Merge node to composite it over another image, the Merge node automatically knows to
use the embedded alpha channel coming into the foreground input to create the desired composite,
as seen in the following screenshot.

A simple node tree for keying; note that only one connection links the DeltaKeyer to the Merge node

Rotoscoping, or manually drawing a mask shape using a Polygon or other Mask node is another
technique used to create the matte channel. There are many ways to configure the node tree for this
task, but the simplest setup is just to connect a Polygon or B-Spline mask node to the Effect Mask
input of a MediaIn or Loader node.

Chapter 18 Understanding Image Channels 411

 TIP: When rotoscoping, it is best to leave the Mask node disconnected from the image while
you draw the shape. This allows you to view the MediaIn node while drawing. Connect the
Matte node once you have finished drawing the shape.

A simple rotoscoping setup using a

Polygon node directly connected to
the Effect Mask input of a MediaIn.

In both cases, you can see how the node tree’s ability to carry a single channel or multiple channels of
image data over a single connection line simplifies the compositing process.

How Channels Propagate During Compositing

Images are combined or composited together using the Merge node. The Merge node takes two
RGBA inputs labeled “Foreground” (green) and “Background” (orange) and combines them into a
single RGB output (or RGBA if both the foreground and background input images have alpha), where
the foreground image is in front (or on top, depending on what you’re working on), and the background
image is, you guessed it, in back.

A simple Merge node composite

Auxiliary channels, on the other hand, are handled in a much more specific way. When you composite
two image layers using the Merge node, auxiliary channels only propagate through the image that’s
connected to the background input. The rationale for this is that in most CGI composites, the
background is most often the CG layer that contains auxiliary channels, and the foreground is a
live-action green screen plate.
Since most compositions use multiple Merge nodes, it pays to be careful about how you connect the
background and foreground inputs of each Merge node to make sure that the correct channels
flow properly.

Chapter 18 Understanding Image Channels 412

 TIP: Merge nodes are also capable of combining the foreground and background inputs
using Z-Depth channels using the “Perform Depth Merge” checkbox, in which case every pair
of pixels are compared. Which one is in front depends on its Z-Depth and not the
connected input.

Rearranging or Combining Channels

Last, but certainly not least, it’s also possible to rearrange and re-combine channels in any way you
need, using one of four different node operations. For example, you might want to combine the red
channel from one image with the blue and green channels of a second image to create a completely
different channel mix. Alternatively, you might want to take the alpha channel from one image and
merge it with the alpha channel of a second image in different ways, adding, subtracting, or using
other intersection operations to create a very specific blend of the two.
The following nodes are used to re-combine channels in different ways:
– Channel Boolean: This is a 3D node used to remap and modify channels of 3D materials using a
variety of simple pre-defined math operations.
– Channel Booleans: Used to shuffle or rearrange RGBA and auxiliary channels within a single input
image, or among two input images, to create a single output image. If you only connect a single
image to this node, it must be connected to the background input to make sure everything works.
– Copy Aux: The Copy Aux node is used to remap channels between RGBA channels and
auxiliary data channels in a single 2D image. The Copy Aux node is mostly a convenience
node, as the copying can also be accomplished with more effort (and flexibility) using a Channel
Booleans node.
– Matte Control: Designed to do any combination of the following: (a) re-combining mattes, masks,
and alpha channels in various ways, (b) modifying alpha channels using dedicated matte controls,
and (c) copying alpha channels into the RGB stream of the image connected to the background
input in preparation for compositing. You can copy specific channels from the foreground input to
the background input to use as an alpha channel, or you can attach masks to the garbage matte
input to use as alpha channels as well.

Understanding Premultiplication
Now that you understand how to direct and recombine RGB images and alpha channels in Fusion, it’s
time to go more deeply into alpha channels to make sure you always combine RGB and alpha
channels correctly for each operation you perform in your composite. This might seem simple, but
small mistakes are easy to make and can result in unsightly artifacts. This is arguably one of the most
confusing areas of visual effects compositing, so don’t skip this section.
When alpha channel and RGB pixels are both contained within a media file, such as a 3D rendered
animation that contains RGB and transparency, or a motion graphics movie file with transparency
baked in, there are two different ways they might be combined, and it’s important to know which
is in use.
– Unpremultipled (Straight): An RGB image unaltered by the semi-transparency information in a
fourth channel (alpha channel)
– Premultiplied: An RGB image that has each channel multiplied by its alpha
channel before compositing.

Chapter 18 Understanding Image Channels 413

The term Premultiplied alpha is a term that has historically been used by editors, visual effects artists,
and motion graphics designers, but it’s imprecise. The alpha channel itself is not multiplied. The R, G,
and B channels are multiplied by the alpha. In the end, the alpha channel stays the same, but the
values contained in the R, G, and B channels are modified.

A RGB image (left) and its alpha channel (right)

Non-premultiplied images, sometimes called “straight” alpha channels, have RGB channels that are
unaltered (not multiplied) by the alpha channel. The result is that the RGB image has no anti-aliased
edges and no semi-transparency. It’s usually obvious where the RGB image ends and the alpha matte
begins. The image below is an example of the ragged edges seen in the RGB channels when using a
non-premultiplied alpha channel. But notice the smooth semi-transparent edges found in the alpha.

A detailed view of a non-premultiplied RGB image (left) and its alpha channel (right)

A premultiplied alpha channel means the RGB pixels are multiplied by the alpha channel. This method
guarantees that the RGB image pixels include semi-transparency where needed, like anti-aliased
edges. Most computer-generated images are premultiplied for convenience, because they’re easier to
review smoothly without actually being placed inside of a composite.

A detailed view of a premultiplied image (left) and its alpha channel (right)

Chapter 18 Understanding Image Channels 414

What does this mean for compositing? The edges of a premultiplied image look much smoother,
making them the preferred choice when compositing foreground over the background in a merge.
Consequently, all alpha channels are made to be premultiplied in compositing operations if they
weren’t already.
On the other hand, it is always preferred to color correct a non-premultiplied RGBA image, because
you don’t want to alter the pixel values of an image after the RGB channels have been multiplied by
the alpha channel.
If you think of this from a mathematical perspective:
– RGB pixel value x 0 = 0: The black transparent areas of an alpha channel have a pixel value of
0. When you take the value of an RGB pixel and multiply it by 0 (n x 0 = 0) then by the laws of
multiplication, the RGB value becomes 0, or fully transparent.
– RGB Pixel value x 1 = RGB Pixel: The solid or opaque white areas have a value of 1.0. When you
take the value of an RGB pixel and multiply it by 1 (n x 1 = n), then the RGB value stays the same,
fully opaque.
– RGB Pixel value x 0.3 = A different color: Along the edges of an alpha channel are gray pixels,
indicating semi-transparency. These semi-transparent pixels have a value falling somewhere
between 1.0 and 0.0. To apply the alpha channel’s anti-aliased edges to the RGB channels, you
multiply the pixel values. The multiplication process mixes some percentage of the transparent
pixels (black) with the RGB pixels. Although this is desired to get good anti-aliased edges, you
can not color correct the image because it alters the smooth semi-transparency you created
once it is done.

RGB pixels are multiplied by varying degrees of

transparency and the result is a different RGB value.

The Rules of Premultiplication

Following from the information presented above, when you’re compositing multiple images together
and one or more has a built-in alpha channel, you want to make sure you follow these general rules to
avoid problems:
– Always use premultiplied images with a Merge node.
– Only color-correct images that are not premultiplied.
– Always filter and transform images that are premultiplied.
– Never double premultiply an image.

Premultiplication and the Merge Node

The foreground input of a Merge node expects a premultiplied RGBA image. It is an additive merge,
meaning the semi-transparent areas of the foreground are added over the background.
However, if the image is not premultiplied, the pixels that should be transparent are still added, which
typically results in an unwanted bright fringe around the edges of your foreground subject.

Chapter 18 Understanding Image Channels 415

If you are compositing with a non-premultiplied alpha, you can fix these bright edges by changing the
Merge to perform a Subtractive merge in the Inspector.

Nonpremultiplied edges in an additive Merge (left) and premultiplied edges in an additive Merge (right)

TIP: When an RGB image and a Mask node are combined using, for instance, a Matte Control
node, if the RGB image is not multiplied by the mask in the Matte control, the checkerboard
background pattern in the viewer will appear only semi-transparent when it should be fully
transparent.

Color Correcting Premultiplied RGBA Images

When you premultiply an image, you tie any changes made to the brightness of the RGB pixels to the
alpha channel pixels as well. If, for example, you raise the brightness of the foreground image in any
way, you’ll also raise the brightness of the alpha channel, which will likely not be desirable as this will
change the transparency the alpha channel creates. The visible result of this when viewing the Merge
node is that the entire background will become brighter (or darker if you’d lowered the RGB brightness)
based on your color adjustment.
For this reason, the rule is always to divide the semi-transparent pixels before performing any color
correction on an image with an alpha channel. You can do this turning on the Pre-Divide/Post Multiply
checkbox in any node that performs color correction. Alternatively, you can use the Alpha Divide and
Alpha Multiply nodes to do the same thing. These methods are covered in more detail later in
this chapter.

Color correcting a premultiplied foreground incorrectly alters the background (left).

Color correcting a nonpremultiplied foreground works correctly (right).

Chapter 18 Understanding Image Channels 416

Double Premultiplied RGBA Means Double Trouble
A common mistake made by many artists is to over-compensate for premultiplication. As important as
it is to premultiply the alpha before compositing in a Merge node, it’s just as important not to double
premultiply the alpha. Performing a premultiply operation two times in a row can create a darken halo
effect around your images. You are effectively multiplying by the gray semi-transparent pixels twice;
this is not optimal.

Double premultiplied image displays dark edges

(left); premultiplied image with correct edges (right)

Premultiplied Alpha Channels and Filtering

When dealing with filtering, the state of the RGBA channels shouldn’t matter for most
composites. However, an exception might be if the filter algorithm you choose includes color
modification. For instance, if a filter attempts to simulate a defocus by blooming highlights like
a real light source, that filter might over-brighten pixels near a transparent edge, which will
result in some manner of artifact when that image is composited.

Alpha Channel Status in MediaIn and Loader Nodes

When using a Loader node to add a clip to the composite, the Import tab in the Inspector includes a
group of checkboxes that let you determine how an embedded alpha channel is handled. There are
checkboxes to make the alpha channel solid (ignore transparency), to Invert the alpha channel, and to
Post-Multiply the RGB channels with the alpha channel, should that be necessary.
When using a MediaIn node, you can modify how an embedded alpha channel is interpreted using
the Clip attributes window. The Clips Attributes window includes an Alpha Mode menu setting to
choose if the alpha channel is ignored, treated as premultiplied, inverted, or treated as
nonpremultiplied (straight).

Controlling Premultiplication in Color Correction Nodes

Most nodes that require you to explicitly deal with the state of premultiplication of an RGBA image
have a “Pre-Divide, Post-Multiply” checkbox. This includes simple color correction nodes such as
Brightness Contrast and Color Curves, as well as the Color Correct node, which has the “Pre-Divide/
Post-Multiply” checkbox in the Options panel of its Inspector settings.

The Pre-Divide/Post-Multiply checkbox of the

Color Curves node, seen in the Inspector

Chapter 18 Understanding Image Channels 417

This checkbox allows you to connect an RGBA premultiplied image to the node and perform a color
correction operation. It takes the RGBA image input, performs a divide operation to remove the
semi-transparency and then performs a multiplication operation before outputting the color corrected
image. This way, the color correction is done using a nonpremultiplied image but the resulting output
is a Merge-friendly premultiplied image.

Controlling Premultiplication With Alpha Divide

and Alpha Multiply
The Alpha Divide and Alpha Multiply nodes, found in the Matte category of the Effects Library, are
available when multiple operations in a row expect a “straight” alpha channel. Instead of performing
repetitive Pre-Divide/Post Multiply operations on each node, you can use these two nodes to book-
end the other nodes. Simply add the Alpha Divide node when you want the RGBA image data not to
be premultiplied, and add the Alpha Multiply node when you want the image data to be premultiplied
again. For example, if you’re using third-party OFX nodes that make color adjustments, you may need
to control premultiplication before and after such an adjustment manually.

A node tree with explicit Alpha Divide and Alpha Multiply nodes

Multi Channel Compositing

If you’ve gone to the cinema and seen any recent superhero movie, you will have seen the results of
sophisticated 3D rendering, combined with a whole lot of compositing. 3D applications can render
very realistic images but the time it takes to render each frame of those realistic images can be
measured in hours, not minutes. Any change to the 3D images, even relatively simple operations like
color adjustments, changes to focus, filtering, or additional masking, means these images will need to
be re-rendered entirely, which means you waiting, often many times over. For efficiencies sake, it’s
much faster to make iterative changes that can be accomplished with 2D image processing operations
in Fusion instead.
To have the flexibility you need to make common changes to 3D images after-the-fact, the various
attributes that make up the 3D scene are separated and rendered as different image sequences, often
referred to as render passes. For example, render passes are often created for attributes like raw
color, shadows, and reflections, that can be recombined as a 2D composite to produce the final result.
Having different attributes rendered into different image sequences gives you a significant amount of
flexibility, since now each image attribute can be color corrected, blurred, or further processed
independently of the other attributes of the image, with fast-processing operations in Fusion.

Chapter 18 Understanding Image Channels 418

The most common render passes that are typically generated come from the RGBA channels of the 3D
scene. These are collectively called beauty passes and can consist of attributes like color, shadows,
lighting, reflections, environment, and others.
Render passes can also contain non-RGB data. Different effects applications have different names for
these passes, such as Data Channels, or AOVs (Arbitrary Output Variables). In Fusion, these channels
are called Auxiliary Channels, and they contain 3D data such as Depth, Normals, Motion Vectors, and
UV Coordinates (to name just a few).
When compositing a 3D render consisting of multiple render passes, the beauty passes are handled
using one technique, and the Auxiliary Channels are handled with another. Since Fusion nodes carry
RGBA channels by default, we’ll cover beauty passes first, and then explain how to work with Auxiliary
Channels later in this chapter.

Compositing with Beauty Passes

Each attribute of a beauty pass can be rendered into individual image sequences, so you end up with
a series of numbered images, one for the diffuse pass, one for the reflection pass, another of the
shadow pass, and so on. Alternatively, all the passes can be contained in a multi-part EXR image
sequence. Multi-part EXRs benefit from requiring a bit less file management, but the passes are
handled similarly in Fusion using either method.
A single MediaIn or Loader node only handles a single beauty pass since only one set of RGBA
channels gets output per node. Setting up your composite in Fusion requires you to use a separate
MediaIn or Loader node for each pass.

A MediaIn or Loader node is needed for each beauty pass.

TIP: The EXR format in Fusion is optimized when multiple Loaders are used to read the same
EXR file. The file is only loaded once to access all the channels.

Beauty Pass Setup

Separated components of 3D renders, such as diffusion, shadows, or reflections, can be rendered
individually using RGB channels. If you’re provided with individual image sequences for each image
component, they’ll import and open in Fusion and can be displayed in the viewer, just like any other
clip. If you’re using multi-part EXR files, then you essentially have multiple RGB images contained
within a single file. However, the MediaIn and Loader nodes can only make use of one set of RGB
channels at a time. If you want to composite multiple RGB beauty passes together with one another,
you must use multiple MediaIn or Loader nodes that point to the same image sequence but are
assigned to different passes contained in the EXR file.

Chapter 18 Understanding Image Channels 419

 Multiple Loader or MediaIn nodes connect to a multi-part EXR image sequence

TIP: It is wise to rename each Loader or MediaIn to represent the beauty pass it contains.

Mapping Each Set of Beauty Passes to a Particular Node

Depending on whether you’re using a MediaIn node or a Loader node, beauty passes can be mapped
to RGB channels using either the Image tab, the Channels tab, or the Format tab in the Inspector.
When using a Loader node, the Loader’s Format tab is used.

Beauty passes mapped to red, green,

and blue channels in a Loader node

The MediaIn’s Image tab includes a Layer menu. Any pass included in a multi-part EXR image
sequence can be selected from this menu and automatically assigned to the RGBA channels.

Chapter 18 Understanding Image Channels 420

 The Layer menu in a MediaIn node showing
headings for combined channel passes

In most cases, the menu shows the combined channel passes, meaning the individual red, green, blue,
and alpha channels cannot be selected. Because the alpha channel is not included in many beauty
passes, you sometimes need to borrow the alpha channel from a different beauty pass. For this
reason, it’s often better to use the Channels tab for mapping the individual channels of a beauty pass
to the channels of the MediaIn node.

The MediaIn node’s Channels tab or the Loader’s

Format tab provides access to individual channels.

The MediaIn node in DaVinci Resolve and the Format tab in a Loader node include the same channel
mapping functionality. The Channels tab and the Format tab include individual RGBA menus at the top
of the tab. You can use these menus to map the RGBA channels from any pass contained in the
multi-part EXR. For instance, if you want to map the Ambient Occlusion pass to the RGB channels,
choose AO. R (Red) from the Red channel menu, AO. G (Green) from the Green channel menu, and AO.
B (Blue) from the Blue channel menu.

A pass’ individual channels mapped to the red,

green, blue, and alpha channels in the MediaIn node

Chapter 18 Understanding Image Channels 421

 TIP: Different 3D applications will label beauty passes in different ways. For instance, the
name for an Ambient Occlusion beauty pass may be AO, AM_OCC, or some other
abbreviation.

The Ambient Occlusion beauty pass does not include an alpha channel. To composite it, you can
reuse the alpha channel pass from another beauty pass. In the image below, the alpha channel is
mapped using the combined render pass’ alpha channel.

The alpha channel from a different beauty pass

combined with the Ambient Occlusion pass

TIP: When using the Format tab in the Loader node, the checkbox next to each channel
needs to be turned on for the corresponding channel to become available in the
node’s output.

Compositing Multiple Beauty Passes in the Node Editor

Once all of your passes are brought in and mapped to RGBA channels, you’ll end up with a series of
MediaIn or Loader nodes. How many MediaIn and Loader nodes is really up to your workflow. There
is no predefined number of passes you will use. Every studio decides for themselves the standard set.
However, there are common render passes that are involved in most composites. The following is a list
of commonly used render passes and their generic names.

Commonly used beauty passes, compared

Chapter 18 Understanding Image Channels 422

Compositing multiple beauty passes into a single output image is relatively straight forward. 3D
rendering applications typically output linear gamma, so no Gamut or other color space conversion
nodes are required if you’re keeping the image in a linear color space for ease of compositing.
The basic compositing is accomplished with either a Merge node or a Channels Booleans node. Both
allow for additive combining of render passes. There’s no strict requirement for compositing each pass
in any particular way, although in most situations a simple additive composite should work just fine.

To begin compositing render passes using a Merge node:

1 Connect a Color pass to the background input of a Merge node.
2 Connect a Direct Lighting pass to the foreground input.
3 Adjust the Alpha gain and Blend parameters to get the look you desire.

Compositing beauty passes starts by connecting a

background and foreground into a Merge node

If you prefer, you can use a Channels Booleans node to make the same composite. In this case, there
is no technical difference between the two nodes.

To composite render passes with a Channels Booleans node:

1 Connect a Color pass to the background input of a Channels Booleans node.
2 Connect a Direct Light pass to the foreground input.
3 Choose Add from the Operations menu
4 Choose Do Nothing from the Alpha To menu.

Channel Booleans is set to Add to combine the

foreground input and the background input.

One of the exceptions to the steps above are Shadow passes, such as Ambient Occlusion. In that
case, a multiply Apply mode is usually employed.

To composite an Ambient Occlusion render pass using a Merge node:

1 Connect last of a sequence of Merge node render passes to the background input of a
Merge node.
2 Connect an Ambient Occlusion pass to the foreground input.
3 Choose Multiply from the Apply Mode menu.
4 Adjust the Gain and Blend parameters to get the look you desire.

Chapter 18 Understanding Image Channels 423

As straightforward as this sounds, compositing using a recipe doesn’t always work for every shot.
When using different images, you may need to experiment with varying techniques of compositing for
the best results.

Embedding Alpha into Beauty Passes

Alpha channels are not included in all beauty passes. If your shot requires that you composite your
assembled beauty passes over live-action or some other background, it may be up to you to add the
alpha channel from a pass that does include it.

To add an alpha channel into your assembled beauty pass composite, do the following:
1 Connect the last Merge or Channel Booleans output into the background input of a Matte
Control node.
2 Connect the render pass that contains the alpha into the green Foreground input of the Matte
Control node.
3 In the Matte Control’s Inspector, choose Combine Alpha from the Combine menu.
4 Choose Copy from the Combine Op menu.

An alpha channel from the color pass added back into the completed beauty pass node tree

TIP: Alpha channels from 3D renderings are typically premultiplied. That being the case, be
sure to turn on the Pre Divide/Post Multiply checkbox on any node that performs color
correction. If using more than one node in a row to perform color correction, use the Alpha
Divide and Alpha Mult nodes instead.

Working with Auxiliary Channels

Auxiliary channels are not RGB images; they are a family of special-purpose 3D image data that
typically describes position, orientation, and object information for use in 2D composites. For example,
Z-Depth channels describe the depth of each pixel of an image along a Z-axis, while an XYZ Normals
channel describes the orientation (facing up, facing down, facing to the left or right) of every pixel
in an image.
Similar to the use of multiple beauty passes, one of the most common reasons to use auxiliary data is
to eliminate the need to re-render computationally expensive 3D imagery, by enabling even more
aspects of rendered images to be manipulated after-the-fact. 3D rendering is computationally
expensive and time-consuming, so outputting descriptive information about a 3D image allows
sophisticated alterations to occur in 2D compositing, which is faster to perform and adjust.
There are two ways of obtaining auxiliary channel data:
– First, auxiliary data may be embedded within a clip rendered from a 3D application, most often
using the EXR file format. In this case, it’s best to consult your 3D application’s documentation to
determine which auxiliary channels can be generated and output.

Chapter 18 Understanding Image Channels 424

 – You may also obtain auxiliary channel data by generating it within Fusion, via 3D operations output
by the Renderer 3D node, by the Optical Flow node, or by the Disparity node.

An RGBA 3D rendered scene that also contains auxiliary channels

TIP: When trying to locate information about auxiliary channels in other software, some 3D
applications refer to auxiliary channels as Arbitrary Output Variables (AOVs), render elements,
or secondaries.

Aux Channel Setup

When using a MediaIn or Loader node that links to a multi-part EXR file with auxiliary channel data, the
Inspector’s Channels or Format tab includes a pre-defined set of auxiliary channels for mapping
purposes. Each pre-defined channel includes a menu that displays every attribute included in a
multi-part EXR. From the menu, you select the render pass that should be assigned to the
corresponding channel. As described earlier, RGB beauty passes like diffuse, shadow, and reflection
are mapped to the red, green, and blue channels. Auxiliary passes include preset mappings.

A multi-part EXR file with embedded render passes

mapped to their aux channels in a MediaIn node

Chapter 18 Understanding Image Channels 425

Displaying Channels in the Viewer
Once you map an auxiliary channel in the Inspector, you can display the data as an RGB image in the
viewer. Clicking the drop down Color menu at the top of the viewer reveals a list of every active
auxiliary channel for the currently viewed node.

Select an aux channel from the color drop-down

menu to display the aux channel in the viewer

TIP: The Color Inspector SubView can be used to read numerical values from all of
the channels.

Auxiliary Channels Explained

Fusion is capable of using auxiliary channels to perform depth-based compositing, to create masks
and mattes based on Object or Material IDs, and for texture replacements. Nodes that work with
auxiliary channel information have been specifically developed to work with this data. The auxiliary
channels that are supported in Fusion are described below.

Z-Depth
Each pixel in a Z-Depth channel contains a value that represents the relative depth of that pixel in the
scene. In the case of overlapping objects in a model, most 3D applications take the depth value from
the object closest to the camera when two objects are present within the same pixel since the closest
object typically obscures the farther object.
When present, Z-Depth can be used to perform depth merging using the Merge node or to control
simulated depth-of-field blurring using the Depth Blur node.

Chapter 18 Understanding Image Channels 426

 The rendered Z-Depth channel for the previous RGBA image

For this example, we’ll examine the case where the Z-Depth channel is provided as a separate file.
The Z- channel can often be rendered as an RGB image. You’ll need to combine the beauty and Z pass
using a Channel Booleans node. When the Z pass is rendered as an image in the RGB channels, the
Channels Booleans node is used to re-shuffle the Lightness of the foreground RGB channel into the
Z-channel.

To combine the Z-pass and beauty pass:

1 Connect the MediaIn node containing the the beauty pass to the background input of the Channel
Booleans node.
2 Connect the MediaIn node containing the Z-Depth pass to the green foreground input of the
Channel Booleans node.
3 Select the Channel Booleans node, and use the Inspector to set the To Red, To Green, To Blue,
and To Alpha menus to Do Nothing.
4 Select the Aux tab, and set the To Z Buffer menu to Lightness FG.
5 Connect the output of the Channels Booleans node into the Depth Blur node.

The Aux tab configured to shuffle the

Foreground Lightness to the Z-Depth channel.

The Depth Blur node is one of the nodes that take advantage of a Z-channel in order to create blurry
depth-of-field simulations. To set this up, the output of the MediaIn node connects to the background
input on the Depth Blur.

The Depth Blur uses the Z-channel that is enabled in the Channel Booleans node.

Chapter 18 Understanding Image Channels 427

The Depth Blur’s controls in the Inspector are very dependent on the type of image you’re using. It can
be easier to begin by adjusting the controls in the Inspector to some better defaults. Start by
increasing the Blur Size to 10. This will make it easier to see even the smallest of changes. Next,
instead of using the Focal Point, you should pick a focal point in the image by dragging the Sample
button into the viewer and selecting a pixel that determines the part of the picture to keep in focus.
The final setup steps are to lower Z Scale to somewhere around 0.2 (if you’re using a floating-point
image), and leave the Depth of Field alone for now. This should show you some blurring in the image.

Start by improving the defaults if your image

is 16- or 32-bit floating point.

Once you see these experimental results, you can return to each parameter and refine it as needed to
achieve the actual look you want.

An image using a Z-Depth channel for blurring

TIP: Z-Depth channels often contain negative values. If this causes problems, you can choose
Normalize Color Range from the viewer’s Options menu to apply a normalization to the
viewer, keeping the image within a range from 0 to 1.

Chapter 18 Understanding Image Channels 428

Z-Coverage
The Z-Coverage channel is a somewhat extinct render pass in most 3D applications. It was a way of
restoring antialiasing to rendered color masks and Z-Depth passes. It indicated pixels in the Z-Depth
that contained two objects. The value was used to indicate, as a percentage, how transparent that
pixel was in the final depth composite. It can still be used today if you are rendering files from one of
the few applications that can produce them.

TIP: The wide adoption of an open-source matte creation technology called Cryptomatte, has
somewhat superseded mattes created from Coverage, Background, Object ID, and Material
ID passes.

Background RGBA
This channel is a somewhat extinct render pass in most 3D applications. It contained the color values
from the objects behind the pixels described in the Z coverage.

Object ID
Most 3D applications are capable of assigning ID values to objects in a scene. Each pixel in the Object
ID channel will be identified by that ID number, allowing for the creation of masks.
If you want to use an Object ID in a comp, like all aux channels you must map the Object ID pass to the
Object ID channel in the MediaIn or Loader Node.

To use an ObjectID pass, do the following:

1 In the MediaIn or Loader node, use the Channels or Format tab to map the Object ID pass to the
Object ID aux channel.
2 In whatever node you want to have affected by the ObjectID matte, select the Settings tab, turn on
the Object ID checkbox, and select the ID number assigned to the object.

The common Settings tab on most

nodes contains ObjectID controls.

Material ID
Most 3D applications are capable of assigning ID values to materials in a scene. Each pixel in the
Material ID channel will be identified by that ID number, allowing for masks based on materials.
You can set up Material IDs using the Settings tab, similarly to how ObjectIDs are set.

Chapter 18 Understanding Image Channels 429

UV Texture
The UV Texture channels contain information about how pixels in an image map to texture
coordinates. This is used to retexture an object in a 2D image. For instance, If you want to apply a logo
onto a rendered object, you can use the UV aux channel with the Texture node.

Texture (left) applied to 2D image (right) using UV channels and texture node.

To use a UV pass, do the following:

1 In the MediaIn or Loader node, use the Channels or Format tab to map the U and V pass to the U
and V aux channels.
2 Connect the output of the MediaIn or Loader node to the background input of the Texture node.
3 Connect the texture image you want to use to the foreground input of the Texture node.
4 If you want to combine the original texture with the new texture, use a merge with the background
input from the original image and the foreground input from the Texture node.
5 Adjust Merge’s Apply mode, Alpha Gain, and blend to get the desired mix of the two textures.

UV channels from a MediaIn node used in a Texture node and merged over the original image

TIP: If you are using a separate UV render pass with the UV data in the RGB channels, map
red to U and green to V in a Channel Booleans node.

X, Y, and Z Normals
The X, Y, and Z Normal channels contain information about each pixel’s orientation (the direction it
faces) in 3D space. The normals are often displayed as lines coming out from your object
perpendicular to the surface, letting you visualize the relationship between the surface and camera.

Chapter 18 Understanding Image Channels 430

 The Normals display the direction of the surface.

The Normals X, Y, and Z channels are often used with a Shader node to perform relighting
adjustments on a 2D rendered image.

To setup a Shader node to use XYZ Normals, do the following:

1 In the MediaIn or Loader node, use the Channels or Format tab to map the individual X, Y, and Z
Normals pass to the X Normal, Y Normal, and Z Normal channels.
2 Connect the output of the MediaIn or Loader node to the background input of the Shader node.
3 Optionally, connect a floating-point EXR image to be used as a reflection image to the Shader
node’s reflection input.
4 Adjust the Shader controls to perform relighting.

Original 2D image (left) and Normals used for relighting (right)

XY Vector and XY BackVector

The Vector channels indicates the pixel’s motion from frame to frame. It can be used to apply motion
blur to an image or to generate optical flow analysis for retiming. The XY Vector points to the next
frame, while the XY BackVector points to the previous frame.

XY Vector pass (left) used with Vector Motion Blur to generate motion blur on spaceship (right)

Chapter 18 Understanding Image Channels 431

Often the vector pass will be rendered in a separate pass as an RGB image. The X and Y vector data is
located in the R and G channels. In order to place them in the vector channels, you can use a Channel
Booleans node.

To use a Motion Vector pass to create motion blur, do the following:

1 Add a MediaIn or Loader node for the image and the Vector render pass.
2 Connect the output of the image into the background of a Channel Booleans node.
3 Connect the output of the Vector render pass to the Channel Boolean’s foreground
4 In the Channel Booleans inspector, set the To Red, To Green, To Blue, and To Alpha all
to Do Nothing.
5 Select to the Aux tab.
6 Turn on Enable Extra Channels.
7 Set the To X Vector drop-down menu to Red FG, and then set the To Y Vector drop-down menu
to Green FG.
8 Connect the Channel Booleans node’s output to the yellow background input on a Vector Motion
Blur node.

The Vector render pass is combined with the beauty image using the Channels
Booleans node, which then feeds the Vector Motion Blur node.

World Position
World Position Pass (WPP) is an auxiliary channel, sometimes referred to as Point Position, XYZ pass, or
WPP. It’s used to represent each pixel’s 3D (XYZ) position as an RGB color value. The result is data that
can be viewed as a very colorful RGB image. Like Z-Depth, this can be used for compositing via depth.
However, it can also be used for masking based on 3D position, regardless of camera transforms.
The colors correspond to a pixel’s position in 3D, so if a pixel sits at 0/0/0 in a 3D scene, the resulting
pixel’s will have an RGB value of 0/0/0 or black. If a pixel sits at 1/0/0 in the 3D scene, the resulting
pixel is fully red. Due to the huge extent, 3D scenes can have the WPP channel should always be
rendered in 32-bit floating-point to provide the accuracy needed.

XYZ Position

Chapter 18 Understanding Image Channels 432

XY Disparity
XY Disparity is the only channel listed here that is not generated in a 3D application. These channels
indicate where each pixel’s corresponding matte can be found in a stereo image. Each eye, left and
right, will use this vector to point to where that pixel would be in the other eye. This can be used for
adjusting stereo effects, or to mask pixels in stereo space.

Combined X and Y Disparity channels

Using Cryptomatte in Fusion

Cryptomatte is an open-source technology that has been widely adopted in 3D applications.
Unlike Z-Depth mattes or Object IDs, Cryptomatte automatically generates anti-aliased ID
mattes from 3D renders with support for motion blur, transparency, and depth of field.
Fusion does not natively support the Cryptomatte forma. However, using a free plug-in from
3rd party developers, you can use Crypotmatte render passes in Fusion.
Crypotmatte for Fusion can be downloaded and installed for free: https://github.com/Psyop/
Cryptomatte
Or to use an easier installer, you can download Reactor, which comes bundled with
Cryptomatte and offers many other free, useful Fusion plug-ins. Reactor can be found at:
https://www.steakunderwater.com

Propagating Auxiliary Channels

Ordinarily, auxiliary channels are propagated along with RGBA image data, from node to node, among
gray-colored nodes, including those in the Blur, Filter, Effect, Transform, and Warp categories.
Basically, most nodes that simply manipulate channel data propagate (and potentially manipulate)
auxiliary channels with no problems.
However, when you composite two image layers using the Merge node, auxiliary channels only
propagate through the image that’s connected to the background input. The rationale for this is that in
most composites that include computer-generated imagery, the background is most often the CG layer
that contains auxiliary channels, while the foreground is a live-action green screen plate with subjects
or elements that are combined against the background, which lack auxiliary channels.

Chapter 18 Understanding Image Channels 433

Nodes That Use Auxiliary Channels
The availability of auxiliary channels opens up a world of advanced compositing functionality.
This section describes every Fusion node that has been designed to work with images that contain
auxiliary channels.
– Copy Aux: The Copy Aux tool can copy auxiliary channels to RGB and then copy them back.
It includes some useful options for remapping values and color depths, as well as removing
auxiliary channels.
– Channel Booleans: The Channel Boolean tool can be used to combine or copy the values from
one channel to another in a variety of ways.
– Custom Tool, Custom Vertex 3D, pCustom: The “Custom” tools can sample data from the
auxiliary channels per pixel, vertex, or particle and use that for whatever processing you
would like.
– Depth Blur: The Depth Blur tool is used to blur an image based on the information present in
the Z-Depth. A focal point is selected from the Z-Depth values of the image and the extent of the
focused region is selected using the Depth of Field control. The Scale value default is based on
an 8-bit image so it is important to lower the scale value when using the Depth Blur with 16- or
32‑bit float files.
– Disparity to Z, Z to Disparity, Z to WorldPos: These tools use the inherent relationships between
depth, position, and disparity to convert from one channel to another.
– Fog: The Fog tool makes use of the Z-Depth to create a fog effect that is thin closer to the camera
and thickens in regions farther away from the camera. You use the Pick tool to select the Depth
values from the image and to define the Near and Far planes of the fog’s effect.
– Lumakeyer: The Lumakeyer tool can be used to perform a key on the Z-Depth channel by
selecting the Z-Depth in the channel drop-down list.
– Merge: In addition to regular compositing operations, Merge is capable of merging two or more
images together using the Z-Depth, Z-Coverage, and BG RGBA buffer data. This is accomplished
by enabling the Perform Depth Merge checkbox from the Channels tab.
– New Eye: For stereoscopic footage, New Eye uses the Disparity channels to create new
viewpoints or to transfer RGBA data from one eye to the other.
– Shader: The Shader tool applies data from the RGBA, UV and the Normal channels to modify the
lighting applied to objects in the image. Control is provided over specular highlights, ambient and
diffuse lighting, and position of the light source. A second image can be applied as a reflection or
refraction map.
– Shadow: The Shadow tool can use the Z-Depth channel for a Z-Map. This allows the shadow to fall
onto the shape of the objects in the image.
– Smooth Motion: Smooth Motion uses Vector and Back Vector channels to blend other channels
temporally. This can remove high frequency jitter from problematic channels such as Disparity.
– SSAO: SSAO is short for Screen Space Ambient Occlusion. Ambient Occlusion is the lighting
caused when a scene is surrounded by a uniform diffuse spherical light source. In the real world,
light lands on surfaces from all directions, not from just a few directional lights. Ambient Occlusion
captures this low frequency lighting, but it does not capture sharp shadows or specular lighting.
For this reason, Ambient Occlusion is usually combined with Specular lighting to create a full
lighting solution. The SSAO tool uses the Z-Depth channel but requires a Camera3D input.
– Stereo Align: For stereoscopic footage, the Disparity channels can be used by Stereo Align to
warp one or both of the eyes to correct misalignment or to change the convergence plane.
– Texture: The Texture tool uses the UV channels to apply an image from the second input as a
texture. This can replace textures on a specific object when used in conjunction with the Object ID
or Material ID masks.

Chapter 18 Understanding Image Channels 434

 – Time Speed and Time Stretcher: These tools can use the Vector and BackVector channels to
retime footage.
– Vector Distortion: The forward XY Vector channels can be used to warp an image with this tool.
– Vector Motion Blur: Using the forward XY Vector channels, the Vector Motion Blur tool can apply
blur in the direction of the velocity, creating a motion blur effect.
– Volume Fog: Volume Fog is a raymarcher that uses the World Position channels to determine ray
termination and volume dataset placement. It can also use cameras and lights from a 3D scene to
set the correct ray start point and Illumination parameters.
– Volume Mask: Volume Mask uses the World Position channels to set a mask in 3D space as
opposed to screen space. This allows a mask to maintain perfect tracking through a camera move.

TIP: The Object ID and Material ID auxiliary channels can be used by some tools in Fusion to
generate a mask. The “Use Object” and “Use Material” settings used to accomplish this are
found in the Settings tab of that node’s controls in the Inspector.

Image Formats That Support Aux Channels

Fusion supports auxiliary channel information contained in a variety of image formats. The number of
channels and methods used are different for each format.
– OpenEXR (*.exr): The OpenEXR file format is the primary format used to contain an arbitrary
number of additional image channels. Many renderers that will write to the OpenEXR format will
allow the creation of channels that contain entirely arbitrary data. For example, a channel with
specular highlights might exist in an OpenEXR. In most cases, the channel will have a custom
name that can be used to map the extra channel to one of the channels recognized by Fusion.
– SoftImage PIC (*.PIC, *.ZPIC and *.Z): The PIC image format (used by SoftImage) is an older image
format that can contain Z-Depth data in a separate file marked by the ZPIC file extension. These
files must be located in the same directory as the RGBA PIC files and must use the same names.
Fusion will automatically detect the presence of the additional information and load the ZPIC
images along with the PIC images.
– Wavefront RLA (*.RLA), 3ds Max RLA (*.RLA) and RPF (*.RPF): This is an older image format
capable of containing any of the image channels mentioned above. All channels are contained
within one file, including RGBA, as well as the auxiliary channels. These files are identified by the
RLA or RPF file extension. Not all RLA or RPF files contain auxiliary channel information, but most
do. RPF files have the additional capability of storing multiple samples per pixel, allowing different
layers of the image to be loaded for very complex depth composites.
– Fusion RAW (*.RAW): Fusion’s native RAW format is able to contain all of the auxiliary channels as
well as other metadata used within Fusion.

Creating Auxiliary Channels in Fusion

The following nodes create auxiliary channels:
– Renderer 3D: Creates these channels in the same way as any other 3D application would, and
you have the option of outputting every one of the auxiliary data channels that the Fusion page
supports.
– Optical Flow: Generates Vector and Back Vector channels by analyzing pixels over consecutive
frames to determine likely movements of features in the image.
– Disparity: Generates Disparity channels by comparing stereoscopic image pairs.

Chapter 18 Understanding Image Channels 435

Chapter 19

Compositing Layers
in Fusion
This chapter is intended to give you a solid base for making the transition from a
layer-based compositing application to Fusion’s node-based interface.
It provides practical information about how to start structuring a node tree for simple
layered composites.

Contents
Applying Effects  437
Adding a Node to the Tree  437
Editing Parameters in the Inspector  437
Replacing Nodes  438
Adjusting Fusion Sliders  439
Compositing Two Clips Together  439
Adding Additional Media to Compositions  439
Automatically Creating Merge Nodes  440
Fixing Problem Edges in a Composite  441
Using Composite Modes in the Merge Node  442
Creating and Using Text  443
Creating Text Using the Text+ Node  443
Styling and Adjusting Text  444
Using Text as a Mask  445
Using Transform Controls in the Merge Node  448
Building a Simple Green-Screen Composite  449
Mapping Timeline Layers to Nodes in Fusion  449
Pulling a Green-Screen Key Using the Delta Keyer  451
Dealing with Spill  454
Masking a Graphic  455

Chapter 19 Compositing Layers in Fusion 436

Applying Effects
Before we dive into multi-layered composites, let’s start by looking at some very simple effects and
build up from there. Opening the Effects Library, then clicking the disclosure control to the left of Tools,
reveals a list of categories containing all the nodes available in Fusion. As mentioned before, each
node does one thing, and by using these nodes in concert you can create extremely complex results
from humble beginnings.
Clicking the Effect category reveals its contents. In this example we’ll use the TV effect.

Browsing the Effect category to find the TV node

Adding a Node to the Tree

Assuming that the MediaIn node in the Fusion page or the Loader node in Fusion Studio is the
currently selected node in the Node Editor, clicking once on the TV node, for example, in the Effects
Library automatically adds that node to the node tree to the right of the selected node. In the Fusion
page, it immediately takes effect in the viewer thanks to the fact that the MediaOut1 node is what’s
loaded in the viewer, since that means that all nodes upstream of the MediaOut1 node will be
processed and shown.

A new node added from the Effects Library

In Fusion Studio, you must press the 1 or 2 key on the keyboard to load the selected node in
the viewer.
There are many other ways of adding nodes to your node tree, but it’s good to know how to browse
the Effects Library as you get started.

Editing Parameters in the Inspector

To improve an effect, you can make an adjustment to a node’s parameters in the Inspector at the right.
The selected node shows its controls in the Inspector, in which most nodes have several tabs of
controls seen as little icons just underneath that node’s title bar.

Chapter 19 Compositing Layers in Fusion 437

The Inspector showing the parameters of the TV effect

Clicking the last panel on any node opens the Settings panel. Every node has a Settings panel, and
this is where the parameters that every node shares, such as the Blend slider and RGBA buttons, are
found. These let you choose which image channels are affected, and let you blend between the effect
and the original image.

The Settings panel, which includes channel limiting

and mask handling controls that every node shares

In the case of the TV effect, for example, the resulting image has a lot of transparency because the
scan lines being added are also being added to the alpha channel, creating alternating lines of
transparency. Turning off the Alpha checkbox results in a more solid image, while opening the Controls
tab (the first tab) and dragging the Scan Lines slider to the right to raise its value to 4 creates a more
visible television effect.

The original TV effect (left), and modifications to the TV effect to make the clip more solid (right)

Replacing Nodes
In the Effect category of the Effects Library, you’ll also find a Highlight node that adds glints to the
highlights of an image.

Chapter 19 Compositing Layers in Fusion 438

Instead of clicking the Highlight node, which would add it after the currently selected node, dragging
and dropping a node from the Effects Library on top of a node in the Node Editor replaces the node in
the Node Editor.

Dragging a node from the Effects Library onto

a node in the Node Editor to replace it

In our example, the Highlight1 node takes the TV node’s place in the node tree, and the new effect can
be seen in the viewer, which in this example consists of star highlights over the lights in the image.
It’s time to use the Inspector controls to customize this effect.

Adjusting Fusion Sliders

When you drag a slider in the Fusion Inspector—in this case, the Number of Points slider—a little dot
appears underneath it. This dot indicates the position of the default value for that slider and also
serves as a reset button if you click it.

Adjusting a slider reveals a reset button underneath it

Each slider is limited to a different range of minimum and maximum values that is particular to the
parameter you’re adjusting. In this case, the Number of Points slider maxes out at 24. However, you
can remap the range of many (but not all) sliders by entering a larger value in the number field to the
right of that slider. Doing so immediately repositions the slider’s controls to the left as the slider’s range
increases to accommodate the value you just entered.

Entering a larger value to expand the range over which a slider will operate

Compositing Two Clips Together

As entertaining as it is adding individual nodes to create simple effects, eventually you’ll need to start
adding additional layers of media in order to merge them together as composites. Let’s turn our
attention to another example in which we need to combine a background clip with a foreground clip
that includes a built-in alpha channel, to see simple layering in action.

Adding Additional Media to Compositions

You’ll often find that even though you start out wanting to do something relatively simple, you end up
adding additional media to create the effect that you need.

Chapter 19 Compositing Layers in Fusion 439

 – In Fusion Studio, you do this by adding additional Loader nodes. If you add a new Loader node to
an empty area of the Node Editor, you’ll add an unconnected Loader2 node (incremented to keep
it unique) that you can then connect how you want.
– In the Fusion page, you can open the Media Pool and drag clips directly to the Node Editor to add
them to your node tree. If you drag a clip from the Media Pool to an empty area of the Node Editor,
you’ll add an unconnected MediaIn2 node (incremented to keep it unique) that you can then
connect in any way you want.

The Media Pool as seen in the Fusion page

Automatically Creating Merge Nodes

If you want to connect the incoming clip immediately to your node tree as the top layer, or foreground,
of a composite, in Fusion Studio select the Loader1 node and then add a second Loader node. In the
Fusion page, drag the new clip from the Media Pool right on top of any connection line.
In both cases, the new MediaIn or Loader node automatically becomes the “foreground input”.

Dragging a node from the Media Pool onto a connection (left), and dropping it to create a Merge node composite (right)

The Node Editor is filled with shortcuts like this to help you build your compositions more quickly.
Here’s one for when you have a disconnected node that you want to composite against another node
with a Merge node. Drag a connection from the output of the node you want to be the foreground
layer, and drop it on top of the output of the node you want to be the background layer. A Merge node
will be automatically created to build that composite. Remember: background inputs are orange, and
foreground inputs are green.

Chapter 19 Compositing Layers in Fusion 440

Dragging a connection from a disconnected node to another node’s output
(left), and dropping it to create a Merge node composite (right)

Adding Clips to a Fusion Composition From the File System

If you drag clips from the file system directly into the Node Editor, they’ll be added to the
DaVinci Resolve Media Pool automatically. So, if you have a library of stock animated
background textures and you’ve just found one you want to use using your file system’s
search tools, you can simply drag it straight into the Node Editor and it’ll be added to the
currently selected bin of the Media Pool.

Fixing Problem Edges in a Composite

Most often, the Merge node does a perfectly good job when handed a foreground image with
premultiplied alpha transparency to composite against a solid background image. However, from time
to time, you may notice a small bit of fringing at the edge of the border of a foreground element and
transparent area, such as seen in the following close-up. This slight lightening at the edge is a tell-tale
sign that the clip probably wasn’t premultiplied. The Merge node expects all foreground images with
alpha channels to be premultiplied. But this is something that’s easily fixed.

A bit of fringing at the edge of a foreground

element surrounded by transparency

Click to select the Merge node for that particular composite, and look for the Subtractive/
Additive slider.

The Subtractive/Additive slider, which can be

used to fix or improve fringing in composites

Drag the slider all the way to the left, to the Subtractive position, and the fringing disappears.

A clip with alpha exhibits fringing (left), and after fixing fringing by
dragging the Subtractive/Additive slider to the left (right)

Chapter 19 Compositing Layers in Fusion 441

The Subtractive/Additive slider, which is only available when the Apply mode is set to Normal, controls
whether the Normal mode performs an Additive merge, a Subtractive merge, or a blend of both. This
slider defaults to Additive merging, which assumes that all input images with alpha transparency are
premultiplied (which is usually the case). If you don’t understand the difference between Additive and
Subtractive merging, here’s a quick explanation:
– An Additive merge, with the slider all the way to the right, is necessary when the foreground image
is premultiplied, meaning that the pixels in the color channels have been multiplied by the pixels
in the alpha channel. The result is that transparent pixels are always black, since any number
multiplied by 0 is always 0. This obscures the background (by multiplying with the inverse of the
foreground alpha), and then simply adds the pixels from the foreground.
– A Subtractive merge, with the slider all the way to the left, is necessary if the foreground image
is not premultiplied. The compositing method is similar to an Additive merge, but the foreground
image is first multiplied by its own alpha to eliminate any background pixels outside the
alpha area.

The Additive/Subtractive slider lets you blend between two versions of the merge operation, one
Additive and the other Subtractive, to find the best combination for the needs of your particular
composite. Blending between the two is occasionally useful for dealing with problem composites that
have edges that are calling attention to themselves as either too bright or too dark.
For example, using Subtractive merging on a premultiplied image may result in darker edges, whereas
using Additive merging with a non-premultiplied image will cause any non-black area outside the
foreground’s alpha to be added to the result, thereby lightening the edges. By blending between
Additive and Subtractive, you can tweak the edge brightness to be just right for your situation.

Using Composite Modes in the Merge Node

To create a more convincing composite in layer-based systems, you often use Blend modes. Blend
modes are located in the Merge node since that is where one layer gets composted over another.
Let’s take an example where you want to use the Screen mode to make a foreground image look more
like a reflection.
The Merge node has a variety of controls built into it for creating just about every compositing effect
you need. Items you may be familiar with as Blend modes are located in the Apply Mode pop-up
menu. You can use these mathematical compositing modes to combine the foreground and
background layers together. A Blend slider allows you to fade the foreground input with the
background.

Adjusting the Apply Mode and Blend slider

of the Merge node in the Inspector

Chapter 19 Compositing Layers in Fusion 442

 NOTE: The Subtractive/Additive slider disappears when you choose any other Apply Mode
option besides Normal, because the math would be invalid. This isn’t unusual; there are a
variety of controls in the Inspector that hide themselves when not needed or when a
particular input isn’t connected.

The Screen node is perfect for simulating reflections, and lowering Blend a bit lets you balance the
foreground and background images. It’s subtle, but helps sell the shot.

TIP: You may have noticed that the Merge node also has a set of Flip, Center, Size,
and Angle controls that you can use to transform the foreground image without needing to
add a dedicated Transform node. It’s a nice shortcut for simplifying node trees large
and small.

Creating and Using Text

In this next example, we’ll look at how to create a simple text object using the Text+ node.
Then, we’ll see how to use the text generator’s alpha channel in another image to create a more
complex composite.

Creating Text Using the Text+ Node

The Text+ node is the primary tool for creating 2D text in the Fusion page. If you are using
DaVinci Resolve, this is also the same Text+ generator available in the Edit page. It is easily accessible
right in the toolbar.
The Text+ node is an incredibly deep tool for creating text effects, with six tabs of controls for
adjusting everything from text styling, to different methods of layout, to a variety of shading controls
including fills, outlines, shadows, and borders. As sophisticated a tool as this is, we’ll only be
scratching the surface in this example.
We’ll start with a MediaIn node that will serve as our background selected in the Node Editor. Clicking
the Text+ button automatically creates a new Text+ node connected as the foreground input of a
Merge node. The same behavior occurs if you are using Fusion Studio, with a Loader node.

Selecting a node you want to append another node to (top) Clicking

the Text+ button on the toolbar automatically creates a Merge
composite with the text as the foreground input connection (bottom)

Chapter 19 Compositing Layers in Fusion 443

Selecting a Text node opens the default Text panel parameters in the Inspector, and it also adds a
toolbar at the top of the viewer with tools specific to that node. Clicking on the first tool at the left lets
you type directly into the viewer, or you can type into the Styled Text field in the Inspector.

The viewer toolbar for the Text node with tools

for text entry, kerning, and outline controls

If you’re viewing the Merge, the text appears in the viewer superimposed against the background clip.
Onscreen controls appear that let you rotate (the circle) and reposition (the red center handle and two
arrows) the text, and we can see a faint cursor that lets us edit and kern the text using other tools in
the viewer toolbar.

Text that’s been typed into the viewer,

with onscreen text transform controls

Styling and Adjusting Text

To style the text, you use the controls in the Inspector, modifying text style controls such as Font, Size,
and Tracking to change the spacing between the letters.

The restyled text

TIP: Holding down the Command key while dragging any control in the Inspector “gears
down” the adjustment so that you can make smaller and more gradual adjustments.

Chapter 19 Compositing Layers in Fusion 444

Selecting the Manual Kerning tool in the viewer toolbar (second tool from the left) reveals small red
dots underneath each letter of text.

The Manual Kerning tool in

the viewer toolbar

Clicking a red dot under a particular letter puts a kerning highlight over that letter.
To make manual kerning adjustments:
1 Option-drag the red dot under any letter of text to adjust that character’s kerning while
constraining letter movement to the left and right. You can also drag letters up and down for other
effects. Depending on your system, the kerning of the letter you’re adjusting might not update until
you drop the red dot in place.
2 If you don’t like what you’ve done, you can open the Advanced Controls in the Inspector and clear
either the kerning of selected letters or all manual kerning before starting over again.

Option-dragging the little red dot revealed by the

Manual Kerning tool to manually adjust kerning

Using Text as a Mask

You can fill the text with a color or gradient, or you can use the text as a matte to cut the letters out of
another image. First, we’ll drag another clip of a chalkboard covered with math from the Media Pool to
the Node Editor as a disconnected MediaIn2 node.

Chapter 19 Compositing Layers in Fusion 445

Disconnecting and Reconnecting Nodes
If we start from the previous example that had a MediaIn as the background and the Text+ as the
foreground to a Merge, we’ll need to do a little rearranging. Clicking the last half of the connection
from the Text1 node to the Merge foreground input disconnects it.

Clicking the second half of a connection to disconnect it (top),

and the result with the text node disconnected (bottom)

Connecting a MediaIn2 or Loader2 node onto the Merge1 node’s foreground input causes the entire
viewer to be filled with the MediaIn2 (assuming we’re still viewing the Merge node). At this point, we
need to insert the Text1 node’s image as an alpha channel into the MediaIn2 node’s connection, and
we can do that using a MatteControl node.

The updated composite, with two video images

connected and the text node disconnected

Using Matte Control Nodes

Select the MediaIn2 node, and click the Matte Control button in the toolbar to add it between the
MediaIn2 and Merge1 nodes. (To tidy up, I’ve moved the nodes around a bit in the screenshot.)
The MatteControl node has numerous uses. Among them is taking one or more alpha channels,
mattes, or images that are connected to the Garbage Matte, Solid Matte, and/or foreground inputs,
combining them, and using the result as an alpha channel for the image that’s connected to the
background input. It’s critical to make sure that the image you want to add an alpha channel to is
connected to the background input of the MatteControl node, or the MatteControl node won’t work.

Chapter 19 Compositing Layers in Fusion 446

The second image properly connected to the
MatteControl node’s background input

With this done, connecting the Text+ node’s output, which has the alpha channel, to the MatteControl
node’s Garbage Matte input, is a shortcut we can use to make a mask, matte, or alpha punch out a
region of transparency in an image.
Keep in mind that it’s easy to accidentally connect to the wrong input. Because inputs rearrange
themselves depending on what’s connected and where the node is positioned (and, frankly, the colors
can be hard to keep track of when you’re first learning), it’s key to make sure that you always check the
tooltips associated with the input you’re dragging a connection over to make sure that you’re really
connecting to the correct one. If you don’t, the effect won’t work, and if your effect isn’t working, the
first thing you should always check is whether you’ve connected the proper inputs.
One alternate method of connecting nodes together is to hold down the Option key while dragging a
connection from one node’s output and dropping it onto the body of another node. This opens a
pop-up menu from which you can choose the specific input you want to connect to, by name. Note
that the menu only appears after you’ve dropped the connection on the node and released your
pointing device’s button.

Option-dragging a node connection to drop onto

another node exposes a node input menu

Once the Text1 node is properly connected to the MatteControl node’s Garbage Matte input, a text-
shaped area of transparency is displayed for the graphic if you load the MatteControl node into
the viewer.

Chapter 19 Compositing Layers in Fusion 447

Connecting the Text node to the MatteControl node’s Garbage Matte input (top),
and the resulting hole punched in the image (bottom)

Customizing Matte Control Nodes

You can use the Inspector to change some parameters to get the result you want. In the Inspector
controls for the MatteControl node, revealing the Garbage Matte controls exposes parameters for
modifying how the Garbage Matte input is applied to the image. For example, you can choose to have
the text mask filled with the image instead of cutting a hole in the image.

Using Transform Controls in the Merge Node

The Merge node includes a set of transform parameters in the Inspector that specifically affect the
foreground input’s image. This makes it quick and easy to adjust a foreground image to match the
background without requiring another node.

The Merge node transform controls that

affect the foreground input’s image

NOTE: When connecting two images of different sizes to a Merge node, the resolution of the
background image defines the output resolution of that node. Keep that in mind when you run
into resolution issues.

You can use the Size slider to resize the foreground.

Chapter 19 Compositing Layers in Fusion 448

The final composite

Building a Simple Green-Screen Composite

Taking another step forward in compositing, the next example shows how you can equate a
multilayered Timeline like the one in DaVinci Resolve’s Edit page to nodes in Fusion’s node tree. We’ll
use DaVinci Resolve, but understanding how layers map to nodes can be helpful for anyone new to
dealing with a node-based interface. In this example, we’ll create a simple composite using a green-
screen key and two other layers to create a news story.

Mapping Timeline Layers to Nodes in Fusion

This composite involves three layers in a Timeline. The Timeline consists of a background graphic on
video track 1, a green-screen clip on video track 2, and a foreground graphic on video track 3.

Background on video track 1 (top left), green-screen clip on video track 2 (bottom),
and graphic file on video track 3 (top right)

Implied in a timeline-based system is that higher numbered video tracks appear as the more forward,
or frontmost, element in the viewer. Video track 1 is the background to all other video tracks. Video
track 3 is in the foreground to both video track 1 and video track 2.

Chapter 19 Compositing Layers in Fusion 449

 TIP: If using DaVinci Resolve, you can bring all three layers from the Edit page into Fusion by
creating a Fusion clip. For more information on creating Fusion Clips, see Chapter 64,
“Getting Clips into Fusion” in the DaVinci Resolve Reference Manual or Chapter 3 in the
Fusion Reference Manual.

A stack of clips to use in a composite (top), and turning that stack into a
Fusion clip in DaVinci Resolve’s Edit page (bottom)

In Fusion, each video clip is represented by a MediaIn in the Fusion page or a Loader in Fusion Studio.
In our example below, the MediaIn2 is video track 2, and MediaIn 1 is video track 1. These two
elements are composited using a Merge node (foreground over background, respectively). The
composite of those two elements becomes the output of the first Merge node, which becomes the
background to a second Merge. There is no loss of quality or precomposing when you chain Merges
together. MediaIn3 represents video track 3 and is the final foreground in the node tree since it is the
topmost layer.

The initial node tree of the three clips we turned into a Fusion clip

Chapter 19 Compositing Layers in Fusion 450

With this node tree assembled to mimic the video layers, we can focus the rest of this example on
adding the nodes we’ll need to each branch of this tree to create the green-screen composite.

Pulling a Green-Screen Key Using the Delta Keyer

To make this composite, you need to create transparency behind the newscaster. When working in a
node tree, you must become accustomed to rearranging existing nodes to make room for new ones.
You’ll often move nodes off to the side or up above to make room for the additional nodes.

Creating space after the MediaIn nodes and selecting

the second one in preparation for adding a node

The DeltaKeyer node is the main tool used for green-screen keying. It attaches to the output of the
node that represents the green screen—in our example, that is the MediaIn2 node. With the MediaIn2
selected, pressing Shift-Space opens the Select Tool dialog where you can search for and insert any
node. Below we have added the DeltaKeyer after the MediaIn2 node but prior to being merged with
the background.

Adding a DeltaKeyer node inline after the MediaIn2 node

The DeltaKeyer node is a sophisticated keyer that is capable of impressive results by combining
different kinds of mattes and a clean-plate layer, but it can also be used very simply if the background
that needs to be keyed is well lit. And once the DeltaKeyer creates a key, it embeds the resulting alpha
channel in its output, so in this simple case, it’s the only node we need to add. It’s also worth noting
that, although we’re using the DeltaKeyer to key a green screen, it’s not limited to keying green or blue
only; the DeltaKeyer can create impressive keys on any color in your image.
With the DeltaKeyer selected, we’ll use the Inspector controls to pull our key by quickly sampling the
shade of green from the background of the image. To sample the green-screen color, drag the
Eyedropper from the Inspector over the screen color in the viewer.

Chapter 19 Compositing Layers in Fusion 451

Dragging the Eyedropper to the viewer samples the screen color

As you drag in the viewer, an analysis of the color picked up by the location of the Eyedropper
appears within a floating tooltip, giving some guidance as to which color you’re really picking.
Meanwhile, if viewing the Merge in a second viewer, we get an immediate preview of the transparency
and the image we’ve connected to the background.

The original image (left), and after sampling the green screen using the Eyedropper from the Inspector (right)

When we’re happy with the preview, releasing the pointer button samples the color, and the Inspector
controls update to display the value we’ve chosen.

The DeltaKeyer Inspector updates with the sampled color

No matter how good the composite may look, once you’ve selected the screen color to pull a key, you
need to load the DeltaKeyer node into the viewer itself. This allows you to evaluate the quality or
density of the alpha channel created by the key. Above the viewer, click the Color button in the viewer
toolbar, or click in the viewer and press C to switch the viewer between the RGB color channels of the
image and the alpha channel.

Chapter 19 Compositing Layers in Fusion 452

 Loading the DeltaKeyer into the viewer and clicking the
Color button to view the alpha channel being produced

Black in a matte represents the transparent areas, while white represents the opaque areas. Gray
areas represent semi-transparency. Unless you are dealing with glass, smoke, or fog, most mattes
should be pure white and pure black with no gray gray areas. If a close examination of the alpha
channel reveals some fringing in the white foreground of the mask, the DeltaKeyer has integrated
controls for post-processing of the key and refining the matte. Following is a quick checklist of the
primary adjustments to make.
After making the screen selection with the Eyedropper, try the following adjustments to
improve the key.
– Adjust the Gain slider to boost the screen color, making it more transparent. This can adversely
affect the foreground transparency, so adjust with care.
– Adjust the Balance slider to tint the foreground between the two non-screen colors. For a green
screen, this pushes the foreground more toward red or blue, shifting the transparency in the
foreground.

Clicking the third of the seven tabs of controls in the DeltaKeyer Inspector opens up a variety of
controls for manipulating the matte.
Initial adjustments in the matte tab may include the following parameters:
– Adjust the lower and upper thresholds to increase the density in black and white areas.
– Very subtly adjust the Clean Foreground and Clean Background sliders to fill small holes in the
black and white matte. The more you increase these parameters, the more harsh the edges of
your matte become.

Adjusting the Clean Foreground slider in the

Matte panel of the DeltaKeyer controls

Chapter 19 Compositing Layers in Fusion 453

 In this case, raising the Clean Foreground slider a bit eliminates the inner fringing we don’t want,
without noticeably compromising the edges of the key.

The original key (left), and the key after using the Clean Foreground slider (right)

With this accomplished, we’re happy with the key, so we load the Merge1 node back into the
viewer, and press C to set the Color control of the viewer back to RGB. We can see the graphic in
the background, but right now it’s too small to cover the whole frame, so we need to make another
adjustment.

The final key is good, but now we need to work on the background

Dealing with Spill

The DeltaKeyer can handle any screen color that may bounce off the background and land on your
subject. This bounce screen light is called spill. However, instead of using one node to handle the
matte creation and the spill correction, consider disabling this in the DeltaKeyer and remove the spill
using a separate color correction node.
To disable spill suppression in the DeltaKeyer, do the following.
– Select the matte tab in the Inspector.
– From the Replace Mode drop-down menu, choose Source. This uses the color from the original
source image, effectively disabling any spill correction in the DeltaKeyer.

Chapter 19 Compositing Layers in Fusion 454

 The Replace Mode menu determines how spill
suppression is handled in the DeltaKeyer

Spill can now be handled using a color correction node placed directly after the DeltaKeyer or
branched from the original MediaIn or Loader node and combined with a MatteControl.

Branching the original image with one branch for the DeltaKeyer and a second for color correction

Masking a Graphic
Next, it’s time to work on the top video track: the news graphic that will appear to the left of the
newscaster. The graphic we will use is actually a sheet of different logos, so we need to cut one out
using a mask and position it into place.

A graphic of multiple logos that must be cropped down to isolate just one

Chapter 19 Compositing Layers in Fusion 455

The easiest way to crop a MediaIn or Loader node is to add one of the mask shapes from the toolbar
directly to it. Selecting the MediaIn or Loader node and clicking the Rectangle mask from the toolbar
will crop, or mask off, the graphic.

Masking the logo using a Rectangle mask connected directly to a Merge node

Now, all we need to do is to use the onscreen controls of the Rectangle mask to crop the area we want
to use, dragging the position of the mask using the center handle, and resizing it by dragging the top/
bottom and left/right handles of the outer border.
As an extra bonus, you can take care of rounded corners when masking a graphic by using the Corner
Radius slider in the Inspector controls for the Rectangle mask to add the same kind of rounding.

Moving and resizing the mask to fit our logo, and rounding
the edges using the Corner Radius Inspector control

For a simple over-the-shoulder graphic, masking the image may be all you need to do, but masking an
image does not change the actual dimensions of the graphic. It only changes the area you see. So,
accurately positioning the graphic based on the center of the composite becomes more difficult, and
any type of match moving would give incorrect results because the graphic has a different resolution
than the background. To fix this resolution mismatch, you can place a Crop node after the MediaIn to
change the actual dimensions of the graphic layer.

Adding a Crop node after the masked MediaIn to center the cropped logo on the background

With the Crop node selected, the viewer toolbar includes a Crop tool.

Selecting the crop tool in the viewer toolbar

Chapter 19 Compositing Layers in Fusion 456

You can crop the image by dragging a bounding box around it. Unlike a mask which creates a small
window you view the image through, a crop effectively changes the resolution of the graphic to the
crop bounding box size.

Dragging a bounding box using the Crop tool (left), and the cropped logo now centered on the frame (right)

NOTE: The Resize, Letterbox, and Scale nodes also change the resolution of an image.

At this point, we’re all set to move the logo into place. Because the logo is the foreground input to
a Merge, you can select the Merge2 node, load it into the viewer, and use the built-in Center X
and Y controls or the oncreen controls to place the logo where you want it and make it a
suitable size.

Placing the logo using the foreground input transform controls of the Merge2 node

Chapter 19 Compositing Layers in Fusion 457

Chapter 20

Rotoscoping
with Masks
This chapter covers how to use masks to rotoscope, one of the most common
tasks in compositing.

Contents
Introduction to Masks and Polylines  459
Mask Nodes  459
Polyline Types  460
Converting Polylines from One Type to Another  461
How to Use Masks with Other Nodes  462
Attaching Masks to an Image for Rotoscoping  462
Combining Multiple Masks  464
Mask Inputs on Other Nodes  464
Creating and Editing Polylines In-Depth  467
The Polyline Toolbar  467
Selecting a Specific Polyline  467
Polyline Creation Modes  468
Protection Modes  469
Closing Polylines  469
Selecting and Adjusting Polylines  470
Polyline Points Selection  470
Moving Polyline Points  470
Smoothing a Polyline Segment  471
Linearizing a Polyline Segment  471
Transforming Individual or Multiple Points  471
Deleting Selected Points  471
Editing Bézier Handles  472

Chapter 20 Rotoscoping with Masks 458

Point Editor  472
Reduce Points  473
Shape Box  473
Showing and Hiding Onscreen Polyline Controls  474
Stop Rendering  475
Roto Assist  475
Creating Softness Using Double Polylines  475
Converting a Single Polyline to a Double Polyline  476
Adding Softness to a Segment  477
Adding Additional Points to the Shape  477
Locking/Unlocking Point Pairs  477
Animating Polyline Masks  477
Removing Animation from a Polyline Mask  478
Adding and Removing Points from an Animated Mask  478
Publishing Specific Control Points  478

Introduction to Masks and Polylines

Polylines are splines that are used whenever a control is animated with a motion path or when a
node’s effect is masked with a drawn shape. They are also used in the Paint and Grid Warp nodes. In a
more basic form, polylines are used to control the animation in the Spline Editor. Since these splines
are used for just about everything, they are extremely flexible, with a considerable amount of controls,
modes, and options. This chapter offers an overview of polylines and their operation, with specific
information on how to use them for masks.

Mask Nodes
Mask nodes create an image that is used to define transparency in another image. Unlike other image
creation nodes in Fusion, mask nodes create a single channel image rather than a full RGBA image.
The most used mask tool, the Polygon mask tool, is located in the toolbar.
For more information on these mask tools, see Chapter 107, “Mask Nodes” in the DaVinci Resolve
Reference Manual or Chapter 46 in the Fusion Reference Manual.

The available nodes in the Mask bin of the Effects Library

Chapter 20 Rotoscoping with Masks 459

Polygon Mask
Polygon masks are user-created Bézier shapes. This is the most common type of polyline and the
basic workhorse of rotoscoping. Polygon mask tools are automatically set to animate as soon as you
add them to the Node Editor.

B-Spline Masks
B-Spline masks are user-created shapes made with polylines that are drawn using the B-Splines. They
behave identically to polyline shapes when linear, but when smoothed the control points influence the
shape through tension and weight. This generally produces smoother shapes while requiring fewer
control points. B-Spline mask tools are automatically set to animate as soon as you add them to the
Node Editor.

Bitmap Masks
The Bitmap mask allows images from the Node Editor to act as masks for nodes and effects. Bitmap
masks can be based on values from any of the color, alpha, hue, saturation, luminance, and the
auxiliary coverage channels of the image. The mask can also be created from the Object or Material ID
channels contained in certain 3D-rendered image formats.

Mask Paint
Mask Paint allows a mask to be painted using Fusion’s built-in vector paint nodes.

Wand Mask
A Wand mask provides a crosshair that can be positioned in the image. The color of the pixel under
the crosshair is used to create a mask, where every contiguous pixel of a similar color is also included
in the mask. This type of mask is ideal for isolating color adjustments.

Ellipse, Rectangle, and Triangle Masks

These are primitive shape masks. For more information on these mask tools, see Chapter 107,
“Mask Nodes” in the DaVinci Resolve Reference Manual or Chapter 46 in the Fusion
Reference Manual.

Ranges Mask
Similar to the Bitmap mask, the Ranges mask allows images from the Node Editor to act as masks for
nodes and effects. Instead of creating a simple luminance-based mask from a given channel, Ranges
allows spline-based selection of low, mid, and high ranges, similar to to the Color Corrector node.

Polyline Types
You can draw polylines using either B-Spline or Bézier spline types. Which you choose depends on
the shape you want to make and your comfort with each spline style.

Bézier Polylines
Bézier polylines are shapes composed of control points and handles. Several points together are used
to form the overall shape of a polyline.

Bézier control point with direction handles

extended to create a smooth curve

Chapter 20 Rotoscoping with Masks 460

Each control point has a pair of handles used to define the exact shape of the polyline segments
passing through each control point. Adjusting the angle or length of the direction handles will affect
whether that segment of the polyline is smooth or linear.

Bézier control point with direction handles aligned to create a linear segment

If you’re familiar with applications such as Adobe Photoshop or Illustrator, you’ll already be familiar with
many of the basic concepts of editing Bézier polylines.

B-Spline Polylines
A B-Spline polyline is similar to a Bézier spline; however, these polylines excel at creating smooth
shapes. Instead of using a control point and direction handles for smoothness, the B-Spline polyline
uses points without direction handles to define a bounding box for the shape. The smoothness of the
polyline is determined by the tension of the point, which can be adjusted as needed.

B-Splines excel at creating smooth curves

Converting Polylines from One Type to Another

Just because you created a shape using a B-Spline or polyline, that doesn’t mean you’re stuck with the
controls you started with. You can convert any shape from B-Spline to Bézier, or Bézier to B-Spline,
as needed.

To switch a shape between Polyline and B-Spline controls:

– Right-click a shape in the viewer and choose Convert Bézier Spline to B-Spline or
Convert B-Spline to Bézier from the spline’s contextual menu (only the appropriate option
will be displayed).

When converting from one type to another, the original shape is preserved. The new polyline
generally has twice as many control points as the original shape to ensure the minimum change to the
shape. While animation is also preserved, this conversion process will not always yield perfect results.
It’s a good idea to review the animation after you convert spline types.

Chapter 20 Rotoscoping with Masks 461

How to Use Masks with Other Nodes
Typically, a node applies its effect to every pixel of an image. However, many nodes have mask inputs
that can be used to limit the effect that node has on the image.

A Blur node with a Polygon

node masking its effect

Masks are single-channel images that can be used to define which regions of an image you want to
affect. Masks can be created using primitive shapes (such as circles and rectangles), complex polyline
shapes that are useful for rotoscoping, or by extracting channels from another image.

A Polygon node’s mask seen in the viewer

Each mask node is capable of creating a single shape. However, Mask nodes are designed to be
added one after the other, so you can combine multiple masks of different kinds to create complex
shapes. For example, two masks can be subtracted from a third mask to cut holes into the resulting
mask channel.
Fusion offers several different ways you can use masks to accomplish different tasks. You can attach
Mask nodes after other nodes in which you want to create transparency, or you can attach Mask nodes
directly to the specialized inputs of other nodes to limit or create different kinds of effects.

Attaching Masks to an Image for Rotoscoping

There are two ways you’ll typically attach a Mask node, such as a Polygon node, so that it adds an
alpha channel to an image for compositing later in the node tree.

Chapter 20 Rotoscoping with Masks 462

Using a MatteControl Node
The MatteControl node is the main node used for combining masks in different ways and inserting the
result into an image stream. The MatteControl node is attached downstream of the node outputting
the image you want to rotoscope. You’ll typically attach a Polygon or B-Spline node to the Garbage
Matte input of the MatteControl node to use the spline as an alpha channel.

Feeding a Polygon node to a MatteControl node to perform rotoscoping

To use this setup, you’ll load the MatteControl node into the viewer and select the Polygon node to
expose its controls so you can draw and modify a spline while viewing the image you’re rotoscoping.
The MatteControl node’s Garbage Matte > Invert checkbox lets you choose which part of the image
becomes transparent.

Connecting a Mask to a MediaIn or Loader Node’s Input

This method is a bit simpler but requires you to know that you can view one node while adjusting
another node, even if that other node is disconnected. If you add an unattached Mask node such as a
Polygon or B-Spline node, and then place a MediaIn or Loader node directly into the viewer while
selecting the Mask node, you can draw a spline to rotoscope the image.

Rotoscoping a MediaIn node using

a disconnected Polygon node

When you’re finished rotoscoping, you simply connect the Polygon node’s output to the Loader node’s
input, and an alpha channel is automatically added to that node.

Connecting a Polygon node to a MediaIn node to use a spline as an alpha channel

TIP: If you connect a Mask node to a MediaIn or Loader node’s effect input without any
shapes drawn, that mask outputs full transparency, so the immediate result is that the image
output by the MediaIn or Loader node becomes completely blank. This is why when you want
to rotoscope by connecting a mask to the input of a MediaIn or Loader node, you need to
work within a disconnected Mask node first. Once the shape you’re drawing has been closed,
connect the Mask node to the MediaIn or Loader’s input, and you’re good to go.

Chapter 20 Rotoscoping with Masks 463

Combining Multiple Masks
Masks are designed to be added one after the other, with each Mask node acting as an additional
layer of masking.

Combining multiple Polygon nodes one after the other in the node tree

When a Mask node’s input is attached to another mask, a Paint Mode drop-down menu appears, which
allows you to choose how you want to combine the two masks.

The Paint Mode parameter in the Polygon node Inspector parameters

The default option is Merge, but you can also choose Subtract, Minimum, Maximum, Multiply, or any
other operation that will give you the mask boolean interaction you need. Additionally, a pair of Invert
and Solid checkboxes let you further customize how to combine the current mask with the one
before it.

The Invert and Solid options

Mask Inputs on Other Nodes

Masks can be used for a variety of reasons, so there are several categories of mask inputs that
different nodes include to accommodate these different uses. Incidentally, in most cases you can
connect either masks or mattes to a mask input to take advantage of that input’s functionality.

TIP: If you select a node with an empty effect mask input, adding a Mask node automatically
connects to the open effect mask input.

Chapter 20 Rotoscoping with Masks 464

Effects Mask Inputs
Almost every node in Fusion has an Effect mask input (colored blue), which lets you choose which
parts of the image will or will not be affected by that node.

A Blur node with a Polygon node masking its effect

While masks (or mattes) are connected via an input, they are actually applied “post effect,” which
means the node first applies its effect to the entire image, and then the mask is used to limit the result
by copying over unaffected image data from the input.

A Polygon node used as a mask to limit the Blur node’s effect

Although many nodes support effects masking, there are a few where this type of mask does not
apply—notably Savers, Time nodes, and Resize, Scale, and Crop nodes.

TIP: Effects masks define the domain of definition (DoD) for that effect,
making it more efficient.

Pre-Masking Inputs
Unlike effect masks, a pre-mask input (the name of which is usually specific to each node using them)
is used by the node before the effect is applied. This usually causes the node to render more quickly
and to produce a more realistic result. In the case of the Highlight and the Glow nodes, a pre-mask
restricts the effect to certain areas of the image but allows the result of that effect to extend beyond
the limits of the mask.

Chapter 20 Rotoscoping with Masks 465

The advantage to pre-masking is that the behavior of glows and highlights in the real world can be
more closely mimicked. For example, if an actor is filmed in front of a bright light, the light will cause a
glow in the camera lens. Because the glow happens in the lens, the luminance of the actor will be
affected even though the source of the glow is only from the light.
In the case of the DVE node, a pre-mask is used to apply a transformation to a selected portion of the
image, without affecting portions of the image outside of the mask. This is useful for applying
transformations to only a region of the image.

Garbage Matte Inputs

Garbage Matte inputs (usually colored gray) are used to exclude light stands, rigging, and boom
microphones that intrude upon masks being pulled via blue-screen and green-screen keys. In the
following example, a lighting stand to the left is removed from the image via a B-Spline node’s mask
connected to the Garbage Matte input of the DeltaKeyer node.

A B-Spline node is connected to the Garbage Matte input of a DeltaKeyer

node to eliminate a light stand at the left of frame

TIP: You can quickly add a mask node to the Effect/Solid/Garbage Matte inputs of a keyer
node by right-clicking the header bar of that node in the Inspector and choosing whichever
mask node you want to use from the Effect Mask, SolidMatte, and GarbageMatte submenus.

You choose whether a garbage matte is applied to a keying node as opaque or transparent in the
Inspector for the node to which it’s connected.

Solid Matte
Solid Matte inputs (colored white) are intended to fill unwanted holes in a matte, often with a less
carefully pulled key producing a dense matte with eroded edges, although you could also use a
polygon or mask paint to serve this purpose. In the example below, a gentle key designed to preserve
the soft edges of the talent’s hair leaves holes in the mask of the woman’s face, but using another
DeltaKeyer to create a solid matte for the interior of the key that can be eroded to be smaller than the
original matte lets you fill the holes while leaving the soft edges alone. This is also sometimes known
as a hold-out matte.

Chapter 20 Rotoscoping with Masks 466

Filling in holes in the mask pulled by the DeltaKeyer1 node (left) with another, harder but eroded
key in DeltaKeyer2 that’s connected to the SolidMatte input of DeltaKeyer1 (right)

Creating and Editing Polylines In-Depth

This section covers the Polygon node’s capabilities in depth.

The Polyline Toolbar

Whenever a node that contains one or more polylines is selected, the polyline is shown on all viewers
and the Polyline toolbar is displayed along the side of each viewer. The toolbar contains several
buttons that make switching polyline modes and options easy to access.

The Polyline toolbar

If you hover the pointer over any of the Polyline toolbar buttons, a tooltip that describes the button’s
function appears. Clicking on a button will affect the currently active polyline or the selected polyline
points, depending on the button.
You can change the size of the toolbar icons, add labels to the buttons, or make other adjustments to
the toolbar’s appearance in order to make polylines easier to use. All the options can by found by
right-clicking on the toolbar and selecting from the options displayed in the contextual menu.

Selecting a Specific Polyline

It is possible to have several polylines in the viewer at once if you select multiple Mask nodes in the
Node Editor, so it’s important to be able to switch between polylines easily.

To make a polyline active, do one of the following:

– Click one of the polyline’s control points or segments.
– Press Tab and Shift-Tab to cycle between available polylines.
– Right-click in the viewer and choose the desired polyline by name from the Controls >
Select menu.

Chapter 20 Rotoscoping with Masks 467

Polyline Creation Modes
There are several different modes available from the toolbar for creating and modifying polylines.
The specific mode used when a polyline is first added will depend on whether it is used as a path
or a mask.
Each of the modes is described in more detail below.

Click Append
This mode is the default mode for mask creation. It’s used to quickly define the rough shape of the
mask, after which you switch to Insert and Modify mode to refine the mask further.

The Click Append toolbar button (Shift-C)

To create a mask using the Click Append mode, do the following:

1 Select Click Append from the toolbar or press Shift-C.
2 Click the pointer where you want to start the shape.
3 Move and click the pointer to append a point to the last one.
4 To close the shape, place the mouse pointer over the first point created and click when the pointer
changes shape.

When a shape is closed, the polyline is automatically switched to Insert and Modify mode.
Although the Click Append mode is rarely used with paths, it can be helpful when you know the
overall shape of a motion path, but you don’t yet know the timing.

TIP: Holding Shift while you draw a mask constrains subsequent points to 45-degree angles
relative to the previous point. This can be very helpful when drawing regular geometry.

Insert and Modify

Masks, which are created in Click Append mode, automatically switch to Insert and Modify mode when
the mask shape is closed. You can also manually switch to this mode by clicking the Insert and Modify
button in the toolbar or using the Shift-I keyboard shortcut. This mode makes it easier to add additional
points and refine the shape of the mask. Dragging the control points or direction handles modifies
existing points on the polyline.

The Insert Modify toolbar button (Shift-I)

Insert and Modify mode is also the default mode for creating motion paths. A new control point is
automatically added to the end of the polyline, extending or refining the path, any time a parameter
that is animated with a motion path is moved.

Chapter 20 Rotoscoping with Masks 468

Draw Append
The Draw Append mode creates a freehand polyline shape directly on the viewer, like drawing with a
pencil or a paintbrush. This mode is ideal to use in conjunction with a tablet and for the creation of
garbage mattes and effect masks around complex shapes.

The Draw Append toolbar button (Shift-D)

Protection Modes
In addition to the modes used to create a polyline, two other modes are used to protect the points
from further changes after they have been created.

Modify Only
Modify Only mode allows existing points on the polyline to be modified, but new points may not be
added to the shape.

The Modify Only toolbar button (Shift-M)

TIP: Even with Modify Only selected, it is still possible to delete points from a polyline.

Done
The Done mode prohibits the creation of any new points, as well as further modification of any existing
points on the polyline.

The Done toolbar button (Shift-N)

Closing Polylines
There are several ways to close a polyline, which will connect the last point to the first.

To close a polyline, do one of the following:

– Hover the pointer over the first point created, and then click on the point.
– Press Shift-O on the keyboard.
– Click the Close button on the polyline toolbar.
– Draw a polyline until you are ready to close the shape, and then right-click and choose
Polygon:Polyline > Closed.

Chapter 20 Rotoscoping with Masks 469

 The Close toolbar button (Shift-O)

All these options are toggles that can also be used to open a closed polygon.

Selecting and Adjusting Polylines

To create the shape you need for a mask or a motion path, you need to know how to manipulate the
splines. Fusion provides a number of simple techniques for selecting, moving, and smoothing a spline,
but also includes more complex adjustment techniques for scale, skewing, and twisting a spline.

Polyline Points Selection

To select one or more control points on a polyline, do one of the following:
– Click directly on the control points.
– Lasso around the points.

To add or remove points from the current selection, do one of the following:
– Hold the Shift key to select a continuous range of points.
– Hold Command and click each control point you want to add or remove.
– Press Command-A to select all the points on the active polyline.

TIP: Once a control point is selected, you can press Page Down or Page Up on the keyboard
to select the next control point in a clockwise or counterclockwise rotation. This can be very
helpful when control points are very close to each other.

Moving Polyline Points

The selected polyline points can be moved using either the keyboard or the mouse.

To move selected control points using the pointer, do one of the following:
– Drag on the selected points anywhere in the viewer.
– Hold Shift while dragging to restrict movement to a single axis.
– Hold Option and drag anywhere in the viewer to move the selected control point.

To move selected control points using the keyboard, do one of the following:
– Press the Up or Down Arrow keys on the keyboard to nudge a point up or down in the viewer.
– Hold Command-Up or Down Arrow keys to move in smaller increments.
– Hold Shift-Up or Down Arrow keys to move in larger increments.

Chapter 20 Rotoscoping with Masks 470

Smoothing a Polyline Segment
If you want to shape the polyline and control its slope, you can choose to smooth a spline segment by
adjusting the Bézier direction handles.

To smooth the selected points on an active polyline, do one of the following:

– Press Shift-S.
– Click the Smooth button on the Polyline toolbar.
– Choose Smooth from the polyline’s contextual menu.

The Smooth button in the toolbar (Shift-S)

Linearizing a Polyline Segment

To make certain that a polyline segment is perfectly straight, that segment must be linearized. A linear
segment aligns the Bézier direction handles with the segment and therefore has no curvatures. The
segment is always drawn in a straight line between two points on the polyline.

To linearize the selected points on an active polyline, do one of the following:

– Press Shift-L.
– Click the Linear button on the polyline’s toolbar.
– Choose Linear from the polyline’s contextual menu.

The Linear button in the toolbar (Shift-L)

Transforming Individual or Multiple Points

Select the points to be transformed, and then do one of the following:
– Hold T and drag to twist.
– Hold S and drag to scale.
– Hold X and drag to scale horizontally only.
– Hold Y and drag to scale vertically only.
– Hold O and drag to offset the points perpendicular to the tangent.
The position of the pointer when the transformation begins becomes the center used for the
transformation.

Deleting Selected Points

You can delete a selected point or group of points by pressing Delete or Backspace, choosing Delete
from the contextual menu, or by clicking the Delete Point button in the toolbar. The shape of the
polyline changes to reflect the removal of these points.

Chapter 20 Rotoscoping with Masks 471

 TIP: Deleting all the points in a polyline does not delete the polyline itself. To delete a
polyline, you must delete the node or modifier that created the polyline.

Editing Bézier Handles

For Bézier polylines, each control point has two direction handles that adjust the slope of a curve
through the control point. These direction handles appear only when the point is selected.
Dragging a direction handle makes adjustments to the curve of the segment that emerges from the
control point. The direction handle on the opposing side of the control point will also move to maintain
the relationship between these two handles.
To break the relationship between direction handles and adjust one independently, hold Command
while dragging a handle. Subsequent changes will maintain the relationship, unless Command is held
during each adjustment.

Hold Command to adjust one handle independently

If you want to adjust the length of a handle without changing the angle, hold Shift while moving a
direction handle.

Point Editor
The Point Editor dialog can be used to reposition control points using precise X and Y coordinates.
Pressing the E key on the keyboard will bring up the Point Editor dialog and allow you to reposition
one or more selected control points.

The Point Editor dialog can be used to position control points

The dialog box contains the X- and Y-axis values for that point. Entering new values in those boxes
repositions the control point. When multiple control points are selected, all the points move to the
same position. This is useful for aligning control points along the X- or Y-axis.
If more than one point is selected, a pair of radio buttons at the top of the dialog box determines
whether adjustments are made to all selected points or to just one. If the Individual option is selected,
the affected point is displayed in the viewer with a larger box. If the selected point is incorrect, you can
use the Next and Previous buttons that appear at the bottom of the dialog to change the selection.

Chapter 20 Rotoscoping with Masks 472

In addition to absolute values for the X- and Y-axis, you can adjust points using relative values from
their current position. Clicking once on the label for the axis will change the value to an offset value.
The label will change from X to X-offset or from Y to Y-offset.

The Point Editor dialog with Offset values

If you are not sure of the exact value, you can also perform mathematical equations in the dialog box.
For example, typing 1.0-5 will move the point to 0.5 along the given axis.

Reduce Points
When freehand drawing a polyline or an editable paint stroke, the spline is often created using more
control points than you need to efficiently make the shape. If you choose Reduce Points from the
polyline’s contextual menu or toolbar, a dialog box will open allowing you to decrease the number of
points used to create the polyline.

The Reduce Points button in the toolbar

The overall shape will be maintained while eliminating redundant control points from the path. When
the value is 100, no points are removed from the spline. As you drag the slider to the left, you reduce
the number of points in the path.

Shape Box
If you have a polyline shape or a group of control points you want to scale, stretch, squish, skew, or
move, you can use the shape box to easily perform these operations.
To enable the shape box, do one of the following:
– Click the Shape Box toolbar button.
– Choose Shape Box from the contextual menu.
– Press Shift-B.

The Shape Box button in the Polyline toolbar

If there are selected points on the polyline when the Shape Box mode is enabled, the shape box is
drawn around those points. Otherwise, you can drag the shape box around the area of control points
you want to include.

Chapter 20 Rotoscoping with Masks 473

If you want to freely resize the shape box horizontally and vertically, you can drag a corner handle.
Dragging a handle on the side of the shape box resizes the polyline along a specific axis.

Dragging a side handle resizes along a specific axis

Holding Command while dragging a shape box handle will apply adjustments from the center of the
shape box, constraining the transformation to the existing proportions of the shape box. Holding Shift
while dragging a corner handle affects only that handle, allowing skewed and non-uniform
transformations.

Hold Shift while dragging a corner to perform non-uniform transformations

Showing and Hiding Onscreen Polyline Controls

It is often difficult to identify individual points when they are placed closely together. You can choose
to display both points and their direction handles, just points, or just handles. These display mode
options are selected using the Show Key Points and Show Handles toolbar buttons, or from the
polyline’s context menu.

The Show Key Points and Show

Handles buttons in the toolbar

You use these options to simplify the screen display when adjusting control points placed closely
together and to avoid accidentally modifying controls and handles that are adjacent to the
intended target.

Chapter 20 Rotoscoping with Masks 474

Stop Rendering
While points along the polyline are being moved, the results are rendered to the viewer to provide
constant interactive feedback. Although extremely useful, there are situations where this can be
distracting and can slow down performance on a complex effect. To disable this behavior so renders
happen only when the points stop moving, you can toggle the Stop Rendering button in the toolbar or
select this option from the polyline contextual menu.

Roto Assist
You can enable the Roto Assist button in the toolbar when you begin drawing your shape to have
points snap to the closest high-contrast edge as you draw the shape. The points that have snapped to
an edge are indicated by a cyan outline.

There are three main Roto Assist options.

– Multiple Points: Allows adding multiple points along an entire edge with a single click instead of
having to add each point individually.
– Distance: Defines the pixel range within which searching for an edge will take place.
– Reset: Used for resetting the snap attribute of the snapped points. After resetting, the points will
become unavailable for tracking.

The Roto Assist options in the toolbar

Creating Softness Using Double Polylines

The standard soft edge control available in all Mask nodes softens the entire mask equally. However,
there are times, particularly with a lot of motion blur, when softening part of the curve while keeping
other portions of the curve sharp is required.
This form of softness is called non-uniform softness, which is accomplished by converting the shape
from a single polyline to a double polyline. The double polyline is composed of two shapes: an inner
and an outer shape. The inner shape is the original shape from the single polyline, whereas the outer
shape is used to determine the spread of the softness. The further the outer shape gets from the inner
shape, the softer that segment of the shape becomes.

Chapter 20 Rotoscoping with Masks 475

A double polyline uses an inner and outer shape for non-uniform softness

Converting a Single Polyline to a Double Polyline

To convert a mask into a double polyline, click the Double Polyline button in the Polyline toolbar or
right-click in the viewer and select Make Outer Polyline from the mask’s contextual menu.
The shape will be converted into an inner and an outer polyline spline. Both polylines start with exactly
the same shape as the original single polyline. This keeps the mask sharp to start with and allows any
animation that may have already been applied to the shape to remain.

Make Double Polyline button

The control points on the outer shape are automatically parented to their matching points on the inner
shape. This means that any changes made to the inner shape will also be made to the outer shape.
The relationship is one-way; adjustments to the outer shape can be made without affecting the
inner shape.

Chapter 20 Rotoscoping with Masks 476

A dashed line drawn between the points indicates the relationship between the points on the inner
and outer shapes.

Adding Softness to a Segment

The outer shape is drawn using a green dashed line instead of a solid line to help distinguish it from
the inner shape.

To select the outer soft edge shape, do one of the following:

– Use the Tab key to cycle between the onscreen controls until the dashed outline is visible
– Right-click over a spline in the view and choose Controls > Select > Polygon: Outer Polygon.

Once the outer polyline is selected, you can drag any of the points away from the inner polyline to add
some softness to the mask.

TIP: Press Shift-A to select all the points on a shape, and then hold O and drag to offset the
points from the inner shape. This gives you a starting point to edit the falloff.

The farther the outer shape segment is from the inner shape, the larger the falloff will be in that area.

Adding Additional Points to the Shape

It is not necessary for every point on the inner shape to have a match on the outer shape, or vice
versa. You can add additional control points to refine the shape of either shape.
Each polyline stores its animation separately; however, if a point is adjusted on the inner shape that is
parented to a point on the outer shape, a keyframe will be set for both splines. Adjusting a parented
point on the outer shape only sets a keyframe for the outer shape’s spline. If a point that is not
parented is adjusted, it will only set a keyframe on the relevant spline. You can disable this behavior
entirely for this polyline by selecting Polygon: Outer Polygon > Follow Inner Polyline from the
contextual menu.

Locking/Unlocking Point Pairs

If you want to parent additional control points, you can select the points, right-click in the viewer, and
choose Lock Point Pairs from the contextual menu for either spline. This will cause the selected point
on the outer shape to become parented to the selected point on the inner shape.
Any animation already applied to either point is preserved when the points become parented.
To unlock a point so it is no longer parented, select the point, right-click in the viewer, and deselect
Lock Point Pairs from the contextual menu.

Animating Polyline Masks

Animating masks is surprisingly easy. When Polygon or B-Spline masks are added to the Node Editor,
the spline’s control points are automatically ready to be animated. All you have to do to animate a
mask is move the playhead to a new frame and then change the shape of the mask. A new keyframe is
added in the Spline Editor and Timeline Editor. This one keyframe controls the position of all control
points for that mask at that frame. Once two or more keyframes have been created, the shape of the
polygon or B-Spline is automatically interpolated from one keyframe to the next.

Chapter 20 Rotoscoping with Masks 477

 TIP: The center point and rotation of a shape are not auto-animated. Only the control points
are automatically animated. To animate the center position or rotation, enable keyframes for
that parameter in the Inspector.

To adjust the overall timing of the mask animation, you edit the Keyframe horizontal position spline
using the Spline Editor or Timeline Editor. Additional points can be added to the mask at any point to
refine the shape as areas of the image become more detailed.

Removing Animation from a Polyline Mask

If you want a Polyline mask to remain static, you can remove the automatic animation setting. In the
Inspector for the mask, right-click in the bottom of the panel where it says Right Click Here For Shape
Animation. From the contextual menu, choose Remove Bézier Spline. If you decide you need to
animate the mask at a later time, right-click in the same area again and choose Animate.

Adding and Removing Points from an Animated Mask

When adding points to an animated mask, the new point is fit into the shape at all keyframes. Deleting
a point removes that point from all keyframes in the animated mask.

Publishing Specific Control Points

Although you can rapidly animate the entire shape of a polyline using a single keyframe, by default the
Spline Editor and Timeline display only one keyframe for the entire shape at any given frame.
This default keyframing behavior is convenient when quickly animating shapes from one form to
another, but it doesn’t allow for specific individual control points that need to be keyframed
independently of all other control points for a particular shape. If you’re working on a complex mask
that would benefit from more precise timing or interpolation of individual control points, you can
expose one or more specific control points on a polyline by publishing them.
Be aware that publishing a control point on a polyline removes that point from the standard animation
spline. From that point forward, that control point can only be animated via its own keyframes on its
own animation spline. Once removed, this point will not be connected to paths, modifiers, expressions,
or trackers that are connected to the main polyline spline.

To publish a selected point or points, do one of the following:

– Click on the Publish Points button in the Polyline toolbar.
– Select Publish Points from the Polyline’s contextual menu.

A new coordinate control is added to the Polyline mask controls for each published point, named Point
0, Point 1, and so on.

The Publish Points controls in the Inspector

Chapter 20 Rotoscoping with Masks 478

The onscreen control indicates published points on the polyline by drawing that control point much
larger. Once a published point is created, it can be connected to a tracker, path, expression, or
modifier by right-clicking on this control and selecting the desired option from the point’s
contextual menu.

The published point in the viewer

Using “Publish to Path” to Preserve Animation

When a point is published, any animation already applied to that point is removed. However, if you
need to keep the animation, you can use the “Publish to Path” option. This Polyline contextual menu
option publishes the selected points and converts their existing animation to a path. You can also use
the Publish to Path button in the Polyline toolbar.

Using “Follow Published Points” to Add Points

There are times when you will need to have control points that lie between two other published points
follow the motion of the published points, while still maintaining their relative offset and shape. For this
reason, points in a Polyline mask can be set to “Follow Published Points” using the Polyline’s
contextual menu.
When a point of an effect mask is set to follow points, the point will be drawn as a diamond shape
rather than a small box.

A control point set to Follow Published Points

When this mode is enabled, the new “following” control points will maintain their position relative to
the motion of any published points in the mask, while attempting to maintain the shape of that segment
of the mask. Unlike published points, the position of the following points can still be animated to allow
for morphing of that segment’s shape over time.

Chapter 20 Rotoscoping with Masks 479

Chapter 21

Paint
This chapter describes how to use Fusion’s non-destructive Paint tool to repair
images, remove objects, and add creative elements.

Contents
Paint Overview  481
Types of Paint Nodes  481
Setting Up the Paint Node  481
Setting the Paint Node’s Resolution  482
Paint Node Workflow  482
Select the Correct Paint Stroke Type  483
Setting the Brush Size  485
Choosing an Apply Mode  485
Editing Paint Strokes  489
Editing Paint Strokes in the Modifiers Tab  489
Deleting Strokes  490
Animating and Tracking Paint Strokes  490
Tracking a Paint Stroke  491
Using the Planar Tracker with the Paint Tool  493
Inverting the Steady Effect to Put the Motion Back In  499
Painting a Clean Plate  500

Chapter 21 Paint 480

Paint Overview
The Paint node is a procedural paint tool, which means that each paint stroke is a live, editable object
that’s drawn with properties that you can mix and match to address a wide variety of painting tasks.
You can use it to paint masks, retouch images, perform beauty work, clone out objects, or even create
motion graphics. Each element of a paint stroke can be altered long after you apply it. Since strokes
are editable, you can apply, change, ignore, delete, and even reorder them in a node tree.

Types of Paint Nodes

There are two types of paint nodes in Fusion. The Paint node is a full-featured creative and retouch
vector-based paint tool that requires an input on which to paint. The Mask Paint node lets you
specifically paint an alpha channel to limit the area of an effect. It allows you to create paint strokes on
an alpha channel without needing to have an input.
– The Paint node is located in the Paint category of the Effects Library.
– The Mask Paint node is located in the Mask category of the Effects Library.

The main difference between these two Paint tools is that the Mask Paint tool only paints on the Alpha
channel, so there are no channel selector buttons. The Paint tool can paint on any or all channels. The
majority of this chapter covers the Paint node, since it shares identical parameters and settings with
the Mask Paint node.

The Paint node available in the Paint category of the Effects Library

Setting Up the Paint Node

The Paint node has two inputs. Typical of most Fusion nodes, the orange input background is the
primary input for connecting the “canvas” or image to paint on, while the second blue input is an
Effect Mask. Unlike the Mask Paint node, the Paint node requires a background input to begin painting.

Chapter 21 Paint 481

Setting the Paint Node’s Resolution
No matter how you use the Paint node in your node tree, the Paint node assumes the resolution of the
background input image as your working resolution for that operation. Although the Paint tool is in fact
resolution independent and you can change this assigned resolution at any time, it’s essential to make
sure you properly set the resolution of the media you’re working with, because it will affect operations
such as motion tracking that you’ll want to use as part of your paint workflow.
Because of this, the Paint node requires a background input to set the resolution of the “canvas” you’ll
be painting upon. To do this, you can set up a Paint node in the node tree in one of two ways: painting
directly on an image or using Paint as the foreground.

Painting Directly on an Image

You can connect the image you want to paint on directly to the background input of the Paint node.
This is the easiest and cleanest node tree construction, but it doesn’t provide much in terms of
compositing flexibility.

The Paint node is inserted directly after the node it is painting on

Using Paint as the Foreground in a Merge Composite

An alternative setup is to use a Background node to set the resolution for the Paint node, compositing
the result over the actual background using a Merge node. Working this way lets you use the Merge
node’s Apply Mode setting (also referred to as composite modes) to control how your paint strokes are
composited against the image, but it does require a bit more setup.

The Paint node is composited over the image you want to paint on using the merge

Setting this up requires some configuration of the nodes. The Background node must be fully
transparent and, unless you are doing something simple like using the Stroke tool set to Color to paint
over an image, you must drag the image you want to clone or smudge into the Source Tool field in the
Paint node’s inspector. These steps are described in more detail later in this chapter.

Paint Node Workflow

You begin painting by first selecting the paint stroke type from the Paint toolbar above the viewer.
There are ten stroke types to choose from as well as two additional tools for selecting and grouping
paint strokes. The stroke types and tools are described in detail in Chapter 112, “Paint Node” in the
DaVinci Resolve Reference Manual or Chapter 51 in the Fusion Reference Manual. The primary tool for

Chapter 21 Paint 482

painting and cloning is the Stroke tool. The Stroke tool is a fully animatable and editable vector-based
paint stroke and initially uses a duration of the entire global range.

The Stroke tool is most often used for cloning, beauty work, and creative paint

Select the Correct Paint Stroke Type

Instead of having multiple dedicated Paint nodes for different operations, the procedural nature of
Fusion’s Paint tool means there is just one tool with a variety of different stroke types. Each stroke type
serves a different purpose. The Paint toolbar above the viewer allows you to choose between different
stroke types and drawing tools. These are grouped into a few categories.

Single-Frame Stroke Types

The Multistroke and Clone Multistroke are explicitly designed for single-frame retouching paint jobs
like removing raindrops in a shot that’s supposed to be a sunny day or removing scratch marks and
dust when restoring vintage content. When handling these types of jobs, these two Multistroke
options are faster than the other stroke types, but they’re not editable later on. This means you must
set up the size and function of the brush as well as the duration for the paint stroke before you paint.
The duration of each of these stoke types is one frame by default, but this can be changed using the
Stroke Duration slider in the Inspector. Multistroke and Clone MultiStroke are basically the same tools,
except Clone Multistroke automatically configures the tool for cloning. In contrast, the Multistroke
requires you to set up the tool for cloning manually.

The Multistroke and Clone Multistroke are

non-editable single frame stroke types

Editable Stroke Types

The Stroke and Polyline are similar in that they can be modified and animated at any time. Also, they
both begin with a duration lasting the entire comp, but that also can be changed using the
Keyframes Editor.

The Stroke and Polyline strokes are

editable and last for the entire comp

The Stroke Tool

One of the most flexible editable stroke types you’ll use for many tasks is the Stroke, because it is fully
animatable and editable. You can animate all elements of the Stroke, and you can use the Write-on/
Write-off parameters to control how the stroke appears onscreen. You can also connect to a tracker
from the Center point of the Stroke if you want to make the stroke follow specific onscreen motion.
By default, the Stroke type does not expose control points for the shape of the path. You can move
and track the center and rotation of the Stroke, but the individual control points that create the spline
are hidden. To reveal the control points, you can open the Stroke Controls at the bottom of the
Inspector and click the Make Editable button.

Chapter 21 Paint 483

The Stroke’s control points can be revealed
using the Make Editable button

Although the Stroke type is the most flexible, that flexibility can come at a performance penalty if
you’re painting hundreds of strokes on a frame. For larger numbers of strokes that do not need to be
animated, it’s better to use Multistroke or Clone Multistroke, as these are more processor efficient.

The Polyline Stroke Tool

The Polyline Stroke acts more like a drawing tool than a paintbrush. It includes the same functionality
as the Stroke tool, except that it is created not by dragging or “painting” like a paintbrush, but by
clicking to create a spline path, as you do with masks and motion paths. Without even creating a stroke
in the viewer, the Polyline Stroke can connect to existing polylines like a mask or a motion path.
If a motion path is published, right-clicking on the Shape Animation label at the bottom of the Polyline
Stroke’s Stroke Controls allows you to use the Connect To menu to assume the shape of a motion path
or mask. You can also use this method if you import SVG graphics and want to “paint-on” the outlines.

The Polyline Stroke Shape Animation label

Shape Drawing Tools

Five shape-based drawing tools allow you to draw shapes and either fill them with a color or clone an
area from a source image. All these tools act similarly to the Stroke and Polyline stroke type in that
they are editable at any time and have a default duration spanning the entire global range of the comp.
However, you can edit the duration at any time in the Keyframes Editor.

The Shape Strokes are used to create shapes or clone areas based on shapes

Chapter 21 Paint 484

All of the Copy [Shape Name] stroke types require that you connect the source node you are cloning
from into the Paint node, and set the Fill Type menu to Image.

The Copy Shapes require a Source to be directly

connected to the Paint tool and set to Image

Setting the Brush Size

After selecting the Stroke type, the Brush size can be set in the Inspector or more intuitively in the
viewer. With the Paint node selected in the Node Editor and the pointer positioned over the viewer,
you can see an outline of the current brush size. To change the brush size, hold down the Command
key and drag. The circle changes size, so you can set it relative to other objects you may be
painting over.

Brush size can be changed interactively in the viewer

Choosing an Apply Mode

The Apply Mode buttons determine the functionality of the paintbrush. There are eight Apply modes
that set the brush to do things like paint a color, clone from a source, smudge an area, or remove
thin wires.

Apply Mode buttons determine the paint brush functionality

Chapter 21 Paint 485

Picking a Paint Color
There are several ways to pick the paint color and opacity for a colored brush stroke. You use the Fill
button in the row of Apply modes when you want to paint with a solid color.

The Color swatch shows the current color Clicking

it opens the OS Color Picker window

To select a color for the paint brush, do one of the following:

– Click the color swatch to open a standard OS Color Picker window.
– Drag the Eyedropper into the viewer.
– Drag inside the color chooser to select a saturation and luminance. Drag on the sidebars to
change the hue and transparency.

When you paint, each stroke is unpremultiplied, so adjusting the Alpha slider in the Inspector does not
affect what you apply to the RGB channels. However, changing opacity affects all four channels.

Cloning from the Frame

Choosing the Clone Apply Mode allows you to paint from one area of an image over another area. This
is the most common use of the Paint tool. It allows you to remove objects or artifacts from a clip by
covering them up with another area of the frame. Depending on the Stroke type chosen, you may
clone on either a single frame or for the entire duration of the clip.

The Clone Apply Mode allows you to sample from one

area and use it as a source to paint over another area

You can use the Clone Apply Mode to clone from the same image connected to the Paint node’s
background input or a different source from the node tree.

Chapter 21 Paint 486

To clone from a different area of the same frame:
1 Select a Stroke type from the Paint toolbar above the viewer.
2 Using the size slider in the Brush controls section of the Inspector, set the size of the brush.
3 From the Apply Mode buttons, select the Clone mode.
4 Option-click over the area in the viewer you want to use as the source. A dot appears showing you
the center of what you’re sampling from.

The Clone source starting area identified by the X

and the paint brush size represented by the circle

5 Paint over the area you want to cover up using the source pixels.

The Clone completed after selecting the source

area and painting over the flag pole

When trying to erase objects or artifacts from a clip using the Clone Apply Mode, it can sometimes
be easier if you sample from a different frame on the same clip. This works well when the object you
are trying to clone out moves during the clip, revealing the area behind the object. Sampling from a
different frame can use the revealed background by offsetting the source frame.

To clone from a different frame of the same clip:

1 Select a Stroke type from the Paint toolbar above the viewer.
2 Set the size of the brush.
3 From the Apply Mode buttons, select the Clone mode.
4 Drag the clip (MediaIn or Loader) from the Node Editor into the Source Tool field in the Inspector.

The Paint Inspector with the MediaIn1 dragged from

the Node Editor into the Source Tool field

Chapter 21 Paint 487

5 Click the Overlay checkbox to see the current frame and the offset frame superimposed.
6 Drag the Time Offset slider to select the source frame you want to use.
7 Option-click over the area in the viewer you want to use as the source or to offset the source
frame’s position.

Overlay shows two frames overlapped with Time Offset,

allowing you to clone from one frame onto another

8 Paint over the area you want to cover up using the source pixels.

The plane is half painted out using on the Overlay with Time Offset

9 Disable the Overlay checkbox.

The Clone Apply Mode can use a

different frame from the same clip

TIP: When using a Clone Apply Mode, you can hold down the O key instead of clicking the
Overlay checkbox in Inspector to see the Overlay. Releasing the O key will return to normal
viewing without the Overlay.

Chapter 21 Paint 488

Editing Paint Strokes
Once you’ve painted using the Stroke or Polyline stroke type, you can change the look of the stroke
by selecting it and updating the parameters in the Inspector. Selecting the stroke requires you to
switch to the selection tool in the Paint toolbar above the viewer. Using the Paint node’s selection tool,
you can either click once on a stroke or drag a bounding box around a stroke to select it for editing.

TIP: To select multiple strokes, you can Shift-click or Command-click to select and deselect
multiple specific strokes, or you can drag a selection box around all strokes you want
to select.

The Stroke or Polyline Stroke type can be edited

by selecting the stroke in the viewer

Although you can make changes in the Tools tab in the Inspector, the Paint node uses both the Tools
tab and the Modifiers tab. In the Tools tab, you can create new brush strokes and select a stroke in the
viewer to edit.  The Modifiers tab presents a list of all the strokes for the selected Paint node, which
makes it easy to modify any previously created paint stroke. 

NOTE: Multistroke and Clone Multistroke each only appear as one item in the Modifiers tab
no matter how many strokes you create using those tools. Those two tools are not editable
after creating them.

The same controls you used in the Tools tab to create the strokes are located in the Modifier’s tab to
modify them. You can also animate each individual stroke.

Editing Paint Strokes in the Modifiers Tab

When you paint a stroke, the settings for that stroke appear in the Inspector’s Modifiers tab. You can
then change the settings in the Tools tab for the next stroke you are about to paint. Each time you
click, drag, and release the pointer button, you create a new stroke. Each stroke is numbered in the
Modifiers tab, where it can be selected and edited.

Chapter 21 Paint 489

 Each stroke is listed in the Modifiers tab for editing

Once you stop painting a stroke, it’s added to the Modifiers tab along with an additional Stroke
modifier that represents the next stroke. For instance, if you paint your first stroke, the Modifiers tab
shows your stroke as Stroke1 and then a stroke 2 as well, which represents the next stroke you create.
You always have one more stroke in the Modifiers tab than strokes in the viewer.

Deleting Strokes
There are two ways you can delete paint strokes.

To delete any individual stroke, do the following:

1 Select the Paint node.
2 Click the Modifiers tab.
3 Right-click over the Stroke header you want to delete and choose Delete from the menu.

To delete all paint strokes you’ve made on every frame, do one of the following:
– Click the reset button in the upper-right corner of the Inspector.
– Delete the Paint node in the Node Editor.

Animating and Tracking Paint Strokes

In some ways, animating paint strokes is no different than animating any other effect in the Inspector.
Each parameter that can be animated includes a gray diamond Keyframe button along the right side.
Clicking the Keyframe button sets a keyframe on the current frame and enables auto-keyframe mode
for the parameter. However, more commonly, the paint stroke is tracked using one of Fusion’s trackers,
or for motion graphics, animated using the Write-On Start and End sliders.

Chapter 21 Paint 490

Animating with Write-On Controls
The Stroke and Polyline stroke types include Write-On controls located in the Stroke controls section
of the Inspector. These Write-On controls animate the appearance of a stroke along the path. You can
animate the Write-On controls using either the Stroke Animation drop-down menu or using the Start
and End sliders.

Stroke Animation Drop-Down Menu

The Stroke Animation drop-down menu includes six options for auto-animating a paint stroke. The first
two options do not truly animate the stroke as much as set a duration. The Limited Duration option
uses the Duration slider to set the number of frames the stroke is onscreen.
To auto-animate the stroke, you can choose one of the three Write options or the Trail option.
Choosing Write On automatically creates a write-on animation. The duration is set by two keyframes
that get added when you choose Write On from the menu. The Start keyframe is set on the frame
where you first created the stroke. The End keyframe is added on the current frame when you choose
Write On from the menu. The remaining options in the menu set their Start and End keyframes similarly
but change the direction of the animation based on the menu selection.

Write-On Start and End Parameters

The Write-On Start and End parameters allow you to manually control the start point and end point
along any stroke’s path, and use keyframes to animate each parameter individually. The Start
parameter determines the point at which the stroke begins, measured as a percentage offset from the
beginning of the stroke’s path. For example, a Start value of 50 moves the starting point of the stroke
to the middle of the stroke’s path. The End parameter works the same way but from the other end of
the stroke. You can animate a stroke onscreen, creating a handwriting effect by setting keyframes for
the End parameter from 0 to 100 over several frames.

Tracking a Paint Stroke

You can animate the position of a paint stroke using any of Fusion’s trackers. For instance, if you
cloned out a flag pole from a clip, but the camera moves, you can track the flag pole and attach the
resulting path to the paint stroke.

Trackers can be attached to the Center parameters of a paint stroke

To attach a tracker to a paint stroke:

1 With the Paint node, select the Stroke brush type and clone out an object on a frame.
2 In the Paint toolbar above the viewer, click the Select tool.
3 Drag a selection box around the stroke to select it.
4 Right-click the center control on the stroke, and then choose Stroke1:Center > Modify With >
Tracker Position.

Chapter 21 Paint 491

 Right-clicking over the Stroke’s Center control in the view allows you to apply a tracker modifier

5 Click the Modifiers tab to view the Tracker controls.

6 From the Node Editor, drag the MediaIn for the image you painted on, and drag it into the Tracker
Source field in the Inspector.

Drag the MediaIn you want to track into the Tracker Source field in the Inspector

7 Click the Track Forward button.

8 After tracking, at the bottom of the Inspector, use the Tracker 1 X Offset/Y Offset controls to
reposition the paint stroke, if necessary.

Tracking a Group of Paint Strokes

You can assign a tracker to multiple strokes by adding the stroke to a group and connecting the
tracker to the group. Instead of connecting each individual stroke, the group’s center is used for all of
the strokes. Assuming the motion of each object is consistent in the same direction, as it would be
objects “nailed to the set”, then applying the tracker to the group makes cloning multiple objects out
with a single paint node very easy.

Chapter 21 Paint 492

To group paint strokes, do the following:
1 Drag a bounding box, Shift-click, or Command-click to select every stroke that you want to
group together.
2 Click the Paint Group button in the Paint toolbar.

Selecting all the strokes and then clicking the Paint Group
button collects all the strokes into a single group

The group’s onscreen controls replace the controls for each paint stroke, and the Modifiers tab in the
Inspector shows the group’s parameters. The individual strokes are still editable by selecting Show
Subgroup Controls in the Modifiers tab of the Inspector. The group then comes with a Center, Angle,
and Size control for connecting to a tracker.

Use the onscreen controls, or in the Modifiers tab, right-click

over the Paint Group’s Center X label to connect the tracker

Using the Planar Tracker with the Paint Tool

Here’s an example that dives deeper into a workflow where we use the Paint tool with the Planar
Tracker for retouching a clip. We’ll eliminate some facial scars on an actor’s forehead in a commercial
by combining the Paint node with the PlanarTracker node, illustrating a common way of using these
two powerful tools together.

The actor has some scars on his forehead

that the director would like painted out

Chapter 21 Paint 493

Because this is a clip in motion, we can’t just paint out the scars on the man’s forehead; we need to
deal with the motion so that the paint work we do stays put on his face. In this case, a common
workflow is to analyze the motion in the image and use it to apply a “steady” operation, pinning down
the area we want to paint in place so we can paint on an unmoving surface.

Setting Up the Planar Tracker for Stabilization

The best way to do this is to use the Planar Tracker, so we’ll add the PlanarTracker node after the
MediaIn1 node, such that the image connects to the background input of the PlanarTracker node. As
always, it’s important to be careful about which input you connect for the effect to work properly.

Adding a PlanarTracker node to analyze and steady

the part of the image we want to paint on

With the PlanarTracker node selected and loaded in the viewer, a viewer toolbar appears with a variety
of tools for drawing shapes and manipulating tracking data. The Planar Tracker works by tracking flat
surfaces that you define by drawing a shape around the feature you want to track. When you first
create a PlanarTracker node, you can immediately begin drawing a shape, so in this case, we draw a
simple polygon over the man’s forehead since that’s the feature we want to steady in preparation for
painting.
We draw a simple box by clicking once each on each corner of the man’s forehead to create control
points, and then clicking the first one we created to close the shape.

Drawing a shape over the man’s forehead to prepare for Planar Tracking

In the Inspector, the PlanarTracker node has tracking transport controls that are similar to those of the
Tracker. However, there are two buttons, Set and Go, underneath the Operation Mode menu, which
defaults to Track, since that’s the first thing we need to do. The Set button lets you choose which
frame to use as the “reference frame” for tracking, so you click the Set button first before clicking the
Track Forward button below.

Chapter 21 Paint 494

Setting a reference frame at the beginning
of the range of frames we want to track

TIP: The Set button lets you supervise a Planar Track in progress and stop it if you see it
slipping, making adjustments as necessary before clicking Set at the new frame to set a new
reference before continuing to track forward towards the end of the clip.

The Pattern controls let you set up how you want to handle the analysis. Of these controls, the Motion
Type menu is perhaps the most important. In this particular case, Perspective tracking is the analysis
we want. Still, in other situations, you may find you get better results with the Translation, Translation/
Rotation, and Translation/Rotation/Scale options.
Once you initiate the track, a series of dots appears within the track region shape you created to
indicate trackable pixels found. A green progress bar at the bottom of the Timeline ruler lets you see
how much of the shot is remaining to track.

Clicking the Track from First Frame button to set the Planar Track in progress; green dots
on the image and a green progress bar let you know the track is happening

NOTE: If nothing happens when you track, or it starts to track and then stops, that’s your cue
that there isn’t enough trackable detail within the shape you’ve drawn for the Planar Tracker
to work, and your best bet is to choose a different location of the image to track.

Chapter 21 Paint 495

Once the track is complete, you can set the Operation Mode of the PlanarTracker node’s controls in
the Inspector to Steady.

Setting the PlanarTracker node to Steady

You’ll immediately see the image warped as much as is necessary to pin the tracked region in place
for whatever operation you want to perform. If you scrub through the clip, you should see that the
image dynamically cornerpin-warps as much as is necessary to keep the forehead region within the
shape you drew pinned in place. In this case, this sets up the man’s head as a canvas for paint.

Steadying the image results in warping as the forehead is pinned in place for painting

At this point, you’re ready to paint out those scars.

Connecting the Paint Node

Although you could paint directly on the image by connecting the Paint node after the Planar Tracker,
it gives you more control over the process if you merge the Paint node over the top of the steadied
image. To do that, you’ll add a Merge after the Planar Tracker and then connect a Background node
into the foreground of the Merge. Insert a Paint node after the background, and you are ready for
this process.

Paint node merged over the top of a steadied Planar Tracker

Making the Background Node Transparent

The Background node must be fully transparent; otherwise, you are painting on the solid color instead
of the Merge background image. To make the background transparent, drag the Alpha channel
sidebar all the way up in the Color Picker.

Chapter 21 Paint 496

The Color Picker in the Background
node set to be transparent

Selecting the Stroke and Clone Mode

With the Paint node selected and set up, the next thing we want to do is to select the Stroke tool. The
Stroke tool is the tool of choice when you want to paint out features or paint in fixes to subjects within
the frame that need to remain in place for the entire shot.

Choosing the Stroke tool from the Paint node’s tools in the viewer toolbar

Next, choose the Clone mode from the Apply Controls. In this example, we’ll clone part of the man’s
face over the scars to get rid of them. Choosing the Clone mode switches the controls of the Paint
node to those used for cloning.

Choosing the Clone mode in the Inspector

Chapter 21 Paint 497

Setting a Clone Source
The Paint node clones from the connected Input image unless you instruct it otherwise. If you are just
painting a color over the background image, then you are good to go. However, if you plan on using
the merge’s steadied background image for cloning with the Paint tool, you must set that image as the
Paint node’s source. To set the clone source for the Paint node, you drag the PlanarTracker node into
the Source Tool field in the Inspector.

Any node can be dragged into the Source

Tool field when cloning with the Paint tool

With the Stroke tool selected in the Paint toolbar, the Clone mode selected in the Inspector controls,
and the Source for cloning added to the Source Tool field, we’re ready to start painting. If we move the
pointer over the viewer, a circle shows us the paint tool, ready to go.
To use the clone brush, first hold down the Option key and click somewhere on the image to identify
the source area of the clone. In this example, we’ll sample from just below the first scar we want to
paint. After Option-clicking to sample the image, you can click to begin painting anywhere in
the frame.

Setting an offset to sample for cloning (left), and dragging to draw a clone stroke (right)

If you don’t like the stroke you’ve created, you can undo with Command-Z and try again. We repeat the
process with the other scar on the man’s forehead, possibly adding a few other small strokes to make
sure there are no noticeable edges, and in a few seconds, we’ve taken care of the issue.

Original image (left), and after painting out two scars on the man’s forehead with the Stroke tool set to Clone

Chapter 21 Paint 498

 TIP: You can adjust the size of the brush right in the viewer, if necessary, by holding down the
Command key and dragging the pointer left and right. You’ll see the brush outline change
size as you do this.

Inverting the Steady Effect to Put the Motion Back In

At this point, scrubbing through the clip shows that the paint strokes we’ve made are indeed sticking
to the man’s forehead as we need them to do. Now we just have to invert the transform that the Planar
Tracker applied to restore the clip back to the way it was, except now with the painted fix attached in
the process. This ends up being a two-part process, but the first part is the simplest.

Scrubbing through the steadied clip shows the paint fix is “sticking” to the man’s forehead

We select and copy the PlanarTracker node coming before the Merge node, and paste a copy of it
after. This copy has all the analysis and tracking data of the original PlanarTracker node.

Pasting a second copy of the PlanarTracker node after the Paint node

With the second PlanarTracker node selected, we go into the Inspector and turn on the Invert Steady
Transform checkbox, which inverts the steady warp transform to restore the image back to the
way it was.

Turning on Invert Steady Transform to try to

restore the image back to the way it was

This is just one example of how to set up a Planar Tracker and Paint node. In some instances, you
made need to do more work with masks and layering, but the above example gives you a good
starting point.

Chapter 21 Paint 499

Painting a Clean Plate
On simple clips, planar tracking the clone paint strokes may work fine. In other cases, you may not be
able to steady the clip, or the strokes might appear like they are “bubbling.” Paint is just a single stroke
repeated over multiple frames. Each painted frame has the potential to appear differently than the one
before or after. Often, the more reliable way to use paint is by creating what is referred to as a “clean
plate.” Create one good painted frame and freeze it. Then use the Planar Tracker to track the cleaned
area over the top of the original. The benefit is that there is only a single paint stroke instead of
repeated paint strokes that potentially show unwanted artifacts. For this next example, we continue
with the previous clip, but now the technique is to freeze, mask, and composite just a single frame of
the fixed forehead over the original clip.

Creating a Freeze Frame and Clean Plate

As in the previous example, you still track the clip using the Planar Tracker. After tracking, you can
branch out from the MediaIn (or Loader node if you are using Fusion Studio) into a Time Stretcher to
create the freeze frame.

Time Remap is used to retime or freeze a clip

Disable the default Keyframe in the Time Stretcher and enter the frame you want to freeze. If you have
already performed a Planar Track, then entering the frame you set as the Reference Frame is usually a
good frame to freeze.
To create the clean plate, you connect the paint node to the output of the Time Stretcher. Clone over
the areas you want to hide, and you now have a single clean frame. Now you need to composite the
clean area over the original.

Isolating the Painted Forehead

First, we need to mask out just the man’s painted forehead. We can do this by connecting a
MatteControl node to the output of the Paint node and then connect a Polygon node to the garbage
matte input. This lets us draw a shape with the Polygon node and use it as a mask to crop out the
man’s painted forehead.

Add a MatteControl node with a garbage mask to cut out the painted forehead

Chapter 21 Paint 500

 TIP: When it comes to using masks to create transparency, there are a variety of ways to
connect one—for example, (a) attach the image to the background input of a Brightness/
Contrast node and attach a Polygon mask node to the effect mask input. On the Brightness/
Contrast node, enable the Alpha channel and lower the Gain slider to darken a hole, or (b)
using Channel Booleans to copy channel data to the alpha from a Polygon node attached to
the foreground input and the image attached to the background input.

Drawing a Polygon Mask

After moving the playhead to the first frame of the clip, we’re ready to draw a mask to isolate the “clean
plate” forehead. Loading the MatteControl1 into the viewer and selecting the Polygon1 node so that we
see its tools in the viewer toolbar sets us up for drawing a polygon.
Drawing shapes using the Polygon node is similar to shape drawing in other spline-based
environments, including the Color page:
– Clicking once draws a corner control point.
– Clicking and dragging creates a Bézier curve.
– Clicking the first control point you created closes a shape.
We click and drag to create a shape that outlines the man’s forehead, and when we close the shape,
we see exactly the opposite of what we want, a hole in the middle of the image.

Drawing a shape to isolate the forehead gives an inverted result

at first when using the Garbage Matte input of the MatteControl
node to attach the Polygon to the MatteControl node

Before fixing this, we drag the Soft Edge slider in the Inspector to the right to blur the edges just a bit.

Inverting the Garbage Input

Selecting the MatteControl1 node, we open the Garbage Matte controls and click the Invert checkbox,
which immediately gives us the result we want, of the forehead in isolation, ready for compositing.

Chapter 21 Paint 501

Inverting the Garbage Matte input (top), and the resulting
inverted mask inverting the forehead (bottom)

Compositing the Painted Forehead Against the Original Image

Almost finished. Next, we’ll add a Merge node to layer the fixed forehead over the original image
being output by the MediaIn node.
We create a Merge node connected to the output of the PlanarTracker node, and then we connect the
MatteControl’s output to the green foreground input of the Merge node. This puts the cropped and
fixed forehead on top of the original image.

The painted forehead composited against the original image

Match Moving the Mask to the Shot

Now we have the best of both worlds: a fixed forehead and the background of the shot looking good.
However, if we select the Polygon node and then scrub forward in the clip, the fixed forehead mask
drifts out of sync with the motion of the shot, so we have one remaining issue. Fortunately, match
moving the mask to move with the shot is really simple.

Because the Polygon isn’t animated to match

the motion of the shot, it goes out of sync

Chapter 21 Paint 502

Selecting the first PlanarTracker node that comes right after the MediaIn node, and choosing Track
from the Operation Mode menu, reveals a Create Planar Transform button at the bottom of the listed
controls. Clicking this button creates a new, disconnected Planar Transform node in the Node Editor,
which has the transforms from the Planar Tracker baked in. Unlike the Planar Tracker, the Planar
Transform is intended for use with masks. With the Planar Transform created, there is no longer a use
for the Planar Tracker, and it can be disconnected or deleted from the Node Editor.
We can insert this new node into the node tree to use it by holding down the Shift key and dragging
the node over the connection between the Polygon node and the MatteControl node, dropping it
when the connection highlights.

Inserting a PlanarTransform node by holding down

the Shift key while dropping over a connection (left),
and after inserting the PlanarTransform node (right)

With the new Planar Transform node inserted, the Polygon automatically moves to match the motion of
the forehead that was tracked by the original PlanarTracker node, and it animates to follow along with
the movement of the shot. At this point, we’re finished!

The final painted image, along with the final node tree

Chapter 21 Paint 503

Chapter 22

Using the
Tracker Node
This chapter shows the many capabilities of the Tracker node in Fusion, starting with
how trackers can be connected in your node trees, and finishing with the different
tasks that can be performed.

Contents
Introduction to Tracking  505
Tracker Node Overview  506
Modes of the Tracker Node  506
Basic Tracker Node Operation  506
Connect to a Tracker’s Background Input  506
Analyze the Image to be Tracked  507
Apply the Tracking Data  508
Viewing Tracking Data in the Spline Editor  510
Tracker Inspector Controls  510
Motion Tracking Workflow In Depth  512
Connect the Image to Track  512
Add Trackers  512
Refine the Search Area  514
Perform the Track Analysis  515
Tips for Choosing a Good Pattern  516
Using the Pattern Flipbooks  517
Using Adaptive Pattern Tracking  517
Dealing with Obscured Patterns  518
Dealing with Patterns That Leave the Frame  518
Setting Up Tracker Offsets  519

Chapter 22 Using the Tracker Node 504

Stabilizing with the Tracker Node  520
Stabilization Using the Tracker Match Move Mode  520
Smoothing Motion  521
Using the Tracker Node for Match Moving  522
Simple Match Moving  522
Corner Positioning Operations  522
Perspective Positioning Operations  523
Connecting to Trackers’ Operations  523
Steady Position  523
Steady Angle  523
Offset Position  524
Unsteady Position  524
Steady Size  524
Using the Outputs of a Tracker  524
Using the Tracker as a Modifier  526
Match Moving Text Example  528
Adding a Layer to Match Move  528
Setting Up Motion Tracking  529
A Simple Tracking Workflow  530
Connecting Motion Track Data to Match Move  533
Offsetting the Position of a Match Moved Image  534

Introduction to Tracking
Tracking is one of the most useful and essential techniques available to a compositor. It can be
roughly defined as the creation of a motion path from analyzing a specific area in a clip over time.
Fusion includes a variety of different tracking nodes that let you analyze different kinds of motion.
Once you have tracked motion on a clip, you can then use the resulting data for stabilization, motion
smoothing, matching the motion of one object to that of another, and a host of other essential tasks.
Types of tracking nodes in Fusion:
– Tracker: Follows a relatively small, identifiable feature or pattern in a clip to derive a
2D motion path. This is sometimes referred to as point tracking.
– Planar Tracker: Follows a flat, unvarying surface area in a clip to derive a 2 ½D motion path
including perspective. A planar tracker is also more tolerant than a point tracker when some
tracked pixels move offscreen or become obscured.
– Camera Tracker: Tracks multiple points or patterns in a clip and performs a more sophisticated
analysis by comparing those moving patterns. The result is a precise recreation of the live-action
camera in virtual 3D space.

Each tracker type has its own chapter in this manual. This chapter covers the tracking techniques with
the Tracker node.

Chapter 22 Using the Tracker Node 505

Tracker Node Overview
The Tracker node is a single node that actually performs tracking, stabilizing, matching moving, and
corner-pinning operations. Since the Tracker node can transform the foreground input, it can be used
to generate tracks and then operate as a Merge in a match move or corner-pin setup. Or you can use it
to produce tracking data only and then publish that data to other nodes in the Node Editor.

Modes of the Tracker Node

The Tracker node is an incredibly flexible tool often used multiple times in a composite to help with
dozens of tasks. However, most of those tasks can be boiled down into just a few operations. The
Tracker node has four operation modes that cover the majority of tracking situations.

Stabilizing
You can use one or more tracked patterns to remove all the motion from the sequence or to smooth
out vibration and shakiness. When you use a single tracker pattern to stabilize, you stabilize only the
X and Y position. Using multiple patterns together, you are able to stabilize position, rotation,
and scaling.

Match Moving
The reverse of stabilizing is match moving, which detects position, rotation, and scaling in a clip using
one or more patterns. Instead of removing that motion, it is applied to another image so that the two
images can be composited together.

Corner Positioning
Corner positioning tracks four patterns that are then used to map the four corners of a new foreground
into the background. This technique is generally used to replace signs or mobile phone screens.
The Planar Tracker node is often a better first choice for these types of tracking tasks.

Perspective Positioning
Perspective positioning again tracks four patterns to identify the four corners of a rectangle.
Each corner is then mapped to a corner of the image, rescaling and warping the image to remove all
apparent perspective. The Planar Tracker node is often a better first choice for removing perspective
from a clip.

Basic Tracker Node Operation

All tracking workflows consist of three fundamental steps.
1 Attach an image you want to track to the yellow background input of the Tracker node.
2 Set the tracking pattern and analyze the clip to create a path.
3 Apply the tracking data to stabilize, match move, corner pin, or remove perspective.

Connect to a Tracker’s Background Input

You start by connecting the output of the image you want to track to a Tracker node’s background
input. The Tracker node analyzes the image that’s attached to its background input.
You can insert the Tracker node serially with other nodes if you intend to use the Tracker node itself to
do a simple stabilization operation or if you want to use it to perform the function of a Merge node in a
match move or corner-pin operation.

Chapter 22 Using the Tracker Node 506

Tracker node connected serially so it can both
track and transform the input images

However, if you’re just using a Tracker node to analyze data for use with multiple nodes elsewhere in
the comp, you could choose to branch it and leave its output disconnected to indicate that Tracker
node is a data repository. Please note that this is not necessary; serially connected Tracker nodes can
be linked to multiple other nodes as well.

Tracker connected as a branch to indicate it is

linked to other nodes and not used directly

Analyze the Image to be Tracked

After constructing the node tree and inserting the Tracker where you want, you can set up the tracker
in the viewer. You identify one or more features in the image that you wish to track (referred to as
patterns) by adding trackers (there’s one by default) and positioning them using the onscreen controls
in the viewer. After the Tracker node analyzes the clip, the resulting tracking data is stored within that
Tracker node. Keyframes, one per frame, indicate the Tracked Center X and Y data that has been
saved, while a motion path shows the path of tracked data in the viewer.

A motion path that indicates the tracked motion, and

tic marks that indicate tracking data keyframes

Chapter 22 Using the Tracker Node 507

Apply the Tracking Data
The resulting tracking data stored within the Tracker node is used to stabilize, match move, corner pin,
or remove perspective in one of two ways.

Method 1: Use the Tracker node to match move and merge

You can connect a foreground image to the Tracker node and apply the motion from the analyzed
background image.

Using a Tracker node in line for a match move

Setting the Operation parameter in the Operation tab in the Inspector to Match Move, Corner Position,
or Perspective Position always applies the motion to the foreground input (if one is connected). This is
an easy workflow for simple situations. In this scenario, you can use the Tracker node to replace a
Merge node since Tracker nodes include all the same functionality as a Merge.

Using a Tracker node to do a match move and merge, all in one

Method 2: Connect specific parameters to the Tracker Node

Alternatively, you can connect the tracking data from the Tracker node to the specific parameters of
other nodes that will actually do the work, for instances where setting up a match move isn’t just a
matter of transforming a foreground image. Each Tracker node and each pattern within the Tracker
node publishes its data for other nodes to use without directly linking to them in the node tree. For
example, in the following node tree, an Ellipse node is being used to isolate a glow effect for the ray
gun prop.

The Tracker set up as a branch and connected using the Connect To menu

Chapter 22 Using the Tracker Node 508

The ellipse needs to follow the motion of the ray gun, so a Tracker node is used to analyze the
movement of the gun tip so that tracking data can be used to animate the ellipse. The ellipse is not
connected to the tracker directly via the foreground input but indirectly through the Connect To
contextual menu.

Applying the light of a ray gun by connecting tracking data to the center position of an Ellipse node

This is made easier by renaming the Tracker you created to something descriptive of what’s
being tracked.

You can rename trackers in the Tracker List by double-

clicking them and typing descriptive text

Once the tip of the ray gun has been tracked, this tracking data is then connected to the Center
parameter of an Ellipse node that’s limiting a Glow effect by right-clicking the label of the Center
parameter in the Inspector, and choosing Tracker1 > Ray Gun Glow: Offset position from the Connect
to submenu of the contextual menu. All the data from every Tracker node in your node tree and every
tracking pattern appears within this submenu, and since we named the Tracker, it’s easy to find.
Choosing Offset position because it will place the center of the ellipse directly over the path. However,
it also gives us the flexibly to offset the ellipse if need be, using the offset controls in the Inspector.

Right-clicking a parameter’s label lets you connect tracking data to it to animate it

You can connect the data from a Tracker node to any other node’s parameter; however, you’ll most
typically connect track data to center, pivot, or corner X/Y style parameters. When you use tracking
data this way, it’s not necessary to connect the output of the Tracker node itself to anything else in
your node tree; the data is passed from the Tracker to the Center parameter by linking it with the
Connect To submenu.

Chapter 22 Using the Tracker Node 509

Viewing Tracking Data in the Spline Editor
Tracking data can be seen in the viewer as a path or as a displacement spline in the Spline Editor.
You can manipulate the tracking data in either place.
The Tracker uses a displacement spline by detail that indicates how far the tracking point is based on
the original location. It is great for modifying velocity, but it doesn’t tell you anything about direction.
If you need to nudge a few points in a certain direction, you can convert the displacement spline to an
X and Y coordinate spline.

Right-clicking a parameter’s label lets you connect tracking data to it to animate it

Right-click in the viewer to bring up a contextual menu. At the very bottom is a reference to the path
the Tracker created, called Tracker1Tracker1Path:Polyline. Choosing it calls up a longer submenu
where you can choose Convert to XY Path.
For more information on Displacement Splines, see Chapter 71, “Animating in Fusion’s Spline Editor” in
the DaVinci Resolve Reference Manual or Chapter 10 in the Fusion Reference Manual.

Tracker Inspector Controls

The layout of the Tracker node’s tabs in the Inspector reflects this workflow. It’s divided into three main
Tracker tabs, as well as the common Settings tab.
– The Tracker Control tab: This is where you create onscreen trackers with which to target patterns,
and where the controls appear that let you perform the required track analysis.

Chapter 22 Using the Tracker Node 510

 The Tracker Control tab

– The Operations tab: This is where you decide how the tracking data is used.

The Tracker Operations tab

– The Display Options tab: This is where you can customize how the onscreen controls
look in the viewer.

The Tracker Display Options tab

Chapter 22 Using the Tracker Node 511

Motion Tracking Workflow In Depth
Tracker nodes serve two purposes. They provide a method to analyze an object you want to follow,
and they serve as a container for the resulting track data. This allows you to use one node for analysis
and to pass on that analysis to any other node that requires it. Following is a more detailed breakdown
of the tracking process.

Connect the Image to Track

Regardless of whether you actually use the Tracker node itself to do anything with the tracking data,
the image you want to track must be connected to the background input (yellow) of a Tracker node for
there to be a successful analysis. While the Tracker node has a foreground input, it is initially ignored
for purposes of tracking analysis, so even if you connect a foreground, the background is the only
input used during the analysis process.

Connecting a MediaIn node’s output to a Tracker node’s background input

Add Trackers
Although each Tracker node starts with a single tracker pattern, a single node is capable of analyzing
multiple tracking patterns that have been added to the Tracker List, enabling you to track multiple
features of an image all at once for later use and to enable different kinds of transforms. Additional
trackers can be added by clicking the Add button immediately above the Tracker List control.

To add an additional tracker, click the Add button in the Inspector

Multiple patterns are useful when stabilizing, match moving, or removing perspective from a clip. They
also help to keep the Node Editor from becoming cluttered by collecting into a single node what
would otherwise require several nodes.

Working in the Tracker list:

– To select a tracker: Click the name of the Tracker you want to select.
– To rename a tracker: You can rename trackers to make it easier to reference them later. For
example, if you’re tracking a car door handle, you can name the Tracker “Car Handle” so it’s easy
to find later. To do so, just double-click the default name of the Tracker in the Tracker list, type a
new one, and press Return.

Chapter 22 Using the Tracker Node 512

 – To delete a tracker: Select a tracker and click the Delete button.
– To disable, suspend, or re-enable trackers: Click the checkbox to the left of each tracker’s
name in the Tracker List. It’s a three-way toggle that switches between Enabled, Suspended, and
Disabled.
– Enabled: An Enabled tracker will re-track its pattern every time the track is performed. Its path
data is available for use by other nodes, and the data is available for stabilization and corner
positioning.
– Suspended: When the checkbox appears gray, it is Suspended. In this state, it does not re-track
its pattern when the track is performed. The data is locked to prevent additional changes, but
the data from the path is still available for other nodes. The data is also available for advanced
tracking modes like stabilization and corner positioning.
– Disabled: A Disabled tracker does not create a path when tracking is performed. Its data is
not available to other nodes or for advanced tracking operations like stabilization and corner
positioning.

Position the Pattern Boxes

A pattern is the region of pixels that are targeted for tracking within an image using a pattern box. The
pattern box is defined in the viewer by a rectangle when a tracker is active. A single Tracker node can
have many pattern boxes, each targeting different patterns. Each tracked pattern will produce its own
motion path.
Clicking any part of a tracker’s onscreen controls will select it. Selected pattern boxes are red, while
deselected pattern boxes are green.

A pattern box positioned over an eye you want to track

When you add a Tracker node to the Node Editor, you start with one pattern box displayed in the
viewer as a small rectangle. When the cursor is placed over the pattern rectangle, the control
expands and two rectangles appear. The outer rectangle has a dashed line, and the inner
rectangle has a solid line. The outer rectangle is the search area, and the inner rectangle is
the pattern.
If you need to select a new pattern, you can move the pattern box by dragging the small (and
easily missed) handle at the top left of the inner pattern box.

To move a pattern in the viewer, drag it from the upper-left corner

Chapter 22 Using the Tracker Node 513

 While moving the pattern box, an overlay pop-up appears, showing a zoomed version of the pixels
contained within the rectangle to help you precisely position the pattern via the crosshairs within.

A zoomed view appears while you drag

a pattern box to help you position it

The pattern rectangle can also be resized by dragging on the edges of the rectangle. You want to
size the pattern box so that it fits the detail you want to track, and excludes area that doesn’t
matter. Ideally, you want to make sure that every pixel of the pattern you’re tracking is on the same
plane, and that no part of the pattern is actually an occluding edge that’s in front of what you’re
really tracking. When you resize the pattern box, it resizes from the center, so one drag lets you
create any rectangle you need.

Resizing a pattern box to fit the tracking point on the ray gun

TIP: The magnified pattern box does not take viewer LUTs into account. When using Log
content, it may make it easier to position the tracker if you temporarily insert a Brightness
Contrast node between the source content and the yellow input of the tracker. You can use
the Brightness Contrast node to temporarily increase the visibility of the region you
are tracking.

Refine the Search Area

A second rectangle with a dotted border surrounds the pattern box. This is the search area. When
progressing from one frame to another while tracking, the Tracker analyzes the region defined by the
search area, which surrounds the last known tracker position in an attempt to relocate the pattern. The
larger the search area, the better chance you have of successfully tracking fast moving objects, but
the longer the track will take. However, there are some ways to optimize tracking for specific content.
For example, tracking a pattern that is moving quickly across the screen from left to right requires a
wide search area but does not require a very tall one, since all movement is horizontal. If the search
area is smaller than the movement of the pattern from one frame to the next, the Tracker will likely fail
and start tracking the wrong pixels, so it’s important to take the speed and direction of the motion into
consideration when setting the search area.

Chapter 22 Using the Tracker Node 514

 You can resize the search area by dragging
the edges of the dotted outline

Perform the Track Analysis

Before you begin analyzing, you’ll need to make sure you’ve set a render range in the Time Ruler that
corresponds to the range of frames during which the pattern is visible. This may be an entire clip or
only a small portion of that clip. Depending on the type of motion you’re tracking, you may want to use
the Adaptive Mode option to aid the analysis (see below for more details).
Once your options are set, you can use any of the tracking transport buttons at the top of the Inspector
to start tracking. Once tracking has started, you cannot work in the Node Editor until it has completed.

The tracking transport buttons and analysis parameters

To begin tracking, do one of the following:

– Click the Track Reverse button to track from the very end of the render range.
– Click the Track Backward from Current Frame button to track backward from the
current playhead position.
– Click the Track Forward button to track from the very start of the render range.
– Click the Track Forward from Current Frame button to track forward from the current
playhead position.
Pattern tracking will stop automatically when it reaches the end of the render range (or the start when
tracking backward), but you can also interrupt it and stop tracking at any time.

To stop tracking, do one of the following:

– Click the Stop Tracking button in the tracker transports.
– Click Stop Render at the bottom of the Fusion window.
– Press the Escape key.
When tracking is complete, the path will be connected to the pattern. The path from that pattern
can now be connected to another node or used for more advanced operations like stabilization
and corner positioning.

Chapter 22 Using the Tracker Node 515

Once the track is complete, assuming it’s good, you can use the various techniques in this chapter to
use the track in your composition.

Tips for Choosing a Good Pattern

The Tracker works by searching each frame for the pixels contained in the pattern. In order for a track
to be successful, a fairly high contrast and unique region of the image must be located in the footage.
This process is known as pattern selection.
The first step in pattern selection is to review the footage to be tracked several times. Watch for
candidate patterns that are visible through the entire range of frames, where the contrast is high and
the shape of the pattern does not change over time. The more unique the pattern, the more likely the
track is to be successful.
In addition to locating high contrast, defined patterns, watch for the frames where the pattern moves
the most. Identifying the maximum range of a pattern’s motion will help to determine the correct size
for the pattern search area.
It is not uncommon to have a scene that requires the use of several different patterns to generate a
single path. This most often occurs because the pattern moves out of frame or is temporarily obscured
by another scene element. Combining patterns into a single pattern is described later in the chapter.

Selecting the Pattern’s Image Channels

When a pattern of pixels is selected, the Tracker automatically selects the color channel used for
tracking the pattern based on an analysis of each channel for contrast, clarity, and reliability. The
channels selected are highlighted in the bars to the right of the Pattern display window in the
node controls.

Highlighted channel bars indicate which

channel is selected for tracking

You can override the automatic channel selection by clicking the buttons beneath the bars for each
channel to determine the channel used for tracking.
You can choose any one of the color channels, the luminance channels, or the alpha channel
to track a pattern.
When choosing a channel, the goal is to choose the cleanest, highest contrast channel for use in the
track. Channels that contain large amounts of grain or noise should be avoided. Bright objects against
dark backgrounds often track best using the luminance channel.

Selecting Patterns for Stabilization

Selecting patterns for stabilization can be a tricky business. The location of the pattern, when it is
selected, is used to determine precisely how the image will be stabilized. At least two patterns are
required to correct for rotation; using three patterns will correct for scaling, and more will usually
improve the quality of the solution.
Try not to select just any potentially valid pattern in the sequence, as some patterns will make the
solution worse rather than better. To help with your selection, use the following guidelines when
selecting patterns for stabilization.

Chapter 22 Using the Tracker Node 516

 – Locate patterns at the same relative depth in the image. Objects further in the background will
move in greater amounts compared to objects in the foreground due to perspective distortion.
This can confuse the stabilization calculations, which do not compensate for depth.
– Locate patterns that are fixed in position relative to each other. Patterns should not be capable of
moving with reference to each other. The four corners of a sign would be excellent candidates,
while the faces of two different people in the scene would be extremely poor choices for patterns.

Using the Pattern Flipbooks

Each pattern has a pair of thumbnail windows shown in the Inspector. The left window shows the
selected pattern, while the right window is updated during the track to show the actual pattern that has
been acquired for each frame.

The Tracker Pattern Selection and Flipbook thumbnails

Each pattern that’s stored is added to a Flipbook. Once the render is complete, you can play this
Pattern Flipbook to help you evaluate the accuracy of the tracked path. If you notice any jumps in the
frames, then you know something probably went wrong.

Using Adaptive Pattern Tracking

Even the most ideal pattern will usually undergo shifts in profile, lighting conditions, and other
variables. These can adversely affect pattern recognition to the point that a pattern becomes
unusable. The Tracker offers three modes of pattern acquisition during tracking that can help to
correct these conditions. The modes can be set using the Adaptive Mode options in the Inspector.

The Adaptive Mode options

None
When the Adaptive mode is set to None, the pattern within the rectangle is acquired when the pattern
is selected, and that becomes the only pattern used during the track.

Every Frame
When Every Frame is chosen, the pattern within the rectangle is acquired when the pattern is
selected, and then reacquired at each frame. The pattern found at frame 1 is used in the search on
frame 2, the pattern found on frame 2 is used to search frame 3, and so on. This method helps the
Tracker adapt to changing conditions in the pattern.
Every Frame tracking is slower and can be prone to drifting from sub-pixel shifts in the pattern from
frame to frame. Its use is therefore not recommended unless other methods fail.

Chapter 22 Using the Tracker Node 517

Best Match Tracking
Best Match tracking works in much the same way as Every Frame tracking; however, it will not
reacquire the pattern if the difference between the original pattern and the new one is too great. This
helps to prevent cases where transient changes in the image cause the Tracker to become confused.
As a comparison between the two Adaptive modes, if a shadow passes over the tracker point, the
Every Frame tracking mode may start tracking the shadow instead of the desired pattern. The Best
Match mode would detect that the change from the previous frame’s pattern was too extreme and
would not grab a new pattern from that frame.
The Adaptive mode is applied to all active patterns while tracking. If you only want some patterns to
use the Adaptive mode, disable all other patterns in the list before tracking.

Dealing with Obscured Patterns

Often, an otherwise ideal pattern can be temporarily obscured (occluded) or blocked from tracking—
for example, when tracking a car that passes behind a telephone pole.
In these situations, you divide the render range up into two ranges, the range before the pattern is
obscured and the range after the pattern becomes visible again. After tracking the two ranges
individually, the Tracker will automatically interpolate between the end of the first range and the start
of the second.
If you need to edit the resulting motion path to account for any non-linear motion that takes place
between the two tracked ranges, you can select the track path to expose a Node toolbar with controls
for adjusting the control points on this path. For example, you can choose Insert and Modify mode to
insert points in the non-tracked range to compensate for any nonlinear motion in the tracked pattern.

Tools for modifying tracker paths in the Node toolbar of the viewer

Dealing with Patterns That Leave the Frame

There are two options when a tracker leaves the frame. If the pattern re-enters the frame, you can treat
it like an obscured pattern. If the pattern does not re-enter the frame, or it is undesirable to hand track
portions of the movement, you can use the Track Center (Append) mode to select a new pattern.

The Track Center (Append) mode pop-up menu

The Track Center (Append) mode selects a new pattern that will continue to add keyframes to the
existing path. The offset between the old pattern and the new pattern is automatically calculated to
create one continuous path.

Chapter 22 Using the Tracker Node 518

To use the Track Center (Append) mode, do the following:
1 When the pattern has become untrackable for some reason, stop analysis and move the playhead
to the last frame that tracked successfully.
2 Choose Track Center (Append) from the Path Center pop-up menu in the Inspector.
3 Now, drag the Pattern selector to a new pattern that can be tracked from that point onward.
4 Restart tracking from the current frame.

When selecting a pattern to use in appending to an existing path, a pattern that is close to the old
pattern and at the same apparent depth in the frame generates the best results. The further away the
new pattern is, the more likely it is that the difference in perspective and axial rotation will reduce
accuracy of the tracked result.

Setting Up Tracker Offsets

Often, it’s impossible to track the thing you want to apply an effect to. For example, the only pattern
available for an accurate track is a button on an actor’s sleeve. However, the effect requires the
person’s hand to be glowing. To cause the glow’s effect mask to be centered on the actor’s hand, it’s
necessary to use the Tracker Offset control.

The Tracker Offset controls in the Inspector

The X and Y Offset controls allow for constant or animated positional offsets to be created relative to
the actual Tracker’s pattern center. The position of the offset in the viewer will be shown by a dashed
line running from the pattern center to the offset position. You can also adjust the offset in the viewer
using the Tracker Offset button. Clicking the button enables you to reposition the path while keeping
the Tracker pattern in place.

The Tracker Offset tool in the Node toolbar of the viewer; a track of the
orange dot is being offset to the center of the ray gun

Once an offset for a pattern is set, you can connect other positional controls to the Tracker’s Offset
menu using the Connect To > Tracker: Offset Position option in the control’s contextual menu. The
path created during the track remains fixed to the center of the pattern.

Chapter 22 Using the Tracker Node 519

Stabilizing with the Tracker Node
When a Tracker node is set to Match Move in the Operations tab, it is capable of a variety of functions.
Applying the motion from the background clip to the foreground clip is the obvious functionality.
However, the Match Move operation is also used for stabilizing footage to to either completely remove
motion from the scene or smooth existing motion.
Here are some common scenarios for stabilization that are handled when the Tracker is set to
Match Move.
– A sequence that should be steady has vibrations or undesirable movement.
– A sequence that requires a smooth camera move suffers from jarring.

The Tracker Operation tab Match Move

set to BG Only for stabilization

Stabilization Using the Tracker Match Move Mode

Stabilizing motion completely removes the appearance of motion from the image. The motion from
frame to frame is calculated, and the contents of the frame are transformed to return the image to a
reference position. This position can be either the start or end of the sequence or a manually selected
frame from the sequence.
Stabilization can correct for position with as little as one pattern. Two or more patterns are required to
correct for rotation or scaling within the image.
When the Operation menu is set to Match Move, choosing BG only from the Merge operation menu
stabilizes the background (yellow input) clip. Only the controls that are applicable for stabilization
operations will appear in the Operation tab.
Several of the stabilization controls are always available, collected under the Match Move Settings
disclosure button. These controls are available at all times because the Steady and Unsteady positions
of a tracker are always published. This makes them available for connection by other controls, even
when the Tracker’s operation is not set to match moving.

The Match Move settings

Chapter 22 Using the Tracker Node 520

Merge
The Merge menu determines to which input connection the Tracking data is applied. When stabilizing
an image to remove all motion, or smooth the motion, the Merge button must be set to BG Only.

Edges
The Edges menu determines whether the edges of an image that leave the visible frame are cropped,
duplicated, or wrapped when the stabilization is applied. Wrapping edges is often desirable for some
methods of match moving, although rarely when stabilizing the image for any other purpose. For more
information on the controls, see Chapter 118, “Tracker Nodes” in the DaVinci Resolve Reference
Manual or Chapter 57 in the Fusion Reference Manual.

Position/Rotation/Scaling
Use the Position, Rotation, and Scaling checkboxes to select what aspects of the motion are corrected.

Match Move Settings

Options for the Match Move settings include Pivot and Reference.

Pivot Type
The Pivot Type for the stabilization is used to calculate the axis of rotation and scaling calculations.
This is usually the average of the combined pattern centers but may be changed to the position of
a single tracker or a manually selected position.

Reference
The Reference controls establish whether the image is stabilized to the first frame in the
sequence, the last frame, or to a manually selected frame. Any deviation from this reference by the
tracked patterns is transformed back to this ideal frame.
As a general rule, when tracking to remove all motion from a clip, set the Merge mode to BG Only,
the Pivot Type to Tracker Average or Selected Tracker, and the Reference control to Start, End, or
Select Time.

Smoothing Motion
When confronted with an image sequence with erratic or jerky camera motion, instead of trying to
remove all movement from the shot, you often need to preserve the original camera movement while
losing the erratic motion.
The Start & End reference option is designed for this technique. Instead of stabilizing to a reference
frame, the tracked path is simplified. The position of each pattern is evaluated from the start of the
path and the end of the path along with intervening points. The result is smooth motion that replaces
the existing unsteady move.

The Reference Intermediate Points slider is displayed when

Start & End is selected to enable the smoothing of motion

To preserve some of the curvature of the original camera motion, you can increase the value of the
Reference Intermediate Points slider that appears when the Start & End reference mode is selected.
When tracking to create smooth camera motion, ensure that the Start & End reference mode is
enabled and set the Merge mode to BG Only. It is recommended to leave the Pivot Type control set to
Tracker Average.

Chapter 22 Using the Tracker Node 521

Using the Tracker Node for Match Moving
A simple match moving example is shown at the beginning of this chapter, but this section presents
additional details that you may not have been aware of. Examples of match moving include:
– A static CG element must be believably added to a moving sequence.
– Two sequences with different motions must be composited together.

Some clips may need to be stabilized so that an element from another source can be added to the
shot. After the element or effect has been composited, the stabilization should be removed to make
the shot look natural again.

Simple Match Moving

Match moving essentially applies the movement from the tracked clip to another clip. There are two
ways to perform match moving. One method involves connecting other nodes, such as Transform or
Merge, to a Tracker’s outputs. The other method is to stabilize an image by trying to remove all motion,
but instead of setting the Merge menu to BG Only, set it to FG Over BG, FG Only, or in rare occasions,
BG Over FG.

Set the Merge menu to BG Only, FG Over BG, or BG Over FG

When using this Merge menu, you connect a foreground image to the Tracker node’s input
connection in the Node Editor.

Connect a foreground image to the Tracker’s foreground input

Enabling the FG Only mode will apply the motion from the background to the foreground, and the
Tracker will only output the modified FG image. This result can later be merged over the original,
allowing further modifications of the foreground to be applied using other nodes before merging the
result over the background clip.

Corner Positioning Operations

The Corner Positioning operation maps the four corners of a foreground image to four patterns within
the Tracker. This operation, or technique, is most commonly used for sign replacements.
The Corner Positioning operation of the Tracker requires the presence of a minimum of four patterns. If
this operation mode is selected and there are not four patterns set up in the Tracker already, additional
patterns will automatically be added to bring the total up to four.
When this mode is enabled, a set of drop-down boxes will appear to select which tracker relates to
each corner of the rectangle. It has no effect when the Merge control option is set to BG Only.

Chapter 22 Using the Tracker Node 522

Perspective Positioning Operations
The Perspective Positioning operation is used to remove perspective from a foreground image or
apply the perspective from one sequence to another. This can be useful when you need to paint out
an area that is distorted by perspective. Removing the perspective flattens the images for painting,
and then another tracker adds the perspective back.
The Perspective Positioning operation of the Tracker requires the presence of a minimum of four
patterns. If this operation mode is selected and there are not four patterns set up in the Tracker
already, additional patterns will automatically be added to bring the total up to four.
When this mode is enabled, a set of drop-down boxes will appear to select which tracker relates to
each corner of the rectangle. It has no effect when the Merge control option is set to BG Only.

Connecting to Trackers’ Operations

One of the most common applications for a tracked pattern is using the tracked position or path to
drive the position of another node’s parameters. For example, tracking an eye in order to color correct
the eye to blue using an effect mask. You start off by tracking the eye, and then create a color
corrector with the desired settings. You create a mask in the shape of the eye and connect the
Tracker’s position to the Center of the mask.
In addition to the path (called Offset Position), each pattern in a tracker publishes four other values for
use as connections that are available to other nodes in the Node Editor.
You connect a node’s position parameters to a tracker by selecting the connection type from the
controls contextual menu (for example, Transform 1: Center > Connect To > Tracker 1 > Offset Position).
There are five connection types automatically published by the tracker to connect to a position
parameter in another node.

Steady Position
Steady Position can be used to stabilize footage in both X and/or Y to remove camera shake and other
unwanted movement. The connection inverts the output of the tracked pattern’s motion. When you
connect a Center parameter to the Steady Position of the Tracker, it will be placed at 0.5/0.5 (the
center of the screen) by default at frame 1. You can change this using the Reference mode in the
Tracker’s Operation tab.

Steady Angle
The Steady Angle mode can be used to stabilize footage in both X and/or Y to remove camera shake
and other unwanted movement. When you connect a control, for example the Angle of a Transform, to
the Steady Angle of the Tracker, it will be placed at 0 degrees by default at frame 1. This can be
changed by means of the Reference mode in the Tracker’s Operation tab. From there on, the resulting
motion of the Steady Angle mode will rotate into the opposite direction of the original motion.
So if the angle at frame 10 is 15 degrees, the result of the Steady Angle will be -15 degrees.
To use Steady Angle, you need at least two tracked patterns in your tracker. With just one point, you
can only apply (Un)Steady Position.

Chapter 22 Using the Tracker Node 523

Offset Position
An Offset Position is essentially the path generated by the tracker. It is the one you select when you
want an object to follow the path. It is available for each single tracker in the Tracker node and refers to
that single tracker only. When you connect the Center X and Y parameters to the offset position of the
Tracker, the node’s center will follow exactly the path of that tracker. Connecting to single trackers is
always useful when you want to match elements with object motion in your footage. For example, you
could track a hand of your actor and attach a ball to the Tracker‘s offset position, so that the ball
follows the exact motion of the hand. Or you could track an element that needs rotoscoping and attach
the mask’s center to the Tracker’s offset position.

Unsteady Position
After using the Steady Position, the Unsteady Position is used to reintroduce the original movement on
an image after an effect or new layer has been added. The resulting motion from Unsteady Position is
basically an offset in the same direction as the original motion.

Steady Size
The Steady Size connection outputs the inverse of the tracked pattern’s scale. When you connect a
parameter, for example the Size of a Transform, to the Steady Size of the Tracker, it will be placed with
a Size of 1 (i.e., the original size) by default at frame 1. This can be changed by means of the Reference
mode in the Tracker’s Operation tab. The resulting size of the Steady Size mode will then counteract
the size changes of the original motion. So if the actual size at frame 10 is 1.15, the result of the Steady
Size will be 1 - (1.15 - 1) = 0.85.
To use Steady Size, you need at least two tracked patterns in your tracker. With just one point, you can
only apply (Un)Steady Position.

Using the Outputs of a Tracker

The tracker outputs described above are published by each tracker pattern created in the Tracker
node. Each Tracker node itself also publishes a Steady Position, Angle, Size and an Unsteady Position.
The values of these Tracker node outputs are calculated using all the patterns in that tracker, as
configured by the Match Move Settings controls in the Tracker’s Operation tab.

The Match Move settings determine the reference frame for

stabilization and the point around which stabilization pivots

Rather than using the Tracker node to perform the Merge operation, an alternative and common way
to use these published outputs is to create a match move by connecting the outputs to multiple nodes.
A tracker is used to track a pattern, and then that data can be connected to multiple other nodes using
the Connect To submenu.

Chapter 22 Using the Tracker Node 524

As an example, to use the Connect To menu to perform a match move, do the following:
1 Track the background clip using at least two tracking patterns in the tracker.
2 In a different branch, add a Transform node to the background clip.

The background branched to go to a Tracker and a Transform

3 Right-click over the Transform’s Center and choose Connect to > Tracker1 > Steady Position.

The tracker publishes its output for other nodes to connect to, as done here to stabilize the clip

4 Connect the foreground to a corner-positioned node, so you can position the corners of the
foreground appropriately over the background.
5 Add another Transform node to the Node Editor after the Merge.

A second Transform after the Merge is used to add back in the original motion with Unsteady Poisition

6 Connect the new Transform’s Center to the Tracker’s Unsteady Position. The image will be
restored to its original state with the additional effect included.

Chapter 22 Using the Tracker Node 525

To better understand how this works, imagine a pattern that is selected at frame 1, at position 0.5, 0.5.
The pattern does not move on frame 2, so its position is still 0.5, 0.5. On the third frame, it moves
10 percent of the image’s width to the right. Now its position is 0.6, 0.5.
If a transform center is connected to the Steady Position output of the Tracker, the Transform node’s
center is 0.5, 0.5 on the first and second frames because there has been no change. On frame 3, the
center moves to 0.4, 0.5. This is the inverse of the horizontal motion that was tracked in the pattern,
moving the image slightly to the right by 10 percent of the image width to counteract the movement
and return the pattern of pixels back to where they were found.

Using the Tracker as a Modifier

Another technique for adding a tracker directly to a control is to add it as a modifier. Choosing the
Tracker from the Modify With contextual menu does not use a Tracker node; it adds a modifier in the
Inspector with a set of parameters almost identical to those found in the Tracker node itself. The
benefit here is that the object that you want to follow the tracked path is automatically connected to
the tracker modifier when you apply it.

Applying the tracker as a modifier

The differences between a Tracker modifier and a Tracker node are as follows:
– The Tracker modifier can only track a single pattern.
– A source image must be set for the Tracker modifier.

The Tracker modifier can only output a single value and cannot be used for complex stabilization
procedures, but it is a nice quick way to apply a tracker to a point that you need to follow.

As an example, to apply the Tracker as a modifier, do the following:

Imagine that you needed to track an actor’s eyes so that an unearthly, alien glow could be applied
to the eyes.
1 Add an ellipse mask node to cover an actor’s eye.
2 In the Inspector, right-click on the mask’s Center parameter and from the contextual menu choose
Ellipse1 Center > Modify With > Tracker Position.

Chapter 22 Using the Tracker Node 526

 Right-click over any Center Coordinate Control and choose
Modify With > Tracker Position to add a Tracker modifier

3 Click the Modifiers tab in the Inspector and drag the MediaIn1 node that you want to track into the
Tracker Source field.

The Modifiers tab includes the tracking controls and a Tracker

Source field to indicate which node to use for tracking

4 Click the Track Forward button to begin tracking the person’s eye.
5 Insert a Soft Glow node directly after the MediaIn and connect the Ellipse Mask to the white
Glow Mask input.

Chapter 22 Using the Tracker Node 527

 A Tracker modifier applied to the Ellipse to create a green glow on an actor’s pupil

You can set a different source image for the Tracker modifier by typing in the name of the node or
dragging and dropping the node from the Node Editor into the Tracker Source field control. If you have
a node (let’s call it node#1) connected to the node that contains the modifier (let’s call it node#2), the
source image for the Tracker modifier will automatically the node #1
For more information on the Tracking parameters, see Chapter 118, “Tracker Nodes” in the
DaVinci Resolve Reference Manual or Chapter 57 in the Fusion Reference Manual.

Match Moving Text Example

This example takes you through a complete motion tracking task, and shows how you can create a
very simple match-moving effect using the Tracker node, which is the Swiss army knife of trackers
in Fusion.

Adding a Layer to Match Move

In the example composition, we have a Text1 node that’s creating a “Switzerland” title that’s
composited over a drone shot flying over and around a mountain bridge. With the Text1 node selected,
the onscreen controls that let you position the text it’s generating are visible in the viewer, and the text
is positioned where we’d like it to start. Note that, with the Text node selected, even the part of the
text that’s offscreen can still be seen as an outline showing us where it is.

Chapter 22 Using the Tracker Node 528

 Some text superimposed against a background, ready to track

Our goal for this composition is to motion track the background image so that the text moves along
with the scene as the camera flies along.

Setting Up Motion Tracking

To set up for the motion track, we’ll begin by creating a disconnected Tracker node, using another
method other than those seen previously. Right-click anywhere in the background of the Node Editor
(preferably where you want the new node to appear), and choose Add Tool > Tracking > Tracker from
the contextual menu to create a new Tracker1 node underneath the MediaIn node (or Loader node if
you are using Fusion Studio).

Creating a new node using the Node Editor contextual menu

Chapter 22 Using the Tracker Node 529

Next, we’ll drag a connection from the MediaIn1 node to the Tracker1 node to automatically connect
the source clip to the Tracker1 background input. This branches the output from the MediaIn1 node to
the Tracker node so that the Tracker1 node processes the image separately from the rest of the node
tree. This is not required, but it’s a nice organizational way to see that the Tracker node is doing an
analysis that must be referred to in a way other than a “physical” connection.

Branching a Tracker node to use to analyze an image

A Simple Tracking Workflow

The Tracker node is the simplest tracking operation the Fusion page has, and while there are several
ways of using it, an extremely common workflow is to use the Tracker node controls to analyze the
motion of a subject in the frame with motion you want to follow, and then use the resulting motion path
data by “connecting” it to the Center parameter of another node that’s capable of transforming the
image you want to match move.

Positioning the Tracker Onscreen Control

When the Tracker node is selected, a single green box appears in the viewer, which is the default
onscreen control for the first default tracker that node contains (seen in the Tracker List of the
Inspector controls). Keep in mind that you only see onscreen controls for nodes that are selected, so if
you don’t see the onscreen tracker controls, you know you need to select the tracker you want to work
with. Loading the tracker you want to work with into the viewer is also the safest way to make sure
you’re positioning the controls correctly relative to the actual image that you’re tracking.
If you position your pointer over this box, the entire onscreen control for that tracker appears, and if
you click the onscreen control to select that tracker, it turns red. As with so many other tracker
interfaces you’ve likely used, this consists of two boxes with various handles for moving and
resizing them:
– The inner box is the “pattern box,” which identifies the “pattern” in the image you’re tracking and
want to follow the motion of. The pattern box has a tiny handle at its upper-left corner that you use
to drag the box to overlap whatever you want to track. You can also resize this box by dragging
any corner, or you can squish or stretch the box by dragging any edge to make the box better fit
the size of the pattern you’re trying to track. The center position of the tracker is indicated via X
and Y coordinates.
– The outer box is the “search box,” which identifies how much of the image the tracker needs to
analyze to follow the motion of the pattern. If you have a slow-moving image, then the default
search box size is probably fine. However, if you have a fast-moving image, you may need to
resize the search box (using the same kind of corner and side handles) to search a larger area,
at the expense of a longer analysis. The name of that tracker is shown at the bottom right of the
search box.

The onscreen controls of a selected tracker seen in isolation

Chapter 22 Using the Tracker Node 530

It’s worth saying a second time that the handle for moving a tracker’s onscreen control is a tiny dot at
the upper-left corner of the inner pattern box. It’s really easy to miss if you’re new to Fusion. You must
click on this dot to drag the tracker around.

The handle for dragging the tracker

boxes to move them around

In this example, we’ll drag the onscreen control so the pattern box overlaps a section of the bridge
right over the leftmost support. As we drag the onscreen control, we see a zoomed-in representation
of the part of the image we’re dragging over to help us position the tracker with greater precision.
For this example, the default sizes of the pattern and search box are fine as is.

The zoomed-in preview that helps you

position the pattern box as you drag it

Using the Tracker’s Inspector Controls to Perform the Analysis

At this point, let’s look at the Tracker node’s controls in the Inspector. There are a lot of controls, but for
this simple example we only care about the main Tracker panel, with the tracking analysis buttons at
the top, the tracking options below those, and the Tracker List underneath those. The Tracker List also
has buttons for adding and deleting trackers; you have the option of adding multiple trackers that can
be analyzed all at once for different workflows, but we don’t need that for now.

Tracker Inspector controls, with the tracking

analysis buttons at top, the tracker options in
the middle, and the Tracker List below

Chapter 22 Using the Tracker Node 531

Additional controls over each tracker and the image channels being analyzed appear at the bottom,
along with offset controls for each tracker, but we don’t need those now (at least not yet).
Again, this track is so simple that we don’t need to change the default behaviors that much, but
because the drone is flying in a circular pattern, the shape of the pattern area is changing as the clip
plays. Fortunately, we can choose Every Frame from the Adaptive Mode menu to instruct the tracker to
update the pattern being matched at every frame of the analysis, to account for this.

Changing the Adaptive Mode of the Tracker node to Every

Frame to account for the camera’s shift of perspective

Now, we just need to use the tracker analysis buttons at the top to begin the analysis. These buttons
work like transport controls, letting you start and stop analysis as necessary to deal with problem
tracks in various ways. Keep in mind that the first and last buttons, Track from Last Frame and Track
from First Frame, always begin a track at the last or first frame of the composition, regardless of the
playhead’s current position, so make sure you’ve placed your tracker onscreen controls appropriately
at the last or first frame.

The analysis buttons, left to right: Track from Last Frame, Track Backward,
Stop Tracking, Track Forward, Track from First Frame

For now, clicking the Track from Beginning button will analyze the entire range of this clip, from the first
frame to the last. A dialog lets you know when the analysis is completed, and clicking the OK button
dismisses it so you can see the nice clean motion path that results.

The analyzed motion path resulting from tracking a

section of the bridge as the camera flies past

Viewing Motion Track Data in the Spline Editor

This is not a necessary part of the tracking workflow, but if you have an otherwise nice track with a few
bumps in it, you can view the motion tracking data in the Spline Editor by viewing that tracker’s
Displacement parameter curve. This curve is editable, so you can massage your tracking data in a
variety of ways, if necessary.

Chapter 22 Using the Tracker Node 532

 Viewing motion tracking analysis data in the Spline Editor

Connecting Motion Track Data to Match Move

Now that we have a successful analysis, it’s time to use it to create the Match Move effect. To make
this process easier, we’ll double-click the tracker’s name in the Tracker List of the Inspector, and enter
a new name that’s easier to keep track of (heh). Adding your own names make that tracker easier to
find in subsequent contextual menus and lets you keep track of which trackers are following which
subjects as you work on increasingly complex compositions.

Renaming a tracker to make it easier to find

Now it’s time to connect the track we’ve just made to the text in order to start it in motion. After loading
the Merge1 node into the viewer to see the text in context with the overall composite we’re creating,
we’ll select the Text1 node to open its parameters in the Inspector, and click the Layout panel icon
(second button from the left) to expose the Layout controls, which are the text-specific transform
controls used to position the text object in the frame. These are the controls that are manipulated
when you use the Text node onscreen controls for repositioning or rotating text.

The Layout controls for a Text node, in the Layout panel

The Center X and Y parameters, while individually adjustable, also function as a single target for
purposes of connecting to tracking to quickly set up match moving animation. You set this up via the
contextual menu that appears when you right-click any parameter in the Inspector, which contains a
variety of commands for adding keyframing, modifiers, expressions, and other automated methods of
animation including connecting to motion tracking.

Chapter 22 Using the Tracker Node 533

If we right-click anywhere on the line of controls for Center X and Y, we can choose Connect To >
Tracker1 > Bridge Track: Offset position from the contextual menu, which connects this parameter to
the tracking data we analyzed earlier.

Connecting the Center X and Y parameter to the Bridge Track: Offset position motion path we analyzed

Immediately, the text moves so that the center position coincides with the center of the tracked motion
path at that frame. This lets us know the center of the text is being match moved to the motion
track path.

The text now aligns with the motion track coordinate

Offsetting the Position of a Match Moved Image

In fact, we want to offset the match-moved text, so it’s higher up in the frame. To do this, we select the
Tracker1 node again and use the Y Offset 1 dial control to move the text up, since now any changes we
make to the Bridge Track dataset now apply to the center of the text that’s connected to it.

Using the X and Y Offset controls in the Tracker1 node to

offset the text layer’s position from the tracked motion path

The offset we create is shown as a dotted red line that lets us see the actual offset being created by
the X and Y Offset controls. In fact, this is why we connected to the Bridge Track: Offset position
option earlier.

Chapter 22 Using the Tracker Node 534

The text offset from the tracked motion path; the offset can be seen as a dotted red line in the viewer

Now, if we play through this clip, we can see the text moving along with the bridge.

Two frames of the text being match moved to follow the bridge in the shot

Chapter 22 Using the Tracker Node 535

Chapter 23

Planar Tracking
This chapter provides an overview of how to use the Planar Tracker node, and how to
use it to make match moves simple. For more information about the Planar Tracker
node, see Chapter 118, “Tracker Nodes” in the DaVinci Resolve Reference Manual or
Chapter 57 in the Fusion Reference Manual.

Contents
Introduction to Tracking  537
Using the Planar Tracker  537
Different Ways of Using the Planar Tracker Node  537
Setting Up to Use the Planar Tracker  538
Check for Lens Distortion  538
A Basic Planar Tracker Match Move Workflow  538
Tips for Choosing Good Planes to Track  541

Chapter 23 Planar Tracking 536

Introduction to Tracking
Fusion includes three different Tracking nodes that let you analyze different kinds of motion. Once you
have tracked motion on a clip, you can then use the resulting data for stabilization, motion smoothing,
matching the motion of one object to that of another, and a host of other essential tasks. Each tracker
type has its own chapter in this manual. This chapter covers the tracking techniques with the Planar
Tracker node.

Using the Planar Tracker

The Planar Tracker node is designed to deal with match moving issues on flat, unvarying surfaces that
commonly arise during post-production. Examples of flat, unvarying surfaces include clips containing a
license plate, a road sign, or a brick wall that often need images merged on top of them, such as
replacing the numbers in the license plate, changing the city’s name in the road sign, or placing a
billboard poster on the empty brick wall.
The Planar Tracker automates this process by analyzing the perspective distortions of a planar surface
on a background plate over time, and then re-applying those same perspective distortions to a
different foreground.

TIP: Part of using the Planar Tracker is also knowing when to give up and fall back to using
Fusion’s Tracker node or to manual keyframing. Some shots are simply not trackable, or the
resulting track suffers from too much jitter or drift. The Planar Tracker is a time-saving node in
the artist’s toolbox and, while it can track most shots, no tracker is a 100% solution.

Different Ways of Using the Planar Tracker Node

Like the other tracking nodes found in Fusion, the Planar Tracker can both analyze and contain the
resulting image tracking data interior to the node, and it can also use that tracking data to transform
either another image, paint strokes, or a polygon mask shape.
The Planar Tracker provides four modes of operation:
– Track: Used to isolate a planar surface and track its movement over time. Then, you can create a
Planar Transform node that uses this data to match move another clip in various ways.
– Steady: After analyzing a planar surface, this mode removes all motion and distortions from the
planar surface, usually in preparation for some kind of paint or roto task, prior to “unsteadying” the
clip to add the motion back.
– Corner Pin: After analyzing a planar surface, this mode computes and applies a matching
perspective distortion to a foreground image you connect to the foreground input of the Planar
Tracker node, and merges it on top of the tracked footage.
– Stabilize: After analyzing a planar surface, allows smoothing of a clip’s translation, rotation, and
scale over time. Good for getting unwanted vibrations out of a clip while retaining the overall
camera motion that was intended.

Chapter 23 Planar Tracking 537

Setting Up to Use the Planar Tracker
Similar to the Tracker node, to do a planar track, you need to connect the output of the image you
want to track to the background input of a Planar Tracker node.

Connecting an image to the background input of a PlanarTracker node

Check for Lens Distortion

If the image has barrel distortion, or any other kinds of lens distortion, it can adversely affect your
track. The more lens distortion in the footage, the more the resulting track will slide and wobble. If you
can see distortion in the image or you’re having problems with the track, you’ll want to try inserting the
Lens Distort node between the image and the Planar Tracker to eliminate this problem.
Fusion’s Lens Distort node can be used to remove or add lens distortion in an image. Connecting the
MediaIn or Loader node to the Lens Distort node displays controls for manually correcting lens
distortion. If you use Synth Eyes, PFTrack or 3D Equalizer software, you can also import lens data from
those applications to make the adjustments more automatic.

A Lens Distort node inserted between a MediaIn1 and Planar Tracker to remove lens distortion

For more information about using the Lens Distort node, see Chapter 121, “Warp Nodes” in the
DaVinci Resolve Reference Manual or Chapter 60 in the Fusion Reference Manual.
If you are using DaVinci Resolve, you can use the Lens Corrections control in the Cut page or Edit
page. This adjustment carries over into the Fusion page. Lens correction in DaVinci Resolve
automatically analyzes the frame in the Timeline viewer for edges that are being distorted by a wide
angle lens. Clicking the Analyze button moves the Distortion slider to provide an automatic correction.
From there, the MediaIn node in the Fusion page will have the correction applied, and you can begin
planar tracking.

A Basic Planar Tracker

Match Move Workflow
Using the Planar Tracker is a process, but it’s straightforward once you’ve learned how to use it. The
following procedure tries to make this process as clear as possible.

To track a surface using the Planar Tracker:

1 Make sure the Operation Mode is set to Track, as you need to analyze an image to track a surface
before you do anything else.
2 With the background input of the Planar Tracker connected to an image, and the Planar Tracker
open in a viewer, move the playhead to a frame of video where the planar surface you want to
track is at its largest, is unoccluded, and is clearly a plane, and then click the Set button in the
Track panel of the Inspector to make this the reference frame that will be used to guide the track.

Chapter 23 Planar Tracking 538

 Clicking the Set button to set the reference frame to use for analysis

3 Next, you’ll need to identify the specific pattern within the image that you want to track. In most
cases, this will probably be a rectangle, but any arbitrary closed polygon can be used. The pixels
enclosed by this region will serve as the pattern that will be searched for on other frames. Please
note that it is important that the pattern is drawn on the reference frame. In this example, we want
to track the wall behind the man, so we draw a polygon around part of the wall that the man won’t
pass over as he moves during the shot.

Drawing a polygon to identify the part of the image you

want to track, which should be a flat, trackable plane

TIP: Do not confuse the pattern you’re identifying with the region you’re planning to
corner pin (which always has four corners and is separately specified in Corner Pin mode.

4 (Optional) If moving objects partially cover up or occlude the planar surface, you may wish to
connect a mask that surrounds and identifies these occlusions to the white “occlusion mask” input
of the Planar Tracker. This lets the Planar Tracker ignore details that will cause problems.

When using the Hybrid Tracker, providing a mask to deal with occluding objects is nearly
mandatory, while with the Point Tracker it is recommended to try tracking without a mask.
5 If necessary, move the playhead back to the reference frame, which in this case was the first
frame. Then, click the Track To End button and wait for the track to complete.

The Analyze buttons of the Planar Tracker

As the clip tracks, you can see track markers and trails (if they’re enabled in the Options tab of the
Inspector) that let you see how much detail is contributing to the track, and the direction of motion
that’s being analyzed.

Chapter 23 Planar Tracking 539

 During tracking, you can see track markers and trails to follow how well the track is going

6 Once the track is complete, play through the clip to visually inspect the track so you can evaluate
how accurate it is. Does it stick to the surface? Switching to Steady mode can help here, as
scrubbing through the clip in Steady mode will help you immediately see unwanted motion in
the track.
7 Since we’re doing a match move, click the Create Planar Transform button to export a Planar
Transform node that will automatically transform either images or masks to follow the analyzed
motion of the plane you tracked.

Clicking Create Planar Transform to create a node

to use to transform other images or masks

In this case, the Planar Transform node will be inserted after a pair of Background and Paint nodes
that are being used to put some irritatingly trendy tech jargon graffiti on the wall. The Planar
Transform will automatically transform the Paint node’s output connected to its background input
to match the movement of the wall.

Adding the PlanarTransform node after a Paint node to match move it to

the background image, combining it via a Merge node

Chapter 23 Planar Tracking 540

The result is a seamless match move of the fake graffiti married to the wall in the original clip.

The final result; the paint layer is match moved to the background successfully

TIP: If you want to composite semi-transparent paint strokes on the wall, or use Apply modes
with paint stroke, you can attach a Paint node to a Background node set to 100 transparency.
The resulting image will be whatever paint strokes you make against transparency and is easy
to composite.

Tips for Choosing Good Planes to Track

The region to track is specified by drawing a polygon on the reference frame. Make sure the region
selected belongs to a physically planar surface in the shot. Sometimes a region that is only
approximately planar can be used. In general, the less planar the surface, the poorer the quality of the
resulting track.
As a rule of thumb, the more pixels in the pattern, the better the quality of the track. In particular, this
means on the reference frame, the pattern to be tracked should:
– Be as large as possible.
– Be as much in frame as possible.
– Be as unoccluded as possible by any moving foreground objects.
– Be at its maximum size (e.g., when tracking an approaching road sign, it is good to pick a later
frame where it is 400 x 200 pixels rather than 80 x 40 pixels).
– Be relatively undistorted (e.g., when the camera orbits around a flat stop sign, it is better to pick
a frame where the sign is face on parallel to the camera rather than a frame where it is at a highly
oblique angle).

If the pattern contains too few pixels or not enough trackable features, this can cause problems with
the resulting track, such as jitter, wobble, and slippage. Sometimes dropping down to a simpler motion
type can help in this situation.

Chapter 23 Planar Tracking 541

Chapter 24

Using Open FX,

Resolve FX, and
Fuse Plug-Ins
Fusion’s capabilities can be extended using different kinds of plug-ins.
All compositions in Fusion Studio and in the Fusion page of DaVinci Resolve support
third-party Open FX plug-ins. Additionally, the Fusion page of DaVinci Resolve
provides access to all of the Resolve FX that come with DaVinci Resolve.

Lastly, you can develop your own plug-ins without using a computer development
environment by scripting Fusion’s native Fuse plug-ins.

Contents
What Are Open FX?  543
What Are Resolve FX?  543
Applying Open FX and Resolve FX Plug-Ins  543
Introduction to Fuse Plug-Ins  544

Chapter 24 Using Open FX, Resolve FX, and Fuse Plug-Ins 542
What Are Open FX?
Fusion is able to use compatible Open FX (OFX) plug-ins that are installed on your computer. Open FX
is an open standard for visual effects plug-ins. It allows plug-ins written to the standard to work on
both DaVinci Resolve and Fusion Studio as well as other applications that support the standard.
OFX plug-ins can be purchased and downloaded from third-party suppliers such as BorisFX,
Red Giant, and RE:Vision Effects. All OFX appear in the Open FX category of the Effects Library,
alongside all other effects that are available in Fusion.

What Are Resolve FX?

The Fusion page is DaVinci Resolve is also able to access Resolve FX. Resolve FX are filter effects
within DaVinci Resolve. Although most Resolve FX have the same capabilities on all DaVinci Resolve
pages, some are specific to the Color page and require the use of the Color page tracker for
full functionality. All Resolve FX appear in the Fusion page Effects Library in the Open FX category
alongside third-party OFX plug-ins. Resolve FX are not available in Fusion Studio. For more
information about Resolve FX, see Chapter 153, “Resolve FX” in the DaVinci Resolve
Reference Manual.

Applying Open FX and Resolve FX Plug-Ins

Resolve FX and OFX plug-ins are applied to the Node Editor just as you would apply native Fusion
nodes. They can help you create fast, interesting looks and effects using the Color page, or
imaginative transitions and effects on your clips on the Edit page. Resolve FX are installed with
DaVinci Resolve,
After installing a set of OFX plug-ins, you can access them or Resolve FX plug-ins in Fusion by
opening the Open FX category in the Effects Library.
To add a plug-in to the Node Editor, either click the Open FX or Resolve FX plug-in name in the Effects
Library or drag and drop the plug-in onto a connection line to insert it into the node tree. If the plug-in
has editable settings, you can adjust these in the Inspector.

Chapter 24 Using Open FX, Resolve FX, and Fuse Plug-Ins 543
Introduction to Fuse Plug-Ins
Fuses are plug-ins developed for Fusion using the Lua built-in scripting language. Being script-based,
Fuses are compiled on-the-fly in Fusion without the need of a computer programming environment.
While a Fuse may be slower than an identical Open FX plug-in created using Fusion’s C++ SDK, a Fuse
will still take advantage of Fusion’s existing nodes and GPU acceleration.

To install a Fuse:
1 Use the .fuse extension at the end of the document name.
2 For DaVinci Resolve, save it in one of the following locations:
– On macOS: Macintosh HD/Users/username/Library/Application Support/Blackmagic Design/
DaVinci Resolve/Fusion/Fuses
– On Windows: C:\Users\username\AppData\Roaming\Blackmagic Design\DaVinci Resolve\
Support\Fusion\Fuses
– On Linux: home/username/.local/share/DaVinciResolve/Fusion/Fuses

For Fusion Studio, save it in one of the following locations:

– On macOS: Macintosh HD/Users/username/Library/Application Support/Blackmagic Design/
Fusion/Fuses/
– On Windows: C:\Users\username\AppData\Roaming\Blackmagic Design\Fusion\Fuses
– On Linux: home/username/.fusion/BlackmagicDesign/Fusion/Fuses

You can open and edit Fuses by selecting the Fuse node in the Node Editor and clicking the Edit
button at the top of the Inspector. The Fuse opens in the text editor specified in the Global
Preferences/Scripting panel.

TIP: Changes made to a Fuse in a text editor do not immediately propagate to other instances
of that Fuse in the composition. Reopening a composition updates all Fuses in the
composition based on the current saved version. Alternatively, you can click the Reload
button in the Inspector to update the selected node without closing and reopening the
composition.

Chapter 24 Using Open FX, Resolve FX, and Fuse Plug-Ins 544
 PART 3

3D Compositing
Chapter 25

3D Compositing
Basics
This chapter covers many of the nodes used for creating 3D composites, the tasks
they perform, and how they can be combined to produce effective 3D scenes.

Contents
An Overview of 3D Compositing  547
3D Compositing Fundamentals  548
Creating a Minimal 3D Scene  549
The Elements of a 3D Scene  550
Geometry Nodes  550
The Merge3D Node  552
The Renderer3D Node  555
Software vs. GPU Rendering  555
Software Renderer  555
OpenGL Renderer  556
OpenGL UV Renderer  556
Loading 3D Nodes into the Viewer  557
Navigating the 3D View  559
Transforming Cameras and Lights Using the Viewers  559
Transparency Sorting  560
Material Viewer  561
Transformations  562
Onscreen Transform Controls  562
Pivot  563
Target  563
Parenting  565

Chapter 25 3D Compositing Basics 546

Cameras  565
Quickly Viewing a Scene Through a Camera  566
Plane of Focus and Depth of Field  566
Importing Cameras  567
Lighting and Shadows  568
Enabling Lighting in the Viewer  568
Enabling Lighting to Be Rendered  568
Cotrolling Lighting within Each 3D Object  569
Lighting Types Explained  569
Materials and Textures  573
Material Components  574
Alpha Detail  576
Illumination Models  577
Textures  578
Reflections and Refractions  579
Bump Maps  581
Projection Mapping  582
Project a Texture onto a Catcher Material  583
Project Using the UVMap Node  584
Geometry  585
Common Visibility Parameters  586
Adding FBX Models  586
Using Text3D  587
Fog 3D and Soft Clipping  591
Material and Object IDs  593
World Position Pass  594
Point Clouds  595

An Overview of 3D Compositing
Traditional image-based compositing is a two-dimensional process. Image layers have only the
amount of depth needed to define one as foreground and another as background. This is at odds with
the realities of production, since all images are either captured using a live-action camera with
freedom in all three dimensions, in a shot that has real depth, or have been created in a true 3D
modeling and rendering application.
Within the Fusion Node Editor, you have a GPU-accelerated 3D compositing environment that includes
support for imported geometry, point clouds, and particle systems for taking care of such things as:
– Converting 2D images into image planes in 3D space
– Creating rough primitive geometry
– Importing mesh geometry from FBX or Alembic scenes
– Creating realistic surfaces using illumination models and shader compositing
– Rendering with realistic depth of field, motion blur, and supersampling

Chapter 25 3D Compositing Basics 547

 – Creating and using 3D particle systems
– Creating, extruding, and beveling 3D text
– Lighting and casting shadows across geometry
– 3D camera tracking
– Importing cameras, lights, and materials from 3D applications such as Maya,
3ds Max, or LightWave
– Importing matched cameras and point clouds from applications such as SynthEyes or PF Track

An example 3D scene in Fusion

3D Compositing Fundamentals
The 3D category of nodes (which includes the Light, Material, and Texture subcategories) work
together to create 3D scenes. Examples are nodes that generate geometry, import geometry, modify
geometry, create lights and cameras, and combine all these elements into a scene. Nearly all these
nodes are collected within the 3D category of nodes found in the Effects Library.

The 3D category of nodes in the Effects Library

Conveniently, at no point are you required to specify whether your overall composition is 2D or 3D,
because you can seamlessly combine any number of 2D and 3D “scenes” together to create a single
output. However, the nodes that create these scenes must be combined in specific ways for this to
work properly.

Chapter 25 3D Compositing Basics 548

Creating a Minimal 3D Scene
Creating a 3D scene couldn’t be easier, but you need to connect the required nodes in the right way.
At minimum, you need only connect a geometry node (such as a Text3D node) to a Renderer3D node
to output a 2D image that can be combined with other 2D images in your composition, as seen below.
However, you’ll only get a simply shaded piece of geometry for your trouble, although you can color
and transform it in the Inspector using controls internal to whichever geometry node you’re using.

A simple 3D scene with a Text3D node connected directly to a Renderer3D node

More realistically, each 3D scene that you want to create will probably have three to five nodes to give
you a better lit and framed result. These include:
– One of the available geometry nodes (such as Text3D or Image Plane 3D)
– A light node (such as DirectionalLight or SpotLight)
– A camera node
– A Merge3D node
– A Renderer3D node

All these should be connected together as seen below, with the resultantly more complex 3D scene
shown below.

The same text, this time lit and framed using Text3D,
Camera, and SpotLight nodes to a Merge3D node

Chapter 25 3D Compositing Basics 549

To briefly explain how this node tree works, the geometry node (in this case Text3D) creates an object
for the scene, and then the Merge3D node provides a virtual stage that combines the attached
geometry with the light and camera nodes to produce a lit and framed result with highlights and
shadows, while the aptly named Renderer3D node renders the resulting 3D scene to produce 2D
image output that can then be merged with other 2D images in your composition.
In fact, these nodes are so important that they appear at the right of the toolbar, enabling you to
quickly produce 3D scenes whenever you require. You might notice that the order of the 3D buttons
on the toolbar, from left to right, corresponds to the order in which these nodes are ordinarily used. So,
if you simply click on each one of these buttons from left to right, you cannot fail to create a properly
assembled 3D scene, ready to work on, as seen in the previous screenshot.

The 3D nodes available from the toolbar include the ImagePlane3D, Shape3D,
Text3D, Merge3D, Camera3D, SpotLight3D, and Renderer3D nodes

The Elements of a 3D Scene

All 3D nodes can be divided into a number of categories.

Geometry Nodes
You can add 3D geometry to a composition using the ImagePlane3D node, the Shape3D node, the
Cube3D node, the Text3D node, or optionally by importing a model via the FBX Mesh 3D node.
Furthermore, you can add particle geometry to scenes from pEmitter nodes. You can connect these to
a Merge3D node either singularly or in multiples to create sophisticated results combining
multiple elements.

A more complex 3D scene combining several geometry nodes including the Text3D, Shape3D, and ImagePlane3D nodes

Texturing Geometry
By itself, geometry nodes can only consist of a simple flat color. However, you can alter the look of 3D
geometry by texturing it using clips (either still images or movies), using material nodes such as the

Chapter 25 3D Compositing Basics 550

Blinn and Phong nodes to create more sophisticated textures with combinations of 2D images and
environment maps, or you can use a preset shader from the Templates > Shader bin of the Effects
Library, which contains materials and texture presets that are ready to use.
If you’re working with simple geometric primitives, you can texture them by connecting either an image
(a still image or movie) or a shader from the Templates bin of the Effects Library directly to the material
input of a Shape3D, Cube3D, or other compatible node, as shown below.

An image connected to the material input of a Shape3D node set to Taurus,

with the image (left), and the shaded taurus (right)

If you’re shading or texturing Text3D nodes, you need to add a texture in a specific way since each
node is actually a scene with individual 3D objects (the characters) working together. In the following
example, the RustyMetal shader preset is applied to a Text3D node using the ReplaceMaterial3D
node. The interesting thing about the ReplaceMaterial3D node is that it textures every geometric
object within a scene at once, meaning that if you put a ReplaceMaterial3D node after a Text3D node,
you texture every character within that node. However, if you place a ReplaceMaterial3D node after a
Merge3D node, then you’ll end up changing the texture of every single geometric object being
combined within that Merge3D node, which is quite powerful.

The geometry created by a Text3D node is textured using a shader connected to a

ReplaceMaterial3D node that’s connected downstream of the object you want to shade

Chapter 25 3D Compositing Basics 551

The Merge3D Node
The Merge3D node combines the output of one or more 3D nodes into a single scene. Unlike the
Merge2D node, the ordering of elements in the scene is not restricted to only background and
foreground inputs. Instead, the Merge3D node lets you connect an unlimited number of inputs, with
the resulting output combined according to each object’s absolute position in 3D space.

Merging many objects together in a 3D scene using the Merge3D node

Combining Objects Directly

While the Merge3D node provides a structured way of combining objects, you can also combine 3D
objects such as Text3D and Shape3D nodes by connecting the output of one 3D object node to the
input of another, as seen in the following screenshot. When you do this, you must use each node’s
internal transform parameters to transform their position, size, and rotation directly, but the transform
control of downstream 3D object nodes also transforms all upstream 3D object nodes. This even
works for lights and the Camera3D node, giving you a fast way of combining a set of objects that
always go together, which you can later connect to a Merge3D node for additional lighting and
eventual connection to a Renderer3D node.

Connecting one Shape3D node to another directly

to combine them Transforming the last downstream
3D object also transforms all upstream objects;
the last Shape3D node is viewed, showing both

Chapter 25 3D Compositing Basics 552

Combining Multiple Merge3D Nodes
Furthermore, Merge3D nodes can be combined with other Merge3D nodes, allowing you to create
composite 3D scenes made up of multiple “sub-scenes,” each put together within individual
Merge3D nodes.

You can build elaborate scenes using multiple Merge3D nodes connected together

Lighting Multiple Merge3D Nodes

Once you’ve combined multiple Merge3D nodes, there’s an easy way to control how lights that are
connected to upstream Merge3D nodes affect the results of other Merge3D nodes connected
downstream. Each Merge3D node’s Controls tab contains a single checkbox, Pass Through Lights,
which enables lighting to pass through the output of an upstream Merge3D node in order to shine
onto objects connected to downstream Merge3D nodes.

You can light downstream Merge3D scenes with lights connected to

upstream Merge3D scenes by turning on Pass Through Lights

This checkbox is disabled by default, which lets you light elements in one Merge3D scene without
worrying about how the lighting will affect geometry attached to other Merge3D nodes further
downstream. For example, you may want to apply a spotlight to brighten the wall of a building in one
Merge3D node without having that spotlight spill over onto the grass or pavement at the foot of the
wall modeled in another Merge3D node. In the example shown below, the left image shows how the
cone and taurus connected to a downstream node remain unlit by the light in an upstream node with
Pass Through Lights disabled, while the right image shows how everything becomes lit when turning
Pass Through Lights on.

The result of lights on the text in one Merge3D node not affecting the cone and taurus
added in a downstream Merge3D node (left) Turning on Pass Through Lights in the upstream
Merge3D node results in those lights also illuminating the downstream shapes (right)

Chapter 25 3D Compositing Basics 553

Transforming Merge3D Scenes
Each Merge3D node includes a Transform tab. These transform parameters adjust the position, scale,
and rotation of all objects being combined within that Merge3D node together, including lighting and
particles. All transformations take place around a common pivot point. This forms the basis of
parenting in the 3D environment.

The Transform tab of a Merge3D node

If you transform a Merge3D node that’s connected to other Merge3D nodes, what happens depends
on which node you’re transforming, an upstream node or the downstream node:
– If you transform a downstream Merge3D node, you also transform all upstream nodes connected
to it as if they were all a single scene.
– If you transform an upstream Merge3D node, this has no effect on downstream Merge3D nodes,
allowing you to make transforms specific to that particular node’s scene.

Transforming Upstream, Lighting Downstream

When building complex scenes using multiple Merge3D nodes being combined together, it’s common
to use one last downstream node to combine light and camera nodes to illuminate the final scene,
while leaving the upstream Merge3D nodes free for controlling object transforms and animation. This
way, you can transform and animate subsets of your overall scene without worrying about accidentally
altering the overall lighting scheme or cameras for that scene, unless you’ve specifically connected
lights or cameras upstream that are meant to be attached to the geometry you’re transforming.

An example of a 3D scene using multiple Merge3D nodes working together; the upstream Merge3D nodes arrange
the 3D objects placed within the scene, while the last Merge3D node (orange) lights and frames the scene.

Chapter 25 3D Compositing Basics 554

The Renderer3D Node
Every 3D node you add outputs a complete 3D scene. This is unlike most traditional 3D modeling and
animation programs, where all objects reside within a global scene environment. This means that the
scenes created by a Camera 3D node and an image plane are separate until they’re combined into the
same scene via a Merge3D node, which itself outputs a complete 3D scene. However, this 3D scene
data can neither be composited with other 2D images in your composition nor rendered out without
first being rendered within the node tree using a Renderer3D node.
To be more specific, 3D nodes that output 3D scenes cannot be connected directly to inputs that
require 2D images. For example, the output of an ImagePlane3D node cannot be connected directly
to the input of a Blur node, nor can the output of a Merge3D node be directly connected to a regular
Merge node. First, a Renderer3D node must be placed at the end of your 3D scene to render it into 2D
images, which may then be composited and adjusted like any other 2D image in your composition.

Output of a Merge3D connected to a Renderer3D node to output 2D image data

The Renderer3D uses one of the cameras in the scene (typically connected to a Merge3D node) to
produce an image. If no camera is found, a default perspective view is used. Since this default view
rarely provides a useful angle, most people build 3D scenes that include at least one camera.
The image produced by the Renderer3D can be any resolution with options for fields processing, color
depth, and pixel aspect.

Software vs. GPU Rendering

The Renderer3D node lets you choose between using a software renderer or an OpenGL renderer,
trading off certain aspects of rendered image quality for speed, and trading off depth of field
rendering for soft shadow rendering, depending on the needs of a particular element of your
composition. To choose which method of rendering to use, there’s a Renderer Type pop-up menu in
the Controls tab of each Renderer3D node’s parameters in the Inspector. The default is
Software Renderer.

The Renderer Type option in the Controls tab of a Renderer3D node

Software Renderer
The software renderer is generally used to produce the final output. While the software renderer is not
the fastest method of rendering, it has twin advantages. First, the software renderer can easily handle
textures much larger than one half of your GPU’s maximum texture size, so if you’re working with
texture images larger than 8K you should choose the software renderer to obtain maximum quality.

Chapter 25 3D Compositing Basics 555

Second, the software renderer is required to enable the rendering of “constant” and “variable” soft
shadows with adjustable Spread, which is not supported by the OpenGL renderer. Soft shadows are
more natural, and they’re enabled in the Shadows parameters of the Controls tab of light nodes; you
can choose Sampling Quality and Softness type, and adjust Spread, Min Softness, and Filter Size
sliders. Additionally, the software renderer supports alpha channels in shadow maps, allowing
transparency to alter shadow density.

When the Renderer3D node “Renderer Type” drop-down is set to OpenGL Renderer, you cannot render
soft shadows or excessively large textures (left). When the Renderer3D node “Renderer Type” drop‑down
is set to Software Renderer, you can render higher-quality textures and soft shadows (right).

OpenGL Renderer
The OpenGL renderer takes advantage of the GPU in your computer to render the image; the textures
and geometry are uploaded to the graphics hardware, and OpenGL shaders are used to produce the
result. This can produce high-quality images that can be perfect for final rendering, and can also be
potentially orders of magnitude faster than the software renderer, but it does pose some limitations on
some rendering effects, as soft shadows cannot be rendered, and the OpenGL renderer also ignores
alpha channels during shadow rendering, resulting in a shadow always being cast from the
entire object.
On the other hand, because of its speed, the OpenGL renderer exposes additional controls for
Accumulation Effects that let you enable depth of field rendering for creating shallow-focus effects.
Unfortunately, you can’t have both soft shadow rendering and depth of field rendering, so you’ll need
to choose which is more important for any given 3D scene you render.

Don’t Forget That You Can Combine Rendered Scenes in 2D

While it may seem like an insurmountable limitation that you can’t output both soft shadows
and depth of field using the same renderer, don’t forget that you can create multiple 3D
scenes each using different renderers and composite them in 2D later on. Furthermore, you
can also render out auxiliary channels that can be used by 2D image processing nodes such
as AmbientOcclusion, DepthBlur, and Fog to create pseudo-3D effects using the
rendered images.

OpenGL UV Renderer
When you choose the OpenGL UV Renderer option, a Renderer3D node outputs an “unwrapped”
version of the textures applied to upstream objects, at the resolution specified within the Image tab of
that Renderer3D node.

Chapter 25 3D Compositing Basics 556

A normally rendered 3D scene (left), and the same scene rendered using
the OpenGL UV Renderer mode of the Renderer3D node (right)

This specially output image is used for baking out texture projections or materials to a texture map for
one of two reasons:
– Baking out projections can speed up a render.
– Baking out projections lets you modify the texture using other 2D nodes within your composition,
or even using third-party paint applications (if you output this image in isolation as a graphics file)
prior to applying it back onto the geometry.

Suppose, for instance, that you have a scene on a street corner, and there’s a shop sign with a phone
number on it, but you want to change the numbers. If you track the scene and have standing geometry
for the sign, you can project the footage onto it, do a UV render, switch the numbers around with a
Paint node, and then apply that back to the mesh with a Texture2D.
The UV renderer can also be used for retouching textures. You can combine multiple DSLR still shots
of a location, project all those onto the mesh, UV render it out, and then retouch the seams and apply
it back to the mesh.
You could project tracked footage of a road with cars on it, UV render out the projection from the
geometry, do a temporal median filter on the frames, and then map a “clean” roadway back down.

Loading 3D Nodes into the Viewer

When you load a 3D node into the viewer, it switches to a 3D Viewer, which lets you pan, zoom, and
rotate the scene in 3D, making it easy to make adjustments in three dimensions.

The 3D Viewer

Chapter 25 3D Compositing Basics 557

The interactive 3D Viewer is highly dependent on the computer’s graphics hardware, relying on
support from OpenGL. The amount of onboard memory, as well as the speed and features of your
workstation’s GPU, make a huge difference in the speed and capabilities of the 3D Viewer.
Displaying a node with a 3D output in any viewer will switch the display type to a 3D Viewer. Initially,
the contents of the scene will be displayed through a default perspective view.

A 3D Viewer’s default perspective view

To change the viewpoint, right-click in the viewer and choose the desired viewpoint from the ones
listed in the Camera submenu. A shortcut to the Camera submenu is to right-click on the axis label
displayed in the bottom corner of the viewer.

Right-click the Axis label of the

viewer to change the viewpoint

In addition to the usual Perspective, Front, Top, Left, and Right viewpoints, if there are cameras and
lights present in the scene as potential viewpoints, those are shown as well. It’s even possible to
display the scene from the viewpoint of a Merge3D or Transform3D by selecting it from the contextual
menu’s Camera > Other submenu. Being able to move around the scene and see it from different
viewpoints can help with the positioning, alignment, and lighting, as well as other aspects of
your composite.

Chapter 25 3D Compositing Basics 558

 The Perspective drop-down menu also shows cameras, lights, and
Merge3D and Transform3D nodes you can switch to.

Navigating the 3D View

For the most part, panning and scaling of the 3D Viewer uses the same controls as the 2D Viewer.
For more information about the options available in the 3D Viewer, see Chapter 68, “Using Viewers”
in the DaVinci Resolve Reference Manual or Chapter 7 in the Fusion Reference Manual.

To pan in a 3D Viewer, do the following:

– Hold the middle mouse button and drag in the viewer.

To dolly (zoom) in the 3D Viewer, do one of the following:

– Hold down the middle and left mouse buttons and drag left or right in the viewer.
– Hold down the Command key and use your pointing device’s scroll control.

To rotate around the 3D Viewer, do the following:

– Hold down the Option key and middle-button-drag left and right in the viewer.

If you want to frame certain objects in the viewer:

1 Select the viewer you want to work in.
2 Do one of the following:
– Press Shift-F to Fit all objects in the viewer.
– Press F to Fit to selection (or Fit All if nothing is selected).
– Press D to Rotate the viewer to look at the center of the currently selected object without
moving the viewer’s position.

Furthermore, selecting a 3D node in the Node Editor also selects the associated object in the
3D Viewer.

Transforming Cameras and Lights Using the Viewers

When the viewer is set to look through a 3D object in the scene, such as a camera or spotlight, the
usual controls for panning and rotating the viewer will now directly affect the position of the camera or
spotlight you’re viewing through. Here’s an example.

To adjust a camera’s position when looking through it in a viewer:

1 Right-click the viewpoint label, and choose a camera from the contextual menu. (Optional) If you’re
in dual-viewer mode, you can load the camera you’ve selected in one viewer into the other viewer
to see its position as you work.
2 Move the pointer into the viewer that’s displaying the camera’s viewpoint.

Chapter 25 3D Compositing Basics 559

3 Hold the middle and left mouse buttons down and drag to zoom the viewer, or middle-click-
drag to pan the viewer, or option-middle-click-drag to rotate the viewer, all while also moving
the camera.

When a viewer is set to display the view of a camera or light, panning, zooming, or rotating the viewer
(seen at right) actually transforms the camera or light you’re viewing through (seen at left)

It is even possible to view the scene from the perspective of a Merge3D or Transform3D node by
selecting the object from the Camera > Others menu. The same transform techniques will then move
the position of the object. This can be helpful when you are trying to orient an object in a certain
direction.

Transparency Sorting
While generally the order of geometry in a 3D scene is determined by the Z-position of each object,
sorting every face of every object in a large scene can take an enormous amount of time. To provide
the best possible performance, a Fast Sorting mode is used in the OpenGL renderer and viewers. This
is set by right-clicking in the viewer and choosing Transparency > Z-buffer. While this approach is much
faster than a full sort, when objects in the scene are partially transparent it can also produce
incorrect results.
The Sorted (Accurate) mode can be used to perform a more accurate sort at the expense of
performance. This mode is selected from the Transparency menu of the viewer’s contextual menu.
The Renderer3D also presents a Transparency menu when the Renderer Type is set to OpenGL.
Sorted mode does not support shadows in OpenGL. The software renderer always uses the Sorted
(Accurate) method.

Transparency Sorting in the viewer contextual menu

Chapter 25 3D Compositing Basics 560

The basic rule is when a scene contains overlapping transparency, use the Full/Quick Sort modes, and
otherwise use the Z-buffer (Fast). If the Full Sort method is too slow, try switching back to
Z-buffer (Fast).

Material Viewer
When you view a node that comes from the 3D > Material category of nodes in the Effects Library, the
viewer automatically switches to display a Material Viewer. This Material Viewer allows you to preview
the material applied to a lit 3D sphere rendered with OpenGL by default.

The Material Viewer mode of the viewer

The type of geometry, the renderer, and the state of the lighting can all be set by right-clicking the
viewer and choosing options from the contextual menu. Each viewer supports A and B buffers to assist
with comparing multiple materials.

Methods of working with the Material Viewer:

– You can change the shape of the previewed geometry by right-clicking the viewer and choosing
an option from the Shape submenu of the contextual menu. The geometry that the material is
applied to is locked to the center of the viewer and scaled to fit. It is not possible to pan or scale
the Material Viewer.
– The Material Viewer can be rotated to provide a different angle on the material by holding Option
while pressing the middle mouse button and dragging to the left and right.
– You can adjust the position of the light used to preview the material by dragging with the middle
mouse button. Or, you can right-click the viewer and choose an option from the Lighting > Light
Position submenu of the contextual menu.
– You can also toggle lighting off and on by right-clicking the viewer and choosing Lighting > Enable
Lighting from the contextual menu.
– You can choose the renderer used to preview the material by right-clicking the viewer and
choosing an option from the Renderer submenu of the contextual menu.

Chapter 25 3D Compositing Basics 561

Transformations
Merge3D, 3D Objects, and Transform3D all have Transform parameters that are collected together into
a Transform tab in the Inspector. The parameters found in this tab affect how the object is positioned,
rotated, and scaled within the scene.

The Transform tab of a Merge3D node

The Translation parameters are used to position the object in local space, the Rotation parameters
affect the object’s rotation around its own center, and the Scale slider(s) affect its size (depending on
whether or not they’re locked together). The same adjustments can be made in the viewer using
onscreen controls.

Onscreen Transform Controls

When an object is selected, it displays onscreen Transform controls in the viewers that allow you to
adjust the object’s position, rotation, and scale. Buttons in the Transform toolbar allow you to switch
modes, or you can use the keyboard shortcuts.

To switch Transform modes, use the following keyboard shortcuts:

– Press Q for Position
– Press W for Rotation
– Press E for Scaling

The Position, Rotation, and Scale

modes in the Transform toolbar

Using Onscreen Transform Controls

In all three modes, red indicates the object’s local X-axis, green the Y-axis, and blue the Z-axis,
respectively (just remember RGB = XYZ). You can drag directly on the red, green, or blue portion of
any onscreen control to constrain the transform to that axis, or if you drag the center of the onscreen
control, you can apply a transform without constraints. Holding Option and dragging in the viewer
allows you to freely translate in all three axes without clicking on a specific control.

Chapter 25 3D Compositing Basics 562

From left to right, the Position, Rotation, and Scale onscreen Transform controls

If the Scale’s Lock XYZ checkbox is enabled in the Inspector, only the overall scale of the object is
adjusted by dragging the red or center onscreen control, while the green and blue portions of the
onscreen control have no effect. If you unlock the parameters, you are able to scale an object along
individual axes separately to squish or stretch the object.

Selecting Objects
With the onscreen controls visible in the viewer, you can select any object by clicking on its center
control. Alternatively, you can also select any 3D object by clicking its node in the Node Editor.

Pivot
In 3D scenes, objects rotate and scale around an axis called a pivot. By default, this pivot goes
through the object’s center. If you want to move the pivot so it is offset from the center of the object,
you can use the X, Y, and Z Pivot parameters in the Inspector.

Target
Targets are used to help orient a 3D object to a specific point in the scene. No matter where the object
moves, it will rotate in the local coordinate system so that it always faces its target, which you can
position and animate.

To enable a target for a 3D object:

1 Select that object’s node.
2 Open the object’s Transform panel in the Inspector.
3 Turn on the Use Target checkbox.

Turning on the Use Target checkbox of a 3D object

4 Use the X/Y/Z Target Position controls in the Inspector or the Target onscreen control in the
viewer to position the target and in turn position the object it’s attached to.

Chapter 25 3D Compositing Basics 563

In the viewer, a line is drawn between the target and the center of the 3D object it’s attached to, to
show the relationship between these two sets of controls. Whenever you move the target, the object
is automatically transformed to face its new position.

A taurus facing its onscreen Target controls

For example, if a spotlight is required in the scene to point at an image plane, enable the spotlight’s
target in the Transform tab and connect the target’s XYZ position to the image plane’s XYZ position.
Now, no matter where the spotlight is moved, it will rotate to face the image plane.

A light made to face the wall using its enabled target control

Chapter 25 3D Compositing Basics 564

Parenting
One of the many advantages of the node-based approach to 3D compositing is that parenting
between objects becomes implicit in the structure of a 3D node tree. The basis for all parenting is the
Merge3D node. If you’re careful about how you connect the different 3D objects you create for your
scene, you can use multiple Merge3D nodes to control which combinations of objects are transformed
and animated together, and which are transformed and animated separately.
For example, picture a scene with two spheres that are both connected to a Merge3D. The Merge3D
can be used to rotate one sphere around the other, like the moon around the earth. Then the
Merge3D can be connected to another Merge3D to create the earth and the moon orbiting
around the sun.

One Merge3D with two spheres parented to another

Merge3D and parenting using three connected spheres

Here are the two simple rules of transforming parented Merge3D nodes:
– Transforms and animation applied to a Merge3D are also applied to all 3D objects connected
to that Merge3D node, including cameras, lights, geometry, and other merge nodes connected
upstream.
– Transforms and animation applied to upstream merge nodes don’t affect downstream
merge nodes.

Cameras
When setting up and animating a 3D scene, the metaphor of a camera is one of the most
comprehensible ways of framing how you want that scene to be rendered out, as well as animating
your way through the scene. Additionally, compositing artists are frequently tasked with matching
cameras from live-action clips, or matching cameras from 3D applications.
To accommodate all these tasks, Fusion provides a flexible Camera3D node with common camera
controls such as Angle of View, Focal Length, Aperture, and Clipping planes, to either set up your own
camera or to import camera data from other applications. The Camera3D node is a virtual camera
through which the 3D environment can be viewed.

Chapter 25 3D Compositing Basics 565

 A camera displayed with onscreen Transform controls in
the viewer; the Focal Plane indicator is enabled in green

Cameras are typically connected and viewed via a Merge3D node; however, you can also connect
cameras upstream of other 3D objects if you want that camera to transform along with that object
when it moves.

Quickly Viewing a Scene Through a Camera

When you’ve added a camera to a scene, you can quickly view the scene “through the camera” by
setting up the following.

To view the scene through the camera:

1 Select the Merge3D node that the camera is connected to, or any node downstream of
that Merge3D.
2 Load the selected Merge3D or downstream node into a viewer.
3 Right-click on the axis label in the bottom corner of the viewer and choose the camera name.

The viewer’s frame may be different from the camera frame, so it may not match the true boundaries of
the image that will be rendered by the Renderer3D node. If there is no Renderer3D node added to
your scene yet, you can use Guides that represent the camera’s framing. For more information about
Guides, see Chapter 68, “Using Viewers” in the DaVinci Resolve Reference Manual or Chapter 7 in the
Fusion Reference Manual.

Plane of Focus and Depth of Field

Cameras have a plane of focus, for when depth of field rendering is available. Here’s the procedure for
enabling depth of field rendering in your scenes.

To render depth of field in a 3D scene:

1 You must add a Renderer3D node at the end of your 3D scene.
2 Select the Renderer3D node, and set the Renderer Type to OpenGL Renderer.

Chapter 25 3D Compositing Basics 566

3 Open the Accumulation Effects disclosure control that appears, and turn on the Enable
Accumulation Effects checkbox in the OpenGL render.

Turning on Enable Accumulation Effects enables additional depth of field controls

Turning on “Enable Accumulation Effects” exposes a Depth of Field checkbox along with Quality and
Amount of DoF Blur sliders that let you adjust the depth of field effect. These controls affect only the
perceived quality of the depth of field that is rendered. The actual depth of field that’s generated
depends solely on the setup of the camera and its position relative to the other 3D objects in
your scene.
When you select your scene’s Camera3D node to view its controls in the Inspector, a new Focal Plane
checkbox appears in the Control Visibility group. Turning this on lets you see the green focal plane
indicator in the 3D Viewer that lets you visualize the effect of the Focal Plane slider, which is located in
the top group of parameters in the Camera3D node’s Controls tab.

Turning on the Focal Plane checkbox

in the Camera3D node

For more information about these specific camera controls, see Chapter 90, “3D Nodes” in the
DaVinci Resolve Reference Manual, or Chapter 29 in the Fusion Reference Manual.

Importing Cameras
If you want to match cameras between applications, you can import camera paths and positions from a
variety of popular 3D applications. Fusion is able to import animation splines from Maya and XSI
directly with their own native spline formats. Animation applied to cameras from 3ds Max and
LightWave are sampled and keyframed on each frame.

To import a camera from another application, do the following:

1 Select the camera in the Node Editor.
2 At the bottom of the Inspector, click the Import Camera button.
3 In the file browser, navigate to and select the scene that contains the camera you want to import.

Chapter 25 3D Compositing Basics 567

A dialog box with several options will appear. When the Force Sampling checkbox is enabled, Fusion
will sample each frame of the motion, regardless of the format.

The Select Camera dialog

TIP: When importing parented or rigged cameras, baking the camera animation in the 3D
application before importing it into Fusion often produces more reliable results.

Lighting and Shadows

You can add light sources to a scene to create very detailed lighting environments and atmosphere.
There are four different types of lights you can use in 3D scenes: ambient, directional, point, and
spotlights.

Enabling Lighting in the Viewer

A scene without lights uses a default directional light, but this automatically disappears once you add a
3D light object. However, even when you add light objects to your scene, lighting and shadows won’t
be visible in the viewer unless you first enable lighting in the viewer contextual menu by right-clicking
anywhere within a viewer and choosing 3D Options > Lighting or Shadows to turn on one or both.

Enabling Lighting to Be Rendered

Lighting effects won’t be rendered in the Renderer3D node until the Enable Lighting and/or Shadows
checkboxes are checked in the Inspector.

The Lighting button under the viewer

NOTE: When lighting is disabled in either the viewer or final renders, the image will appear to
be lit by a 100% ambient light.

Chapter 25 3D Compositing Basics 568

Cotrolling Lighting within Each 3D Object
All nodes that create or merge geometry also include lighting options that are used to choose how
each object is affected by light:
– Merge3D nodes have a Pass Through Lights checkbox that determines whether lights attached to
an upstream Merge3D node also illuminate objects attached to downstream Merge3D nodes.
– ImagePlane3D, Cube3D, Shape3D, Text3D, and FBXMesh3D nodes have a set of Lighting controls
that let you turn three controls on and off: Affected by Lights, Shadow Caster, and Shadow
Receiver.

3D objects have individual lighting controls that let you

control how each object interacts with light and shadows

Lighting Types Explained

Here’s a more detailed explanation of each type of light in Fusion.

Ambient Light
You use ambient light to set a base light level for the scene, since it produces a general uniform
illumination of the scene. Ambient light exists everywhere without appearing to come from any
particular source; it cannot cast shadows and will tend to fill in shadowed areas of a scene.

Directional Light
A directional light is composed of parallel rays that light up the entire scene from one direction,
creating a wall of light. The sun is an excellent example of a directional light source.

Point Light
A point light is a well defined light that has a small clear source, like a light bulb, and shines from that
point in all directions.

Spotlight
A spotlight is an advanced point light that produces a well defined cone of light with falloff. This is the
only light that produces shadows.

From left to right: Directional light, point light, and spotlight

Chapter 25 3D Compositing Basics 569

All of the Light nodes display onscreen controls in the viewer, although not all controls affect every
light type. In the case of the ambient light, the position has no effect on the results. The directional
light can be rotated, but position and scale will be ignored. The point light ignores rotation. Both
position and rotation apply to the spotlight.

Lighting Hierarchies
Lights normally do not pass through a Merge, since the Pass Through Lights checkbox is off by default.
This provides a mechanism for controlling which objects are lit by which lights. For example, in the
following two node trees, two shapes and an ambient light are combined with a Merge3D node, which
is then connected to another Merge3D node that’s also connected to a plane and a spotlight. At the
left, the first Merge3D node of this tree has Pass Through Lights disabled, so you can only see the two
shapes lit. At the right, Pass Through Lights has been enabled, so both the foreground shapes and the
background image plane receive lighting.

Pass Through Lights is disabled, so only the front two shapes are illuminated (left) Pass Through
Lights is enabled, so all shapes connected to both Merge3D nodes are illuminated (right)

Lighting Options
Most nodes that generate geometry have additional options for lighting. These options are used to
determine how each individual object reacts to lights and shadows in the scene.

3D objects have individual lighting controls that let you

control how each object interacts with light and shadows

– Affected By Lights: If the Affected By Lights checkbox is enabled, lights in the scene will affect
the geometry.
– Shadow Caster: When enabled, the object will cast shadows on other objects in the scene.
– Shadow Receiver: If this checkbox is enabled, the object will receive shadows.

Chapter 25 3D Compositing Basics 570

Shadows
The only light that can cast shadows is the spotlight. Spotlight nodes cast shadows by default,
although these shadows will not be visible in the viewer until shadows are enabled using the viewer
toolbar button. Shadows will not appear in the output of the Renderer3D unless the Shadows option is
enabled for that renderer. If you want to prevent a spotlight from casting shadows, you can disable the
Enable Shadows checkbox in the node’s Inspector.

An image with spotlight casting a variable soft shadow

For more information on shadow controls, see the “Spotlight” section of Chapter 91, “3D Light Nodes”
in the DaVinci Resolve Reference Manual or Chapter 30 in the Fusion Reference Manual.

Shadow Maps
A shadow map is an internal depth map that specifies each pixel’s depth in the scene. This information
is used to assemble the shadow layer created from a spotlight. All the controls for the shadow map are
found in the Spotlight Inspector.
The quality of the shadow produced depends greatly on the size of the shadow map. Larger maps
generate better-looking shadows but will take longer to render. The wider the cone of the spotlight, or
the more falloff in the cone, the larger the shadow map will need to be to produce useful quality
results. Setting the value of the Shadow Map Size control sets the size of the depth map in pixels.
Generally, through trial and error, you’ll find a point of diminishing returns where increasing the size of
the shadow map no longer improves the quality of the shadow. It is not recommended to set the size
of the shadow maps any larger than they need to be.
The Shadow Map Proxy control is used to set a percentage by which the shadow map is scaled for fast
interactive previews, such as Autoproxy and LoQ renders. A value of .4, for example, represents a
40% proxy.

Shadow Softness
By default, the spotlight generates shadows without soft edges, but there are options for constant and
variable soft shadows. Hard-edged shadows will render significantly faster than either of the Soft
Shadow options. Shadows without softness will generally appear aliased, unless the shadow map size
is large enough. In many cases, softness is used to hide the aliasing rather than increasing the shadow
map to preserve memory and avoid exceeding the graphics hardware capabilities.

Soft Shadow controls in the Control panel

Setting the spotlight’s shadow softness to None will render crisp and well-defined shadows.
The Constant option will generate shadows where the softness is uniform across the shadow,

Chapter 25 3D Compositing Basics 571

regardless of the shadow’s distance from the casting geometry. The Variable option generates
shadows that become softer as they get farther from the geometry that is casting the shadow. This is a
more realistic effect, but the shadows are somewhat harder to control. When this option is selected,
additional controls for adjusting the falloff of the shadow will appear, as well as sliders for the minimum
and maximum softness.

Hard shadow cast by a spotlight

Selecting the Variable option reveals the Spread, Min Softness, and Filter Size sliders. A side
effect of the method used to produce variable softness shadows is that the size of the blur applied to
the shadow map can become effectively infinite as the shadow’s distance from the geometry
increases. These controls are used to limit the shadow map by clipping the softness calculation to a
reasonable limit.
The filter size determines where this limit is applied. Increasing the filter size increases the maximum
possible softness of the shadow. Making this smaller can reduce render times but may also limit the
softness of the shadow or potentially even clip it. The value is a percentage of the shadow map size.
For more information, see “Spotlight” in Chapter 91, “3D Light Nodes” in the DaVinci Resolve
Reference Manual or Chapter 30 in the Fusion Reference Manual.

Multiplicative and Additive Bias

Shadows are essentially textures applied to objects in the scene that occasionally result in “fighting.”
Z-fighting results when portions of an object that should be receiving shadows instead render over the
top of the shadow because they effectively exist in the same exact location in 3D space.

Results of shadow map Z-fighting (top), and the

corrected shadow shown using Biasing (bottom)

Chapter 25 3D Compositing Basics 572

Two Biasing sliders in the Shadows group of Spotlight parameters work by adding a small depth offset
to move the shadow away from the surface it is shadowing, eliminating the Z-fighting. When too little
bias is added, the objects can self shadow themselves. When too much is added, the shadow can
become separated from the surface.

The Multiplicative and Additive Bias sliders, and the Non-Transmissive

Materials checkbox, all in the Spotlight Inspector controls

The goal is to adjust the Multiplicative Bias slider until the majority of the Z-fighting is resolved, and
then adjust the Additive Bias slider to eliminate the rest. The softer the shadow, the higher the bias will
probably have to be. You may even need to animate the bias to get a proper result for some
particularly troublesome frames.

Force All Materials Non-Transmissive

How light passes through a semi-transparent material plays an important role in determining the
appearance of the shadow an object casts. Normally, this transmittance behavior is defined in each
object’s Materials tab. However, selecting Force All Materials Non-Transmissive in the Spotlight
Inspector overrides this, causing the shadow map produced by the node to ignore
transmittance entirely.

Materials and Textures

To render a 3D scene, the renderer must take into account the shape of the object as well as its
appearance. The geometry of an object defines the shape of the object, while the material applied to
the object defines its appearance. Fusion provides a range of options for applying materials and
textures to geometry, so you can give your 3D objects the surface qualities you want.
Nodes that describe the geometry’s response to light are called illumination models. Blinn, Cook-
Torrance, Ward, and Phong are the included illumination models. These nodes are found in the 3D >
Material category of nodes in the Effects Library.
Most materials also accept textures, which are typically 2D images. Textures are used to refine the
look of an object further, by adding photorealistic details, transparency, or special effects. More
complex textures like bump maps, 3D textures, and reflection maps are also available in the 3D >
Texture category.
Materials can also be combined to produce elaborate and highly detailed composite materials.
Each node that creates or loads geometry into a 3D scene also assigns a default material. The default
material is the Blinn illumination model, but you can override this material using one of several nodes
that output a 3D material. Some of these materials provide a greater degree of control over how the
geometry reacts to light, providing inputs for diffuse and specular texture maps, bump mapping, and
environmental maps, which mimic reflection and refraction.

Chapter 25 3D Compositing Basics 573

Material examples from the bin

Material Components
All the standard illumination models share certain characteristics that must be understood.

Diffuse
The Diffuse parameters of a material control the appearance of an object where light is absorbed or
scattered. This diffuse color and texture are the base appearance of an object, before taking into
account reflections. The opacity of an object is generally set in the diffuse component of the material.

Alpha
The Alpha parameter defines how much the object is transparent to diffuse light. It does not affect
specular levels or color. However, if the value of alpha, either from the slider or a Material input from
the diffuse color, is very close to or at zero, those pixels, including the specular highlights, will be
skipped and disappear.

Opacity
The Opacity parameter fades out the entire material, including the specular highlights. This value
cannot be mapped; it is applied to the entire material.

Specular
The Specular parameters of a material control the highlight of an object where the light is reflected to
the current viewpoint. This causes a highlight that is added to the diffuse component. The more
specular a material is, the glossier it appears. Surfaces like plastics and glass tend to have white

Chapter 25 3D Compositing Basics 574

specular highlights, whereas metallic surfaces like gold have specular highlights that tend to inherit
their color from the material color.
Specularity is made up of color, intensity, and exponent. The specular color determines the color of
light that reflects from a shiny surface. Specular intensity is how bright the highlight will be.

Three spheres, left to right: diffuse only, specular only, and combined

The specular exponent controls the falloff of the specular highlight. The larger the value, the sharper
the falloff and the smaller the specular component will be.

Left to right: white, complimentary, and matching specular colors

Transmittance
When using the software renderer, the Transmittance parameters control how light passes through a
semi-transparent material. For example, a solid blue pitcher will cast a black shadow, but one made of
translucent blue plastic would cast a much lower density blue shadow. The transmittance parameters
are essential to creating the appearance of stained glass.

TIP: You can adjust the opacity and transmittance of a material separately. It is possible to
have a surface that is fully opaque yet transmits 100% of the light arriving upon it, so in a
sense it is actually a luminous/emissive surface.

Transmissive surfaces can be further limited using the Alpha and Color Detail control.

Chapter 25 3D Compositing Basics 575

Attenuation
The transmittance color determines how much color is passed through the object. For an object to
have fully transmissive shadows, the transmittance color must be set to to RGB = (1, 1, 1), which means
100% of green, blue, and red light passes through the object. Setting this color to RGB = (1, 0, 0) means
that the material will transmit 100% of the red arriving at the surface but none of the green or blue light.

Alpha Detail
When the Alpha Detail slider is set to 0, the non-zero portions of the alpha channel of the diffuse color
are ignored and the opaque portions of the object casts a shadow. If it is set to 1, the alpha channel
determines how dense the object casts a shadow.

NOTE: The OpenGL renderer ignores alpha channels for shadow rendering, resulting in a
shadow always being cast from the entire object. Only the software renderer supports alpha
in the shadow maps.

The following examples for Alpha Detail and Color Detail cast a shadow using this image. It is a
green-red gradient from left to right. The outside edges are transparent, and inside is a small semi-
transparent circle.

Alpha Detail set to 1; the alpha channel determines the density of the shadow

Alpha Detail set to 0; a dense-colored shadow results

Color Detail
Color Detail is used to color the shadow with the object’s diffuse color. Increasing the Color Detail
slider from 0 to 1 brings in more diffuse color and texture into the shadow.

Chapter 25 3D Compositing Basics 576

 TIP: The OpenGL renderer will always cast a black shadow from the entire object, ignoring
the color. Only the software renderer supports color in the shadow maps.

Color Detail set to 0; no color is visible in the shadow.

Saturation
Saturation will allow the diffuse color texture to be used to define the density of the shadow without
affecting the color. This slider lets you blend between the full color and luminance only.

Transmittance and Shadows

The transmittance of an object’s material plays an important role in determining the appearance of the
shadow it casts. Normally, the transmittance behavior is defined in each object’s Materials tab as
explained above. However, selecting Force All Materials Non-Transmissive in the Spotlight Inspector
overrides this, causing the shadow map produced by the spotlight to ignore transmittance entirely.

Illumination Models
Now that you understand the different components that make up a material or shader, we’ll look at
them more specifically. Illumination models are advanced materials for creating realistic surfaces like
plastic, wood, or metal. Each illumination model has advantages and disadvantages, which make it
appropriate for particular looks. An illumination model determines how a surface reacts to light, so
these nodes require at least one light source to affect the appearance of the object. Four different
illumination models can be found in the Nodes > 3D > Material menu.

Illumination models left to right: Standard, Blinn, Phong, Cook-Torrance, and Ward.

Standard
The Standard material provides a default Blinn material with basic control over the diffuse, specular,
and transmittance components. It only accepts a single texture map for the diffuse component with the
alpha used for opacity. The Standard Material controls are found in the Material tab of all nodes that
load or create geometry. Connecting any node that outputs a material to that node’s Material Input will
override the Standard material, and the controls in the Material tab will be hidden.

Chapter 25 3D Compositing Basics 577

Blinn
The Blinn material is a general purpose material that is flexible enough to represent both metallic and
dielectric surfaces. It uses the same illumination model as the Standard material, but the Blinn material
allows for a greater degree of control by providing additional texture inputs for the specular color,
intensity, and exponent (falloff), as well as bump map textures.

Phong
The Phong material produces the same diffuse result as Blinn, but with wider specular highlights at
grazing incidence. Phong is also able to make sharper specular highlights at high exponent levels.

Cook-Torrance
The Cook-Torrance material combines the diffuse illumination model of the Blinn material with a
combined microfacet and Fresnel specular model. The microfacets need not be present in the mesh or
bump map; they are represented by a statistical function, Roughness, which can be mapped. The
Fresnel factor attenuates the specular highlight according to the Refractive Index, which can
be mapped.

Ward
The Ward material shares the same diffuse model as the others but adds anisotropic highlights, ideal
for simulating brushed metal or woven surfaces, as the highlight can be elongated in the U or V
directions of the mapping coordinates. Both the U and V spread functions are mappable.
This material does require properly structured UV coordinates on the meshes it is applied to.

Textures
Texture maps modify the appearance of a material on a per-pixel basis. This is done by connecting an
image or other material to the inputs on the Material nodes in the Node Editor. When a 2D image is
used, the UV mapping coordinates of the geometry are used to fit the image to the geometry, and
when each pixel of the 3D scene is rendered, the material will modify the material input according to
the value of the corresponding pixel in the map.

TIP: UV Mapping is the method used to wrap a 2D image texture onto 3D geometry. Similar to
X and Y coordinates in a frame, U and V are the coordinates for textures on 3D objects.

Texture maps are used to modify various material inputs, such as diffuse color, specular color, specular
exponent, specular intensity, bump map, and others. The most common uses of texture maps is the
diffuse color/opacity component.

The Fast Noise texture used to control the

roughness of a Cook-Torrance material

A node that outputs a material is frequently used, instead of an image, to provide other shading
options. Materials passed between nodes are RGBA samples; they contain no other information about
the shading or textures that produced them.

Chapter 25 3D Compositing Basics 578

The Texture2D node is used to translate
a texture in the UV space of the object, as
well as set the filtering and wrap mode

Composite Materials
Building complex materials is as easy as connecting the output of a Material node to one of the
Material inputs of another Material or Texture node. When a Material input is supplied just as with
a 2D image, its RGBA values are used per pixel as a texture. This allows for very direct compositing
of shaders.
For instance, if you want to combine an anisotropic highlight with a Blinn material, you can take the
output of the Blinn, including its specular, and use it as the diffuse color of the Ward material. Or, if you
do not want the output of the Blinn to be relit by the Ward material, you can use the Channel Boolean
material to add the Ward material’s anisotropic specular component to the Blinn material with a greater
degree of control.

Combining an anisotropic highlight with a Blinn

material using the Channel Boolean material

Reflections and Refractions

Environment maps can be applied with the Reflect material in the 3D > Material category. This node
can be used to simulate reflections and refractions on an object. Reflections are direct-bounce light
that hits an object, while refractions simulate the distortion of light seen through semi-
translucent surfaces.
The reflections and refractions use an environment mapping technique to produce an approximation
that balances realistic results with greater rendering performance. Environment maps assume an
object’s environment is infinitely distant from the object and rendered into a cubic or spherical texture
surrounding the object.
The Nodes > 3D > Texture > Cube Map and Sphere Map nodes can be used to help create
environment maps, applying special processing and transforms to create the cubic or spherical
coordinates needed.

Chapter 25 3D Compositing Basics 579

Sphere map example

To produce reflections with real-time interactive feedback at a quality level appropriate for production
environment maps, you make some trade-offs on functionality when compared with slower but
physically accurate raytraced rendering. Environment-mapped reflections and refractions do not
provide self-reflection or any other kind of interaction between different objects. In particular, this
infinite distance assumption means that objects cannot interact with themselves (e.g., the reflections
on the handle of a teapot will not show the body of the teapot). It also means that objects using the
same cube map will not inter-reflect with each other. For example, two neighboring objects would not
reflect each other. A separate cube map must be rendered for each object.
The Reflect node outputs a material that can be applied to an object directly, but the material does not
contain an illumination model. As a result, objects textured directly by the Reflect node will not
respond to lights in the scene. For this reason, the Reflect node is usually combined with the Blinn,
Cook-Torrance, Phong, or Ward nodes.

Reflection
Reflection outputs a material making it possible to apply the reflection or refraction to other materials
either before or after the lighting model with different effects.

A Blinn material connected to a A Reflect is connected to the Diffuse

background material input of the Color component of the Blinn, causing the
Reflect This causes the reflection reflection to be multiplied by the diffuse
to be added to the Blinn output color and modulated by the lighting

Refraction
Refraction occurs only where there is transparency in the background material, which is generally
controlled through the Opacity slider and/or the alpha channel of any material or texture used for the
Background Material Texture input. The Reflect node provides the following material inputs:

Chapter 25 3D Compositing Basics 580

 – Background Material: Defines both the opacity for refraction and the base color for reflection.
– Reflection Color Material: The environment reflection.
– Reflection Intensity Material: A multiplier for the reflection.
– Refraction Tint Material: The environment refraction.
– Bump Map Texture: Normal perturbing map for environment reflection/refraction vectors.

Working with reflection and refraction can be tricky. Here are some techniques to make it easier:
– Typically, use a small amount of reflection, between 0.1 and 0.3 strength. Higher values are used
for surfaces like chrome.
– Bump maps can add detail to the reflections/refractions. Use the same bump map in the
Illumination model shader that you combine with Reflect.
– When detailed reflections are not required, use a relatively small cube map, such
as 128 x 128 pixels, and blur out the image.
– The alpha of refracted pixels is set to 1 even though the pixels are technically transparent.
Refracted pixels increase their alpha by the reflection intensity.
– If the refraction is not visible even when a texture is connected to the Refraction Tint Material
input, double-check the alpha/opacity values of the background material.

Bump Maps
Bump mapping helps add details and small irregularities to the surface appearance of an object. Bump
mapping modifies the geometry of the object or changes its silhouette.

Split screen of a sphere—half with bump map, half without

To apply a bump map, you typically connect an image containing the bump information to the
BumpMap node. The bump map is then connected to the Bump input of a Material node. There are
two ways to create a bump map for a 3D material: a height map and a bump map.

Image connected to a BumpMap connected to a CookTorrance material node

Using a Height Map

A height map is an image where the value of a pixel represents the height. It is possible to select
which color channel is used for bump creation. White means high and black means low; however, it is
not the value of a pixel in the height map that determines the bumpiness, but rather how the value
changes in the neighborhood of a pixel.

Chapter 25 3D Compositing Basics 581

Using a Bump Map
A bump map is an image containing normals stored in the RGB channels.

TIP: Normals are generated by 3D modeling and animation software as a way to trick the eye
into seeing smooth surfaces, even though the geometry used to create the models uses only
triangles to build the objects.

Normals are 3 float values (nx, ny, nz) whose components are in the range [–1, +1]. Because you can
store only positive values in Fusion’s integer images, the normals are packed from the range [–1, +1] to
the range [0, 1] by multiplying by 0.5 and adding 0.5. You can use Brightness Contrast or a Custom
node to do the unpacking.
If you were to connect a bump map directly to the bump map input of a material, it will result in
incorrect lighting. Fusion prevents you from doing this, however, because Fusion uses a different
coordinate system for doing the lighting calculation. You first must use a BumpMap that expects a
packed bump map or height map and will do the conversion of the bump map to work correctly.
If your bump mapping doesn’t appear correct, here are a few things to look for:
– Make sure you have the nodes connected correctly. The height/bump map should connect into a
BumpMap and then, in turn, should connect into the bump map input on a material.
– Change the precision of the height map to get less banding in the normals. For low frequency
images, float32 may be needed.
– Adjust the Height scale on the BumpMap. This scales the overall effect of the bump map.
– Make sure you set the type to HeightMap or BumpMap to match the image input. Fusion cannot
detect which type of image you have.
– Check to ensure High Quality is on (right-click in the transport controls bar and choose High
Quality from the contextual menu). Some nodes like Text+ produce an anti-aliased version in High
Quality mode that will substantially improve bump map quality.
– If you are using an imported normal map image, make sure it is packed [0–1] in RGB and that it is in
tangent space. The packing can be done in Fusion, but the conversion to tangent space cannot.

Projection Mapping
Projection is a technique for texturing objects using a camera or projector node. This can be useful for
texturing objects with multiple layers, applying a texture across multiple separate objects, projecting
background shots from the camera’s viewpoint, image-based rendering techniques, and much more.
There are three ways to do projection mapping in Fusion.

Using the Projector/Camera Tool to Project Light

When lighting is enabled, a Camera 3D or Projector 3D can act as a light with all the lighting features.
When Camera Projection is enabled or you use a projector, you can choose whether the projection
behaves like a spotlight or an ambient light; however, alpha channels cannot be projected.
Overlapping projections add together like any other light node. An internal clipping plane (at around
0.01 distance from camera) limits how close the projector or camera can get to the receivers
of the projection.

Chapter 25 3D Compositing Basics 582

Camera node used for a projection map

Project a Texture onto a Catcher Material

If you do not want to work with light sources, you can use the projector or camera as a texture
projector. To work without lighting, a catcher is required in order to receive the texture and apply it to a
material. Only objects using this material will receive the projection. This offers some advantages, like
the projection of alpha channels, and texturing other channels like specular color or roughness. If the
software renderer is used, overlapping projections can be combined in various ways (mean, median,
blend, and so on) via the Catcher node. When using the OpenGL renderer, one catcher per projector
is used, and the results can be combined using another material. Similar to the Light Projection
technique, an internal clipping plane (at around 0.01 distance from camera) limits how close the
projector/camera can get to the projection receivers.

Camera projection used with a Catcher node (example from an older version of Fusion)

Chapter 25 3D Compositing Basics 583

Project Using the UVMap Node
This mode requires a camera and a UVMap3D node downstream of the objects to which the texture is
being projected. In the Inspector, when the UVMap Map mode is set to Camera, it gathers the
information from the camera and creates new UVs for the input objects, which are used for texturing.
Because the UVs are stored in the vertices of the mesh, the object must be tessellated sufficiently.
Textures are assigned to the object like any other texturing technique. The UVs can be locked to the
vertices at a chosen frame using the Ref Time slider. This locking only works as long as vertices are
not created, destroyed, or reordered (e.g., projection locking will not work on particles because they
get created/destroyed, nor will they work on a Cube3D when it’s subdivision level slider animated).

TIP: Projected textures can be allowed to slide across an object. If the object moves relative
to the Projector 3D, or alternatively, by grouping the two together with a Merge3D, they can
be moved as one and the texture will remain locked to the object.

In the following section of a much larger composition, an image (the Loader1 node) is projected into 3D
space by mapping it onto five planes (Shape3D nodes renamed ground, LeftWall, RightWall, Building,
and Background), which are positioned as necessary within a Merge3D node to apply reflections onto
a 3D car to be composited into that scene.

Excerpt of a composition that’s projecting an image of a street scene into 3D space

The output of the Merge3D node used to assemble those planes into a scene is then fed to a UV Map
node, which in conjunction with a Camera3D node correctly projects all of these planes into 3D space
so they appear as they would through that camera in the scene. Prior to this UVMap projection, you
can see the planes arranged in space at left, where each plane has the scene texture mapped to it. At
right is the image after the UVMap projection, where you can see that the scene once again looks
“normal,” with the exception of a car-shaped hole introduced to the scene.

Five planes positioning a street scene in 3D space in preparation for UV Projection (left), and the UV Map
node being used to project these planes so they appear as through a camera in the scene (right)

Chapter 25 3D Compositing Basics 584

However, this is now a 3D scene, ready for a digital car to be placed within it, receiving reflections and
lighting and casting shadows into the scene as if it were there.

The new 3D scene casting reflections and lighting onto

a 3D car, and receiving shadows caused by the car

Geometry
There are five nodes used for creating geometry in Fusion. These nodes can be used for a variety of
purposes. For instance, the Image Plane 3D is primarily used to place image clips into a 3D scene,
while the Shapes node can add additional building elements to a 3D set, and Text 3D can add three-
dimensional motion graphics for title sequences and commercials. Although each node is covered in
more detail in the “3D Nodes” chapter, a summary of the 3D creation nodes is provided below.

Cube 3D
The Cube 3D creates a cube with six inputs that allow mapping of different textures to each of the
cube’s faces.

Image Plane 3D
The Image Plane 3D is the basic node used to place a 2D image into a 3D scene with an automatically
scaled plane.

Shape 3D
This node includes several basic primitive shapes for assembling a 3D scene. It can create planes,
cubes, spheres, cylinders, cones, and toruses.

Text 3D
The Text 3D is a 3D version of the Text+ node. This version supports beveling and extrusion but does
not have support for the multi-layered shading model available from Text+.

Particles
When a pRender node is connected to a 3D view, it will export its particles into the 3D environment.
The particles are then rendered using the Renderer3D instead of the Particle renderer. For more
information, see Chapter 113, “Particle Nodes” in the DaVinci Resolve Reference Manual or Chapter 52
in the Fusion Reference Manual.

Chapter 25 3D Compositing Basics 585

Common Visibility Parameters
Visibility parameters are found in the Controls tab of most 3D geometry-producing nodes, exposed
via a disclosure control. These parameters let you control object visibility in the viewers and in the
final render.

A 3D geometry node’s visibility parameters

Visible
If the Visibility checkbox is not selected, the object will not be visible in a viewer, nor will it be
rendered into the output image by a Renderer3D. A non-visible object does not cast shadows. This is
usually enabled by default, so objects that you create are visible in both the viewers and final renders.

Unseen by Cameras
If the Unseen by Cameras checkbox is selected, the object will be visible in the viewers but invisible
when viewing the scene through a camera, so the object will not be rendered into the output image by
a Renderer3D. Shadows cast by an Unseen object will still be visible.

Cull Front Face/Back Face

Use these options to cull (exclude) rendering of certain polygons in the geometry. If Cull Back Face is
selected, all polygons with normals pointing away from the view will not be rendered and will not cast
shadows. If Cull Front Face is selected, all polygons with normals pointing away from the view will
likewise be excluded. Selecting both checkboxes has the same effect as deselecting the Visible
checkbox.

Ignore Transparent Pixels in Aux Channels

For any piece of geometry, the Renderer3D rejects transparent pixels in the auxiliary image channels.
The reason this is the default is to prevent aux channels (e.g., normals, Z-channel, UVs) from filling in
where there should be blank space or full transparency. For example, suppose in post you want to add
some fog to the rendered image. If you had fully transparent geometry in the foreground affecting the
Z-channel, you would get incorrect fog rendering. By deselecting this checkbox, the transparency will
not be considered and all the aux channels will be filled for all the pixels. This could be useful if you
wanted to replace texture on a 3D element that is fully transparent in certain areas with a texture that is
transparent in different areas; it would be useful to have the whole object set aux channels (in
particular UVs).

Adding FBX Models

The Filmbox FBX format is a scene interchange format that facilitates moving 3D scene information
from one application to another. Fusion’s FBX format support extends model import support to other
3D files such as Collada and OBJ.

Importing an FBX Scene

To import an entire FBX scene, you add an FBXMesh3D node to your node tree. After being prompted
to choose a scene or object file, Fusion imports it to create a composition with the same lights,
cameras, materials, and geometry found in an FBX file.

Chapter 25 3D Compositing Basics 586

An imported model, via the FBXMesh3D node

FBX Scene Import Dialog

The FBX Mesh node is used to import mesh geometry from an FBX file. The first texture applied to a
mesh will also be imported, if available.
Since different 3D applications use different units to measure their 3D scenes, the imported geometry
may be enormous compared to the rest of the scene, because Fusion treats its scale of measurement
as equal to its own system. For example, if your 3D application defaults to using millimeters as its
scale, an object that was 100 millimeters in size will import as a massive 100 units.
You can use the Size slider in the FBX Mesh Inspector parameters to reduce the scale of such files to
something that matches Fusion’s 3D scene.

FBX Exporter
You can export a 3D scene from Fusion to other 3D packages using the FBX Exporter node. On
render, it saves geometry, cameras lights, and animation into different file formats such as .dae or .fbx.
The animation data can be included in one file, or it can be baked into sequential frames. Textures and
materials cannot be exported.

Using Text3D
The Text3D node is probably the most ubiquitous node employed by motion graphics artists looking
to create titles and graphics from Fusion. It’s a powerful node filled with enough controls to create
nearly any text effect you might need, all in three dimensions. This section seeks to get you started
quickly with what the Text3D node is capable of. For more information, see Chapter 90, “3D Nodes” in
the DaVinci Resolve Reference Manual or Chapter 29 in the Fusion Reference Manual.

Chapter 25 3D Compositing Basics 587

Assembling Text Objects
Each Text3D node is a self-contained scene within which each character of text is an individual object.
Because of this, the ideal way to combine numerous text objects that you might want to animate or
style independently from one another is to connect as many Text3D objects as you want to be able to
independently animate or style to one or more Merge3D nodes.

Merging multiple text objects to create an intricately styled scene

TIP: If you click the Text icon in the toolbar to create a Text3D node, and then you click it
again while the Text3D node you just created is selected, a Merge3D node is automatically
created and selected to connect the two. If you keep clicking the Text icon, more Text3D
nodes will be added to the same selected Merge3D node.

Entering Text
When you select a Text3D node and open the Inspector, the Text tab shows a “Styled Text” text entry
field at the very top into which you can type the text you want to appear onscreen. Below, a set of
overall styling parameters are available to set the Font, Color, Size, Tracking, and so on. All styling you
do in this tab affects the entire set of text at once, which is why you need multiple text objects if you
want differently styled words in the same scene.

The text entry and styling parameters in the Text tab

Chapter 25 3D Compositing Basics 588

Near the bottom of the Text tab are the Extrusion parameters, available within a disclosure control.

The Extrusion parameters near the bottom of the Text tab

By default, all text created with the Text3D node is flat, but you can use the Extrusion Style, Extrusion
Depth, and various Bevel parameters to give your text objects thickness.

Unextruded text (left), and Extruded text (right)

Positioning and Transforming Text

By default, every new Text3D node is positioned at 0, 0, 0, so when you add multiple Text3D nodes,
they’re all in the same place. Fortunately, every Text3D node has built-in transform controls in the
Transform tab.

Text3D nodes also have Transform parameters built-in

Chapter 25 3D Compositing Basics 589

Additionally, selecting a Text3D node exposes all the onscreen transform controls discussed
elsewhere in this chapter. Using these controls, you can position and animate each text object
independently.

Repositioned text objects to create a title sequence

Combining Text3D nodes using Merge3D nodes doesn’t just create a scene; it also enables you to
transform your text objects either singly or in groups:
– Selecting an individual Text3D node or piece of text in the viewer lets you move that one text
object around by itself, independently of other objects in the scene.
– Selecting a Merge3D node exposes a transform control that affects all objects connected to that
Merge3D node at once, letting you transform the entire scene.

Layout Parameters
The Layout tab presents parameters you can use to choose how text is drawn: on a straight line, a
frame, a circle, or a custom spline path, along with contextual parameters that change depending on
which layout you’ve selected (all of which can be animated).

Text using two different layouts

“Sub” Transforms
Another Transform tab (which the documentation has dubbed the “Sub” Transform tab) lets you apply a
separate level of transform to either characters, words, or lines of text, which lets you create even
more layout variations. For example, choosing to Transform by Words lets you change the spacing
between words, rotate each word, and so on. You can apply simultaneous transforms to characters,
words, and lines, so you can use all these capabilities at once if you really need to go for it. And, of
course, all these parameters are animatable.

Chapter 25 3D Compositing Basics 590

Transforming individual words in two different ways

Shading
The Shading tab lets you shade or texture a text object using standard Material controls.

Shading controls for text objects

Fog 3D and Soft Clipping

The Fog3D node helps to create atmospheric depth cues.

Split screen with and without fog

Chapter 25 3D Compositing Basics 591

The Fog3D node works well with depth of field and antialiasing supported by the OpenGL renderer.
Since it is not a post-processing node (like the VolumeFog node found in the Nodes > Position menu
or Fog node in Nodes > Deep Pixel), it does not need additional channels like Position or Z-channel
color. Furthermore, it supports transparent objects.
The SoftClip node uses the distance of a pixel from the viewpoint to affect opacity, allowing objects to
gradually fade away when too close to the camera. This prevents objects from “popping off” should
the camera pass through them. This is especially useful with particles that the camera may be
passing through.
Geometry nodes such as the Shape3D node use a Matte Objects checkbox to enable masking out
parts of the 3D scene. Effectively, everything that falls behind a matte object doesn’t get rendered.
However, matte objects can contribute information into the Z-channel and the Object ID channel,
leaving all other channels at their default values. They do not remove or change any geometry; they
can be thought of as a 3D garbage matte for the renderer.

Circle shape used as a Matte

object to see the floor

Matte Object Parameters

Opening the Matte disclosure control reveals the Is Matte option, which when turned on enables two
more options.

Matte parameters in the Shape3D node;

enabling Is Matte reveals additional options

Chapter 25 3D Compositing Basics 592

Is Matte
Located in the Controls tab for the geometry, this is the main checkbox for matte objects. When
enabled, objects whose pixels fall behind the matte object’s pixels in Z do not get rendered.

Opaque Alpha
When the Is Matte checkbox is enabled, the Opaque Alpha checkbox is displayed. Enabling this
checkbox sets the alpha value of the matte object to 1. Otherwise the alpha, like the RGB, will be 0.

Infinite Z
When the Is Matte checkbox is enabled, the Infinite Z checkbox is displayed. Enabling this checkbox
sets the value in the Z-channel to infinite. Otherwise, the mesh will contribute normally to the
Z-channel.
Matte objects cannot be selected in the viewer unless you right-click in the viewer and choose 3D
Options > Show Matte Objects in the contextual menu. However, it’s always possible to select the
matte object by selecting its node in the node tree.

Material and Object IDs

Most nodes in Fusion that support effect masking can use Object ID and Material ID auxiliary channels
to generate a mask. The parameters used to accomplish this are found in the Common Controls tab of
each node.

Material ID parameters in a Shape3D node’s Inspector controls

The Material ID is a value assigned to identify what material is used on an object. The Object ID is
roughly comparable to the Material ID, except it identifies objects and not materials.
Both the Object ID and Material ID are assigned automatically in numerical order, beginning with 1. It is
possible to set the IDs to the same value for multiple objects or materials even if they are different.
Override 3D offers an easy way to change the IDs for several objects. The Renderer will write the
assigned values into the frame buffers during rendering, when the output channel options for these
buffers are enabled. It is possible to use a value range from 0 to 65534. Empty pixels have an ID of 0,
so although it is possible to assign a value of 0 manually to an object or material, it is not advisable
because a value of 0 tells Fusion to set an unused ID when it renders.

Object ID for ground plane and object

set to the same numeric value

Chapter 25 3D Compositing Basics 593

World Position Pass
The World Position Pass, or WPP, is a render pass generated from 3D applications. Each pixel is
assigned the XYZ position where the pixel was generated in the world coordinates. So if the face from
which the pixel was derived in the scene sits at (0,0,0), the resulting pixel will have a Position value of
(0,0,0). If we visualize this as RGB, the pixel will be black. If a face sits at (1,0,0) in the original scene, the
resulting RGB pixel will be red. Due to the huge range of possible positions in a typical 3D scene, and
7⁄8 of those possible positions containing negative coordinates, the Position channel is always
rendered in 32-bit float.

A World Position Pass rendering of a scene with its center at (0,0,0) The actual image is on the top left

3D Scene Input
Nodes that utilize the World Position channel are located under the Position category. VolumeFog and
Z to WorldPos require a camera input matching the camera that rendered the Position channels, which
can either be a Camera3D or a 3D scene containing a camera. Just as in the Renderer3D, you can
choose which camera to use if more than one are in the scene. The VolumeFog can render without a
camera input from the Node Editor if the world space Camera Position inputs are set to the correct
value. VolumeMask does not use a camera input. Nodes that support the World Position Pass, located
under the Position category, offer a Scene input, which can be either a 3D Camera or a 3D scene
containing a camera.
There are three Position nodes that can take advantage of World Position Pass data.
– Nodes > Position > Volume Fog
– Nodes > Position > Volume Mask
– Nodes > Position > Z to World
– The “Dark Box”

Empty regions of the render will have the Position channel incorrectly initialized to (0,0,0). To get the
correct Position data, add a bounding sphere or box to your scene to create distant values and allow
the Position nodes to render correctly.

Chapter 25 3D Compositing Basics 594

Without a bounding mesh to generate Position values, the fog fills in the background incorrectly

Point Clouds
The Point Cloud node is designed to work with locator clouds generated from 3D tracking software.
3D camera tracking software, such as SynthEyes and PF Track, will often generate hundreds or even
thousands of tracking points. Seeing these points in the scene and referencing their position in 3D and
screen space is important to assist with lining up live action and CG, but bringing each point in as an
individual Locator3D would impact performance dramatically and clutter the node tree.

Point cloud in the viewer

The Point Cloud node can import point clouds written into scene files from match moving or 3D
scanning software.

To import a point cloud, do the following:

1 Add the PointCloud3D node to your composition.
2 Click the Import Point Cloud button in the Control panel.
3 Browse to the scene file and select a cloud to import from the scene.

The entire point cloud is imported as one object, which is a significantly faster approach.

Chapter 25 3D Compositing Basics 595

Finding, Naming, and Publishing Points
Many 3D trackers allow for the naming of individual tracking points, as well as setting tracking points
on points of interest. The Point Cloud 3D will quickly find these points and publish them. A published
point in the cloud can be used to drive the animation of other parameters.

To find a point in the point cloud, do the following:

1 Right-click anywhere within a viewer.
2 Choose Find from the Point Cloud’s submenu in the contextual menu.
3 Type the name of the point and click OK.

Finding a point cloud using the viewer contextual menu

If a point that matches the name you entered is found, it will be selected in the point cloud and
highlighted yellow.

TIP: The Point Cloud Find function is a case-sensitive search. A point named “tracker15” will
not be found if the search is for “Tracker15”.

Chapter 25 3D Compositing Basics 596

Renaming a Point in the Cloud
You can use the Point Cloud contextual menu to rename a selected point. This works only for a single
point. A group of points cannot be renamed.

Publishing a Point
If you want to use a point’s XYZ positions for connections to other controls in the scene, you can
publish the point. This is useful for connecting objects to the motion of an individual tracker. To publish
a point, right-click it and choose Publish from the contextual menu.

Publishing a point using the viewer contextual menu

Chapter 25 3D Compositing Basics 597

Chapter 26

3D Camera Tracking
This chapter presents an overview of using the Camera Tracker node and the workflow
it involves. Camera tracking is used to create a virtual camera in Fusion’s 3D
environment based on the movement or a live-action camera in a clip. You can then
use the virtual camera to composite 3D models, text, or 2D images into a live-action
clip that has a moving camera.

For more information on other types of tracking in Fusion, see Chapter 83, “Using the
Tracker Node” in the DaVinci Resolve Reference Manual or Chapter 22 in the Fusion
Reference Manual.

Contents
Introduction to Tracking  599
What Is 3D Camera Tracking?  599
How Camera Tracking Works  599
The Camera Tracking Workflow  600
Clips That Don’t Work Well for Camera Tracking  600
Outputting from the Camera Tracker  601
2D View  602
3D View  602
Auto-Tracking in the Camera Tracker  603
Increasing Auto-Generated Tracking Points  603
Masking Out Objects  604
Matching the Live-Action Camera  606
Running the Solver  606
How Do You Know When to Stop?  607
Using Seed Frames  608
Cleaning Up Camera Solves  608

Chapter 26 3D Camera Tracking 598

Exporting a 3D Scene for Efficiency  611
Unalign the 3D Scene Transforms  611
Setting the Ground Plane  612
Setting the Origin  612
Setting the Scale  613
Realign the Scene  613
Viewing the Exported Results  613

Introduction to Tracking
Tracking is one of the most useful and essential techniques available to a compositor. It can roughly be
defined as the creation of a motion path from analyzing a specific area in a clip over time. Fusion
provides a variety of different tracking nodes that let you analyze different kinds of motion.
Each tracker type has its own chapter in this manual. This chapter covers the tracking techniques with
the Camera Tracker node.

What Is 3D Camera Tracking?

Camera tracking is used for match moving, and it’s a vital link between 2D scenes and 3D scenes,
allowing compositors to integrate 3D CGI elements into live-action clips. The Camera Tracker node
calculates the path of a live-action camera and generates a virtual camera in 3D space. This virtual
camera is intended to be identical to the actual camera that shot the scene, not only in terms of motion
but in matching the lens focal length as well. The calculated position and movement of the virtual
camera is central to realistically compositing 3D elements with live action.

An example of 3D elements integrated in a live-action scene

How Camera Tracking Works

Camera tracking begins by tracking the movement of fixed features from one frame to the next. To put
it another way, camera tracking algorithms follow features that are “nailed to the set.” Objects in the
scene that move independently of the camera movement in the shot, such as cars driving or people
walking, cause poor tracks, so masks can be used to restrict the features that are tracked in order to
improve the results. Additionally, it is helpful to provide specific camera metadata, such as the sensor
size and the focal length of the lens. This information guides the scene reconstruction calculation,
called a solver, toward generating a more accurate virtual camera.

Chapter 26 3D Camera Tracking 599

The Camera Tracker’s purpose is to create a 3D animated camera and point cloud of the scene.
A point cloud is a large group of points generated by the solver that roughly recreates the 3D positions
of the tracked features in a scene. The point cloud can then be used as a guide when integrating other
2D or 3D elements alongside live-action features.

The Camera Tracking Workflow

Camera tracking has two main phases:
1 Tracking, which is the analysis of a scene.
2 Solving, which calculates the virtual 3D scene.

Once you complete these steps, an animated camera and point cloud are exported from the Inspector
into a 3D composite. The Camera Tracker encompasses this complete workflow within one tool. Five
tabs at the top of the Inspector are roughly laid out in the order in which you’ll use them. These five
tabs are:
– Track: Used to track a clip.
– Camera: Configures the basic Camera parameters.
– Solve: Calculates the 3D placement of the 2D tracking points and reconstructs the camera.
– Export: Generates a Camera 3D node, a Point Cloud node, and a 3D scene in the node tree.
– Options: Used to customize the look of the onscreen overlays.

The Camera Tracker tab

Clips That Don’t Work Well for Camera Tracking

Even though the Camera Tracker is somewhat automatic, it sometimes needs your help. If you can
identify potential issues before you even track or solve the shot, you can save yourself much time.
Certain types of clips will cause more significant problems for camera tracking than others. Some are
fixable, while for others you just have to admit defeat and figure out another solution. Here is a list of
the types of shots to look out for, as they can be big headaches for camera tracking:
– Lack of depth: Camera tracking requires parallax in a clip in order to work. You must be able to
identify objects further away and objects that are nearer as the camera moves. If everything is at
the same distance from the camera, there is no way to calculate depth. In this case, it’s better to
skip the Camera Tracker node and find another solution.
– Locked-off shots: If the camera does not move, there is no way to calculate which objects are
closer and which are nearer. Again, don’t spend too much time in this situation; it is better to skip
the Camera Tracker node and find another solution.

Chapter 26 3D Camera Tracking 600

 – Tripod pans: Similar to a locked-off shot, there is no way to calculate which objects are closer and
which are nearer from a pan that remains centered on a locked-off tripod. Skip the Camera Tracker
node and find another solution.
– No detail: Clips like green screens without tracking markers lack enough detail to track. If you are
lucky enough to be involved in the shooting of these types of shots, including tracker markers
makes it much easier to get a good track. Without detail, camera tracking will fail and you will need
to find a more manual solution.
– Motion blur: Fast camera motion or slow shutter speeds can introduce motion blur, which will
make it difficult to find patterns to track. It’s worth trying shots like these to see if there are enough
details to get a good solve, but know when give up and turn to another solution.
– Rolling shutter: CMOS-based cameras sometimes introduce distortion due to the shutter
capturing different lines at slightly different times. This distortion can create significant problems
for camera tracking. Sometimes it is possible to create motion vectors with the Optical Flow node
to create new in-between frames without the wobble distortion of the rolling shutter. Then you can
use the corrected image to connect to the Camera Tracker.
– Parallax issues: When objects at different distances in a shot overlap in the frame, the overlapping
area can be misinterpreted as a corner. Having a tracker assigned to an overlapping angle like this
will cause errors as the parallax starts to shift and the overlapping area slides. This can be solved
in Fusion by removing that tracker before running the solver.
– Moving objects: It’s difficult to capture a shot where objects in the clip do not move. People, cars,
animals, or other object may move in and out of a shot. These objects move independent of the
camera movement and must be eliminated or they will cause solving errors. You can fix these
issues by masking out objects that are not “nailed to the set.” The masks are then connected to
the Track Mask input on the Camera Tracker node.

TIP: Some shots that cannot be tracked using Fusion’s Camera Tracker can be performed in
dedicated 3D camera-tracking software like 3D Equalizer and PF Track. Camera tracking data
from these applications can then be imported in the Camera3D node in Fusion.

Outputting from the Camera Tracker

Unlike most Fusion nodes, the Camera Tracker node has two outputs:
– The primary output is a 2D view used when you are setting up the Track, refining the camera, and
performing your initial solve.
– There is also a 3D output used after your initial solve for viewing the camera path and point cloud
in 3D space. This view can be helpful when you are refining tracks to increase the accuracy of the
solve and aligning your ground plane. It can be used simultaneously with the 2D output in side-by-
side views.

Note that the selection of tracks in the 2D view and their corresponding locators (in the point cloud) in
the 3D view are synchronized. There are also viewer menus available in both the 2D and 3D views to
give quick control of the functionality of this tool.

Chapter 26 3D Camera Tracking 601

2D View
The 2D view is the primary display for the node. Viewing the node displays the image being tracked as
well as overlay tracker markers and their motion paths. A dedicated toolbar gives you access to the
common features used to track and solve a clip.

The Camera Tracker 2D output with toolbar and auto-track points

3D View
The second output of the Camera Tracker node displays a 3D scene. To view this, connect this 3D
output to a 3D Transform or Merge 3D node and view that tool.

After solving, connecting the second output of the Camera

Tracker node to a Merge 3D displays the point cloud in 3D

Chapter 26 3D Camera Tracking 602

After an initial solve, the 3D output displays the point cloud and the camera, along with the image
connected to it. Selecting points displays the Camera Tracker toolbar above the viewer, which gives
control of various functions, such as renaming, deleting, and changing the colors of points in the
point cloud.

3D output of a point cloud and a solved camera path

Auto-Tracking in the Camera Tracker

Tracking is the term used to describe the task of observing or analyzing a sequence of frames (or clip).
The Camera Tracker node must take into account the movement of the source footage before it can
determine the location and movement of the virtual camera. The Camera Tracker tool automatically
searches for features that are high-contrast patterns within the clip and assigns trackers to those
features. Having a wide distribution of tracking points across the frame and having points with long
durations results in the best track.

Increasing Auto-Generated Tracking Points

Unlike the Tracker node, setting tracking points is entirely automatic in the Camera Tracker, but the
Detection Threshold and Minimum Feature Separation sliders let you adjust the criteria by which
tracking points are found. Lowering these parameters lets you increase the number of tracking points
that will be found. This can be useful if the scene has too few points, which will prevent the solver from
generating an accurate camera and point cloud. However, make these adjustments with care, since
adding too many points may generate redundant trackers that slow down the entire process with
minimal benefit.

Previewing Tracking Points

You can view the tracking points currently generated on the clip by viewing the Camera Tracker node
and turning on the Preview AutoTrack Locations checkbox. This causes green tracking points to be
displayed in the viewer as you play through the clip. Using this preview, you can decide whether you
need to make adjustments to the Detection Threshold or Minimum Feature Separation to increase or
decrease the number of tracking points that are found automatically.

Chapter 26 3D Camera Tracking 603

Green tracker marks are added automatically to the features in an image

Bi-Directional Tracking
When performing a track, you can enable the Bidirectional Tracking checkbox, which first tracks
forward from the start of the clip, and then tracks a second pass in reverse. This two-pass approach
can potentially extend the duration of any given point by re-analyzing points initially identified in the
forward pass. There is very little reason not to have this enabled unless you are very short on time.
Bidirectional tracking takes longer, but it’s usually worth it, and the process is reasonably quick
considering the benefit.

Choosing a Tracking Algorithm

There are three available choices for the algorithm to use when tracking. The three options in the
New Track Defaults section of the Inspector include:
– Optical Flow: Usually your best choice, unless you have a great deal of criss-crossing
objects in a clip.
– Tracker: A good second choice when Optical Flow can’t be used due to motion estimation errors
like criss-crossing objects.
– Planar: Mostly used in simpler clips, where the majority of the image consists of planar surfaces
such as the facades of buildings.

Masking Out Objects

When tracking a clip, the Camera Tracker automatically generates trackers on feature details.
However, not all features that stand out in a clip are appropriate for camera tracking. You only want to
track features that are “nailed to the set.” In other words, objects that move independently of the
camera motion, like moving cars and people, cause inaccuracies when camera tracking. You must
eliminate these types of objects from the analysis.
The primary way of avoiding these problem areas is by masking. You connect a mask to the Camera
Tracker node’s Track Mask input to identify areas of a scene that the Camera Tracker can analyze. For
example, if you have a clip of an airport runway along a shoreline, the waves of the water and moving
clouds in the sky must be masked since they move independently of the camera.

Chapter 26 3D Camera Tracking 604

When creating a mask, the fixed areas of the image to be analyzed for tracking should be
encompassed in the white portion of the mask. All moving objects that need to be ignored should be
encompassed in the black portion. The mask should then be attached to the Camera Tracker Track
Mask input.

Masks used to omit the moving clouds and waves from being tracked by the Camera Tracker

By doing this, the tracker ignores the waves of the water and moving clouds. Unlike drawing a mask
for an effect, the mask in this case does not have to be perfect. You are just trying to identify the rough
area to occlude from the tracking analysis.

The original image to be tracked (left), and the occlusion mask of the clouds and water (right)

TIP: If there’s a lot of motion in a shot, you can use the Tracker or Planar Tracker nodes to
make your occlusion mask follow the area you want to track. Just remember that, after using
the PlanarTracker or PlanarTransform node to transform your mask, you need to use a Bitmap
node to turn it back into a mask that can be connected to the Camera Tracker node’s Track
Mask input.

Chapter 26 3D Camera Tracking 605

Matching the Live-Action Camera
Once you have completed tracking, the next stage of this workflow requires the controls in the
Camera tab. This is where you define the actual camera used on set, primarily the film gate size and
focal length. This information should have been logged on the set to make available for post-
production. When using camera-original media, you can sometimes locate this information in the file
metadata.

To locate camera metadata, do the following:

– If you are using DaVinci Resolve, select the MediaIn node with the camera clip, open the Metadata
Editor, and view the Camera metadata preset.
– If you are using Fusion Studio, display the metadata subview from the viewer toolbar.

If the actual values are not known, try a best guess. The solver attempts to find a camera near these
parameters, and it helps the solver by giving parameters as close to the live action as possible. The
more accurate the information you provide, the more accurate the solver calculation. At a minimum, try
to at least choose the correct camera model from the Film Gate menu. If the film gate is incorrect, the
chances that the Camera Tracker correctly calculates the lens focal length become very low.

The Camera tab in the Camera Tracker tool

Unlike the Track and Solve tabs, the Camera tab does not include a button at the top of the Inspector
that executes the process. There is no process to perform on the Camera tab once you configure the
camera settings. After you set the camera settings to match the live-action camera, you move to the
Solve tab.

Running the Solver

The next step in this workflow involves the controls found in the Solve tab. Solving is a compute-
intensive process in which the Camera Tracker analyzes the currently existing tracks to create a 3D
scene. It generates a virtual camera that matches the live action and a point cloud consisting of 3D
locators that recreate the tracked features in 3D space. The analysis is based on parallax in the frame,
which is the perception that features closer to the camera move quicker than features further away.
This is much like when you look out the side window of a car and can see objects in the distance
move more slowly than items near the roadside.
The trackers found in the Track phase of this workflow have a great deal to do with the success or
failure of the solver, making it critical to deliver the best set of tracking points from the very start.
Although the masking you create to occlude objects from being tracked helps to omit problematic
tracking points, you almost always need to further filter and delete poor quality tracks in the Solver
tab. That’s why, from a user’s point of view, solving should be thought of as an iterative process.

Chapter 26 3D Camera Tracking 606

To solve a camera’s motion:
1 Click the Solve button to run the solver.
2 Filter out and delete poor tracks.
3 Rerun the solver.

The Solver tab after it has run and produced an

average solve error of 04367 pixels

How Do You Know When to Stop?

At the end of the solve process, an Average Solve Error (sometimes called a reprojection error)
appears at the top of the Inspector. This is the crucial value that tells you how well the calculation has
gone. A good Average Solve Error for HD content is below 1.0.
You can interpret a value of 1.0 as a pixel offset; at any given time, the track could be offset by 1 pixel.
The higher the resolution, the lower the solve error should be. If you are working with 4K material, your
goal should be to achieve a solve error below 0.5.

Tips for Solving Camera Motion

When solving camera movement, it’s important to provide accurate live-action camera
information, such as focal length and film gate size, which can significantly improve the
accuracy of the camera solve. For example, if the provided focal length is too far away from
the correct physical value, the solver can fail to converge, resulting in a useless solution.
Additionally, for the solver to accurately triangulate and reconstruct the camera and point
cloud, it is important to have:

– A good balance of tracks across objects at different depths, with not too many tracks
in the distant background or sky (these do not provide any additional perspective
information to the solver).
– Tracks distributed evenly over the image and not highly clustered on a few objects or
one side of the image.
– The track starts and ends staggered over time, with not too many tracks ending on
the same frame.

Chapter 26 3D Camera Tracking 607

Using Seed Frames
The solver works by first constructing a partial solution between two seed frames. These seed frames
are selected automatically. However, automatic selection adds time to the process. The time is
reported in the solve summary at the top of the Inspector once you run the solver. You can select your
own seed frames to speed the process and potentially get a better solve on trickier clips. The solver
uses these seed frames to create an intermediate solution and then extends this forward and
backward for the duration of the clip.
Selecting appropriate seed frames is not necessarily recommended unless you have some experience
with camera tracking. Keeping the default Auto Select Seed Frames checkbox enabled in the Solve
Options section of the Solver tab selects the best frames in most cases. However, you can disable the
checkbox and use the Seed Frame 1 and Seed Frame 2 slider to select frames you believe achieve
better results.
When choosing seed frames, it is important to satisfy two main characteristics:
– Seed frames should have lots of tracks in common.
– Seed frames should be far apart in perspective.

Sometimes There’s Nothing You Can Do

Some shots that do not have enough camera motion to triangulate feature locations cannot be
reconstructed with any useful accuracy. Ensuring that a shot is camera-trackable begins on
set, with proper usage of track markers, and by ensuring that camera moves have enough
perspective shifts for the solver to glean useful data from.

Cleaning Up Camera Solves

Sometimes the first solve will be good enough. Other times, it may take a few hours of cleaning up
tracks to get a good solve, and sometimes it is impossible. With experience, one gets a feel for which
tracks should be deleted and which should be kept, and which shots will be easy, difficult, or
impossible to solve.
Be aware that deleting too many tracks can cause the Average Solve Error to increase, as the solver
has too little information to work with. In particular, if there are fewer than eight tracks on any frame,
mathematically there is not enough information to solve the clip. However, it is strongly recommended
to use a lot more than eight tracks to get a robust and accurate solve.

IMPORTANT
If you are not familiar with camera tracking, it may be tempting to try to directly edit the
resulting 3D splines in the Spline Editor in order to improve a solved camera’s motion path.
This option should be used as an absolute last resort. It’s preferable, instead, to modify the 2D
tracks being fed into the solver.

Chapter 26 3D Camera Tracking 608

How to Judge Track Accuracy
The automatic color coding of tracking markers makes deleting false or poor tracks easier. After the
solver runs, each tracker is assigned a solve error color that indicates which 3D locators match their
2D tracking points well, and which match up poorly.
– Green: Good. Tracked very well.
– Yellow: Moderate confidence. Usually an acceptable track.
– Orange: Low Confidence. May be acceptable in some situations.
– Red: No Confidence. The tracks have not solved well.

Hovering the pointer over any tracking point displays a large metadata tooltip that includes the solve
error for the point. For a more visual representation of the accuracy, you can enable the display of
3D locators in the viewer by clicking the Reprojection Locators button in the viewer toolbar.

After a solve, the Camera Tracker toolbar can display Reprojection locators

When the tracking points are converted into a point cloud by the solver, it creates 3D reprojection
locators for each tracking point. These Reprojection locators appear as small X marks near the
corresponding tracking point. The more the two objects overlap, the lower the solve error.

Reprojection locators displayed with tracking points, and tooltip

The goal when filtering the trackers is to remove all red tracker marks and keep all the green marks.
Whether you decide to keep both the yellow and orange or just the yellow is more a question of how
many marks you have in the clip. You produce a better solve if you retain only the yellow marks;
however, if you do not have enough marks to calculate the 3D scene, you will have to keep some of
the better orange marks as well.

Chapter 26 3D Camera Tracking 609

 Tips for What to Keep and What to Delete
Understanding what false tracks look like, and then manually cleaning the track data to
reduce it to a precise set of clear tracks, will result in a more accurate solve. When cleaning up
any track—particularly yellow and orange color coded tracks—keep the following in mind:

– Keep all tracks with motion that’s completely determined by the motion of the
live‑action camera.
– Delete tracks on moving objects or people and tracks that have parallax issues.
– Delete tracks that are reflected in windows or water.
– Delete tracks of highlights that move over a surface.
– Delete tracks that do not do a good job of following a feature.
– Delete tracks that follow false corners created by the superposition of
foreground and background layers.
– Consider deleting tracks that correspond to locators that the solver has reconstructed
at an incorrect Z-depth.

Deleting Tracks
You can manually delete tracks in the viewer or use filters to select groups of tracks. When deleting
tracks in the viewer, it is best to modify the viewer just a bit to see the tracks more clearly. From the
Camera Tracker toolbar above the viewer, clicking the Track Trails button hides the trails of the
tracking points. This cleans up the viewer to show points only, making it easier to make selections.
At the right end of the toolbar, clicking the Darken Image button slightly darkens the image, again
making the points stand out a bit more in the viewer.

Hiding trails and darkening the viewer can

make it easier to see and select poor tracks

To begin deleting poor-quality tracks, you can drag a selection box around a group of tracks you want
to remove and then either click the Delete Tracks button in the Camera Tracker toolbar or press
Command-Delete.
You can hold down the Command key to select discontiguous tracking marks that are not near each
other. If you accidentally select tracks you want to keep, continue holding the Command key and drag
over the selected tracks to deselect them.
When deleting tracks, take note of the current Average Solve Error at the top of the Inspector and then
rerun the solver. It is better to delete small groups of tracks and then rerun the solver than to delete
one or two large sections. As mentioned previously, deleting too many tracks can have adverse
effects and increase the Average Solve Error.

Using Filters to Delete Problem Tracks

The Solve tab includes filters that can be used to select groups of similar tracks by track length, track
error, and solve error. These can be used to quickly select and delete poorly performing tracks that
may be misleading to the resulting camera, leaving a concise list of accurate tracks.

Chapter 26 3D Camera Tracking 610

 Tracks can be selected using filters and deleted
using the Operations On Selected Tracks buttons

For instance, it is generally best to run the solver using tracks with longer durations. Since shorter
tracks tend to be less accurate when calculating the camera, you can remove them using the Filter
section in the Inspector.
Increasing the Minimum Track Length parameter sets a threshold that each tracker must meet. Tracks
falling below the threshold appear red. You can then click the Select Tracks Satisfying Filters button to
select the shorter tracks and click Delete from the Options section in the Inspector.

Exporting a 3D Scene for Efficiency

The Camera Tracker saves all its 2D tracks into the composition, sometimes resulting in a rather large
file on disk. If you are dealing with a large clip with many 2D tracks over a long duration, the saved
composition can reach over a gigabyte in size. Using a Camera Tracker node in a composition can
make it cumbersome to load and operate. While it is possible to use the Camera Tracker node directly
to composite via the 3D output, you’ll achieve better performance by exporting. Once the quality of
the solve is satisfactory, the Export tab can generate a “low memory” alternative by producing
individual Camera 3D, Point Cloud, Ground Plane, and 3D Renderer nodes.
Before you can export the 3D scene, you must provide a bit more information about it. You’ll do this
using controls found in the Export tab. Cameras do not include tiltmeters, so clips do not contain
metadata that indicates how the camera is tilted or oriented. This is critical information when recreating
the virtual camera. It is also useful to determine the location for the center of this 3D scene. The Export
tab provides various translation, rotation, and scale controls to set these options.

Unalign the 3D Scene Transforms

By default, the Export tab is set to Aligned in the 3D Scene Transform section. The Aligned setting
locks the orientation and scale of the 3D scene to prevent accidentally changing it. So, before you can
set the ground plane and origin location, you must change the Camera Tracker to be unaligned using
the 3D Scene Transform menu in the Export tab. After you have gone through the Export settings and
configured them how you want, you must set the menu back to Aligned before exporting.

Set the 3D Scene Transform menu to

Unaligned before setting the ground plane
Chapter 26 3D Camera Tracking 611
Setting the Ground Plane
The Camera Tracker has no idea if the camera is on its side or tilted in some way. So, it is up to you to
indicate where the ground plane is in a clip. After choosing Unaligned from the 3D Scene Transform
menu, you can begin identifying the ground plane.

Drag a selection box around marks that represent the ground

To set the ground plane, do the following:

1 Move to a frame with lots of green 3D locators where you can see a large part of the ground.
2 In the viewer, drag a selection box around the marks located on the ground in the clip.
3 In the Inspector Orientation section, click the Set from Selection button.

TIP: In some cases, the clip you are tracking may not have the ground in the frame.
If necessary, you can set the Selection menu to XY, which indicates you are selecting points
on a wall.

Setting the Origin

When it comes time to add and position new objects into your 3D scene, you can make it easier by
setting the origin or center location.

To set the origin of the 3D scene, do the following:

1 Move to a frame that clearly shows the area you want to select as the center of the scene.
2 In the viewer, either select a single point or drag a selection box around a few marks located
where you want to position the center of the 3D scene.
3 In the inspector Origin section, click the Set from Selection button.

Chapter 26 3D Camera Tracking 612

Setting the Scale
The Camera Tracker has no idea of the size of the 3D scene, so the scale parameter is used to scale
the scene output. This makes it possible to match the scale of two or more clips.

Realign the Scene

Before exporting the scene from the Camera Tracker, you must set the 3D scene Transform menu
back to Aligned. Now you’re ready to export.

Viewing the Exported Results

Clicking the Export button at the top of the Inspector creates a functional 3D scene with five new
nodes automatically added to the node tree.
– Camera 3D
– Point Cloud
– Ground Plane
– Merge 3D
– Camera Tracker Renderer (3D Renderer)

Five nodes created as a result of exporting from the Camera Tracker

To work with the 3D scene, you can select the Merge 3D and load it into one of the viewers, and then
select the Camera Tracker Renderer and load that into a second viewer.

Chapter 26 3D Camera Tracking 613

Viewing the Merge 3D shows the point cloud, ground plane, and camera

When the Merge 3D is selected, a toolbar above the viewer can add 3D test geometry like an
image plane or cube to verify the precision of the 3D scene and camera. You can then connect
actual 3D elements into the Merge 3D as you would any manually created 3D scene. The point
cloud can help align and guide the placement of objects, and the CameraTracker Renderer is a
Renderer 3D node with all the same controls.

Use the point cloud to accurately place different elements into a 3D scene

At this point, there is no need for the Camera Tracker node unless you find that you need to rerun
the solver. Otherwise, you can save some memory by deleting the Camera Tracker node.

Chapter 26 3D Camera Tracking 614

Chapter 27

Particle Systems
This chapter is designed to give you a brief introduction to the creation of fully
3D particle systems, one of Fusion’s most powerful features. Once you understand
these basics, for more Information on each particle system node that’s available,
see Chapter 113, “Particle Nodes” in the DaVinci Resolve Reference Manual or
Chapter 52 in the Fusion Reference Manual.

Contents
Introduction to Particle Systems  616
Anatomy of a Simple Particle System  617
Particle System Distribution  619
Particle Nodes Explained by Type  620
Emitters  620
Forces  621
Compositing  621
Rendering  621
Example Particle Systems  622

Chapter 27 Particle Systems 615

Introduction to Particle Systems
Particle systems are computer simulations that use customizable rules to automatically generate and
animate large numbers of elements to simulate smoke, dust, fire, leaves, sparks, or any other animated
system of shapes. As Fusion is a full-featured 3D compositing environment, particle systems can be
created in 2D or 3D, which makes them incredibly flexible and capable of producing all kinds of visual
effects or abstract animated content for use in motion graphics.

A 3D particle system, also created entirely within Fusion

The three most fundamental nodes required for creating particle systems are found on the toolbar.
As with the 3D nodes to the right, these are arranged, from left to right, in the order in which they must
be connected to work, so even if you can’t remember how to hook up a simple particle system, all you
need to do is click the three particle system nodes from left to right to create a functional
particle system.

The pEmitter, pMerge, and pRender particle

system nodes available from the toolbar

However, these three nodes are only the tip of the iceberg. Opening the Particle category in the
Effects Library reveals many, many particle nodes designed to work together to create increasingly
complex particle interactions.

A sample of the nodes available in the

Particles bin of the Effects Library

Chapter 27 Particle Systems 616

All particle nodes begin with the letter “p,” and they’re designed to work together to produce
sophisticated effects from relatively simple operations and settings. The next section shows different
ways particle nodes can be connected to produce different effects.

Anatomy of a Simple Particle System

The simplest particle system you can create is a pEmitter node connected to a pRender node. The
pEmitter node includes the core controls for creating various kinds of particles in different ways, while
the pRender node is required to render a 2D or 3D result that can be composited with other scenes
within your composition.

The minimum node tree required to create a simple particle system

If your needs are more complicated, you can combine two or more pEmitter nodes using a pMerge
node (the particle system version of a Merge node), to create compound particle systems where
multiple types of particles combine with one another to create a result.

Compositing two pEmitter nodes to create a compound particle

system, combining two kinds of particles together

If you’re trying to create particle systems with more natural effects, you can add “forces” to each
emitter. These forces are essentially physics or behavioral simulations that automatically cause the
particles affected by them to be animated with different kinds of motion, or to be otherwise affected
by different objects within scenes.

Customizing the effect of pEmitter nodes using different forces to add complexity to the particle animation

You can also attach the following types of nodes to a pEmitter node to deeply customize a
particle system:
– Attach a 2D image to a pEmitter node to create highly customized particle shapes. Make sure your
image has an appropriate alpha channel.
– Attach a Shape3D or other 3D geometry node to a pEmitter node to create a more specific region
of emission (by setting Region to Mesh in the Region tab).

Chapter 27 Particle Systems 617

Customizing pEmitter nodes using mesh geometry to define regions and 2D images to define particle shape

The above examples assume that you’ll output 2D renders to combine into the rest of a
2D composition. However, because particle systems are fully 3D, you also have the option of
outputting your particle system in such a way as to be used from within other 3D scenes
in your composition.

Connecting a particle system to a Merge3D node so the particles

are subject to lighting and shadows within a 3D scene

The Output Mode of the pRender node, at the very top of the controls exposed in the Inspector, can
be set to either 2D or 3D, depending on whether you want to combine the result of the particle system
with 2D layers or with objects in a 3D scene.

Choosing whether a particle system’s output is 2D

or 3D in the pRender node’s Inspector controls

If you connect a pRender node to a Merge3D node, the Output Mode is locked to 3D, meaning that 3D
geometry is output by the pRender node for use within the Merge3D node’s scene. This means that
the particles can be lit, they can cast shadows, and they can interact with 3D objects within that scene.

Chapter 27 Particle Systems 618

The result of using a particle system within a 3D scene

NOTE: Once you set the pRender node to either 2D or 3D and make any change to the
nodes in the Inspector, you cannot change the output mode.

Particle System Distribution

To adjust the distribution of particles being emitted, select the pEmitter node to expose its controls in
the Inspector, then open the Velocity controls in the Controls tab, and use the Angle, Angle Variance,
Angle Z, and Angle Z Variance controls to adjust the direction and width over which particles are
emitted. All these controls can be animated.

A pEmitter node’s Velocity Angle and

Angle Variance controls let you adjust the
direction and width of particle distribution

Particle systems can be positioned and rotated by loading the pEmitter nodes that generate particles
into a viewer and using the onscreen 3D position and Rotation controls provided to move the particle
system around.

Chapter 27 Particle Systems 619

A pEmitter node loaded into the viewer with the rotation onscreen controls enabled

Alternatively, you can use the controls of the pEmitter’s Region tab in the Inspector to adjust
Translation, Rotation, and Pivot. All these controls can be animated.

A pEmitter node’s Region controls open in the Inspector

Particle Nodes Explained by Type

This section introduces the four types of particle system nodes available in the Effects Library.

Emitters
pEmitter nodes are the source of all particles. Each pEmitter node can be set up to generate a single
type of particle with enough customization so that you’ll never create the same type of particle twice.
Along with the pRender node, this is the only other node that’s absolutely required to create a
particle system.

Chapter 27 Particle Systems 620

pEmitter nodes have four parameters tabs:
– Controls: The primary controls governing how many particles are generated (Number), how long
they live (Lifespan), how fast they move (Velocity) and how widely distributed they are (Angle and
Angle Variance), their rotation (Rotation Mode with X, Y, and Z controls), and whether there’s spin
(Spin X, Y, and Z controls). For each parameter of particle generation, there’s an accompanying
Variance control that lets you make that parameter less uniform and more natural by introducing
random variation.
– Sets: This tab contains settings that affect the physics of the particles emitted by the node. These
settings do not directly affect the appearance of the particles. Instead, they modify behaviors such
as velocity, spin, quantity, and lifespan.
– Style: While the Controls tab has a simple control for choosing a color for particles, the Style
tab has more comprehensive controls including color variance and Color Over Life controls.
Additionally, size controls including Size Over Life, fade controls, and blur controls let you create
sophisticated particle animations with a minimum of adjustments, while Merge controls give you
an additional level of control over how overlapping particles combine visually. A set of controls at
the bottom lets you choose how animated effects are timed.
– Region: The Region tab lets you choose what kind of geometric region is used to disperse
particles into space and whether you’re emitting particles from the region’s volume or surface.
The Winding Rule and Winding Ray Direction controls determine how the mesh region will handle
particle creation with geometric meshes that are not completely closed, as is common in many
meshes imported from external applications. Tweaking these last parameters is common when
using imported mesh geometry as a region for emitting particles, since even geometry that
appears closed will frequently appear to “leak” particles thanks to improperly welded vertices.

Forces
Many of the particle nodes found in the Particles bin of the Effects Library are “forces” that enhance a
particle simulation by simulating the effect of various forces acting upon the particles generated by
an emitter.
Some forces, including pDirectionalForce, pFlock, pFriction, pTurbulence, and pVortex, are rules that
act upon particles without the need for any other input. These are simply “acts of nature” that cause
particles to behave in different ways.
Other forces, such as pAvoid, pBounce, pFollow, and pKill, work in conjunction with 3D geometry in a
scene such as shapes or planes to cause things to happen when a particle interacts or comes near
that geometry. Note that some of the particles described previously can also use geometry to direct
their actions, so these two categories of forces are not always that clear-cut.

Compositing
The pMerge node is a simple way to combine multiple emitters so that different types of particles work
together to create a sophisticated result. The pMerge node has no parameters; you simply connect
emitters to it, and they’re automatically combined.

Rendering
The pRender node is required whether you’re connecting a particle system’s output to a 2D Merge
node or to a Merge3D node for integration into a 3D scene. Along with the pEmitter node, this is the
only other node that’s absolutely required to create a particle system.
– Controls: The main controls that let you choose whether to output 2D or 3D image data, and
whether to add blur or glow effects to the particle systems, along with a host of other details
controlling how particles will be rendered.
– Scene: These controls let you transform the overall particle scene all at once.

Chapter 27 Particle Systems 621

 – Grid: The grid is a helpful, non-rendering guide used to orient 2D particles in 3D space. The grid
is never output in renders. The width, depth, number of lines, and grid color can be set using the
controls found in this tab.
– Image: Controls the output of the pRender node, with controls over the process mode, resolution,
and color space settings of the output.

Example Particle Systems

The Templates category in the Inspector in the Fusion page of DaVinci Resolve or in the Bins window
in Fusion Studio includes over 20 different examples of particle systems creating a variety of effects.
One of the best ways of learning how to create and customize particle systems is to open these and
investigate how they’re made.

Different particle system presets in the Templates category of the Bins window in Fusion Studio

Simply drag and drop any of the particle presets into the Node Editor, load the last node into the
viewer, and you’ll see how things are put together.

The Blowing Leaves preset from the Templates category

Chapter 27 Particle Systems 622

 PART 4

Advanced
Compositing
Techniques
Chapter 28

Optical Flow
and Stereoscopic
Nodes
This chapter covers the numerous stereoscopic and optical flow-based nodes
available in Fusion and their related workflows.

Contents
Overview  625
Stereoscopic Overview  625
Optical Flow Overview  626
Toolset Overview  626
Working with Aux Deep Channels  627
Optical Flow Workflows  628
OpticalFlow  628
TimeSpeed, TimeStretcher  628
SmoothMotion  628
Repair Frame, Tween  628
Advanced Optical Flow Processing  628
Stereoscopic Workflows  629
Stereo Camera  629
Stereo Materials  630
Disparity  631
NewEye, StereoAlign  631
DisparityToZ, ZToDisparity  631
Separate vs. Stack  631

Chapter 28 Optical Flow and Stereoscopic Nodes 624

Setting Up Stereo in the Node Editor  631
About the Disparity Channel  632
Viewing of Disparity and Vector Channels  633
Stereo and Optical Flow Best Practices  633
Semi-Transparent Objects  633
Motion Blur  633
Depth of Field  633
Where to Calculate Disparity and Optical Flow?  634
Cropping the Source  634
Nodes with Multiple Outputs  634
Picking from Aux Channels  634
Vector and Disparity Channels  635

Overview
Fusion includes 3D stereoscopic and optical flow-based nodes, which can work together or
independently of each other to create, repair, and enhance 3D stereoscopic shots.

Stereoscopic comp displayed in the viewers.

Stereoscopic Overview
All stereoscopic features are fully integrated into Fusion’s 3D environment. Stereoscopic images can
be created using a single camera, which supports eye separation and convergence distance, and a
Renderer 3D for the virtual left and right eye. It is also possible to combine two different cameras for a
stereo camera rig.
Stereoscopic nodes can be used to solve 3D stereoscopic shooting issues, like 3D rig misalignment,
image mirror polarization differences, camera timing sync issues, color alignment, convergence, and
eye separation issues. The stereo nodes can also be used for creating depth maps.

Chapter 28 Optical Flow and Stereoscopic Nodes 625

Optical Flow Overview
Optical Flow analyzes the motion in a clip and generates motion vectors between neighboring frames.
It generates X and Y vectors from the previous frame to the current frame (Back Vectors) and to the
next frame in sequence (Forward Vectors). Once calculated, optical flow data can be used by other
nodes to create smooth slow motion and variable retiming of clips, repair missing frames, and even
correct disparity in stereo 3D clips.

NOTE: The stereoscopic nodes in the Fusion page work independently of the stereoscopic
tools in the other DaVinci Resolve pages.

Toolset Overview
Here is an overview of the available nodes.

Optical Flow Nodes

– Optical Flow > OpticalFlow: Analyzes motion between neighboring frames in a sequence to
generate motion vectors, which can then be used by other nodes for retiming, motion blur, and
other effects.
– Miscellaneous > TimeSpeed: Retimes a clip at a constant speed using Flow Interpolation mode.
– Miscellaneous > TimeStretcher: Retimes a clip at variable speeds using Flow Interpolation mode.
– Optical Flow > RepairFrame: Generates a new frame using the motion vectors between two
neighboring frames.
– Optical Flow > SmoothMotion: Smoothes the color or aux channels using motion vectors.
– Optical Flow > Tween: Interpolates between two non-sequential images to generate a new frame.
– Color > CopyAux: Copies aux channels, including motion vectors, into RGBA more efficiently than
Channel Booleans.

Stereoscopic Nodes
– Stereo > Anaglyph: Combines stereo images to create a single anaglyph image for viewing.
– Stereo > Combiner: sStacks a separate stereo images into a single stacked pair,
so they can be processed together.
– Stereo > Disparity: Generates disparity between left/right images.
– Stereo > DisparityToZ: Converts disparity to Z-depth.
– Stereo > Global Align: Shifts each stereo eye manually to do basic alignment of stereo images.
– Stereo > NewEye: Replaces left and/or right eye with interpolated eyes.
– Stereo > Splitter: Separates a stacked stereo image into to left and right images.
– Stereo > StereoAlign: Adjusts vertical alignment, convergence, and eye separation.
– Stereo > ZToDisparity: Converts Z-depth to disparity.

Chapter 28 Optical Flow and Stereoscopic Nodes 626

Working with Aux Deep Channels
Certain image formats can contain channels other than RGBA color, called aux deep channels. Stereo
Disparity and OpticalFlow deal directly with auxiliary deep channels.

Aux channels supported in Fusion include:

– RGBA: These are the standard colors.
– Z: The eyespace Z coordinate is almost always negative because in eyespace, Fusion’s camera
sits at (0, 0, 0) looking down the Z-axis. Z values start at Z = 0 at the camera focal point and
progressively become more negative for objects deeper in the scene.
– Coverage: The percentage of the pixel covered by the frontmost pixel, used for antialiased
Z-compositing.
– Object ID: These are user-assigned integers to meshes.
– Material ID: These are user-assigned integers to materials.
– Texture Coords: Normalized texture coordinates stored as (u, v) pairs.
– Normal Vector: Normal vector (nx, ny, nz) where the components are typically in the range [–1, +1].
– Background Color: The color of the pixel if the frontmost layer were removed,
used for antialiased Z-compositing.
– Vector: The forward motion vector is an offset (vx, vy) that compares every pixel’s
position in one frame to the same pixel’s position in the next frame.
– Back Vector: The backward motion vector is an offset (vx, vy) that compares every
pixel’s position in one frame to the same pixel’s position in the previous frame.
– World Position: The position (wx, wy, wz) of the pixel in world coordinates.
– Disparity: An offset (dx, dy) that maps a pixel in the Left > Right or Right > Left frames.

Some extra channels are used by specific Fusion nodes.

For example:
– Merge can use the Z channel to perform a depth merge. If the Coverage and BackgroundColor
channels are present, it can do a better job on antialiased edges during the Z merge.
– Most image-processing nodes (e.g., BrightnessContrast) have options on their common controls
tab to limit their processing by MaterialID and ObjectID.
– The Fog and DepthBlur nodes make use of the Z channel.
– The Texture node makes use of the TexCoord channel.
– The Shader node makes use of the Normal channel.

There are a couple of ways to retrieve or generate those extra channels within Fusion.
For example:
– The Renderer3D node is capable of generating most of these channels.
– The OpticalFlow node generates the Vector and BackVector channels, and then TimeStretcher
and TimeSpeed can make use of these channels.
– The Disparity node generates the Disparity channels, and then DisparityToZ, NewEye, and
StereoAlign nodes can make use of the Disparity channels.
– The OpenEXR format can be used to import or export aux channels into Fusion by specifying a
mapping from EXR attributes to Fusion Aux channels using CopyAux.

Chapter 28 Optical Flow and Stereoscopic Nodes 627

Optical Flow Workflows
The Optical Flow analysis is a non real-time process, and depending on your computer, the clip’s
resolution, and the duration of the clip, it can take some time. Because of this, the general idea is that
you pre-generate the motion vectors, either by performing the analysis overnight or using a render
farm, and save results into an OpenEXR sequence. The Optical Flow toolset is designed around four
types of nodes that either generate, destroy, pass through, or construct the motion vectors.

OpticalFlow
The Optical Flow node generates the Vector and BackVector data. Typically, for optimal performance,
you connect the Optical Flow output to a Saver to save the image as OpenEXR files with the motion
vectors stored in an aux channel.

TimeSpeed, TimeStretcher
You can create smooth constant or variable slow-motion effects using the TimeSpeed or
TimeStretcher nodes. When Optical Flow motion vectors are available in the aux channel of an image,
enabling Flow mode in the TimeSpeed or TimeStretcher Interpolation settings will take advantage of
the Vector and BackVector channels. For the Flow mode to work, there must be either an upstream
OpticalFlow node generating the hidden channels or an OpenEXR Loader bringing these channels in.
These nodes use the Vector/BackVector data to do interpolation on the motion channel and then
destroy the data on output since the input Vector/BackVector channels are invalid. For more detail on
TimeSpeed or TimeStretcher, see Chapter 110, “Miscellaneous Nodes” in the DaVinci Resolve
Reference Manual and Chapter 49 in the Fusion Reference Manual.

SmoothMotion
SmoothMotion can be used to smooth the Vector and BackVector channels or smooth the disparity in
a stereo 3D clip. This node passes through, modifies, or generates new aux channels, but does not
destroy them.

Repair Frame, Tween

The Tween and Repair Frame nodes are different from standard optical flow nodes because they have
the OpticalFlow analysis and motion vector generation built in. Tween will compare two frames and
create an in-between frame, which is good for recreating a missing or flawed frame. Repair Frame will
look at frames on either side of the current frame and repair scratches, dust marks, and so on.
Because these nodes work with flow values between non-sequential frames, they cannot use the
optical flow stored in the input image’s Vector/BackVector channels, but rather must regenerate the
flow of each frame, do their processing, and then destroy the flow channels. This being the case, these
nodes are computationally expensive. For more detail on Tween or Repair Frame, see Chapter 111,
“Optical Flow” in the DaVinci Resolve Reference Manual and Chapter 50 in the Fusion
Reference Manual.

Advanced Optical Flow Processing

The Optical Flow, Repair Frame, and Tween nodes include a faster GPU-based Optical Flow algorithm.
When you add the Optical Flow, Repair Frame, or Tween node to a comp, the Inspector includes a
Method drop-down menu where you can choose Advanced to enable the GPU-based algorithm. This
Advanced method is the same Optical Flow algorithm used in other DaVinci Resolve pages.
By choosing Classic from the Method drop-down menu in the Inspector, you can use the older
CPU-based algorithm to maintain compatibility with comps created in previous versions. This method
may also be better suited for some Stereo3D processing.

Chapter 28 Optical Flow and Stereoscopic Nodes 628

Stereoscopic Workflows
Disparity is the difference between the left and right image. The Disparity map is used by nodes to
align and massage the stereo pair of images.

The Disparity node analyzes a stereo pair of

images and generates an X&Y disparity map.

The workflow is to load a left and right stereo image pair and process those in the Disparity node.
Once the Disparity map is generated, other nodes can process the images.

TIP: When connectng stereo pairs in the node tree, make sure that the left and right images
are connected to the left and right inputs of the Disparity node.

Disparity generation, like Optical Flow, is computationally expensive, so the general idea is that you
can pre-generate these channels, either overnight or on a render farm, and save them into an
EXR sequence.
The toolset is designed around this philosophy.

Stereo Camera
There are two ways to set up a stereoscopic camera. The common way is to simply add a Camera 3D
and adjust the eye separation and convergence distance parameters.

Chapter 28 Optical Flow and Stereoscopic Nodes 629

Stereoscopic cameras can be done with a single
camera or two connected cameras.

The other way is to connect another camera to the RightStereoCamera input port of the Camera 3D.
When viewing the scene through the original camera or rendering, the connected camera is used for
creating the right-eye content.

Stereo Materials
Using the Stereo Mix material node, it is possible to assign different textures per eye.

Material viewer showing stereoscopic material.

Chapter 28 Optical Flow and Stereoscopic Nodes 630

Disparity
The Disparity node does the heavy lifting of generating disparity maps. This generates the Disparity
channel and stores it in the hidden aux channels of their output image.

NewEye, StereoAlign
NewEye and StereoAlign use and destroy the Disparity channel to do interpolation on the
color channel.
The hidden channels are destroyed in the process because, after the nodes have been applied, the
original Disparity channels would be invalid.
For these nodes to work, there must be either an upstream Disparity node generating the hidden
channels or an OpenEXR Loader bringing these channels in.

DisparityToZ, ZToDisparity
These nodes pass through, modify, or generate new aux channels, but do not destroy any.

TIP: If the colors between shots are different, use Color Corrector or Color Curves to do a
global alignment first before calculating the Disparity map. Feed the image you will change
into the orange input and the reference into the green input. In the Histogram section of the
Color Corrector, select Match, and also select Snapshot Match Time. In the Color Curves’
Reference section, select Match Reference.

Separate vs. Stack

Stereo nodes can work in Separate or Stack modes. When in Stack mode, the left/right eyes are
stacked horizontally or vertically, forming one image with double width or height, respectively.
The advantage to using Stack mode is that you do not have to have duplicate branches of the Node
Editor for the left and right eyes. As a consequence, you will see Stereo nodes with two inputs and two
outputs labeled as “Left” and “Right.”
When in Stack mode, the stack should be connected to the left eye input and the Left output should
be used for connecting further nodes. In Stack mode, the respective Right eye inputs and outputs
are hidden.

Setting Up Stereo in the Node Editor

The disparity generation is the first operation. This can be configured in the Node Editor in two
different ways.

Chapter 28 Optical Flow and Stereoscopic Nodes 631

Two stereoscopic workflows.

In the above example, the workflow on the right takes the left and right eye, generates the disparity,
and then NewEye is used to generate a new eye for the image right away.
The example on the left renders the frames with disparity to intermediate EXR images. These images
are then loaded back into Stereo nodes and used to create the NewEye images.
By using Render nodes to compute the disparity first, the later processing of the creative operations
can be a much faster and interactive experience.
Although not shown in the above diagram, it is usually a good idea to color correct the right eye to be
similar to the left eye before disparity generation, as this helps with the disparity-tracking algorithm.
The color matching does not need to be perfect—for example, it can be accomplished using the
“Match” option in a Color Corrector’s histogram options.

About the Disparity Channel

The Disparity channel stores the displacement vectors that match pixels in one eye to the other eye.
The left image’s Disparity channel will contain vectors that map left>right and the right image’s
Disparity channel will contain vectors that map right>left. For example:
(xleft, yleft) + (Dleft. x, Dleft. y) -> (xright, yright) (xright, yright) +
(Dright. x, Dright. y) -> (xleft, yleft)

You would expect for non-occluded pixels that Dleft = -Dright, although, due to the disparity
generation algorithm, this is only an approximate equality.

NOTE: Disparity stores both X and Y values because rarely are left/right images perfectly
registered in Y, even when taken through a carefully set up camera rig.

Both Disparity and Optical Flow values are stored as un-normalized pixel shifts. In particular, note that
this breaks from Fusion’s resolution-independent convention. After much consideration, this
convention was chosen so the user wouldn’t have to worry about rescaling the Disparity/Flow values
when cropping an image or working out scale factors when importing/exporting these channels to
other applications. Because the Flow and Disparity channels store things in pixel shifts, this can cause
problems with Proxy and AutoProxy. Fusion follows the convention that, for proxied images, these
channels store unscaled pixel shifts valid for the full-sized image. So if you wish to access the Disparity
values in a script or via a probe, you need to remember to always scale them by (image. Width/image.
OriginalWidth, image. Height/ image. OriginalHeight).

Chapter 28 Optical Flow and Stereoscopic Nodes 632

Viewing of Disparity and Vector Channels
Aux channels can be displayed directly in the viewers through the Channel viewer button’s menu.
The CopyAux node is used to copy those channels directly into the RGB channels for viewing or
further processing. The advantage of using the CopyAux node is that it does static normalization,
which reduces a lot of flicker that the viewer’s time-variant normalization causes. When viewing long
sequences of aux channels, the CopyAux node has the option to kill off aux channels and keep only
the current RGB channels, freeing up valuable memory so you can cache more frames.

TIP: Although you can use the Channel Booleans to copy any aux channel into RGBA, it
involves a few additional clicks when compared to CopyAux.

One thing to be aware of is that aux channels tend to consume a lot of memory. A float-32 1080p
image containing just RGBA uses about 32 MB of memory, but with all the aux channels enabled it
consumes around 200 MB of memory.

Stereo and Optical Flow Best Practices

How you create your composition, the images you are using, and the type of shot you are working on
can all have an impact on the success of the Disparity generation and Optical Flow analysis. Below,
we’ll look at some of the situations to be aware of and how you can avoid some pitfalls when dealing
with optical flow.

Semi-Transparent Objects
The Optical Flow and Disparity generation algorithms Fusion uses assume there is only one layer per
pixel when tracking pixels from frame to frame. In particular, transparent objects and motion blur will
cause problems. For example, a shot flying through the clouds with the semi-transparent clouds in the
foreground and a distant landscape background will confuse the Optical Flow/Stereo algorithms, as
they do not recognize overlapping objects with different motions. Usually the optical flow will end up
tracking regions of one object or the other. If the transparent object and the background are near the
same depth and consequently have the same disparity, then it is not a problem.

Motion Blur
Motion blur is also a serious problem for the reason explained in the previous point. The Disparity and
Optical Flow algorithms are unsure whether to assign a pixel in the motion blur to the moving object or
the background pixel. Because the algorithms used are global in nature, not only the vectors on the
motion blur will be wrong, but it will confuse the algorithm on regions close to the motion blur.

Depth of Field
Depth of field is also another problem related to the above two problems. The problem occurs when
you have a defocused foreground object over a background object that is moving (Optical Flow case)
or shifts between L/R (Stereo Disparity case). The blurred edges will confuse the tracking because
they can’t figure out that the edges are actually two separate objects.

Chapter 28 Optical Flow and Stereoscopic Nodes 633

Where to Calculate Disparity and Optical Flow?
Where you choose to generate optical flow or disparity in your composition can drastically affect
the results.
For example, if you have composited a lens flare in, it is better to compute OpticalFlow/Disparity
before that, since the semi-transparent lens flare will confuse the tracking algorithms.
If you are color correcting the left/right eyes to match or for deflickering, it is better to apply the
OpticalFlow/Disparity afterward, since it will be easier for the tracking algorithm to find matches if the
color matches between frames.
If you are removing lens distortion, think carefully about whether you want to do it before or after
Disparity computation. If you do it after, your Disparity map will also act as a lens distortion map,
combining the two effects as one.
As a general rule of thumb, it is best to use OpticalFlow/Disparity before any compositing operations
except an initial color matching correction and a lens distortion removal.

Cropping the Source

As a general tip, if you are cropping down your input images for any reason, it is probably better to
compute the optical flow or disparity before the crop and then afterward crop the flow/disparity along
with the color.
The reason is that flow/disparity matching works well when there is common pixel data to match in
both frames, but when there are pixels that show up in just one frame (or one eye), then the Disparity/
OpticalFlow nodes must make a guess and fill in the data. The biggest occlusions going from L <–> R
are usually pixels along the L/R edges of the images that get moved outside. This is similar for optical
flow when you have a moving camera.
Another thing to be aware of are black borders around the edges of your frames, which you should
crop away.

Nodes with Multiple Outputs

Many of the stereo nodes in the Fusion toolset have multiple outputs. This can cause some confusion
to new users. One particularly confusing thing is that when you drag a Stereo node to the view, it will
always display the left output. There is no way to view the right output without connecting another
node like BC (BrightnessContrast) to the right output and viewing that.

Picking from Aux Channels

Some nodes, like StereoAlign, allow one to drag pick from the Z or Disparity auxiliary channels. You
must pick from a node upstream of the StereoAlign, not from the output of the StereoAlign. If you try to
pick a disparity from the output of a StereoAlign node, you will get nothing because StereoAlign
consumes/destroys the Disparity aux channel (and even if it did not destroy the Disparity channel, you
would still be picking the wrong value since you would be picking from the aligned result).

The typical workflow for picking is:

1 View StereoAlign in the left view.
2 View the node upstream of StereoAlign in the right view.
3 Pick the Disparity value from the left eye in the right view.

Chapter 28 Optical Flow and Stereoscopic Nodes 634

Although this picking functionality does not operate any differently from normal picking of color
channels, this issue may cause some confusion. If it helps, the analogous workflow mistake with color
nodes would be a user trying to pick a gradient color for a Background node from a view showing the
Background node itself (you are trying to pick a color for a node from its own output).
Another issue that you need to be aware of is which eye you are picking. To avoid problems, it’s a
good idea to always pick from the left eye. The reason is that the Disparity channels for the left and
right eyes are different, and when you pick from a horizontal/vertical stereo stack, Fusion has no way
of knowing whether you picked the Disparity value from the left or right eye.
The above are not hard and fast rules; rather, they are guidelines to prevent foot shootings. If you
understood the above reasoning fully, you’ll realize there are exceptions, like picking disparity from the
left output of DisparityToZ and Z from the left/right output of ZToDisparity, where everything is okay.

Vector and Disparity Channels

The Vector and BackVector channels store the forward and reverse optical flow.
The Vector channel might be better named “forward vector” or “forward flow,” since the name “Vector”
to describe a channel is “not technically correct,” as the more mathematically-inclined user might
recognize that all the channels except the scalar channels Z/ID are technically “vector” channels. A
frames Vector aux channel will store the flow forward from the current frame to the next frame in the
sequence, and the BackVector aux channel will store the flow backward from the current frame to the
previous frame. If either the previous or next frames do not exist (either not on disk or the global range
of a Loader does not allow OpticalFlow to access them), Fusion will fill the corresponding channels
with zeros (transparent black).
The Disparity channel stores the displacement vectors that match pixels in one eye to the other eye.
The left image’s Disparity channel will contain vectors that map left > right and the right image’s
Disparity channel will contain vectors that map right > left.

For example:
(xleft, yleft) + (Dleft. x, Dleft. y) -> (xright, yright) (xright, yright) +
(Dright. x, Dright. y) -> (xleft, yleft)

You would expect for non-occluded pixels that Dleft = -Dright, although due to the disparity generation
algorithm, this is only an approximate equality. Note that Disparity stores both X and Y values because
rarely are left/right images perfectly registered in Y, even when taken through a carefully set up
camera rig.
Disparity and Optical Flow values are stored as un-normalized pixel shifts. In particular, note that this
breaks from Fusion’s resolution-independent convention. After much consideration, this convention
was chosen so the user wouldn’t have to worry about rescaling the Disparity/Flow values when
cropping an image or working out scale factors when importing/exporting these channels to other
applications. Because the Flow and Disparity channels store things in pixel shifts, this can cause
problems with Proxy and AutoProxy. The convention that Fusion follows is that, for proxied images,
these channels store unscaled pixel shifts valid for the full-sized image. So if you wish to access the
disparity values in a script or via a probe, you need to remember to always scale them by (image.
Width/image. OriginalWidth, image. Height/image. OriginalHeight).
When using Vector and BackVector aux channels, remember that all nodes expect these aux channels
to be filled with the flow between sequential frames.

Chapter 28 Optical Flow and Stereoscopic Nodes 635

More precisely, if you have sequence of three frames A, B, C, then:

B Vector will contain the flow B>C

B BackVector will contain the flow B>A

A Vector will contain the flow A>B

A BackVector is written with zeros as there is no frame before A

C Vector is written with zeros as there is no frame D to flow C>D

C BackVector will contain the flow C>B

When working with these channels, it is the user’s responsibility to follow these rules (or for clever
users to abandon them). Nodes like TimeStretcher will not function correctly since they still expect the
channels to contain flow forward/back by 1 frame.

NOTE: Currently DoD/RoI is not supported for all Fusion nodes.

Chapter 28 Optical Flow and Stereoscopic Nodes 636

 PART 5

Fusion
Page Effects
Chapter 29

3D Nodes
This chapter covers, in great detail, the nodes used for creating 3D composites.
The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Alembic Mesh 3D [ABC]  639 Renderer 3D [3RN]  691
Bender 3D [3BN]  642 Replace Material 3D [3RPL]  700
Camera 3D [3CM]  644 Replace Normals 3D [3RPN]  702
Cube 3D [3CB]  653 Replicate 3D [3REP]  704
Custom Vertex 3D [3CV]  655
Ribbon 3D [3RI]  710
Displace 3D [3DI]  660
Shape 3D [3SH]  712
Duplicate 3D [3DP]  662
Soft Clip [3SC]  715
FBX Exporter 3D [FBX]  667
Spherical Camera [3SC]  717
FBX Mesh 3D [FBX]  670
Text 3D [3TXT]  720
Fog 3D [3FO]  672
Transform 3D [3XF]  729
Image Plane 3D [3IM]  675
Triangulate 3D [3TRI]  732
Locator 3D [3LO]  677
Merge 3D [3MG]  679 UV Map 3D [3UV]  733

Override 3D [3OV]  681 Weld 3D [3WE]  736

Point Cloud 3D [3PC]  683 Modifier  739

Projector 3D [3PJ]  687 The Common Controls  740

Chapter 29 3D Nodes 638

Alembic Mesh 3D [ABC]

The Alembic Mesh 3D node

Alembic Mesh Node Introduction

At times, you may need to import 3D geometry from applications like Blender, Cinema4D, or Maya.
One of the formats you can use for importing 3D geometry is the Alembic file format. This file type is a
3D scene interchange format that contains baked animation with its geometry. In other words, it
eliminates the animation calculation times by embedding fixed, uneditable animation with 3D
geometry. The animation is typically embedded using a point cache, which saves the dynamic data
such as velocity after it has been calculated. Alembic objects can contain mesh geometry, cameras,
points, UVs, normals, and baked animation.
You can import Alembic files (.abc) into Fusion in two ways:
– Choose File > Import > Alembic Scene in Fusion or Fusion > Import > Alembic Scene in
DaVinci Resolve’s Fusion page.
– Add an AlembicMesh3D node to the Node Editor.

The first method is the preferred method; both Alembic and FBX nodes by themselves import the
entire model as one object. However, the Import menu breaks down the model, lights, camera, and
animation into a string of individual nodes. This makes it easy to edit and modify and use subsections
of the imported Alembic mesh. Also, transforms in the file are read into Fusion splines and into the
Transform 3D nodes, which get saved with the comp. Later, when reloading the comp, the transforms
are loaded from the comp and not the Alembic file. Fusion handles the meshes differently, always
reloading them from the Alembic file.
Arbitrary user data varies depending on the software creating the Alembic file, and therefore this type
of metadata is mostly ignored.

Alembic Import Dialog

An Alembic Import dialog is displayed once you select the file to import.

Alembic Import options

Chapter 29 3D Nodes 639

The top half of the Import dialog displays information about the selected file including the name of the
plug-in/application that created the Alembic file, the version of the Alembic software developer kit
used during the export, the duration of the animation in seconds, if available, and the frame rate(s)
in the file.
Various objects and attributes can be imported by selecting the checkboxes in the Import section.
– Hierarchy: When enabled, the full parenting hierarchy is recreated in Fusion using multiple
Transform 3D nodes. When disabled, the transforms in the Alembic file are flattened down into
the cameras and meshes. The flattening results in several meshes/cameras connected to a single
Merge node in Fusion. It is best to have this disabled when the file includes animation. If enabled,
the many rigs used to move objects in a scene will result in an equally large number of nodes in
Fusion, so flattening will reduce the number of nodes in your node tree.
– Orphaned Transforms: When the hierarchy option is enabled, an Orphaned Transforms setting
is displayed. Activating this Orphan Transforms setting imports transforms that parent a mesh or
camera. For example, if you have a skeleton and associated mesh model, the model is imported as
an Alembic mesh, and the skeleton as a node tree of Merge3Ds. If this is disabled, the Merge3Ds
are not created.
– Cameras: When enabled, importing a file includes cameras along with Aperture, Angles of View,
Plane of Focus, as well as Near and Far clipping plane settings. The resolution Gate Fit may be
imported depending on whether the application used to export the file correctly tagged the
resolution Gate Fit metadata. If your camera does not import successfully, check the setting for the
Camera3D Resolution Gate Fit. Note that 3D Stereoscopic information is not imported.
– InverseTransform: Imports the Inverse Transform (World to Model) for cameras.
– Points: Alembic files support a Points type. This is a collection of 3D points with position
information. Some 3D software exports particles as points. However, keep in mind that while
position is included, the direction and orientation of the particles are lost.
– Meshes: This setting determines whether importing includes 3D models from the Alembic file. If it
is enabled, options to include UVs and normals are displayed.

Animation
This section includes one option for the Resampling rate. When exporting an Alembic animation, it is
saved to disk using frames per second (fps). When importing Alembic data into Fusion, the fps are
detected and entered into the Resample Rate field unless you have changed it previously in the
current comp. Ideally, you should maintain the exported frame rate as the resample rate, so your
samples match up with the original. The Detected Sampling Rates information at the top of the dialog
can give an idea of what to pick if you are unsure. However, using this field, you can change the frame
rate to create effects like slow motion.
Not all objects and properties in a 3D scene have an agreed upon universal convention in the Alembic
file format. That being the case, Lights, Materials, Curves, Multiple UVs, and Velocities are not currently
supported when you import Alembic files.
Since the FBX file format does support materials and lights, we recommend the use of FBX for lights,
cameras, and materials. Use Alembic for meshes only.

Inputs
The AlembicMesh3D node has two inputs in the Node Editor. Both are optional since the node is
designed to use the imported mesh.
– SceneInput: The orange input can be used to connect an additional 3D scene or model. The
imported Alembic objects combine with the other 3D geometry.
– MaterialInput: The optional green input is used to apply a material to the geometry by
connecting a 2D bitmap image. It applies the connected image to the surface of the geometry
in the scene.

Chapter 29 3D Nodes 640

Basic Node Setup
The AlembicMesh3D node is designed to be part of a larger 3D scene. Typically, when imported, a 3D
geometry model is represented by one node, and any transforms are in another node. The nodes
imported as part of the Alembic file connect into a Merge 3D node along with a camera, lights, and
other elements that may be required for the scene.

Alembic node structure

Inspector

Alembic mesh 3D controls

Controls Tab
The first tab in the Inspector is the Controls tab. It includes a series of unique controls specific to the
Alembic Mesh 3D node as well as six groupings of controls that are common to most 3D nodes. The
“Common Controls” section at the end of this chapter includes detailed descriptions of the
common controls.
Below are descriptions of the Alembic Mesh 3D specific controls.

Filename
The complete file path of the imported Alembic file is displayed here. This field allows you to change
or update the file linked to this node.

Chapter 29 3D Nodes 641

Object Name
This text field shows the name of the imported Alembic mesh, which is also used to rename the
Alembic Mesh 3D node in the Node Editor.
When importing with the Alembic Mesh 3D node, if this text field is blank, the entire contents of the
Alembic geometry are imported as a single mesh. When importing geometry using File > Import >
Alembic Scene, this field is set by Fusion.

Wireframe
Enabling this option causes the mesh to display only the wireframe for the object in the viewer. When
enabled, there is a second option for wireframe anti-aliasing. You can also render these wireframes
out to a file if the Renderer 3D node has the OpenGL render type selected.

Common Controls
Controls, Materials, Transform, and Settings Tabs
The controls for Visibility, Lighting, Matte, Blend Mode, Normals/Tangents, and Object ID in the
Controls tab are common in many 3D nodes. The Materials tab, Transforms tab and Settings tab in the
Inspector are also duplicated in other 3D nodes. These common controls are described in detail at the
end of this chapter in “The Common Controls” section.

Bender 3D [3BN]

The Bender 3D node

Bender 3D Introduction
The Bender 3D node is used to bend, taper, twist, or shear 3D geometry based on the geometry’s
bounding box. It works by connecting any 3D scene or object to the orange input on the Bender 3D
node, and then adjusting the controls in the Inspector. Only the geometry in the scene is modified.
Any lights, cameras, or materials are passed through unaffected.
The Bender node does not produce new vertices in the geometry; it only alters existing vertices in the
geometry. So, when applying the Bender 3D node to primitives, like the Shape 3D, or Text 3D nodes,
increase the Subdivision setting in the primitive’s node to get a higher-quality result.

Inputs
The following inputs appear on the Bender 3D node in the Node Editor.
– SceneInput: The orange scene input is the required input for the Bender 3D node. You use this
input to connect another node that creates or contains a 3D scene or object.

Basic Node Setup

The Bender 3D node works by connecting a 3D node that contains geometry, like an image plane 3D,
Shape 3D or Text 3D. The element you connect to the Bender 3D node will be distorted based on the
controls in the Inspector. The Bender 3D node is designed to be part of a larger 3D scene, with the
output typically connecting into a Merge 3D.

Chapter 29 3D Nodes 642

Bender 3D node structure

Inspector

Bender 3D controls

Controls Tab
The first tab in the Inspector is the Controls tab. It includes all the controls for the Bender 3D node.

Bender Type
The Bender Type menu is used to select the type of deformation to apply to the geometry. There are
four modes available: Bend, Taper, Twist, and Shear.

Amount
Adjusting the Amount slider changes the strength of the deformation.

Axis
The Axis control determines the axis along which the deformation is applied. It has a different meaning
depending on the type of deformation. For example, when bending, this selects the elbow in
conjunction with the Angle control. In other cases, the deform is applied around the specified axis.

Angle
The Angle thumbwheel control determines what direction about the axis a bend or shear is applied.
It is not visible for taper or twist deformations.

Chapter 29 3D Nodes 643

Range
The Range control can be used to limit the effect of a deformation to a small portion of the geometry.
The Range control is not available when the Bender Type is set to Shear.

Group Objects
If the input of the Bender 3D node contains multiple 3D objects, either through a Merge 3D or strung
together, the Group Objects checkbox treats all the objects in the input scene as a single object, and
the common center is used to deform the objects, instead of deforming each component object
individually.

Common Controls
Settings
The Settings tab in the Inspector is common to all 3D nodes. This common tab is described in detail at
the end of this chapter in “The Common Controls” section.

Camera 3D [3CM]

The Camera 3D node

Camera 3D Node Introduction

The Camera 3D node generates a virtual camera for viewing the 3D environment. It closely emulates
the settings used in real cameras to make matching live-action or 3D-rendered elements as seamless
as possible. Adding any cameras to a 3D composite allows you to frame the elements in a composite
how you want and animate the camera during a scene to create moving camera shots.

Camera Projection
The Camera 3D node can also be used to perform Camera Projection by projecting a 2D image
through the camera into 3D space. Projecting a 2D image can be done as a simple Image Plane
aligned with the camera, or as an actual projection, similar to the behavior of the Projector 3D node,
with the added advantage of being aligned precisely with the camera. The Image Plane, Projection,
and Materials tabs do not appear until you connect a 2D image to the magenta image input on the
Camera 3D node in the Node Editor.

Stereoscopic
The Camera node has built-in stereoscopic features. They offer control over eye separation and
convergence distance. The camera for the right eye can be replaced using a separate camera node
connected to the green left/right stereo camera input. Additionally, the plane of focus control for depth
of field rendering is also available here.
If you add a camera by dragging the camera icon from the toolbar onto the 3D view, it automatically
connects to the Merge 3D you are viewing. Also, the current viewer is set to look through the
new camera.
Alternatively, it is possible to copy the current viewer to a camera (or spotlight or any other object) by
selecting the Copy PoV To option in the viewer’s contextual menu, under the Camera submenu.

Chapter 29 3D Nodes 644

Inputs
There are three optional inputs on the Camera 3D node in the Node Editor.
– SceneInput: The orange input is used to connect a 3D scene or object. When connected, the
geometry links to the camera’s field of view. It acts similarly to an image attached to the Image
Plane input. If the camera’s Projection tab has projection enabled, the image attached to the
orange image input projects on to the geometry.
– ImageInput: The optional magenta input is used to connect a 2D image. When camera
projection is enabled, the image can be used as a texture. Alternatively, when the camera’s
image plane controls are used, the parented planar geometry is linked to the camera’s
field of view.
– RightStereoCamera: The green input should be connected to another Camera 3D node when
creating 3D stereoscopic effects. It is used to override the internal camera used for the right
eye in stereoscopic renders and viewers.

Basic Node Setup

The output of a camera 3D node should be connected to a Merge 3D node. You then view the Merge
3D node and select the camera from the viewer’s right-click menu or by right-clicking over the axis
label in the viewer.

Camera node connected to and viewed through the Merge 3D

Displaying a camera node directly in the viewer shows only an empty scene; there is nothing for the
camera to see. To view the scene through the camera, view the Merge 3D node where the camera is
connected, or any node downstream of that Merge 3D. Then right-click on the viewer and select
Camera > [Camera name] from the contextual menu. Right-clicking on the axis label found in the lower
corner of each 3D viewer also displays the Camera submenu.
The aspect of the viewer may be different from the aspect of the camera, so the camera view may not
match the actual boundaries of the image rendered by the Renderer 3D node. Guides can be enabled
to represent the portion of the view that the camera sees and assist you in framing the shot. Right-click
on the viewer and select an option from the Guides > Frame Aspect submenu. The default option uses
the format enabled in the Composition > Frame Format preferences. To toggle the guides on or off,
select Guides > Show Guides from the viewers’ contextual menu, or use the Command-G (macOS) or
Ctrl-G (Windows) keyboard shortcut when the viewer is active.

Chapter 29 3D Nodes 645

Inspector

Camera 3D controls

Controls Tab
The Camera3D Inspector includes six tabs along the top. The first tab, called the Controls tab, contains
some of the most fundamental camera settings, including the camera’s clipping plains, field of view,
focal length, and stereoscopic properties. Some tabs are not displayed until a required connection is
made to the Camera 3D node.

Projection Type
The Projection Type menu is used to select between Perspective and Orthographic cameras.
Generally, real-world cameras are perspective cameras. An orthographic camera uses parallel
orthographic projection, a technique where the view plane is perpendicular to the viewing direction.
This produces a parallel camera output that is undistorted by perspective.
Orthographic cameras present controls only for the near and far clipping planes, and a control to set
the viewing scale.

Chapter 29 3D Nodes 646

Near/Far Clip
The clipping planes are used to limit what geometry in a scene is rendered based on an object’s
distance from the camera’s focal point. Clipping planes ensure objects that are extremely close to the
camera, as well as objects that are too far away to be useful, are excluded from the final rendering.
The default perspective camera ignores this setting unless the Adaptive Near/Far Clip checkbox
located under the Near/Far Clip control is disabled.
The clip values use units, so a far clipping plane of 20 means that any object more than 20 units from
the camera is invisible to the camera. A near clipping plane of 0.1 means that any object closer than 0.1
units is also invisible.

NOTE: A smaller range between the near and far clipping planes allows greater accuracy in
all depth calculations. If a scene begins to render strange artifacts on distant objects, try
increasing the distance for the Near Clip plane.

Adaptive Near/Far Clip

When selected, the renderer automatically adjusts the camera’s near/far clipping plane to match the
extents of the scene. This setting overrides the values of the Near and Far clip range controls
described above. This option is not available for orthographic cameras.

Viewing Volume Size

When the Projection Type is set to Orthographic, the viewing volume size adjustment appears. It
determines the size of the box that makes up the camera’s field of view.
The Z-distance of an orthographic camera from the objects it sees does not affect the scale of those
objects, only the viewing size does.

Angle of View Type

Use the Angle of View Type buttons to choose how the camera’s angle of view is measured. Some
applications use vertical measurements, some use horizontal, and others use diagonal measurements.
Changing the Angle of View type causes the Angle of View control below to recalculate.

Angle of View
Angle of View defines the area of the scene that can be viewed through the camera. Generally, the
human eye can see more of a scene than a camera, and various lenses record different degrees of the
total image. A large value produces a wider angle of view, and a smaller value produces a narrower, or
more tightly focused, angle of view.
Just as in a real-world camera, the angle of view and focal length controls are directly related. Smaller
focal lengths produce a wider angle of view, so changing one control automatically changes the
other to match.

Focal Length
In the real world, a lens’ Focal Length is the distance from the center of the lens to the film plane. The
shorter the focal length, the closer the focal plane is to the back of the lens. The focal length is
measured in millimeters. The angle of view and focal length controls are directly related. Smaller focal
lengths produce a wider angle of view, so changing one control automatically changes the
other to match.
The relationship between focal length and angle of view is angle = 2 * arctan[aperture / 2 /
focal_length].
Use the vertical aperture size to get the vertical angle of view and the horizontal aperture size to get
the horizontal angle of view.

Chapter 29 3D Nodes 647

Plane of Focus (For Depth of Field)
Like a focal point on a real-world camera, this setting defines the distance from the camera to an
object. It is used by the OpenGL renderer in the Renderer 3D node to calculate depth of field.

Stereo
The Stereo section includes options for setting up 3D stereoscopic cameras. 3D stereoscopic
composites work by capturing two slightly different views, displayed separately to the left and right
eyes. The mode menu determines if the current camera is a stereoscopic setup or a mono camera.
When set to the default mono setting, the camera views the scene as a traditional 2D film camera.
Three other options in the mode menu determine the method used for 3D stereoscopic cameras.

Toe In
In a toe-in setup, both cameras are rotating in on a single focal point. Though the result is
stereoscopic, the vertical parallax introduced by this method can cause discomfort by the audience.
Toe-in stereoscopic works for convergence around the center of the images but exhibits keystoning,
or image separation, to the left and right edges. This setup is can be used when the focus point and
the convergence point need to be the same. It is also used in cases where it is the only way to match a
live-action camera rig.

Toe In 3D camera setup

Off Axis
Regarded as the correct way to create stereo pairs, this is the default method in Fusion. Off Axis
introduces no vertical parallax, thus creating stereo images with less eye strain. Sometimes called a
skewed-frustum setup, this is akin to a lens shift in the real world. Instead of rotating the two cameras
inward as in a toe-in setup, Off Axis shifts the lenses inward.

Off axis 3D camera setup

Chapter 29 3D Nodes 648

Parallel
The cameras are shifted parallel to each other. Since this is a purely parallel shift, there is no
Convergence Distance control that limits your control over placing objects in front of or behind the
screen. However, Parallel introduces no vertical parallax, thus creating less strain on the eyes.

Parallel 3D camera setup

Rig Attached To
This drop-down menu allows you to control which camera is used to transform the stereoscopic setup.
Based on this menu, transform controls appear in the viewer either on the right camera, left camera, or
between the two cameras. The ability to switch the transform controls through rigging can assist in
matching the animation path to a camera crane or other live-action camera motion. The Center option
places the transform controls between the two cameras and moves each evenly as the separation and
convergence are adjusted. Left puts the transform controls on the left camera, and the right camera
moves as the separation and convergence are adjusted. Right puts the transform controls on the right
camera, and the left camera moves as adjustments are made to separation and convergence.

Eye Separation
Eye Separation defines the distance between both stereo cameras. Setting Eye Separation to a value
larger than 0 shows controls for each camera in the viewer when this node is selected. Note that there
is no Convergence Distance control in Parallel mode.

Convergence Distance
This control sets the stereoscopic convergence distance, defined as a point located along the Z-axis
of the camera that determines where both left- and right-eye cameras converge. The Convergence
Distance controls are only available when setting the Mode menu to Toe-In or Off Axis.

Film Back
Film Gate
The size of the film gate represents the dimensions of the aperture. Instead of setting the aperture’s
width and height, you can choose it using the list of preset camera types in the Film Gate menu.
Selecting one of the options automatically sets the aperture width and aperture height to match.

Aperture Width/Height
The Aperture Width and Height sliders control the dimensions of the camera’s aperture or the portion
of the camera that lets light in on a real-world camera. In video and film cameras, the aperture is the
mask opening that defines the area of each frame exposed. The Aperture control uses inches as its
unit of measurement.

Resolution Gate Fit

Determines how the film gate is fitted within the resolution gate. This only has an effect when the
aspect of the film gate is not the same aspect as the output image.

Chapter 29 3D Nodes 649

 NOTE: This setting corresponds to Maya’s Resolution Gate. The modes Overscan, Horizontal,
Vertical, and Fill correspond to Inside, Width, Height, and Outside.

– Inside: The image source defined by the film gate is scaled uniformly until one of its dimensions
(X or Y) fits the inside dimensions of the resolution gate mask. Depending on the relative
dimensions of image source and mask background, either the image source’s width or height may
be cropped to fit the dimension of the mask.
– Width: The image source defined by the film gate is scaled uniformly until its width (X) fits the
width of the resolution gate mask. Depending on the relative dimensions of image source and
mask, the image source’s Y-dimension might not fit the mask’s Y-dimension, resulting in either
cropping of the image source in Y or the image source not covering the mask’s height entirely.
– Height: The image source defined by the film gate is scaled uniformly until its height (Y) fits the
height of the resolution gate mask. Depending on the relative dimensions of image source and
mask, the image source’s X-dimension might not fit the mask’s X-dimension, resulting in either
cropping of the image source in X or the image source not covering the mask’s width entirely.
– Outside: The image source defined by the film gate is scaled uniformly until one of its dimensions
(X or Y) fits the outside dimensions of the resolution gate mask. Depending on the relative
dimensions of image source and mask, either the image source’s width or height may be cropped
or not fit the dimension of the mask.
– Stretch: The image source defined by the film gate is stretched in X and Y to accommodate the
full dimensions of the generated resolution gate mask. This might lead to visible distortions of the
image source.

Control Visibility
This section allows you to selectively activate the onscreen controls that are displayed along with
the camera.
– Show View Controls: Displays or hides all camera onscreen controls in the viewers.
– Frustum: Displays the actual viewing cone of the camera.
– View Vector: Displays a white line inside the viewing cone, which can be used to determine the
shift when in Parallel mode.
– Near Clip: The Near clipping plane. This plane can be subdivided for better visibility.
– Far Clip: The Far clipping plane. This plane can be subdivided for better visibility.
– Focal Plane: The plane based on the Plane of Focus slider explained in the Controls tab above.
This plane can be subdivided for better visibility.
– Convergence Distance: The point of convergence when using Stereo mode. This plane can be
subdivided for better visibility.

Import Camera
The Import Camera button displays a dialog to import a camera from another application.
It supports the following file types:

*LightWave Scene .lws

*Max Scene .ase

*Maya Ascii Scene .ma

*dotXSI .xsi

Chapter 29 3D Nodes 650

 NOTE: FBX cameras can be imported using DaVinci Resolve’s Fusion > Import > FBX Scene
menu or File > Import > FBX Scene in Fusion Studio.

Image Tab
When a 2D image is connected to the magenta image input on the Camera3D node, an Image tab is
created at the top of the inspector. The connected image is always oriented so it fills the camera’s
field of view.

Camera 3D image plane tab

Except for the controls listed below, the options in this tab are identical to those commonly found in
other 3D nodes. For more detail on visibility, lighting, matte, blend mode, normals/tangents, and
Object ID, see “The Common Controls” section at the end of this chapter.

Enable Image Plane

Use this checkbox to enable or disable the usage of the Image Plane.

Fill Method
This menu configures how to scale the image plane if the camera has a different aspect ratio.
– Inside: The image plane is scaled uniformly until one of its dimensions (X or Y) fits the inside
dimensions of the resolution gate mask. Depending on the relative dimensions of image source
and mask background, either the image source’s width or height may be cropped to fit the
dimensions of the mask.
– Width: The image plane is scaled uniformly until its width (X) fits the width of the mask. Depending
on the relative dimensions of image source and the resolution gate mask, the image source’s
Y-dimension might not fit the mask’s Y-dimension, resulting in either cropping of the image source
in Y or the image source not covering the mask’s height entirely.
– Height: The image plane is scaled uniformly until its height (Y) fits the height of the mask.
Depending on the relative dimensions of image source and the resolution gate mask, the image
source’s X-dimension might not fit the mask’s X-dimension, resulting in either cropping of the
image source in X or the image source not covering the mask’s width entirely.
– Outside: The image plane is scaled uniformly until one of its dimensions (X or Y) fits the outside
dimensions of the resolution gate mask. Depending on the relative dimensions of image source
and mask, either the image source’s width or height may be cropped or not fit the respective
dimension of the mask.
– Depth: The Depth slider controls the image plane’s distance from the camera.

Chapter 29 3D Nodes 651

 NOTE: The Camera Z position has no effect on the image plane’s distance from the camera.

Materials Tab
The options presented in the Materials tab are identical to those commonly found in other 3D nodes.
For more detail on Diffuse, Specular, Transmittance, and Martial ID controls, see the “Common
Controls” section at the end of this chapter.

Projection Tab
When a 2D image is connected to the camera node, a fourth projection tab is displayed at the top of
the Inspector. Using this Projection tab, it is possible to project the image into the scene. A projection
is different from an image plane in that the projection falls onto the geometry in the scene exactly as if
there were a physical projector present in the scene. The image is projected as light, which means the
Renderer 3D node must be set to enable lighting for the projection to be visible.

Camera 3D projection tab

Enable Camera Projection

Select this checkbox to enable projection of the 2D image connected to the magenta input on the
Camera node.

Projection Fit Method

This menu can be used to select the method used to match the aspect of the projected image to the
camera’s field of view.

Projection Mode
– Light: Defines the projection as a spotlight.
– Ambient Light: Defines the projection as an ambient light.
– Texture: Allows a projection that can be relighted using other lights. Using this setting requires a
Catcher node connected to the applicable inputs of the specific material.

Chapter 29 3D Nodes 652

Common Controls
Transform and Settings Tabs
The options presented in the Transform and Settings tabs are commonly found in other 3D nodes.
For more detail on the controls found in these tabs, see “The Common Controls” section at the end of
this chapter.

Tips for Camera 3D

Camera Projection: When importing a camera from a 3D application that is also used as a
projector, make sure that the Fit Resolution Gate options on the Controls tab as well as the
Projection tab are in sync. Only the first one automatically sets to what the 3D app was using.
The latter might have to be adjusted manually.
Image Plane: The camera’s image plane isn‘t just a virtual guide for you in the viewers.
It‘s actual geometry that you can also project on to. To use a different image on the image
plane, you need to insert a Replace Material node after your Camera node.
Parallel Stereo: There are three ways you can achieve real Parallel Stereo mode:
– Connect an additional external (right) camera to the green Right Stereo Camera
input of your camera.
– Create separate left and right cameras.
– When using Toe-In or Off Axis, set the Convergence Distance slider to a very large
value of 999999999.

Rendering Overscan: If you want to render an image with overscan, you also must modify
your scene‘s Camera3D. Since overscan settings aren’t exported along with camera data from
3D applications, this is also necessary for cameras you’ve imported via .fbx or .ma files. The
solution is to increase the film back’s width and height by the factor necessary to account for
extra pixels on each side.

Cube 3D [3CB]

The Cube 3D node

Cube 3D Node Introduction

The Cube 3D node is a basic primitive geometry type capable of generating a simple cube.
The node also provides six additional image inputs that can be used to map a texture onto the six
faces of the cube. Cubes are often used as shadow casting objects and for environment maps. For
other basic primitives, see the Shape 3D node in this chapter.

Inputs
The following are optional inputs that appear on the Cube3D node in the Node Editor:

Chapter 29 3D Nodes 653

 – SceneInput: The orange scene input is used to connect another node that creates or contains
a 3D scene or object. The additional geometry gets added to the Cube3D.
– NameMaterialInput: These six inputs are used to define the materials applied to the six faces
of the cube. You can connect either a 2D image or a 3D material to these inputs. Textures
or materials added to the Cube3D do not get added to any 3D objects connected to the
Cube’s SceneInput.

Basic Node Setup

The output of a Cube 3D node typically connects to a Merge 3D node, integrating it into a larger
scene. When 3D tracking, the Cube 3D is often used as a placeholder for proper geometry that is not
available at the current time.

Cube 3D node connected to a 3D scene exported from the Camera Tracker node

Inspector

Cube 3D controls

Chapter 29 3D Nodes 654

Controls Tab
The first tab in the Inspector is the Controls tab. It includes the primary controls for determining the
overall size and shape of the Cube 3D node.

Lock Width/Height/Depth
This checkbox locks the Width, Height, and Depth dimensions of the cube together. When selected,
only a Size control is displayed; otherwise, separate Width, Height, and Depth sliders are shown.

Size or Width/Height/Depth
If the Lock checkbox is selected, then only the Size slider is shown; otherwise, separate sliders are
displayed for Width, Height, and Depth. The Size and Width sliders are the same control renamed, so
any animation applied to Size is also applied to Width when the controls are unlocked.

Subdivision Level
Use the Subdivision Level slider to set the number of subdivisions used when creating the
image plane.
The 3D viewers and renderer use vertex lighting, meaning all lighting is calculated at the vertices on
the 3D geometry and then interpolated from there. Therefore, the more subdivisions in the mesh, the
more vertices are available to represent the lighting. For example, make a sphere and set the
subdivisions to be small so it looks chunky. With lighting on, the object looks like a sphere but has
some amount of fracturing resulting from the large distance between vertices. When the subdivisions
are high, the vertices are closer and the lighting becomes more even. So, increasing subdivisions can
be useful when working interactively with lights.

Cube Mapping
Enabling the Cube Mapping checkbox causes the cube to wrap its first texture across all six faces using
a standard cubic mapping technique. This approach expects a texture laid out in the shape of a cross.

Wireframe
Enabling this checkbox causes the mesh to render only the wireframe for the object when rendering
with the OpenGL renderer in the Renderer 3D node.

Common Controls
Controls, Materials, Transform, and Settings Tabs
The remaining controls for Visibility, Lighting, Matte, Blend Mode, Normals/Tangents, and Object ID are
common to many 3D nodes. The same is true of the Materials, Transform, and Settings tabs. Their
descriptions can be found in “The Common Controls” section at the end of this chapter.

Custom Vertex 3D [3CV]

The Custom Vertex 3D node

Custom Vertex 3D Node Introduction

The Custom Vertex 3D node is an advanced custom node for 3D geometry that performs per vertex
manipulations. If you have moderate experience with scripting or C++ programming, you should find
the structure and terminology used by the Custom node familiar.

Chapter 29 3D Nodes 655

Using scripting math functions and lookup tables from images, you can move vertex positions on 3D
geometry. Vertices can be more than just positions in 3D space. You can manipulate normals, texture
coordinates, vectors, and velocity.
For example, Custom Vertex 3D can be used to make a flat plane wave like a flag, or create
spiral models.
Besides providing a 3D scene input and three image inputs, the Inspector includes up to eight number
fields and as many as eight XYZ position values from other controls and parameters in the node tree.

NOTE: Modifying the X, Y, and Z positions of a 3D object does not modify the normals/
tangents. You can use a ReplaceNormals node afterward to recompute the normals/tangents.

TIP: Not all geometry has every attribute. For example, most Fusion geometry does not have
vertex colors, with the exception of particles and some imported FBX/Alembic meshes. No
geometry currently has environment coordinates, and only particles have velocities. If an
attribute is not present on the input geometry, it is assumed to have a default value.

Inputs
The Custom Vertex 3D node includes four inputs. The orange scene input is the only one of the four
that is required.
– SceneInput: The orange scene input takes 3D geometry or a 3D scene from a 3D node output.
This is the 3D scene or geometry that is manipulated by the calculations in the Custom Vertex
3D node.
– ImageInput1, ImageInput2, ImageInput3: The three image inputs using green, magenta, and
teal colors are optional inputs that can be used for compositing.

NOTE: Missing attributes on the input geometry are created if the expression for an attribute
is nontrivial. The values for the attributes are given as in the above point. For example, if the
input geometry does not have normals, then the values of (nx, ny, nz) is always (0,0,1). To
change this, you could use a ReplaceNormals node beforehand to generate them.

Basic Node Setup

The object you want to manipulate connects to the orange scene input of the Custom Vertex 3D node.
The output typically connects to a Merge 3D node, integrating it into a larger scene.

Custom Vertex 3D node manipulating an Image Plane 3D node

Chapter 29 3D Nodes 656

Inspector

Custom Vertex 3D node Vertex tab

Vertex Tab
Using the fields in the Vertex tab, vertex calculations can be performed on the Position, Normals,
Vertex Color, Texture Coordinates, Environment Coordinates, UV Tangents, and Velocity attributes.
The vertices are defined by three XYZ Position values in world space as px, py, pz. Normals, which
define as a vector the direction the vertex is pointing as nx, ny, nz.
Vertex color is the Red, Green, Blue, and Alpha color of the point as vcr, vcg, vcb, vca.

Numbers Tab

Custom Vertex 3D node Numbers tab

Numbers 1-8
Numbers are variables with a dial control that can be animated or connected to modifiers exactly as
any other control might. The numbers can be used in equations on vertices at current time: n1, n2,
n3, n4,… or at any time: n1_at(float t), n2_at(float t), n3_at(float t), n4_at(float t), where t is the time you
want. The values of these controls are available to expressions in the Setup and Intermediate tabs.
They can be renamed and hidden from the viewer using the Config tab.

Chapter 29 3D Nodes 657

Points Tab

Custom Vertex 3D node Points tab

Points 1-8
The point controls represent points in the Custom Vertex 3D tool, not the vertices. These eight point
controls include 3D X,Y,Z position controls for positioning points at the current time: (p1x, p1y, p1z, p2x,
p2y, p2z) or at any time: p1x_at(float t), p1y_at(float t), p1z_at(float t), p2x_at(float t), p2y_at(float t), p2z_
at(float t), where t is the time you want. For example, you can use a point to define a position in 3D
space to rotate the vertices around. They can be renamed and hidden from the viewer using the
Config tab. They are normal positional controls and can be animated or connected to modifiers as any
other node might.

LUT Tab

Custom Vertex 3D node LUT tab

LUTs 1-4
The Custom Vertex 3D node provides four LUT splines. A LUT is a lookup table that will return a value
from the height of the LUT spline. For example, getlut1(float x), getlut2(float x),...
where x = 0 … 1 accesses the LUT values.
The values of these controls are available to expressions in the Setup and Intermediate tabs using the
getlut# function. For example, setting the R, G, B, and A expressions to getlut1(r1), getlut2(g1),
getlut3(b1), and getlut4(a1) respectively, would cause the Custom Vertex 3D node to mimic the Color
Curves node.
These controls can be renamed using the options in the Config tab to make their meanings more
apparent, but expressions still see the values as lut1, lut2,...lut8.

Chapter 29 3D Nodes 658

Setup Tab

Custom Vertex 3D node Setup tab

Setups 1-8
Up to eight separate expressions can be calculated in the Setup tab of the Custom Vertex 3D node.
The Setup expressions are evaluated once per frame, before any other calculations are performed.
The results are then made available to the other expressions in the node as variables s1, s2,
s3, and s4.
Think of them as global setup scripts that can be referenced by the intermediate and channel scripts
for each vertex.
For example, Setup scripts can be used to transform vertex from model to world space.

NOTE: Because these expressions are evaluated once per frame only and not for each pixel,
it makes no sense to use per-pixel variables like X and Y or channel variables like r1, g1, b1,
and so on. Allowable values include constants, variables like n1…n8, time, W and H, and so on,
and functions like sin() or getr1d().

Intermediate Tab

Custom Vertex 3D Node Intermediate tab

Intermediates 1-8
An additional eight expressions can be calculated in the Intermediate tab. The Intermediate
expressions are evaluated once per vertex, after the Setup expressions are evaluated. Results are
available as variables i1, i2, i3, i4, i5, i6, i7, i8, which can be referenced by channel scripts. Think of
them as “per vertex setup” scripts.
For example, you can run the script to produce the new vertex (i.e., new position, normal, tangent,
UVs, etc.) or transform from world space back to model space.

Chapter 29 3D Nodes 659

Config Tab

Custom Vertex 3D node Config tab

Random Seed
Use this to set the seed for the rand() and rands() functions. Click the Reseed button to set the seed to
a random value. This control may be needed if multiple Custom Vertex 3D nodes are required with
different random results for each.

Number Controls
There are eight sets of Number controls, corresponding to the eight sliders in the Numbers tab.
Disable the Show Number checkbox to hide the corresponding Number slider, or edit the Name for
Number text field to change its name.

Point Controls
There are eight sets of Point controls, corresponding to the eight controls in the Points tab. Disable the
Show Point checkbox to hide the corresponding Point control and its crosshair in the viewer. Similarly,
edit the Name for Point text field to change the control’s name.

Common Controls
Settings Tab
The Settings tab controls are common to many 3D nodes, and their descriptions can be found in “The
Common Controls” section at the end of this chapter.

Displace 3D [3DI]

The Displace 3D node

Displace 3D Node Introduction

The Displace 3D node is used to displace the vertices of an object along their normals based on a
reference image. The texture coordinates on the geometry are used to determine where to sample
the image.
When using Displace 3D, keep in mind that it only displaces existing vertices and does not subdivide
surfaces to increase detail. To obtain a more detailed displacement, increase the subdivision amount

Chapter 29 3D Nodes 660

for the geometry that is being displaced. Note that the pixels in the displacement image may contain
negative values.

TIP: Passing a particle system through a Displace 3D node disables the Always Face Camera
option set in the pEmitter. Particles are not treated as point-like objects; each of the four
particle vertices are individually displaced, which may or may not be the preferred outcome.

Inputs
The following two inputs appear on the Displace 3D node in the Node Editor:
– SceneInput: The orange scene input is the required input for the Displace 3D node. You use
this input to connect another node that creates or contains a 3D scene or object.
– Input: This green input is used to connect a 2D image that is used to displace the object
connected to the Scene input. If no image is provided, this node effectively passes the scene
straight through to its output. So, although not technically a required input, there isn’t much use
for adding this node unless you connect this input correctly.

Basic Node Setup

The output of a Displace 3D node typically connects to a Merge 3D node, integrating it into a larger
scene. The 3D geometry you want to displace is connected to the orange input, and in this example, a
Fast Noise node is used to displace the geometry.

An image on an Image Plane 3D is displaced by a Fast Noise node

Inspector

Displace 3D controls

Chapter 29 3D Nodes 661

Controls Tab
The Displace 3D Inspector includes two tabs along the top. The primary tab, called the Controls tab,
contains the dedicated Displace 3D controls.

Channel
Determines which channel of the connected input image is used to displace the geometry.

Scale and Bias

Use these sliders to scale (magnify) and bias (offset) the displacement. The bias is applied first and the
scale afterward.

Camera Displacement
– Point to Camera: When the Point to Camera checkbox is enabled, each vertex is displaced toward
the camera instead of along its normal. One possible use of this option is for displacing a camera’s
image plane. The displaced camera image plane would appear unchanged when viewed through
the camera but is deformed in 3D space, allowing one to comp-in other 3D layers that correctly
interact in Z.
– Camera: This menu is used to select which camera in the scene is used to determine the camera
displacement when the Point to Camera option is selected.

Common Controls
Settings Tab
The Settings tab controls are common to many 3D nodes, and their descriptions can be found in “The
Common Controls” section at the end of this chapter.

Duplicate 3D [3DP]

The Duplicate 3D node

Duplicate 3D Node Introduction

Similar to the 2D version called the Duplicate node, the Duplicate 3D node can be used to duplicate
any geometry in a scene, applying a successive transformation to each, and creating repeating
patterns and complex arrays of objects. The options in the Jitter tab allow non-uniform
transformations, such as random positioning or sizes.

Inputs
The Duplicate 3D node has a single input by default where you connect a 3D scene. An optional Mesh
input appears based on the settings of the node.
– SceneInput: The orange Scene Input is a required input. The scene or object you connect to
this input is duplicated based on the settings in the Control tab of the Inspector.
– MeshInput: A green optional mesh input appears when the Region’s tab Region menu is set to
mesh. The mesh can be any 3D model, either generated in Fusion or imported.

Chapter 29 3D Nodes 662

Basic Node Setup
The output of a Duplicate 3D node typically connects to a Merge 3D node, integrating it into a larger
scene. The 3D geometry you want duplicated, in this case a Cube 3D, is connected to the
orange input.

A Cube 3D is duplicated

Inspector

Duplicate 3D controls

Controls Tab
The Controls tab includes all the parameters you can use to create, offset, and scale copies of the
object connected to the scene input on the node.

Chapter 29 3D Nodes 663

Copies
Use this range control to set the number of copies made. Each copy is a copy of the last copy, so if this
control is set to [0,3], the parent is copied, then the copy is copied, then the copy of the copy is copied,
and so on. This allows some interesting effects when transformations are applied to each copy using
the controls below.
Setting the First Copy to a value greater than 0 excludes the original object and shows only
the copies.

Time Offset
Use the Time Offset slider to offset any animations that are applied to the source geometry by a set
amount per copy. For example, set the value to -1.0 and use a cube set to rotate on the Y-axis as the
source. The first copy shows the animation from a frame earlier; the second copy shows animation
from a frame before that, etc. This can be used with great effect on textured planes—for example,
where successive frames of a clip can be shown.

Transform Method

– Linear: When set to Linear, transforms are multiplied by the number of the copy, and the total
scale, rotation, and translation are applied in turn, independent of the other copies.
– Accumulated: When set to Accumulated, each object copy starts at the position of the previous
object and is transformed from there. The result is transformed again for the next copy

Transform Order
With this menu, the order in which the transforms are calculated can be set. It defaults to Scale-
Rotation-Transform (SRT).
Using different orders results in different positions of your final objects.

Translation
The X, Y, and Z Offset sliders set the offset position applied to each copy. An X offset of 1 would offset
each copy 1 unit along the X-axis from the last copy.

Rotation
The buttons along the top of this group of rotation controls set the order in which rotations are applied
to the geometry. Setting the rotation order to XYZ would apply the rotation on the X-axis first, followed
by the Y-axis rotation, then the Z-axis rotation.
The three Rotation sliders set the amount of rotation applied to each copy.

Chapter 29 3D Nodes 664

Pivot
The pivot controls determine the position of the pivot point used when rotating each copy.

Scale
– Lock: When the Lock XYZ checkbox is selected, any adjustment to the duplicate scale is applied
to all three axes simultaneously. If this checkbox is disabled, the Scale slider is replaced with
individual sliders for the X, Y, and Z scales.
– Scale: The Scale controls tell Duplicate how much scaling to apply to each copy.

Jitter Tab
The options in the Jitter tab allow you to randomize the position, rotation, and size of all the copies
created in the Controls tab.

Duplicate 3D Jitter tab

Random Seed
The Random Seed slider is used to generate a random starting point for the amount of jitter applied to
the duplicated objects. Two Duplicate nodes with identical settings but different random seeds
produce two completely different results.

Randomize
Click the Randomize button to auto generate a random seed value.

Jitter Probability
Adjusting this slider determines the percentage of copies that are affected by the jitter. A value of 1.0
means 100% of the copies are affected, while a value of 0.5 means 50% are affected.

Time Offset
Use the Time Offset slider to offset any animations that are applied to the source geometry by a set
amount per copy. For example, set the value to –1.0 and use a cube set to rotate on the Y-axis as the
source. The first copy shows the animation from a frame earlier; the second copy shows animation
from a frame before that, etc. This can be used with great effect on textured planes—for example,
where successive frames of a clip can be shown.

Chapter 29 3D Nodes 665

Translation Jitter
Use these three controls to adjust the amount of variation in the X, Y, and Z translation of the
duplicated objects.

Rotation Jitter
Use these three controls to adjust the amount of variation in the X, Y, and Z rotation of the
duplicated objects.

Pivot Jitter
Use these three controls to adjust the amount of variation in the rotational pivot center of the
duplicated objects. This affects only the additional jitter rotation, not the rotation produced by the
Rotation settings in the Controls tab.

Scale Jitter
Use this control to adjust the amount of variation in the scale of the duplicated objects. Disable the
Lock XYZ checkbox to adjust the scale variation independently on all three axes.

Region Tab
The options in the Region tab allow you to define an area in the viewer where the copies can appear
or are prevented from appearing. Like most parameters in Fusion, this area can be animated to cause
the copied object to pop on and off the screen based on the region’s shape and setting.

Duplicate 3D Region tab

Region Tab
The Region section includes two settings for controlling the shape of the region and the affect the
region has on the duplicate objects.
– Region Mode: There are three options in the Region Mode menu. The default, labeled “Ignore
region” bypasses the node entirely and causes no change to the copies of objects from how they
are set in the Controls and Jitter tabs. The menu option labeled “When inside region” causes the
copied objects to appear only when their position falls inside the region defined in this tab. The
last menu option, “When not Inside region” causes the copied objects to appear only when their
position falls outside the region defined in this tab.

Chapter 29 3D Nodes 666

 – Region: The Region menu determines the shape of the region. The five options include cube,
sphere, and rectangle primitive shapes. The mesh option allows you to connect a 3D model into
the green mesh input on the node. The green input appears only after the Region menu is set
to Mesh. The All setting refers to the entire scene. This allows the copies to pop on and off if the
Region mode is animated. When the Region menu is set to Mesh, four other options are displayed.
These are described below.
– Winding Rule: Using four common techniques, the Winding Rule menu determines how the
mesh of polygons is determined as an area of volume and consequently how copies locate the
vertices in the mesh. Complex overlapping regions of a mesh can cause an irregular fit. Trying
a different technique from this menu can sometimes create a better match between the mesh
and how the copies interpret the mesh shape.
– Winding Ray Direction: A 3D model is a mesh of vertices made up of flat polygons. When
making this a volume for a region, the Winding Ray Direction is used to determine in which
direction the volume of each polygon (like depth extrude) is aligned.
– Limit by Object ID: When a scene with multiple meshes is connected to the green Mesh input
on the node, all the meshes are used as the region. Enabling this checkbox allows you to use
the Object ID slider to select the ID for the mesh you want to use as the Region.
– Object ID: When the Limit by Object ID checkbox is enabled, this slider selects the number ID
for the mesh object you want to use for the Region.

Common Controls
Settings Tab
The Settings tab controls are common to many 3D nodes, and their descriptions can be found in
“The Common Controls” section at the end of this chapter.

FBX Exporter 3D [FBX]

The FBX Exporter node

FBX Exporter Node Introduction

The FBX Exporter node provides a method of exporting a Fusion 3D scene to the FBX scene
interchange format. Each node in Fusion is a single object in the exported file. Objects, lights, and
cameras use the name of the node that created them. The node can be set to export a single file for
the entire scene, or to output one frame per file.
Setting the Preferences > Global > General > Auto Clip Browse option in the Fusion Studio application,
or the Fusion > Fusion Settings > General > Auto Clip Browse option in DaVinci Resolve to Enabled
(default), and then adding this node to a composition automatically displays a file browser allowing you
to choose where to save the file.
Once you have set up the node, the FBX Exporter is used similarly to a Saver node: clicking the
Render button in the toolbar renders out the file.
Besides the FBX format, this node can also export to the 3D Studio’s .3ds, Collada’s .dae, Autocad’s .
dxf, and the Alias .obj formats.

Chapter 29 3D Nodes 667

Inputs
The FBX Exporter node has a single orange input.
– Input: The output of the 3D scene that you want to export connects to the orange input on the
FBX Exporter node.

Basic Node Setup

The input to the FBX Exporter 3D node is any 3D scene you want to export. Below, the node is placed
as a separate branch off of the Duplicate 3D node. Only the objects generated by the Duplicate 3D
node are exported.

An FBX Exporter 3D branched off from the Duplicate 3D node

Inspector

FBX Exporter controls

Controls Tab
The Controls tab includes all the parameters you used to decide how the FBX file is created and what
elements in the scene get exported.

Filename
This Filename field is used to display the location and file that is output by the node. You can click the
Browse button to open a file browser dialog and change the location where the file is saved.

Chapter 29 3D Nodes 668

Format
This menu is used to set the format of the output file.
Not all features of this node are supported in all file formats. For example, the .obj format does not
handle animation.

Version
The Version menu is used to select the available versions for the chosen format. The menu’s contents
change dynamically to reflect the available versions for that format. If the selected format provides
only a single option, this menu is hidden.
Choosing Default for the FBX formats uses FBX2011.

Frame Rate
This menu sets the frame rate that is in the FBX scene.

Scale Units By
This slider changes the working units in the exported FBX file. Changing this can simplify workflows
where the destination 3D software that you have uses a different scale.

Geometry/Lights/Cameras
These three checkboxes determine whether the node attempts to export the named scene element.
For example, deselecting Geometry and Lights but leaving Cameras selected would output only the
cameras currently in the scene.

Render Range
Enabling this checkbox saves the Render Range information in the export file, so other applications
know the time range of the FBX scene.

Reduce Constant Keys

Enabling this option automatically removes keyframes if the adjacent keyframes have the same value.

File Per Frame (No Animation)

Enabling this option forces the node to export one file per frame, resulting in a sequence of numbered
files. This disables the export of animation. Enable this checkbox to reveal the Sequence Start Frame
control where you can set the first frame in the sequence to a custom value.

Sequence Start Frame

Enabling this checkbox displays a thumbwheel control to set a specific start frame for the number
sequence applied to the rendered filenames. For example, if Global Start is set to 1 and frames 1–30
are rendered, files are normally numbered 0001–0030. If the Sequence Start frame is set to 100, the
rendered output is numbered from 100–131.

Common Controls
Settings Tab
The Settings tab controls are common to many 3D nodes, and their descriptions can be found in “The
Common Controls” section at the end of this chapter.

Chapter 29 3D Nodes 669

FBX Mesh 3D [FBX]

The FBX Mesh node

FBX Mesh 3D Node Introduction

The FBX Mesh 3D node is used to import polygonal geometry from scene files that are saved in the
FilmBox (FBX) format. It is also able to import geometry from OBJ, 3DS, DAE, and DXF scene files.
This provides a method for working with more complex geometry than is available using Fusion‘s
built-in primitives.
When importing geometry with this node, all the geometry in the FBX file is combined into one mesh
with a single pivot and transformation. The FBX Mesh node ignores any animation applied to
the geometry.
Alternatively, in Fusion Studio, the File > Import > FBX Scene or in DaVinci Resolve, the Fusion > Import
> FBX Scene menu can be used to import an FBX scene. This option creates individual nodes for each
camera, light, and mesh in the file. This menu option can also be used to preserve the animation of
the objects.
Setting the Preferences > Global > General > Auto Clip Browse option in Fusion Studio, or the Fusion >
Fusion Settings > General > Auto Clip Browse option in DaVinci Resolve to Enabled (default), and then
adding this node to a composition automatically displays a file browser allowing you to choose the file
to import.

Inputs
– SceneInput: The orange scene input is an optional connection if you wish to combine other 3D
geometry nodes with the imported FBX file.
– Material Input: The green input is the material input that accepts either a 2D image or a 3D
material. If a 2D image is provided, it is used as a diffuse texture map for the basic material tab
in the node. If a 3D material is connected, then the basic material tab is disabled.

Basic Node Setup

The FBX Mesh 3D node can be used as a stand-alone node without any other nodes connected to it.
The output is connected to a Merge 3D, integrating the FBX model into a lager scene. Below, the FBX
Mesh 3D node also has a chrome material connected to its material input.

An FBX Mesh 3D node with chrome material applied

Chapter 29 3D Nodes 670

Inspector

FBX Mesh 3D controls

Controls Tab
Most of the Controls tab is taken up by common controls. The FBX-specific controls included on this
tab are primarily information and not adjustments.

Size
The Size slider controls the size of the FBX geometry that is imported. FBX meshes have a tendency to
be much larger than Fusion’s default unit scale, so this control is useful for scaling the imported
geometry to match the Fusion environment.

FBX File
This field displays the filename and file path of the currently loaded FBX mesh. Click the Browse
button to open a file browser that can be used to locate a new FBX file. Despite the node’s name, this
node is also able to load a variety of other formats.

FBX ascii (*.fbx)

FBX 5.0 binary (*.fbx)

Chapter 29 3D Nodes 671

 Autocad DXF (*.dxf)

3D Studio 3Ds (*.3ds)

Alias OBJ (*.obj)

Collada DAE (*.dae)

Object Name
This input shows the name of the mesh from the FBX file that is being imported. If this field is blank,
then the contents of the FBX geometry are imported as a single mesh. You cannot edit this field; it is
set by Fusion when using the File > Import > FBX Scene menu.

Take Name
FBX files can contain multiple instances of an animation, called Takes. This field shows the name of the
animation take to use from the FBX file. If this field is blank, then no animation is imported. You cannot
edit this field; it is set by Fusion when using the File > Import > FBX Scene menu.

Wireframe
Enabling this checkbox causes the mesh to render only the wireframe for the object. Only the OpenGL
renderer in the Renderer 3D node supports wireframe rendering.

Common Controls
Controls, Materials, Transform, and Settings Tabs
The remaining controls for Visibility, Lighting, Matte, Blend Mode, Normals/Tangents, and Object ID
are common to many 3D nodes. The same is true of the Materials, Transform, and Settings tabs.
Their descriptions can be found in “The Common Controls” section at the end of this chapter.

Fog 3D [3FO]

The Fog 3D node

Fog 3D Node Introduction

The Fog 3D node applies fog to the scene based on a depth cue. It is the 3D version of the Fog node
in the Deep Pixel category. It is designed to work completely in 3D space and takes full advantage of
anti-aliasing and depth of field effects during rendering.
The Fog 3D node essentially retextures the geometry in the scene by applying a color correction
based on the object’s distance from the camera. An optional density texture image can be used to
apply variation to the correction.

Inputs
The Fog 3D node has two inputs in the Node Editor, only one of which is required for the Fog 3D to
project onto a 3D scene.

Chapter 29 3D Nodes 672

 – SceneInput: The required orange-colored input accepts the output of a 3D scene on which the
fog is “projected.”
– DensityTexture: This optional green-colored input accepts a 2D image. The color of the
fog created by this node is multiplied by the pixels in this image. When creating the image
for the density texture, keep in mind that the texture is effectively projected onto the scene
from the camera.

Basic Node Setup

The Fog 3D node is placed after the Merge 3D node that contains the scene. Viewing the Fog node
will show the fog applied to the objects in the 3D scene based on their Z position.

A Fog 3D node placed after the Merge 3D scene

Inspector

Fog node controls

Controls Tab
The Controls tab includes all the parameters you use to decide how the Fog looks and projects onto
the geometry in the scene.

Chapter 29 3D Nodes 673

Enable
Use this checkbox to enable or disable parts of the node from processing. This is not the same as the
red switch in the upper-left corner of the inspector. The red switch disables the tool altogether and
passes the image on without any modification. The Enable checkbox is limited to the effect part of the
tool. Other parts like scripts in the Settings tab still processes as normal.

Show Fog in View

By default, the fog created by this node is visible only when the scene is viewed using a camera node.
When this checkbox is enabled, the fog becomes visible in the scene from all points of view.

Color
This control can be used to set the color of the fog. The color is also multiplied by the density texture
image, if one is connected to the green input on the node.

Radial
By default, the fog is created based on the perpendicular distance to a plane (parallel with the near
plane) passing through the eye point. When the Radial option is checked, the radial distance to the eye
point is used instead of the perpendicular distance. The problem with perpendicular distance fog is
that when you move the camera about, as objects on the left or right side of the frustum move into the
center, they become less fogged although they remain the same distance from the eye. Radial fog
fixes this. Radial fog is not always desirable, however. For example, if you are fogging an object close
to the camera, like an image plane, the center of the image plane could be unfogged while the edges
could be fully fogged.

Radial vs. Perpendicular fog effect

Type
This control is used to determine the type of falloff applied to the fog.
– Linear: Defines a linear falloff for the fog.
– Exp: Creates an exponential nonlinear falloff.
– Exp2: Creates a stronger exponential falloff.

Near/Far Fog Distance

This control expresses the range of the fog in the scene as units of distance from the camera. The
Near Distance determines where the fog starts, while the Far Distance sets the point where the fog
has its maximum effect. Fog is cumulative, so the farther an object is from the camera, the thicker the
fog should appear.

Common Controls
Settings Tab
The Settings tab controls are common to many 3D nodes, and their descriptions can be found in
“The Common Controls” section at the end of this chapter.

Chapter 29 3D Nodes 674

Image Plane 3D [3IM]

The Image Plane 3D node

Image Plane 3D Node Introduction

The Image Plane node produces 2D planar geometry in 3D space. The node is most commonly used
to represent 2D images on “cards” in the 3D space. The aspect of the image plane is determined by
the aspect of the image connected to the material input. If you do not want the aspect ratio of the
image to modify the “card” geometry, then use a Shape 3D node instead.

Inputs
Of the two inputs on this node, the material input is the primary connection you use to add an image to
the planar geometry created in this node.
– SceneInput: This orange input expects a 3D scene. As this node creates flat, planar geometry,
this input is not required.
– MaterialInput: The green-colored material input accepts either a 2D image or a 3D material. It
provides the texture and aspect ratio for the rectangle based on the connected source such as
a Loader node in Fusion Studio or a MediaIn node in DaVinci Resolve. The 2D image is used as
a diffuse texture map for the basic material tab in the Inspector. If a 3D material is connected,
then the basic material tab is disabled.

Basic Node Setup

The Image Plane 3D node is primarily used to bring a video clip into a 3D composite. The MediaIn or
Loader node is connected to the Image Plane 3D node, and the Image Plane 3D is then connected to
a Merge 3D node. Viewing the Merge 3D node will show all the Image Plane 3D nodes and other
elements connected to it.

Multiple Image Plane 3D nodes connected to a Merge 3D

Chapter 29 3D Nodes 675

Inspector

Image Plane 3D controls

Controls Tab
Most of the Controls tab is taken up by common controls. The Image Plane specific controls at the top
of the Inspector allow minor adjustments.

Lock Width/Height
When checked, the subdivision of the plane is applied evenly in X and Y. When unchecked, there are
two sliders for individual control of the subdivisions in X and Y. This defaults to on.

Subdivision Level
Use the Subdivision Level slider to set the number of subdivisions used when creating the image
plane. If the Open GL viewer and renderer are set to Vertex lighting, the more subdivisions in the
mesh, the more vertices are available to represent the lighting. So, high subdivisions can be useful
when working interactively with lights.

Wireframe
Enabling this checkbox causes the mesh to render only the wireframe for the object when using the
OpenGL renderer.

Common Controls
Controls, Materials, Transform, and Settings Tabs
The remaining controls for Visibility, Lighting, Matte, Blend Mode, Normals/Tangents, and Object ID
are common to many 3D nodes. The same is true of the Materials, Transform, and Settings tabs.
Their descriptions can be found in “The Common Controls” section at the end of this chapter.

Chapter 29 3D Nodes 676

Locator 3D [3LO]

The Locator 3D node

Locator 3D Node Introduction

The Locator 3D node’s purpose is to transform a point in 3D space to 2D coordinates that other nodes
can use as part of expressions or modifiers.
When the Locator is provided with a camera and the dimensions of the output image, it transforms the
coordinates of a 3D control into 2D screen space. The 2D position is exposed as a numeric output that
can be connected to/from other nodes. For example, to connect the center of an ellipse to the
2D position of the Locator, right-click on the Mask center control and select Connect To > Locator 3D
> Position.

Inputs
Two inputs accept 3D scenes as sources. The orange scene input is required, while the green Target
input is optional.
– SceneInput: The required orange scene input accepts the output of a 3D scene. This scene
should contain the object or point in 3D space that you want to covert to 2D coordinates.
– Target: The optional green target input accepts the output of a 3D scene. When provided, the
transform center of the scene is used to set the position of the Locator. The transformation
controls for the Locator become offsets from this position.

Basic Node Setup

The scene provided to the Locator’s input must contain the camera through which the coordinates are
projected. So, the best practice is to place the Locator after the Merge that introduces the camera to
the scene.
If an object is connected to the Locator node’s target input, the Locator is positioned at the object’s
center, and the Transformation tab’s translation XYZ sliders function in the object’s local coordinate
space instead of global scene space. This is useful for tracking an object’s position despite any
additional transformations applied further downstream.

Locator 3D connected after a Merge 3D with the SpotLight as the target

Chapter 29 3D Nodes 677

Inspector

Locator 3D controls

Controls Tab
Most of the controls for the locator 3D are cosmetic, dealing with how the locator appears and whether
it is rendered in the final output. However, the Camera Settings are critical to getting the results you’re
looking for.

Size
The Size slider is used to set the size of the Locator’s onscreen crosshair.

Color
A basic Color control is used to set the color of the Locator’s onscreen crosshair.

Matte
Enabling the Is Matte option applies a special texture to this object, causing this object to not only
become invisible to the camera, but also making everything that appears directly behind the camera
invisible as well. This option overrides all textures. For more information, see Chapter 86,
“3D Compositing Basics” in the DaVinci Resolve Reference Manual or Chapter 25 in the Fusion
Reference Manual.
– Is Matte: When activated, objects whose pixels fall behind the matte object’s
pixels in Z do not get rendered.
– Opaque Alpha: Sets the Alpha value of the matte object to 1. This checkbox is visible only when
the Is Matte option is enabled.
– Infinite Z: Sets the value in the Z-channel to infinity. This checkbox is visible only when the Is
Matte option is enabled.

Chapter 29 3D Nodes 678

Sub ID
The Sub ID slider can be used to select an individual subelement of certain geometry, such as an
individual character produced by a Text 3D node or a specific copy created by a Duplicate 3D node.

Make Renderable
Defines whether the Locator is rendered as a visible object by the OpenGL renderer. The software
renderer is not currently capable of rendering lines and hence ignores this option.

Unseen by Camera
This checkbox control appears when the Make Renderable option is selected. If the Unseen by
Camera checkbox is selected, the Locator is visible in the viewers but not rendered into the output
image by the Renderer 3D node.

Camera
This drop-down control is used to select the Camera in the scene that defines the screen space used
for 3D to 2D coordinate transformation.

Use Frame Format Settings

Select this checkbox to override the width, height, and pixel aspect controls, and force them to use the
values defined in the composition’s Frame Format preferences instead.

Width, Height, and Pixel Aspect

In order for the Locator to generate a correct 2D transformation, it must know the dimensions and
aspect of the image. These controls should be set to the same dimensions as the image produced by
a renderer associated with the camera specified above. Right-clicking on these controls displays a
contextual menu containing the frame formats configured in the composition’s preferences.

Common Controls
Transform and Settings tabs
The remaining Transform and Settings tabs are common to many 3D nodes. Their descriptions can be
found in “The Common Controls” section at the end of this chapter.

Merge 3D [3MG]

The Merge 3D node

Merge 3D Introduction
The Merge 3D node is the primary node in Fusion that you use to combine separate 3D elements into
the same 3D environment.
For example, in a scene created with an image plane, a camera, and a light, the camera would not be
able to see the image plane and the light would not affect the image plane until all three objects are
introduced into the same environment using the Merge 3D node.
The Merge provides the standard transformation controls found on most nodes in Fusion’s 3D suite.
Unlike those nodes, changes made to the translation, rotation, or scale of the Merge affect all the
objects connected to the Merge. This behavior forms the basis for all parenting in
Fusion’s 3D environment.

Chapter 29 3D Nodes 679

Inputs
The Merge node displays only two inputs initially, but as each input is connected a new input appears
on the node, assuring there is always one free to add a new element into the scene.
– SceneInput[#]: These multicolored inputs are used to connect image planes, 3D cameras,
lights, entire 3D scenes, as well as other Merge 3D nodes. There is no limit to the number of
inputs this node can accept. The node dynamically adds more inputs as needed, ensuring that
there is always at least one input available for connection.

Basic Node Setup

The Merge 3D is the hub of a 3D composite. All elements in a 3D scene connect into a Merge 3D.
Multiple Merge 3D nodes can be strung together to control lighting or for neater organization. The last
Merge 3D in a string must connect to a Renderer 3D to be output as a 2D image.

Merge 3D with a connected Image Plane, FBX Mesh object, SpotLight, and camera

Inspector

Merge 3D controls

Controls Tab
The Controls tab is used only to pass through any lights connected to the Merge 3D node.

Pass Through Lights

When the Pass Through Lights checkbox is selected, lights are passed through the Merge into its
output to affect downstream elements. Normally, the lights are not passed downstream to affect the
rest of the scene. This is frequently used to ensure projections are not applied to geometry introduced
later in the scene.

Common Controls
Transform and Settings Tabs
The remaining controls for the Transform and Settings tabs are common to most 3D nodes. Their
descriptions can be found in “The Common Controls” section at the end of this chapter.

Chapter 29 3D Nodes 680

Override 3D [3OV]

The Override 3D node

Override 3D Node Introduction

The Override node lets you change object-specific options for every object in a 3D scene
simultaneously. This is useful, for example, when you wish to set every object in the input scene to
render as a wireframe. Additionally, this node is the only way to set the wireframe, visibility, lighting,
matte, and ID options for 3D particle systems and the Text 3D node.

Inputs
– SceneInput: The orange Scene input accepts the output of a Merge 3D node or any node
creating a 3D scene.

Basic Node Setup

The Override 3D node is frequently used in conjunction with the Replace Material node to produce
isolated passes. For example, in the node tree below, a scene branches out to an Override node that
turns off the Affected by Lights property of each node, then connects to a Replace Material node that
applies a Falloff shader to produce a falloff pass of the scene.

Override 3D connected to a Replace Material node

Chapter 29 3D Nodes 681

Inspector

Override 3D controls

Controls Tab
The function of the controls found in the Controls tab is straightforward. First, you select the option to
override using the Do [Option] checkbox. That reveals a control that can be used to set the value of
the option itself. The individual options are not documented here; a full description of each can be
found in any geometry creation node in this chapter, such as the Image Plane, Cube, or Shape nodes.

Do [Option]
Enables the override for this option.

[Option]
If the Do [Option] checkbox is enabled, then the control for the property itself becomes visible. The
control values of the properties for all upstream objects are overridden by the new value.

Common Controls
Settings Tabs
The Settings tab includes controls common to most 3D nodes. Their descriptions can be found in “The
Common Controls” section at the end of this chapter.

Chapter 29 3D Nodes 682

Point Cloud 3D [3PC]

The PointCloud 3D node

Point Cloud 3D Node Introduction

A Point Cloud is generally many null objects created by 3D tracking or modeling software.
When produced by 3D tracking software, the points typically represent each of the patterns tracked to
create the 3D camera path. These point clouds can be used to identify a ground plane and to orient
other 3D elements with the tracked image. The Point Cloud 3D node creates a point cloud either by
importing a file from a 3D tracking application or generating it when you use the Camera Tracker node.

NOTE: A null object is an invisible 3D object that has all the same transform properties of a
visible 3D object.

Inputs
The Point Cloud has only a single input for a 3D scene.
– SceneInput: This orange input accepts a 3D scene.

Basic Node Setup

The Point Cloud 3D node is viewed and connected through a Merge 3D node, integrating it into the
larger 3D scene.

Point Cloud 3D connected and viewed through a Merge 3D

Chapter 29 3D Nodes 683

Inspector

Point Cloud 3D controls

Controls Tab
The Controls tab is where you can import the point cloud from a file and controls its appearance in
the viewer.

Style
The Style menu allows you to display the point cloud as cross hairs or points in the viewer.

Lock X/Y/Z
Deselect this checkbox to provide individual control over the size of the X, Y, and Z arms of the points
in the cloud.

Size X/Y/Z
These sliders can be used to increase the size of the onscreen crosshairs used to represent
each point.

Density
This slider defines the probability of displaying a specific point. If the value is 1, then all points are
displayed. A value of 0.2 shows only every fifth point.

Color
Use the standard Color control to set the color of onscreen crosshair controls.

Import Point Cloud

The Import Point Cloud button displays a dialog to import a point cloud from another application.

Chapter 29 3D Nodes 684

Supported file types are:

Alias's Maya .ma

3DS Max ASCII Scene Export .ase

NewTek's LightWave .lws

Softimage XSI's .xsi.

Make Renderable
Determines whether the point cloud is visible in the OpenGL viewer and in final renderings made by
the OpenGL renderer. The software renderer does not currently support rendering of visible
crosshairs for this node.

Unseen by Camera
This checkbox control appears when the Make Renderable option is selected. If the Unseen by
Cameras checkbox is selected, the point cloud is visible in the viewers but not rendered into the
output image by the Renderer 3D node.

Common Controls
Transform and Settings Tabs
The remaining Transform and Settings tabs are common to many 3D nodes. Their descriptions can be
found in “The Common Controls” section at the end of this chapter.

Onscreen Contextual Menu

The Point Cloud 3D contextual menu options

Frequently, one or more of the points in an imported point cloud is manually assigned to track the
position of a specific feature. These points usually have names that distinguish them from the rest of
the points in the cloud. To see the current name for a point, hover the mouse pointer directly over a
point, and after a moment a small tooltip appears with the name of the point.
When the Point Cloud 3D node is selected, a submenu is added to the viewer’s contextual menu with
several options that make it simple to locate, rename, and separate these points from the rest of the
point cloud.

Chapter 29 3D Nodes 685

The contextual menu contains the following options:
– Find: Selecting this option from the viewer contextual menu opens a dialog to search for and
select a point by name. Each point that matches the pattern is selected.
– Rename: Rename any point by selecting Rename from the contextual menu. Type the new name
into the dialog that appears and press Return. The point now has that name, with a four-digit
number added to the end. For example, the Name window is window0000, and multiple points
would be window0000, window0001, and so on. Names must be valid Fusion identifiers (i.e., no
spaces allowed, and the name cannot start with a number).
– Delete: Selecting this option deletes the currently selected points.
– Publish: Normally, the exact position of a point in the cloud is not exposed. To expose the
position, select the points, and then select the Publish option from this contextual menu.
This adds a coordinate control to the control panel for each published point that displays the
point’s current location.

Additional Toolbar and Shortcuts

Delete Selected Points Del

Select All Shift+A

Find Points Shift+F

Rename Selected Points F2

Create New Point Shift+C

Toggle Names on None/Selected/Published/All Points Shift+N

Toggle Locations on None/Selected/Published/All Points Shift+L

Publish Selected Points Shift+P

Unpublish Selected Points Shift+U

Create a Shape at Selected Points Shift+S

Create and Fit an ImagePlane to Selected Points Shift+I

Create a Locator at Selected Points Shift+O

Chapter 29 3D Nodes 686

Projector 3D [3PJ]

The Projector 3D node

Projector 3D Node Introduction

The Projector 3D node is used to project an image upon 3D geometry. This can be useful in many
ways: texturing objects with multiple layers, applying a texture across multiple separate objects,
projecting background shots from the camera’s viewpoint, image-based rendering techniques, and
more. The Projector node is just one of several nodes capable of projecting images and textures.
Each method has advantages and disadvantages. For more information, see Chapter 86, “3D
Compositing Basics,” in the DaVinci Resolve Reference Manual or Chapter 25 in the Fusion
Reference Manual.
Projected textures can be allowed to “slide“ across the object if the object moves relative to the
Projector 3D, or, alternatively, by grouping the two with a Merge 3D so they can be moved as one and
the texture remains locked to the object.
The Projector 3D node’s capabilities and restrictions are best understood if the Projector is considered
to be a variant on the SpotLight node. The fact that the Projector 3D node is actually a light has several
important consequences when used in Light or Ambient Light projection mode:
– Lighting must be turned on for the results of the projection to be visible.
– The light emitted from the projector is treated as diffuse/specular light. This means that it is
affected by the surface normals and can cause specular highlights. If this is undesirable, set the
Projector 3D to project into the Ambient Light channel.
– Enabling Shadows causes Projector 3D to cast shadows.
– Just as with other lights, the light emitted by a Projector 3D only affects objects that feed into the
first Merge 3D that is downstream of the Projector 3D node in the node tree.
– Enabling Merge 3D’s Pass Through Lights checkbox allows the projection to light objects further
downstream.
– The light emitted by a Projector 3D is controlled by the Lighting options settings on objects and
the Receives Lighting options on materials.
– Alpha values in the projected image do not clip geometry in Light or Ambient Light mode. Use
Texture mode instead.
– If two projections overlap, their light contributions are added.

To project re-lightable textures or textures for non-diffuse color channels (like Specular Intensity or
Bump), use the Texture projection mode instead:
– Projections in Texture mode only strike objects that use the output of the Catcher node for all or
part of the material applied to that object.
– Texture mode projections clip the geometry according to the Alpha channel of the
projected image.

See the section for the Catcher node for additional details.

Chapter 29 3D Nodes 687

Camera Projection vs. Projection 3D Node
The Camera 3D node also provides a projection feature, and should be used when the projection is
meant to match a camera, as this node has more control over aperture, film back, and clip planes. The
Projector 3D node was designed to be used as a custom light in 3D scenes for layering and texturing.
The projector provides better control over light intensity, color, decay, and shadows.

Inputs
The Projector 3D has two inputs: one for the scene you are projecting on to and another for the
projected image.
– SceneInput: The orange scene input accepts a 3D scene. If a scene is connected to this input,
then transformations applied to the spotlight also affect the rest of the scene.
– ProjectiveImage: The white input expects a 2D image to be used for the projection. This
connection is required.

Basic Node Setup

As an example, the Projector 3D node below is used to project a texture (MediaIn2) onto 3D primitives
as a way to create a simple 3D set. All the set elements are connected into a Merge 3D, which outputs
the projected set into a larger scene with camera, lights, and other elements. As an alternative, the
Projector 3D node could be inserted between the two Merge 3D nodes; however, the transform
controls in the Projector 3D node would then affect the entire scene.

Projector 3D texturing groups of shapes to construct a set

Inspector

Projector 3D controls

Chapter 29 3D Nodes 688

Controls Tab
Enabled
When this checkbox is enabled, the projector affects the scene. Disable the checkbox to turn off the
projector. This is not the same as the red switch in the upper-left corner of the Inspector. The red
switch disables the tool altogether and passes the image on without any modification. The Enabled
checkbox is limited to the effect part of the tool. Other parts, like scripts in the Settings tab still process
as normal.

Color
The input image is multiplied by this color before being projected into the scene.

Intensity
Use this slider to set the Intensity of the projection when the Light and Ambient Light projection modes
are used. In Texture mode, this option scales the Color values of the texture after multiplication by
the color.

Decay Type
A projector defaults to No Falloff, meaning that its light has equal intensity on geometry, despite the
distance from the projector to the geometry. To cause the intensity to fall off with distance, set the
Decay type to either Linear or Quadratic modes.

Angle
The Cone Angle of the node refers to the width of the cone where the projector emits its full intensity.
The larger the angle, the wider the cone angle, up to a limit of 90 degrees.

Fit Method
The Fit Method determines how the projection is fitted within the projection cone.
The first thing to know is that although this documentation may call it a “cone,” the Projector 3D and
Camera 3D nodes do not project an actual cone; it’s more of a pyramid of light with its apex at the
camera/projector. The Projector 3D node always projects a square pyramid of light—i.e., its X and Y
angles of view are the same. The pyramid of light projected by the Camera 3D node can be non-
square depending on what the Film Back is set to in the camera. The aspect of the image connected
into the Projector 3D/Camera 3D does not affect the X/Y angles of the pyramid, but rather the image is
scaled to fit into the pyramid based upon the fit options.
When both the aspect of the pyramid (AovY/AovX) and the aspect of the image (height * pixelAspectY)/
(width * pixelAspectX) are the same, there is no need for the fit options, and in this case the fit options
all do the same thing. However, when the aspect of the image and the pyramid (as determined by the
Film Back settings in Camera 3D) are different, the fit options become important.
For example, Fit by Width fits the width of the image across the width of the Camera 3D pyramid. In
this case, if the image has a greater aspect ratio than the aspect of the pyramid, some of the projection
extends vertically outside of the pyramid.
There are five options:
– Inside: The image is uniformly scaled so that its largest dimension fits inside the cone. Another
way to think about this is that it scales the image as big as possible subject to the restriction that
the image is fully contained within the pyramid of the light. This means, for example, that nothing
outside the pyramid of light ever receives any projected light.
– Width: The image is uniformly scaled so that its width fits inside the cone. Note that the image
could still extend outside the cone in its height direction.
– Height: The image is uniformly scaled so that its height fits inside the cone. Note that the image
could still extend outside the cone in its width direction.

Chapter 29 3D Nodes 689

 – Outside: The image is uniformly scaled so that its smallest dimension fits inside the cone. Another
way to think about this is that it scales the image as small as possible subject to the restriction
that the image covers the entire pyramid (i.e., the pyramid is fully contained within the image). This
means that any pixel of any object inside the pyramid of light always gets illuminated.
– Stretch: The image is non-uniformly scaled, so it exactly covers the cone of the projector.

Projection Mode
– Light: Projects the texture as a diffuse/specular light.
– Ambient Light: Uses an ambient light for the projection.
– Texture: When used in conjunction with the Catcher node, this mode allows re-lightable texture
projections. The projection strikes only objects that use the catcher material as part of their
material shaders.
One useful trick is to connect a Catcher node to the Specular Texture input on a 3D Material node
(such as a Blinn). This causes any object using the Blinn material to receive the projection as part
of the specular highlight. This technique can be used in any material input that uses texture maps,
such as the Specular and Reflection maps.

Shadows
Since the projector is based on a spotlight, it is also capable of casting shadows using shadow maps.
The controls under this reveal are used to define the size and behavior of the shadow map.
– Enable Shadows: The Enable Shadows checkbox should be selected if the light is to produce
shadows. This defaults to selected.
– Shadow Color: Use this standard Color control to set the color of the shadow.
This defaults to black (0, 0, 0).
– Density: The Shadow Density determines the transparency of the shadow. A density of 1.0
produces a completely transparent shadow, whereas lower values make the shadow transparent.
– Shadow Map Size: The Shadow Map Size control determines the size of the bitmap used to create
the shadow map. Larger values produce more detailed shadow maps at the expense of memory
and performance.
– Shadow Map Proxy: The Shadow Map Proxy determines the size of the shadow map used for
proxy and auto proxy calculations. A value of 0.5 would use a 50% shadow map.
– Multiplicative/Additive Bias: Shadows are essentially textures applied to objects in the scene,
so there is occasionally Z-fighting, where the portions of the object that should be receiving the
shadows render over the top of the shadow instead.
– Multiplicative and Additive Bias: Bias works by adding a small depth offset to move the shadow
away from the surface it is shadowing, eliminating the Z-fighting. Too little bias and the objects can
self-shadow themselves. Too much bias and the shadow can become separated from the surface.
Adjust the multiplicative bias first, then fine tune the result using the additive bias control.
– Force All Materials Non-Transmissive: Normally, an RGBAZ shadow map is used when rendering
shadows. By enabling this option, you are forcing the renderer to use a Z-only shadow map.
This can lead to significantly faster shadow rendering while using a fifth as much memory. The
disadvantage is that you can no longer cast “stained-glass”-like shadows.
– Shadow Map Sampling: Sets the quality for sampling of the shadow map.

Chapter 29 3D Nodes 690

 – Softness: Soft edges in shadows are produced by filtering the shadow map when it is
sampled. Fusion provides three separate filtering methods that produce different effects when
rendering shadows.
– None: Shadows have a hard edge. No filtering of the shadow map is done at all. The advantage
of this method is that you only have to sample one pixel in the shadow map, so it is fast.
– Constant: Shadow edges have a constant softness. A filter with a constant width is used when
sampling the shadow map. Adjusting the Constant Softness slider controls the size of the
filter. Note that the larger you make the filter, the longer it takes to render the shadows. If the
Softness is set to constant, then a Constant slider appears. It can be used to set the overall
softness of the shadow.
– Variable: The softness of shadow edges grows the farther away the shadow receiver is from
the shadow caster. The variable softness is achieved by changing the size of the filter based
on the distance between the receiver and caster. When this option is selected, the Softness
Falloff, Min Softness and Max Softness sliders appear.

Softness Falloff The Softness Falloff slider appears when the Softness is set to variable.
This slider controls how fast the softness of shadow edges grows with distance.
More precisely, it controls how fast the shadow map filter size grows based on
the distance between shadow caster and receiver. Its effect is mediated by the
values of the Min and Max Softness sliders.

Min Softness The Min Softness slider appears when the Softness is set to variable. This slider
controls the Minimum Softness of the shadow. The closer the shadow is to the
object casting the shadow, the sharper it is up to the limit set by this slider.

Max Softness The Max Softness slider appears when the Softness is set to variable.
This slider controls the Maximum Softness of the shadow. The farther the
shadow is from the object casting the shadow, the softer it is up to the limit set
by this slider.

Common Controls
Transform and Settings Tabs
The remaining Transform and Settings tabs are common to many 3D nodes. Their descriptions can be
found in “The Common Controls” section at the end of this chapter.

Renderer 3D [3RN]

The Renderer 3D node

Renderer 3D Node Introduction

The Renderer 3D node converts the 3D environment into a 2D image using either a default
perspective camera or one of the cameras found in the scene. Every 3D scene in a composition
terminates with at least one Renderer 3D node. The Renderer node includes a software and OpenGL
render engine to produce the resulting image. Additional render engines may also be available via
third-party plug-ins.

Chapter 29 3D Nodes 691

The software render engine uses the system’s CPU only to produce the rendered images. It is usually
much slower than the OpenGL render engine, but produces consistent results on all machines, making
it essential for renders that involve network rendering. The Software mode is required to produce soft
shadows, and generally supports all available illumination, texture, and material features.
The OpenGL render engine employs the GPU processor on the graphics card to accelerate the
rendering of the 2D images. The output may vary slightly from system to system, depending on the
exact graphics card installed. The graphics card driver can also affect the results from the OpenGL
renderer. The OpenGL render engines speed makes it possible to provide customized supersampling
and realistic 3D depth of field options. The OpenGL renderer cannot generate soft shadows. For soft
shadows, the software renderer is recommended.
Like most nodes, the Renderer’s motion blur settings can be found under the Common Controls tab.
Be aware that scenes containing particle systems require that the Motion Blur settings on the pRender
nodes exactly match the settings on the Renderer 3D node.
Otherwise, the subframe renders conflict producing unexpected (and incorrect) results.

NOTE: The Open GL renderer respects the Color Depth option in the Image tab of the
Renderer 3D node. This can cause slowdowns on certain graphics cards when rendering to
int16 or float32.

Inputs
The Renderer 3D node has two inputs on the node. The main scene input takes in the Merge 3D or
other 3D nodes that need to be converted to 2D. The effect mask limits the Renderer 3D output.
– SceneInput: The orange scene input is a required input that accepts a 3D scene that you want
to convert to 2D.
– EffectMask: The blue effects mask input uses a 2D image to mask the output of the node.

Basic Node Setup

All 3D scenes must end with a Renderer 3D node. The Renderer 3D node is used to convert a 3D
scene into a 2D image. Below, the Renderer 3D node takes the output of a Merge 3D node, and
renders the 3D scene into a 2D image.

Renderer 3D connected directly after a Merge 3D, rendering the 3D scene to a 2D image

Chapter 29 3D Nodes 692

Inspector

Render 3D controls

Controls Tab
Camera
The Camera menu is used to select which camera from the scene is used when rendering. The Default
setting uses the first camera found in the scene. If no camera is located, the default perspective view
is used instead.

Eye
The Eye menu is used to configure rendering of stereoscopic projects. The Mono option ignores the
stereoscopic settings in the camera. The Left and Right options translate the camera using the stereo
Separation and Convergence options defined in the camera to produce either left- or right-eye
outputs. The Stacked option places the two images one on top of the other instead of side by side.

Reporting
The first two checkboxes in this section can be used to determine whether the node prints warnings
and errors produced while rendering to the console. The second set of checkboxes tells the node
whether it should abort rendering when a warning or error is encountered. The default for this node
enables all four checkboxes.

Renderer Type
This menu lists the available render engines. Fusion provides three: the software renderer, OpenGL
renderer, and the OpenGL UV render engine. Additional renderers can be added via third-
party plug-ins.

Chapter 29 3D Nodes 693

All the controls found below this drop-down menu are added by the render engine. They may change
depending on the options available to each renderer. So, each renderer is described in its own
section below.

Software Controls
Output Channels
Besides the usual Red, Green, Blue, and Alpha channels, the software renderer can also embed the
following channels into the image. Enabling additional channels consumes additional memory and
processing time, so these should be used only when required.
– RGBA: This option tells the renderer to produce the Red, Green, Blue, and Alpha color channels of
the image. These channels are required, and they cannot be disabled.
– Z: This option enables rendering of the Z-channel. The pixels in the Z-channel contain a value that
represents the distance of each pixel from the camera. Note that the Z-channel values cannot
include anti-aliasing. In pixels where multiple depths overlap, the frontmost depth value is used for
this pixel.
– Coverage: This option enables rendering of the Coverage channel. The Coverage channel
contains information about which pixels in the Z-buffer provide coverage (are overlapping with
other objects). This helps nodes that use the Z-buffer to provide a small degree of anti-aliasing.
The value of the pixels in this channel indicates, as a percentage, how much of the pixel is
composed of the foreground object.
– BgColor: This option enables rendering of the BgColor channel. This channel contains the color
values from objects behind the pixels described in the Coverage channel.
– Normal: This option enables rendering of the X, Y, and Z Normals channels. These three channels
contain pixel values that indicate the orientation (direction) of each pixel in the 3D space. A color
channel containing values in a range from [–1,1] represents each axis.
– TexCoord: This option enables rendering of the U and V mapping coordinate channels. The pixels
in these channels contain the texture coordinates of the pixel. Although texture coordinates are
processed internally within the 3D system as three-component UVW, Fusion images store only UV
components. These components are mapped into the Red and Green color channel.
– ObjectID: This option enables rendering of the ObjectID channel. Each object in the 3D
environment can be assigned a numeric identifier when it is created. The pixels in this floating-
point image channel contain the values assigned to the objects that produced the pixel. Empty
pixels have an ID of 0, and the channel supports values as high as 65534. Multiple objects can
share a single Object ID. This buffer is useful for extracting mattes based on the shapes of objects
in the scene.
– MaterialID: This option enables rendering of the Material ID channel. Each material in the 3D
environment can be assigned a numeric identifier when it is created. The pixels in this floating-
point image channel contain the values assigned to the materials that produced the pixel. Empty
pixels have an ID of 0, and the channel supports values as high as 65534. Multiple materials
can share a single Material ID. This buffer is useful for extracting mattes based on a texture; for
example, a mask containing all the pixels that comprise a brick texture.

Lighting
– Enable Lighting: When the Enable Lighting checkbox is selected, objects are lit by any lights in
the scene. If no lights are present, all objects are black.
– Enable Shadows: When the Enable Shadows checkbox is selected, the renderer produces
shadows, at the cost of some speed.

Chapter 29 3D Nodes 694

OpenGL Controls

Render 3D Open GL controls

Output Channels
In addition to the usual Red, Green, Blue, and Alpha channels, the OpenGL render engine can also
embed the following channels into the image. Enabling additional channels consumes additional
memory and processing time, so these should be used only when required.
– RGBA: This option tells the renderer to produce the Red, Green, Blue, and Alpha color channels of
the image. These channels are required, and they cannot be disabled.
– Z: This option enables rendering of the Z-channel. The pixels in the Z-channel contain a value that
represents the distance of each pixel from the camera. Note that the Z-channel values cannot
include anti-aliasing. In pixels where multiple depths overlap, the frontmost depth value is used for
this pixel.
– Normal: This option enables rendering of the X, Y, and Z Normals channels. These three channels
contain pixel values that indicate the orientation (direction) of each pixel in the 3D space. A color
channel containing values in a range from [–1,1] is represented by each axis.

Chapter 29 3D Nodes 695

 – TexCoord: This option enables rendering of the U and V mapping coordinate channels. The pixels
in these channels contain the texture coordinates of the pixel. Although texture coordinates are
processed internally within the 3D system as three-component UVW, Fusion images store only UV
components. These components are mapped into the Red and Green color channels.
– ObjectID: This option enables rendering of the ObjectID channel. Each object in the 3D
environment can be assigned a numeric identifier when it is created. The pixels in this floating-
point image channel contain the values assigned to the objects that produced the pixel. Empty
pixels have an ID of 0, and the channel supports values as high as 65534. Multiple objects can
share a single Object ID. This buffer is useful for extracting mattes based on the shapes of objects
in the scene.
– MaterialID: This option enables rendering of the Material ID channel. Each material in the 3D
environment can be assigned a numeric identifier when it is created. The pixels in this floating-
point image channel contain the values assigned to the materials that produced the pixel. Empty
pixels have an ID of 0, and the channel supports values as high as 65534. Multiple materials
can share a single Material ID. This buffer is useful for extracting mattes based on a texture—for
example, a mask containing all the pixels that comprise a brick texture.

Anti-Aliasing
Anti-aliasing can be enabled for each channel through the Channel menu. It produces an output image
with higher quality anti-aliasing by brute force, rendering a much larger image, and then rescaling it
down to the target resolution. Rendering a larger image in the first place, and then using a Resize node
to bring the image to the desired resolution can achieve the exact same results. Using the
supersampling built in to the renderer offers two distinct advantages over this method.
The rendering is not restricted by memory or image size limitations. For example, consider the steps to
create a float-16 1920 x 1080 image with 16x supersampling. Using the traditional Resize node would
require first rendering the image with a resolution of 30720 x 17280, and then using a Resize to scale
this image back down to 1920 x 1080. Simply producing the image would require nearly 4 GB of
memory. When anti-aliasing is performed on the GPU, the OpenGL renderer can use tile rendering to
significantly reduce memory usage.
The GL renderer can perform the rescaling of the image directly on the GPU more quickly than the
CPU can manage it. Generally, the more GPU memory the graphics card has, the faster the operation
is performed.
Interactively, Fusion skips the anti-aliasing stage unless the HiQ button is selected in the Time Ruler.
Final quality renders always include supersampling, if it is enabled.
Because of hardware limitations, point geometry (particles) and lines (locators) are always rendered at
their original size, independent of supersampling. This means that these elements are scaled down
from their original sizes, and likely appear much thinner than expected.

Anti-Aliasing of Aux Channels in the OpenGL Renderer

The reason Fusion supplies separate anti-aliasing options for color and aux channels in the Anti-
Aliasing preset is that supersampling of color channels is quite a bit slower than aux channels. You may
find that 1 x 3 LowQ/HiQ Rate is sufficient for color, but for world position or Z, you may require 4 x 12 to
get adequate results. The reasons color anti-aliasing is slower are that the shaders for RGBA can be
10x to even 100x or 1000x more complex, and color is rendered with sorting enabled, while aux
channels get rendered using the much faster Z-buffer method.

TIP: For some things, sometimes using an SS Z-buffer improves quality, but for other things
like using the merge’s PerformDepthMerge option, it may make things worse.

Chapter 29 3D Nodes 696

Do not mistake anti-aliasing with improved quality. Anti-aliasing an aux channel does not mean it’s
better quality. In fact, anti-aliasing an aux channel in many cases can make the results much worse.
The only aux channels we recommend you enable anti-aliasing on are WorldCoord and Z.

TIP: We strongly recommend disabling Anti-Aliasing on Material ID and Object ID channels,

TexCoord, Normal, BackVector, and Vector channels. The issue arises when you have
multiple 3D surfaces with radically different TexCoord values in one pixel. The anti-aliasing
does not restrict itself to sampling the main surface but samples both surfaces. For example,
if one surface has TexCoords that are approximately (u,v) = (0, 0) within that pixel, and the
other surface has (0.5, 0.5), you get a blending of these two. The blended area of the texture
could have colors like (0, 0) or (0.5, 0.5), resulting in an oddly colored pixel artifact being
output from the 2D Texture node. The same problem can happen for normals.

Enable (LowQ/HiQ)
These two check boxes are used to enable anti aliasing of the rendered image.

Supersampling LowQ/HiQ Rate

The LowQ and HiQ rate tells the OpenGL render how large to scale the image. For example, if the rate
is set to 4 and the OpenGL renderer is set to output a 1920 x 1080 image, internally a 7680 x 4320
image is rendered and then scaled back to produce the target image. Set the multiplier higher to get
better edge anti-aliasing at the expense of render time. Typically 8 x 8 supersampling (64 samples per
pixel) is sufficient to reduce most aliasing artifacts.
The rate doesn’t exactly define the number of samples done per destination pixel; the width of the
reconstruction filter used may also have an impact.

Filter Type
When downsampling the supersized image, the surrounding pixels around a given pixel are often used
to give a more realistic result. There are various filters available for combining these pixels. More
complex filters can give better results but are usually slower to calculate. The best filter for the job
often depends on the amount of scaling and on the contents of the image itself.

The functions of these filters are shown in the image above. From left to right these are:

Box This is a simple interpolation scale of the image.

Bi-Linear (triangle) This uses a simplistic filter, which produces relatively clean and fast results.

This filter produces a nominal result. It offers a good compromise between

Bi-Cubic (quadratic)
speed and quality.

This produces better results with continuous tone images but is slower
Bi-Spline (cubic) than Quadratic. If the images have fine detail in them, the results may be
blurrier than desired.

This produces good results with continuous tone images which are scaled down,
Catmul-Rom
producing sharp results with finely detailed images.

Gaussian This is very similar in speed and quality to Quadratic.

This is similar to Catmull-Rom but produces better results with finely detailed
Mitchell
images. It is slower than Catmull-Rom.

Chapter 29 3D Nodes 697

 This is very similar to Mitchell and Catmull-Rom but is a little cleaner
Lanczos
and also slower.

This is an advanced filter that produces very sharp, detailed results, however,
Sinc
it may produce visible `ringing' in some situations.

Bessel This is similar to the Sinc filter but may be slightly faster.

Window Method
The Window Method menu appears only when the reconstruction filter is set to Sinc or Bessel.

Hanning This is a simple tapered window.

Hamming Hamming is a slightly tweaked version of Hanning.

Blackman A window with a more sharply tapered falloff.

Accumulation Effects
Accumulation effects are used for creating depth of field effects. Enable both the Enable Accumulation
Effects and Depth of Field checkboxes, and then adjust the quality and Amount sliders.
The blurrier you want the out-of-focus areas to be, the higher the quality setting you need.
A low amount setting causes more of the scene to be in focus.
The accumulation effects work in conjunction with the Focal plane setting located in the Camera 3D
node. Set the Focal Plane to the same distance from the camera as the subject you want to be in
focus. Animating the Focal Plane setting creates rack of focus effects.

Lighting
– Enable Lighting: When the Enable Lighting checkbox is selected, any lights in the scene light
objects. If no lights are present, all objects are black.
– Enable Shadows: When the Enable Shadows checkbox is selected, the renderer produces
shadows, at the cost of some speed.

Texturing
– Texture Depth: Lets you specify the bit depth of texture maps.
– Warn about unsupported texture depths: Enables a warning if texture maps are in an
unsupported bit depth that Fusion can’t process.

Lighting Mode
The Per-vertex lighting model calculates lighting at each vertex of the scene’s geometry. This
produces a fast approximation of the scene’s lighting but tends to produce blocky lighting on poorly
tessellated objects. The Per-pixel method uses a different approach that does not rely on the detail in
the scene’s geometry for lighting, so it generally produces superior results.
Although the per-pixel lighting with the OpenGL renderer produces results closer to that produced by
the more accurate software renderer, it still has some disadvantages. The OpenGL renderer is less
capable of dealing correctly with semi-transparency, soft shadows, and colored shadows, even with
per-pixel lighting. The color depth of the rendering is limited by the capabilities of the graphics card in
the system.

Chapter 29 3D Nodes 698

Transparency
The OpenGL renderer reveals this control for selecting which ordering method to use when
calculating transparency.
– Z Buffer (fast): This mode is extremely fast and is adequate for scenes containing only opaque
objects. The speed of this mode comes at the cost of accurate sorting; only the objects closest
to the camera are certain to be in the correct sort order. So, semi-transparent objects may not be
shown correctly, depending on their ordering within the scene.
– Sorted (accurate): This mode sorts all objects in the scene (at the expense of speed)
before rendering, giving correct transparency.
– Quick Mode: This experimental mode is best suited to scenes that almost exclusively
contain particles.

Shading Model
Use this menu to select a shading model to use for materials in the scene. Smooth is the shading
model employed in the viewers, and Flat produces a simpler and faster shading model.

Wireframe
Renders the whole scene as wireframe. This shows the edges and polygons of the objects. The edges
are still shaded by the material of the objects.

Wireframe Anti-Aliasing
Enables anti-aliasing for the Wireframe render.

OpenGL UV Renderer
The OpenGL UV renderer is a special case render engine. It is used to take a model with existing
textures and render it out to produce an unwound flattened 2D version of the model. Optionally,
lighting can be baked in. This is typically done so you can then paint on the texture and reapply it.

Render 3D Open GL UV controls

Chapter 29 3D Nodes 699

 Below are some issues to be aware of when using the OpenGL UV renderer.

Baked-in lighting: After you have baked lighting into a model’s texture, you need to be careful
to turn lighting off on the object later when you render it with the baked-in lighting texture.
Single textures/multiple destinations: Beware of cases where a single area of the texture map
is used on multiple areas of the model. This is often done to save texture memory and decrease
modeling time. An example is the texture for a person where the artist mirrored the left side
mesh/uvs/texture to produce the right side. Trying to bake in lighting in this case won’t work.
Unwrapped more the one mesh: Unwrapping more than one mesh at once can cause
problems. The reason is that most models are authored so they make maximum usage of (u,v)
in [0,1] x [0,1], so that in general models overlap each other in UV space.
Seams: When the UV gutter size is left at 0, this produces seams when the model is retextured
with the unwrapped texture.
UV Gutter Size: Increase this value to hide seams between faces.

Common Controls
Image and Settings Tabs
The remaining controls for the Image and Settings tabs are common to many 3D nodes.
Their descriptions can be found in “The Common Controls” section at the end of this chapter.

Replace Material 3D [3RPL]

The Replace Material 3D node

Replace Material 3D Node Introduction

The Replace Material 3D node replaces the material applied to all the geometry in the input scene with
its own material input. Any lights or cameras in the input scene are passed through unaffected.
The scope of the replacement can be limited using Object and Material identifiers in the Inspector.
The scope can also be limited to individual channels, making it possible to use a completely different
material on the Red channel, for example.
Since the Text 3D node does not include a material input, you can use the Replace Material to add
material shaders to the text.

Inputs
The Replace Material node has two inputs: one for the 3D scene, object, or 3D text that contains the
original material, and a material input for the new replacement material.
– SceneInput: The orange scene input accepts a 3D scene or 3D text that you want to replace
the material.
– MaterialInput: The green material input accepts either a 2D image or a 3D material. If a 2D
image is provided, it is used as a diffuse texture map for the basic material built into the node. If
a 3D material is connected, then the basic material is disabled.

Chapter 29 3D Nodes 700

Basic Node Setup
The Replace Material 3D node is inserted directly after the 3D object or scene whose material you
want to replace. Below, it is used to replace the default material on a Text 3D node with a
chrome shader.

Replace Material 3D used to replace the default material of a Text 3D node

Inspector

Replace Material 3D controls

Controls Tab
Enable
This checkbox enables the material replacement. This is not the same as the red switch in the upper-
left corner of the Inspector. The red switch disables the tool altogether and passes the image on
without any modification. The enable checkbox is limited to the effect part of the tool. Other parts, like
scripts in the Settings tab, still process as normal.

Replace Mode
The Replace Mode section offers four methods of replacing each RGBA channel:
– Keep: Prevents the channel from being replaced by the input material.
– Replace: Replaces the material for the corresponding color channel.
– Blend: Blends the materials together.
– Multiply: Multiplies the channels of both inputs.

Limit by Object ID/Material ID

When enabled, a slider appears where the desired IDs can be set. Other objects keep their materials.
If both options are enabled, an object must satisfy both conditions.

Chapter 29 3D Nodes 701

Common Controls
Material and Settings Tabs
The remaining controls for the Material and Settings tabs are common to many 3D nodes. Their
descriptions can be found in “The Common Controls” section at the end of this chapter.

Replace Normals 3D [3RPN]

The Replace Normals 3D node

Replace Normals Node Introduction

In 3D modeling, normals are vectors used to determine the direction light reflects off surfaces. The
Replace Normals node is used to replace the normals/tangents on incoming geometry, effectively
adjusting the surface of an object between smooth and flat. All geometry connected to the scene
input on the node is affected. Lights/Cameras/PointClouds/Locators/Materials, and other non-mesh
nodes are passed through unaffected. The normals/tangents affected by this node are Per-vertex
normals/tangents, not Per-face normals/tangents. The input geometry must have texture coordinates
in order for tangents to be computed. Sometimes geometry does not have texture coordinates, or the
texture coordinates were set to All by the FBX import because they were not present on the FBX.

Inputs
The Replace Normals node has a single input for the 3D scene or incoming geometry.
– SceneInput: The orange scene input accepts a 3D scene or 3D geometry that contains the
normal coordinates you want to modify.

Basic Node Setup

The Replace Normals 3D node is inserted directly after the 3D object or scene whose normals you
want to modify. Below, it is used to smooth the material on an imported 3D model.

Replace Normals 3D used to smooth normals on 3D geometry

Chapter 29 3D Nodes 702

Inspector

Replace Normals 3D controls

Control Tab
The options in the Control tab deal with repairing 3D geometry and then recomputing
normals/tangents.

Pre-Weld Position Vertices

Sometimes position vertices are duplicated in a geometry, even though they have the same position,
causing normals/tangents to be miscomputed. The results of pre-welding are thrown away; they do
not affect the output geometry’s position vertices.

Recompute
Controls when normals/tangents are recomputed.
– Always: The normals on the mesh are always recomputed.
– If Not Present: The normals on the mesh are recomputed only if they are not present.
– Never: The normals are never computed. This option is useful when animating.

Smoothing Angle
Adjacent faces with angles in degrees smaller than this value have their adjoining edges smoothed
across. A typical value one might choose for the Smoothing Angle is between 20 and 60 degrees.
There is special case code for 0.0f and 360.0f (f stands for floating-point value). When set to 0.0f,
faceted normals are produced; this is useful for artistic effect.

Ignore Smooth Groups

If set to False, two faces that have different Smooth Groups are not smoothed across (e.g., the faces of
a cube or the top surfaces of a cylinder have different Smooth Groups). If you check this On and set
the smoothing angle large enough, the faces of a cube are smoothed across. There is currently no
way to visualize Smooth Groups within Fusion.

Flip Normals
Flipping of tangents can sometimes be confusing. Flip has an effect if the mesh has tangent vectors.
Most meshes in Fusion don’t have tangent vectors until they reach a Renderer 3D, though. Also, when
viewing tangent vectors in the viewers, the tangent vectors are created if they don’t exist. The
confusing thing is if you view a Cube 3D that has no tangent vectors and press the FlipU/FlipV button,
nothing happens. This is a result of there being no tangent vectors to create, but later the GL renderer
can create some (unflipped) tangent vectors.

Chapter 29 3D Nodes 703

 There are five items you should be aware of when dealing with normals.

#1 The FBX importer recomputes the normals if they don’t exist, but you can get a higher-
quality result from the Replace Normals node.
#2 Bump maps can sometimes depend on the model’s normals. Specifically, when you
simplify a complex high polygon model to a low polygon model + bump map, the normals and
bump map can become “linked.” Recomputing the normals in this case can make the model
look funny. The bump map was intended to be used with the original normals.
#3 Most primitives in Fusion are not generated with tangents; when needed, they are
generated on the fly by a Renderer 3D and cached.
#4 Tangents currently are only needed for bump mapping. If a material needs bump mapping,
then tangents are created. These tangents are created with some default settings (e.g.,
Smoothing Angle, and so on). If you don’t want Fusion automatically creating tangents, you
can use the Replace Normals node to create them manually.
#5 All computations are done in the local coordinates of the geometries instead of in the
coordinate system of the Replace Normals 3D node. This can cause problems when there is a
non-uniform scale applied to the geometry before Replace Normals 3D is applied.

Common Controls
Settings Tab
The Settings tab is common to many 3D nodes. The description of these controls can be found in
“The Common Controls” section at the end of this chapter.

Replicate 3D [3REP]

The Replicate 3D node

Replicate 3D Node Introduction

The Replicate 3D node replicates input geometry at positions of destination vertices. The vertices can
be mesh vertices as well as particle positions. For each copy of the replicated input geometry, various
transformations can be applied. The options in the Jitter tab allow non-uniform transformations, such
as random positioning or sizes.

Inputs
There are two inputs on the Replicate 3D node: one for the destination geometry that contains the
vertices, and one for the 3D geometry you want to replicate.
– Destination: The orange destination input accepts a 3D scene or geometry with vertex
positions, either from the mesh or 3D particle animations.
– Input[#]: The input accepts the 3D scene or geometry for replicating. Once this input is
connected, a new input for alternating 3D geometry is created.

At least one connected input is required.

Chapter 29 3D Nodes 704

Basic Node Setup
In the example below, a Replicate 3D node is inserted directly after the pRender node. A spaceship
FBX node is connected to the green input representing the object that will be replicated based on the
particles. Each particle cell takes on the shape of the 3D geometry connected to the input.

Replicate 3D used to create an armada of swarming spaceships

Inspector

Replicated 3D Jitter controls

Controls Tab
Step
Defines how many positions are skipped. For example, a step of 3 means that only every third vertice
of the destination mesh is used, while a step of 1 means that all positions are used.
The Step setting helps to keep reasonable performance for big destination meshes. On parametric
geometry like a torus, it can be used to isolate certain parts of the mesh.

Chapter 29 3D Nodes 705

Point clouds are internally represented by six points once the Make Renderable option has been set.
To get a single point, use a step of 6 and set an X offset of –0.5 to get to the center of the point cloud.
Use –0.125 for Locator 3Ds. Once these have been scaled, the offset may differ.

Input Mode
This menu defines in which order multiple input scenes are replicated at the destination. No matter
which setting you choose, if only one input scene is supplied this setting has no effect.
– When set to Loop, the inputs are used successively. The first input is at the first position, the
second input at the second position, and so on. If there are more positions in the destination
present than inputs, the sequence is looped.
– When set to Random, a definite but random input for each position is used based on the seed in
the Jitter tab. This input mode can be used to simulate variety with few input scenes.
– The Death of Particles setting causes the input geometries’ IDs to change; therefore, their copy
order may change.

The Input Mode Loop vs. Random order

Time Offset
Use the Time Offset slider to offset any animations that are applied to the input geometry by a set
amount per copy. For example, set the value to –1.0 and use a cube set to rotate on the Y-axis as the
source. The first copy shows the animation from a frame earlier; the second copy shows animation
from a frame before that, etc.
This can be used with great effect on textured planes—for example, where successive frames of a
video clip can be shown.

Alignment
Alignment specifies how to align the copies in respect of the destination mesh normal or
particle rotation.
– Not Aligned: Does not align the copy. It stays rotated in the same direction as its input mesh.

Replicate 3D Not Aligned layout

Chapter 29 3D Nodes 706

 – Aligned: This mode uses the point’s normal and tries to reconstruct an upvector. It works best with
organic meshes that have unwelded vertices, like imported FBX meshes, since it has the same
rotations for vertices at the same positions. On plane geometric meshes, a gradual shift in rotation
is noticeable. For best results, it is recommended to use this method at the origin before any
transformations.

Replicate 3D Aligned layout

– Aligned TBN: This mode results in a more accurate and stable alignment based on the tangent,
binormal, and normal of the destination point. This works best for particles and geometric shapes.
On unwelded meshes, two copies of multiple unwelded points at the same position may lead to
different alignments because of their individual normals.

Replicate 3D Aligned TBN layout

Color
Affects the diffuse color or shader of each copy based on the input’s particle color.
– Use Object Color: Does not use the color of the destination particle.
– Combine Particle Color: Uses the shader of any input mesh and modifies the diffuse color to
match the color from the destination particle.
– Use Particle Color: Replaces the complete shader of any input mesh with a default shader. Its
diffuse color is taken from the destination particle.

Chapter 29 3D Nodes 707

 Replicate 3D Color options

Translation
These three sliders tell the node how much offset to apply to each copy. An X Offset of 1 would offset
each copy one unit, one unit along the X-axis from the last copy.

Rotation Order
These buttons can be used to set the order in which rotations are applied to the geometry. Setting the
rotation order to XYZ would apply the rotation on the X-axis first, followed by the Y-axis rotation, and
then the Z-axis rotation.

XYZ Rotation
These three rotation sliders tell the node how much rotation to apply to each copy.

XYZ Pivot
The pivot controls determine the position of the pivot point used when rotating each copy.

Lock XYZ
When the Lock XYZ checkbox is selected, any adjustment to the scale is applied to all three axes
simultaneously.
If this checkbox is disabled, the Scale slider is replaced with individual sliders for the X, Y,
and Z scales.

Scale
The Scale control sets how much scaling to apply to each copy.

Chapter 29 3D Nodes 708

Replicated 3D Jitter controls

Jitter Tab
The Jitter tab can be used to introduce randomness to various parameters.

Random Seed/Randomize
The Random Seed is used to generate the jitter applied to the replicated objects. Two Replicate nodes
with identical settings but different random seeds will produce two completely different results. Click
the Randomize button to assign a Random Seed value.

Time Offset
Use the Time Offset slider to offset any animations that are applied to the source geometry. Unlike
Time Offset on the Controls tab, Jitter Time Offset is random, based on the Random Seed setting.

Translation XYZ Jitter

Use these three controls to adjust the variation in the translation of the replicated objects.

Rotation XYZ Jitter

Use these three controls to adjust the variation in the rotation of the replicated objects.

Pivot XYZ Jitter

Use these three controls to adjust the variation in the rotational pivot center of the replicated objects.
This affects only the additional jitter rotation, not the rotation produced by the rotation settings in the
Controls tab.

Scale XYZ Jitter

Use this control to adjust the variation in the scale of the replicated objects. Uncheck the Lock XYZ
checkbox to adjust the scale variation independently on all three axes.

Chapter 29 3D Nodes 709

Common Controls
Settings Tab
The Settings tab is common to many 3D nodes. The description of these controls can be found in
“The Common Controls” section at the end of this chapter.

Ribbon 3D [3RI]

The Ribbon 3D node

Ribbon 3D Node Introduction

Ribbon 3D generates an array of subdivided line segments or a single line between two points. It is
quite useful for motion graphics, especially in connection with Replicate 3D to attach other geometry
to the lines, and with Displace3D for creating lightning bolt-like structures. The array of lines is, by
default, assigned with texture coordinates, so they can be used with a 2D texture. As usual, UVMap 3D
can be used to alter the texture coordinates. This node relies heavily on certain OpenGL features and
does not produce any visible result when the Renderer 3D node is set to use the software renderer.
Furthermore, the way lines are drawn is completely up to the graphics card capabilities, so the ribbon
appearance may vary based on your computer’s graphics card.

Inputs
There are two inputs on the Ribbon 3D node: one for the destination geometry that contains the
vertices, and one for the 3D geometry you want to replicate.
– 3D Scene: The orange input accepts a 3D scene or geometry.
– Material: The input accepts the 2D texture for the ribbon.

Neither connected input is required.

Basic Node Setup

In the example below, a Ribbon 3D node is used to generate lines. A gradient background is
connected to “colorize” the lines. Additional nodes are then used after the Ribbon 3D to bend and
distort the lines.

Ribbon 3D generates lines distorted by additional nodes

Chapter 29 3D Nodes 710

Inspector

Ribbon 3D controls

Controls Tab
The Controls tab determines the number of ribbon strands, their size, length, and spacing.

Number of Lines
The number of parallel lines drawn between the start point and end point.

Line Thickness
Line thickness is allowed in the user interface to take on a floating-point value, but some graphics
cards allow only integer values. Some cards may only allow lines equal to or thicker than one, or max
out at a certain value.

Subdivision Level
The number of vertices on each line between start point and end points. The higher the number, the
more precise and smoother 3D displacement appears.

Ribbon Width
Determines how far the lines are apart from each other.

Start
XYZ control to set the start point of the ribbon.

End
XYZ control to set the end point of the ribbon.

Ribbon Rotation
Allows rotation of the ribbon around the virtual axis defined by start point and end points.

Anti-Aliasing
Allows you to apply anti-aliasing to the rendered lines. Using anti-aliasing isn’t necessarily
recommended. When activated, there may be be gaps between the line segments. This is especially
noticeable with high values of line thickness. Again, the way lines are drawn is completely up to the
graphics card, which means that these artifacts can vary from card to card.

Chapter 29 3D Nodes 711

Common Controls
Controls, Materials, and Settings Tabs
The controls for Visibility, Lighting, Matte, Blend Mode, Normals/Tangents, and Object ID in the
Controls tab are common in many 3D nodes. The Materials tab and Settings tab in the Inspector are
also duplicated in other 3D nodes. These common controls are described in detail at the end of this
chapter in “The Common Controls” section.

Shape 3D [3SH]

The Shape 3D node

Shape 3D Node Introduction

The Shape 3D node is used to produce several basic primitive 3D shapes, including planes, cubes,
spheres, and cylinders.

Inputs
There are two optional inputs on the Shape 3D. The scene input can be used to combine additional
geometry with the Shape 3D, while the material input can be used to texture map the Shape
3D object.
– SceneInput: Although the Shape 3D creates its own 3D geometry, you can use the orange
scene input to combine an additional 3D scene or geometry.
– MaterialInput: The green input accepts either a 2D image or a 3D material. If a 2D image is
provided, it is used as a diffuse texture map for the basic material built into the node. If a 3D
material is connected, then the basic material is disabled.

Basic Node Setup

In the example below, four Shape 3D nodes are used to create the primitives of a 3D set. Two of the
Shape 3D nodes are connected creating a more complex primitive shape. Those shapes can then be
used with a Projector 3D to texture them with a realistic material.

Shape 3D nodes combined with Projector 3D to create a realistic 3D set

Chapter 29 3D Nodes 712

Inspector

Shape 3D controls

Controls Tab
The Controls tab allows you to select a shape and modify its geometry. Different controls appear
based on the specific shape that you choose to create.

Shape
This menu allows you to select the primitive geometry produced by the Shape 3D node. The
remaining controls in the Inspector change to match the selected shape.
– Lock Width/Height/Depth: [plane, cube] If this checkbox is selected, the width, height, and depth
controls are locked together as a single size slider. Otherwise, individual controls over the size of
the shape along each axis are provided.
– Size Width/Height/Depth: [plane, cube] Used to control the size of the shape.

Cube Mapping
When Cube is selected in the shape menu, the Cube uses cube mapping to apply the Shape node’s
texture (a 2D image connected to the material input on the node).

Radius
When a Sphere, Cylinder, Cone, or Torus is selected in the shape menu, this control sets the radius of
the selected shape.

Top Radius
When a cone is selected in the Shape menu, this control is used to define a radius for the top of a
cone, making it possible to create truncated cones.

Start/End Angle
When the Sphere, Cylinder, Cone, or Torus shape is selected in the Shape menu, this range control
determines how much of the shape is drawn. A start angle of 180° and end angle of 360° would only
draw half of the shape.

Start/End Latitude
When a Sphere or Torus is selected in the Shape menu, this range control is used to crop or slice the
object by defining a latitudinal subsection of the object.

Chapter 29 3D Nodes 713

Bottom/Top Cap
When Cylinder or Cone is selected in the Shape menu, the Bottom Cap and Top Cap checkboxes are
used to determine if the end caps of these shapes are created or if the shape is left open.

Section
When the Torus is selected in the Shape menu, Section controls the thickness of the tube making up
the torus.

Subdivision Level/Base/Height
The Subdivision controls are used to determine the tessellation of the mesh on all shapes. The higher
the subdivision, the more vertices each shape has.

Wireframe
Enabling this checkbox causes the mesh to render only the wireframe for the object.

Common Controls
Controls, Materials, Transform and Settings Tabs
The controls for Visibility, Lighting, Matte, Blend Mode, Normals/Tangents, and Object ID in the
Controls tab are common in many 3D nodes. The Materials tab, Transforms tab, and Settings tab in the
Inspector are also duplicated in other 3D nodes. These common controls are described in detail at the
end of this chapter in “The Common Controls” section.

Sphere Map vs. Connecting the Texture to a Sphere Directly

You can connect a LatLong (equirectangular) texture map directly to a sphere instead of piping it
through the Sphere Map node first. This results in a different rendering if you set the start/end angle
and latitude to less than 360°/180°. In the first case, the texture is squashed. When using the Sphere
Map node, the texture is cropped. Compare:

Spherical mapping differences

NOTE: If you pipe the texture directly into the sphere, it is also mirrored horizontally. You can
change this by using a Transform node first.

Chapter 29 3D Nodes 714

Soft Clip [3SC]

The Soft Clip node

Soft Clip Node Introduction

The Soft Clip node is used to fade out geometry and particles that get close to the camera. This helps
avoid the visible “popping off” that affects many particle systems and 3D flythroughs.
This node is very similar to the Fog 3D node, in that it is dependent on the geometry’s distance from
the camera.

Inputs
The Soft Clip includes only a single input for a 3D scene that includes a camera connected to it.
– SceneInput: The orange scene input is a required connection. It accepts a 3D scene input that
includes a Camera 3D node.

Basic Node Setup

The Soft Clip node is usually placed just before the Renderer 3D node to ensure that downstream
adjustments to lighting and textures do not affect the result. It can be placed in any part of the 3D
portion of the node tree if the soft clipping effect is only required for a portion of the scene.

Soft Clip placed between a Merge 3D and a Renderer 3D node

Chapter 29 3D Nodes 715

Inspector

Soft Clip controls

Controls Tab
The Controls tab determines how an object transitions between opaque and transparent as it moves
closer to the camera.

Enable
This checkbox can be used to enable or disable the node. This is not the same as the red switch in the
upper-left corner of the Inspector. The red switch disables the tool altogether and passes the image
on without any modification. The Enable checkbox is limited to the effect of the tool. Other parts, like
scripts in the Settings tab, still process as normal.

Smooth Transition
By default, an object coming closer and closer to the camera slowly fades out with a linear
progression. With the Smooth Transition checkbox enabled, the transition changes to a nonlinear
curve, arguably a more natural-looking transition.

Radial
By default, the soft clipping is done based on the perpendicular distance to a plane (parallel with the
near plane) passing through the eye point. When the Radial option is checked, the Radial distance to
the eye point is used instead of the Perpendicular distance. The problem with Perpendicular distance
soft clipping is that when you move the camera about, as objects on the left or right side of the frustum
move into the center, they become less clipped, although they remain the same distance from the eye.
Radial soft clip fixes this. Sometimes Radial soft clipping is not desirable.
For example, if you apply soft clip to an object that is close to the camera, like an image plane, the
center of the image plane could be unclipped while the edges could be fully clipped because they are
farther from the eye point.

Show In Display Views

Normally, the effect is only visible when the scene is viewed using a Camera node. When enabled, the
soft clip becomes visible in the scene from all points of view.

Transparent/Opaque Distance
Defines the range of the soft clip. The objects begin to fade in from an opacity of 0 at the Transparent
distance and are fully visible at the Opaque distance. All units are expressed as distance from the
camera along the Z-axis.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 29 3D Nodes 716

Spherical Camera [3SC]

The Spherical Camera node

Spherical Camera Node Introduction

The Spherical Camera allows the 3D Renderer node to output an image covering all viewing angles,
laid out in several different formats. This image may be used, for example, as a skybox texture or
reflection map or viewed in a VR headset. The Image Width setting in the 3D Renderer sets the size of
each square cube face, so the resulting image may be a multiple of this size horizontally and vertically.

Inputs
The Spherical camera node has two inputs.
– Image: This orange image input requires an image in a spherical layout, which can be any of
LatLong (2:1 equirectangular), Horizontal/Vertical Cross, or Horizontal/Vertical Strip.
– Stereo Input: The green input for a right stereo camera if you are working in stereo VR.

Neither input is required.

Basic Node Setup

In many ways, the Spherical Camera is set up identically to the regular Camera 3D node. The output of
the camera connects into a Merge 3D. Typically, the Merge 3D has an image from a LatLong or H
Cross/V Cross formatted image either directly or through a Panomap node. The image is wrapped
around a sphere, and the camera is placed inside the sphere.

Spherical Camera placed inside a sphere

Chapter 29 3D Nodes 717

Inspector

Spherical Camera controls

Controls Tab
Layout
– VCross and HCross: VCross and HCross are the six square faces of a cube laid out in a cross,
vertical or horizontal, with the forward view in the center of the cross, in a 3:4 or 4:3 image.
– VStrip and HStrip: VStrip and HStrip are the six square faces of a cube laid vertically or
horizontally in a line, ordered as Left, Right, Up, Down, Back, Front (+X, -X, +Y, -Y, +Z, -Z), in a 1:6 or
6:1 image.
– LatLong: LatLong is a single 2:1 image in equirectangular mapping.

Near/Far Clip
The clipping plane is used to limit what geometry in a scene is rendered based on the object’s
distance from the camera’s focal point. This is useful for ensuring that objects that are extremely close
to the camera are not rendered and for optimizing a render to exclude objects that are too far away to
be useful in the final rendering.
The default perspective camera ignores this setting unless the Adaptively Adjust Near/Far Clip
checkbox control below is disabled.
The values are expressed in units, so a far clipping plane of 20 means that any objects more than 20
units from the camera are invisible to the camera. A near clipping plane of 0.1 means that any objects
closer than 0.1 units are also invisible.

Adaptively Adjust Near/Far Clip

When selected, the Renderer automatically adjusts the camera’s near/far clipping plane to match the
extents of the scene. This setting overrides the values of the Near and Far clip range control described
above. This option is not available for orthographic cameras.

Chapter 29 3D Nodes 718

Viewing Volume Size
The Viewing Volume Size control appears only when the Projection Type is set to Orthographic. It
determines the size of the box that makes up the camera’s field of view. The Z distance of an
orthographic camera from the objects it sees does not affect the scale of those objects, only the
viewing size does.

NOTE: A smaller range between the near and far clipping plane allows greater accuracy in all
depth calculations. If a scene begins to render strange artifacts on distant objects, try
increasing the distance for the near clip plane. Use the vertical aperture size to get the
vertical angle of view and the horizontal aperture size to get the horizontal angle of view.

Plane of Focus (for Depth of Field)

This value is used by the OpenGL renderer to calculate depth of field. It defines the distance to a
virtual target in front of the camera.

Stereo Method
This control allows you to adjust your stereoscopic method to your preferred working model.

Toe In
Both cameras point at a single focal point. Though the result is stereoscopic, the vertical parallax
introduced by this method can cause discomfort by the audience.

Off Axis
Often regarded as the correct way to create stereo pairs, this is the default method in Fusion. Off Axis
introduces no vertical parallax, thus creating less stressful stereo images.

Parallel
The cameras are shifted parallel to each other. Since this is a purely parallel shift, there is no
Convergence Distance control. Parallel introduces no vertical parallax, thus creating less stressful
stereo images.

Eye Separation
Defines the distance between both stereo cameras. If the Eye Separation is set to a value larger
than 0, controls for each camera are shown in the viewer when this node is selected. There is no
Convergence Distance control in Parallel mode.

Convergence Distance
This control sets the stereoscopic convergence distance, defined as a point located along the Z-axis
of the camera that determines where both left and right eye cameras converge.

Control Visibility
Allows you to selectively activate the onscreen controls that are displayed along with the camera.
– Frustum: Displays the actual viewing cone of the camera.
– View Vector: Displays a white line inside the viewing cone, which can be used to determine the
shift when in Parallel mode.
– Near Clip: The Near clipping plane. This plane can be subdivided for better visibility.
– Far Clip: The Far clipping plane. This plane can be subdivided for better visibility.
– Plane of Focus: The camera focal point according to the Plane of Focus slider explained above.
This plane can be subdivided for better visibility.
– Convergence Distance: The point of convergence when using Stereo mode. This plane can be
subdivided for better visibility.

Chapter 29 3D Nodes 719

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Text 3D [3TXT]

The Text 3D node

Text 3D Node Introduction

The Text3D node is a 3D version of the 2D Text+ node. The controls for this node are mostly identical
to the controls for the 2D version in almost all respects, except Text 3D supports only one
shading element.
The Text 3D node was based on a tool that predates the Fusion 3D environment. So, some of the
controls found in the basic primitive shapes and geometry loaders, such as many of the material,
lighting, and matte options, are not found in this node’s controls. The Text 3D node has a built-in
material, but unlike the other 3D nodes it does not have a material input. The Shading tab contains
controls to adjust the diffuse and specular components. To replace this default material with a more
advanced material, follow the Text Plus node with a Replace Material 3D node. The Override 3D node
can be used to control the lighting, visibility, and matte options for this node.
When network rendering a comp that contains Text 3D nodes, each render machine is required to
have the necessary fonts installed or the network rendering fails. Fusion does not share or copy fonts
to render slaves.

Inputs
– SceneInput: The orange scene input accepts a 3D scene that can be combined with the 3D
text created in the node.
– ColorImage: The green color image input accepts a 2D image and wraps it around the text as
a texture. This input is visible only when Image is selected in the Material Type menu located in
the Shading tab.
– BevelTexture: The magenta bevel texture input accepts a 2D image and wraps it around the
bevel as a texture. This input is visible only when one Material is disabled in the Shader tab and
Image is selected in the Bevel Type menu.

Basic Node Setup

The Text 3D node generates text, so most often this node starts a branch of your node tree. However,
to apply more realistic materials, a replace Material node is often added after the Text 3D, prior to
connecting into a Merge 3D node.

Chapter 29 3D Nodes 720

Text 3D with Replace Material node added to apply alternative material

Inspector

Text 3D controls

Text Tab
The Text 3D text tab in the Inspector is divided into three sections: Text, Extrusion, and Advanced
Controls. The Text section includes parameters that are familiar to anyone who has used a word
processor. It includes commonly used text formatting options. The Extrusion section includes controls
to extrude the text and create beveled edges for the text. The Advanced controls are used for
kerning options.

Styled Text
The Edit box in this tab is where the text to be created is entered. Any common character can be
typed into this box. The common OS clipboard shortcuts (Command-C or Ctrl-C to copy, Command-X
or Ctrl-X to cut, Command-V or Ctrl-V to paste) also work; however, right-clicking on the Edit box
displays a custom contextual menu with several modifiers you can add for more animation and
formatting options.

Chapter 29 3D Nodes 721

Font
Two Font menus are used to select the font family and typeface such as Regular, Bold, and Italic.

Color
This control sets the basic tint color of the text. This is the same Color control displayed in the Material
type section of the Shader tab.

Size
This control is used to increase or decrease the size of the text. This is not like selecting a point size in
a word processor. The size is relative to the width of the image.

Tracking
The Tracking parameter adjusts the uniform spacing between each character of text.

Line Spacing
Line Spacing adjusts the distance between each line of text. This is sometimes called leading in
word-processing applications.

V Anchor
The Vertical Anchor controls consist of three buttons and a slider. The three buttons are used to align
the text vertically to the top, middle, or bottom baseline of the text. The slider can be used to
customize the alignment. Setting the Vertical Anchor affects how the text is rotated but also the
location for line spacing adjustments. This control is most often used when the Layout type is set to
Frame in the Layout tab.

V Justify
The Vertical Justify slider allows you to customize the vertical alignment of the text from the V Anchor
setting to full justification so it is aligned evenly along the top and bottom edges. This control is most
often used when the Layout type is set to Frame in the Layout tab.

H Anchor
The Horizontal Anchor controls consist of three buttons and a slider. The three buttons justify the text
alignment to the left edge, middle, or right edge of the text. The slider can be used to customize the
justification. Setting the Horizontal Anchor affects how the text is rotated but also the location for
tracking (leading) spacing adjustments. This control is most often used when the Layout type is set to
Frame in the Layout tab.

H Justify
The Horizontal Justify slider allows you to customize the justification of the text from the H Anchor
setting to full justification so it is aligned evenly along the left and right edges. This control is most
often used when the Layout type is set to Frame in the Layout tab.

Direction
This menu provides options for determining the direction in which the text is to be written.

Line Direction
These menu options are used to determine the text flow from top to bottom, bottom to top, left to right,
or right to left.

Write On
This range control is used to quickly apply simple Write On and Write Off animation to the text. To
create a Write On effect, animate the End portion of the control from 1 to 0 over the length of time
required. To create a Write Off effect, animate the Start portion of the range control from 0 to 1.

Chapter 29 3D Nodes 722

Extrusion Depth
An extrusion of 0 produces completely 2D text. Any value greater than 0 extrudes the text to generate
text with depth.

Text 3D extrusion and bevel controls

Bevel Depth
Increase the value of the Bevel Depth slider to bevel the text. The text must have extrusion before this
control has any effect.

Bevel Width
Use the Bevel Width control to increase the width of the bevel.

Smoothing Angle
Use this control to adjust the smoothing angle applied to the edges of the bevel.

Front/Back Bevel
Use these checkboxes to enable beveling for the front and back faces of the text separately

Custom Extrusion
In Custom mode, the Smoothing Angle controls the smoothing of normals around the edges of a text
character. The spline itself controls the smoothing along the extrusion profile. If a spline segment is
smoothed, for example by using the shortcut Shift-S, the normals are smoothed as well. If the control
point is linear, the shading edge is sharp. The first and last control points on the spline define the
extent of the text.

Custom Extrusion Subdivisions

Controls the number of subdivisions within the smoothed portions of the extrusion profile.

TIP: Splines can also be edited from within the Spline Editor panel. It provides a larger
working space for working with any spline including the Custom Extrusion.
Extrusion profile spline control: Do not try to go to zero size at the Front/Back face.
This results in Z-fighting resulting from self-intersecting faces. To avoid this problem, make
sure the first and last point have their profiles set to 0.

Force Monospaced
This slider control can be used to override the kerning (spacing between characters) that is defined in
the font. Setting this slider to zero (the default value) causes Fusion to rely entirely on the kerning
defined with each character. A value of one causes the spacing between characters to be completely
even, or monospaced.

Chapter 29 3D Nodes 723

Text 3D Advanced Controls can be used to manually kern letters.

Use Font Defined Kerning

This enables kerning as specified in the True Type font and is on by default.

Manual Font Kerning

Manual Font Kerning is only performed using the Text+ node. To perform manual kerning on Text3D,
create the text using the Text+ node and kern it in that tool. Then, right-click over the tool’s name in the
Inspector and choose Copy. Once the settings are copied, select the Text 3D node and choose Paste
Settings from the Inspector’s contextual menu. Once the manual kerning is pasted in the Text 3D
node, the two buttons in the Inspector clear either the selected character’s kerning or all the kerning
adjustment in the current text.

Layout Tab
The Layout Tab is used to position the text in one of four different layout types.

Text 3D Layout tab for changing the layout of the text block

Layout Type
This menu selects the layout type for the text.
– Point: Point layout is the simplest of the layout modes. Text is arranged around an adjustable
center point.
– Frame: Frame layout allows you to define a rectangular frame used to align the text. The alignment
controls are used to justify the text vertically and horizontally within the boundaries of the frame.
– Circle: Circle layout places the text around the curve of a circle or oval. Control is offered over
the diameter and width of the circular shape. When the layout is set to this mode, the Alignment
controls determine whether the text is positioned along the inside or outside of the circle’s edge,
and how multiple lines of text are justified.
– Path: Path layout allows you to shape your text along the edges of a path. The path can be
used simply to add style to the text, or it can be animated using the Position on Path control that
appears when this mode is selected.

Chapter 29 3D Nodes 724

Center X, Y, and Z
These controls are used to position the center of the layout. For instance, moving the center X, Y,
and Z parameters when the layout is set to Frame moves the position of the frame the text is within.

Size
This slider is used to control the scale of the layout element. For instance, increasing size when the
layout is set to Frame increases the frame size the text is within.

Width and Height

The Width and Height controls are visible when the Layout mode is set to Circle or Frame. The Width
and Height controls are visible only when the Layout mode is set to Frame. They are used to adjust the
dimensions and aspect of the Layout element.

Rotation Order
These buttons allow you to select the order in which 3D rotations are applied to the text.

X, Y, and Z
These angle controls can be used to adjust the angle of the Layout element along any axis.

Fit Characters
This menu control is visible only when the Layout type is set to Circle. This menu is used to select how
the characters are spaced to fit along the circumference.

Position on Path
The Position on Path control is used to control the position of the text along the path. Values less than
zero or greater than one cause the text to move beyond, continuing in the same direction set by the
last two points on the path.

Right-Click Here for Shape Animation

This label appears only when the Layout type is set to Path. It is used to provide access to a contextual
menu that provides options for connecting the path to other paths in the node tree, and animating the
spline points on the path over time.

Transform Tab
There are actually two Transform tabs in the Text 3D Inspector. The first Transform tab is unique to the
Text 3D tool, while the second is the common Transform tab found on many 3D nodes. The Text
3D-specific Transform tab is described below since it contains some unique controls for this node.

Text 3D Transform tab

Transform
This menu determines the portion of the text affected by the transformations applied in this tab.
Transformations can be applied to line, word, and character levels simultaneously. This menu is only
used to keep the number of visible controls to a reasonable number.

Chapter 29 3D Nodes 725

 – Characters: Each character of text is transformed along its own center axis.
– Words: Each word is transformed separately on the word’s center axis.
– Lines: Each line of the text is transformed separately on that line’s center axis.

Spacing
The Spacing slider is used to adjust the amount of space between each line, word, or character.
Values less than one usually cause the characters to begin overlapping.

Pivot X, Y, and Z
This provides control over the exact position of the axis. By default, the axis is positioned at the
calculated center of the line, word, or character. The pivot control works as an offset, such that a value
of 0.1, 0.1 in this control would cause the axis to be shifted downward and to the right for each of the
text elements. Positive values in the Z-axis slider move the axis further along the axis (away from the
viewer). Negative values bring the axis of rotation closer.

Rotation Order
These buttons are used to determine the order in which transforms are applied. X, Y, and Z would
mean that the rotation is applied to X, then Y, and then Z.

X, Y, and Z
These controls can be used to adjust the angle of the text elements in any of the three dimensions.

Shear X and Y
Adjust these sliders to modify the slanting of the text elements along the X- and Y-axis.

Size X and Y
Adjust these sliders to modify the size of the text elements along the X- and Y-axis.

Shading
The Shading tab for the Text 3D node controls the overall appearance of the text and how lights affect
its surface.

Text 3D Shading tab

Chapter 29 3D Nodes 726

Opacity
Reducing the material’s opacity decreases the color and Alpha values of the specular and diffuse
colors equally, making the material transparent and allowing hidden objects to be seen through
the material.

Use One Material

Deselecting this option reveals a second set of Material controls for the beveled edge of the text.

Type
To use a solid color texture, select the Solid mode. Selecting the Image mode reveals a new external
input on the node that can be connected to another 2D image.

Specular Color
Specular Color determines the color of light that reflects from a shiny surface. The more specular a
material is, the glossier it appears. Surfaces like plastics and glass tend to have white specular
highlights, whereas metallic surfaces like gold have specular highlights that tend to inherit their color
from the material color. The basic shader material does not provide an input for textures to control the
specularity of the object. Use nodes from the 3D Material category when more precise control is
required over the specular appearance.

Specular Intensity
Specular Intensity controls the strength of the specular highlight. If the specular intensity texture port
has a valid input, then this value is multiplied by the Alpha value of the input.

Specular Exponent
Specular Exponent controls the falloff of the specular highlight. The greater the value, the sharper the
falloff, and the smoother and glossier the material appears. The basic shader material does not
provide an input for textures to control the specular exponent of the object. Use nodes from the 3D
Material category when more precise control is required over the specular exponent.

Image Source
This control determines the source of the texture applied to the material. If the option is set to Tool,
then an input appears on the node that can be used to apply the output of a 2D node as the texture.
Selecting Clip opens a file browser that can be used to select an image or image sequence from disk.
The Brush option provides a list of clips found in the Fusion\brushes folder.

Bevel Material
This option appears only when the Use One Material checkbox control is selected. The controls under
this option are an exact copy of the Material controls above but are applied only to the beveled edge
of the text.

Position, Rotation, Shear, and Size

These transform controls act similarly to the transform controls in the Transform tab when a single
shading element is enabled from the top of the Shading tab. However, when two or more shading
elements are enabled, these transform controls are applied to the currently selected shading element.
This allows you to independently control the position, rotation, shearing, and size of borders, fill colors,
and shadows.

Uncapped 3D Text

To hide the front face of extruded text, uncheck Use One Material on the Shading tab and
reduce the first material’s color to black, including its Alpha value.

Chapter 29 3D Nodes 727

Common Controls
Transform and Settings Tabs
The Transform and Settings tabs in the Inspector are duplicated in other 3D nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Text 3D Modifiers
Right-clicking within the Styled Text box displays a menu with the following text modifiers. Only one
modifier can be applied to a Text 3D Styled Text box. Below is a brief list of the text specific modifiers,
but for more information see Chapter 122, “Modifiers” in the DaVinci Resolve Reference Manual or
Chapter 61 in the Fusion Reference Manual.

Animate
Use the Animate command to set to a keyframe on the entered text and animate the content over time.

Character Level Styling

The Text 3D node doesn’t support Character Level Styling directly. However, you can create a Text+
node first and modify its text field with a Character Level Styling modifier. Then either connect the Text
3D’s text field to the modifier that is now available or copy the Text+ node and paste its settings to the
Text 3D node (right-click > Paste Settings)

Comp Name
Comp Name puts the name of the composition in the Styled Text box and is generally used as a quick
way to create slates.

Follower
Follower is a text modifier that can be used to ripple animation applied to the text across each
character in the text. See “Text Modifiers” at the end of this chapter.

Publish
Publish the text for connection to other text nodes.

Text Scramble
A text modifier ID is used to randomize the characters in the text. See “Text Modifiers” at the end of
this chapter.

Text Timer
A text modifier is used to count down from a specified time or to output the current date and time. See
“Text Modifiers” at the end of this chapter.

Time Code
A text modifier is used to output Time Code for the current frame. See “Text Modifiers” at the end of
this chapter.

Connect To
Use this option to connect the text generated by this Text node to the published output of
another node.

Chapter 29 3D Nodes 728

Transform 3D [3XF]

The Transform 3D node

Transform 3D Node Introduction

The Transform 3D node can be used to translate, rotate, or scale all the elements within a scene
without requiring a Merge 3D node. This can be useful for hierarchical transformations or for offsetting
objects that are merged into a scene multiple times. Its controls are identical to those found in other
3D nodes’ Transformation tabs.

Inputs
The Transform node has a single required input for a 3D scene or 3D object.
– Scene Input: The orange scene input is connected to a 3D scene or 3D object to apply a
second set of transformation controls.

Basic Node Setup

The Transform 3D node adds 3D position, rotation, and pivot control onto any existing transforms in
the 3D node prior to it. You can combine multiple Transform 3D nodes together to build parenting or
hierarchical movement.

Transform 3D nodes strung together to create a parenting hierarchy

Chapter 29 3D Nodes 729

Inspector

Transform 3D controls

Controls Tab
The Controls tab is the primary tab for the Transform 3D node. It includes controls to translate, rotate,
or scale all elements within a scene without requiring a Merge 3D node.

Translation
– X, Y, Z Offset: Controls are used to position the 3D element in 3D space.

Rotation
– Rotation Order: Use these buttons to select the order used to apply the rotation along each axis
of the object. For example, XYZ would apply the rotation to the X-axis first, followed by the Y-axis,
and then the Z-axis.
– X, Y, Z Rotation: Use these controls to rotate the object around its pivot point. If the Use Target
checkbox is selected, then the rotation is relative to the position of the target; otherwise, the
global axis is used.

Pivot Controls
– X, Y, Z Pivot: A pivot point is the point around which an object rotates. Normally, an object rotates
around its own center, which is considered to be a pivot of 0,0,0. These controls can be used to
offset the pivot from the center.

Scale
– X, Y, Z Scale: If the Lock X/Y/Z checkbox is checked, a single scale slider is shown. This adjusts
the overall size of the object. If the Lock checkbox is unchecked, individual X, Y, and Z sliders are
displayed to allow scaling in any dimension.

NOTE: If the Lock checkbox is checked, scaling of individual dimensions is not possible, even
when dragging specific axes of the Transformation widget in Scale mode.

Chapter 29 3D Nodes 730

Use Target
Selecting the Use Target checkbox enables a set of controls for positioning an XYZ target. When Use
Target is enabled, the object always rotates to face the target. The rotation of the object becomes
relative to the target.

Import Transform
Opens a file browser where you can select a scene file saved or exported by your 3D application.
It supports the following file types:

LightWave Scene .lws

Max Scene .ase

Maya Ascii Scene .ma

dotXSI .xsi

The Import Transform button imports only transformation data. For 3D geometry, lights and cameras,
consider using the File > FBX Import option from the menus.

Import Transform Browser

Onscreen Transformation Controls

Onscreen Transformation controls provide an alternative way of using the controls in the Inspector.
The viewer includes modes for transformation, rotation, and scaling. To change the mode of the
onscreen controls, select one of the three buttons in the toolbar along the side of the viewer.
The modes can also be toggled using the keyboard shortcut Q for translation, W for rotation, and E
for scaling. In all three modes, an individual axis of the control may be dragged to affect just that axis,
or the center of the control may be dragged to affect all three axes.
The scale sliders for most 3D nodes default to locked, which causes uniform scaling of all three axes.
Unlock the Lock X/Y/Z Scale checkbox to scale an object on a single axis only.

Transform 3D onscreen
transformation controls

Chapter 29 3D Nodes 731

Triangulate 3D [3TRI]

The Triangulate 3D node

Triangulate 3D Node Introduction

The Triangulate 3D node is a unique node in that it has no controls. This node turns polygon shapes
into triangles. For instance, a quad that is four points becomes two triangles. It is used to convert
complex polygon shapes into a mesh for easier processing.

Inputs
The Triangulate 3D node has a single required input for a 3D scene or 3D object.
– Scene Input: The orange scene input is connected to the 3D scene or 3D object you
want to triangulate.

Basic Node Setup

The Triangulate 3D node is placed after the geometry you want to triangulate.

Triangulate 3D nodes connected after an imported model

Inspector

Triangulate 3D controls

Controls Tab
There are no controls for this node.

Chapter 29 3D Nodes 732

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

UV Map 3D [3UV]

The UVMap 3D node

UV Map 3D Node Introduction

The UV Map 3D node replaces the UV texture coordinates on the geometry in the scene. These
coordinates tell Fusion how to apply a texture to an object. While it is possible to adjust the global
properties of the selected mapping mode, it is not possible to manipulate the UV coordinates of
individual vertices directly from within Fusion. The onscreen controls drawn in the viewers are for
reference only and cannot be manipulated.

Camera Projections with UV Map 3D

The Camera Mapping mode makes it possible to project texture coordinates onto geometry through a
camera. Once you select Camera from the Mapping mode menu, then connect the Camera 3D node
that you want to use to create the UV coordinates.
Note that this does not directly project an image through the camera. The image to be projected
should be connected to the diffuse texture input of whatever material is assigned to the objects. When
the texture is applied, it uses the UV coordinates created by the camera. Because this is a texture
projection and not light, the Alpha channel of the texture correctly sets the opacity of the geometry.
See the Camera 3D and Projector 3D nodes for alternate approaches to projection.
The projection can optionally be locked to the vertices as it appears on a selected frame.
This fails if the number of vertices in the mesh changes over time, as Fusion must be able to match up
the mesh at the reference time and the current time. To be more specific, vertices may not be created
or destroyed or reordered. So, projection locking does not work for many particle systems, or for
primitives with animated subdivisions, or with duplicate nodes using non-zero time offsets.

NOTE: The UV Map 3D node does not put a texture or material on the mesh; it only modifies
the texture coordinates that the materials use. This may be confusing because the material
usually sits upstream, as seen in the Basic Node Setup example below.

Chapter 29 3D Nodes 733

Inputs
The UV Map 3D node has two inputs: one for a 3D scene or 3D object and another optional input for a
Camera 3D node.
– Scene Input: The orange scene input is connected to the 3D scene or 3D object you want to
triangulate.
– CameraInput: This input expects the output of the Camera 3D node. It is only visible when the
Camera Map mode menu is set to Camera.

Basic Node Setup

The UV Map 3D node is placed after all the geometry and set to Camera Map. Connecting a camera to
the UV map allows you to line up the texture based on a centered camera position and 3D geometry.

UV Map 3D is placed after the Merge 3D, with a camera connected to line up the texture

Inspector

UV Map 3D controls

Controls Tab
The UV Map 3D Controls tab allows you to select Planar, Cylindrical, Spherical, XYZ, and Cubic
mapping modes, which can be applied to basic Fusion primitives as well as imported geometry. The
position, rotation, and scale of the texture coordinates can be adjusted to allow fine control over the

Chapter 29 3D Nodes 734

texture’s appearance. An option is also provided to lock the UV produced by this node to animated
geometry according to a reference frame. This can be used to ensure that textures applied to
animated geometry do not slide.

Map Mode
The Map mode menu is used to define how the texture coordinates are created. You can think of this
menu as a way to select the virtual geometry that projects the UV space on the object.
– Planar: Creates the UV coordinates using a plane.
– Cylindrical: Creates the UV coordinates using a cylindrical-shaped object.
– Spherical: The UVs are created using a sphere.
– XYZ to UVW: The position coordinates of the vertices are converted to UVW coordinates directly.
This is used for working with procedural textures.
– CubeMap: The UVs are created using a cube.
– Camera: Enables the Camera input on the node. After connecting a camera to the node, the
texture coordinates are created based on camera projection.

Orientation X/Y/Z
Defines the reference axis for aligning the Map mode.

Fit
Clicking this button fits the Map mode to the bounding box of the input scene.

Center
Clicking this button moves the center of the Map mode to the bounding box center of the input scene.

Lock UVs on Animated Objects

If the object is animated, the UVs can be locked to it by enabling this option. The option also reveals
the Ref Time slider, where it is possible to choose a reference frame for the UV mapping. Using this
feature, it is not required to animate the UV map parameters. It is enough to set up the UV map at the
reference time.

Size X/Y/Z
Defines the size of the projection object.

Center X/Y/Z
Defines the position of the projection object.

Rotation/Rotation Order
Use these buttons to select which order is used to apply the rotation along each axis of the object. For
example, XYZ would apply the rotation to the X-axis first, followed by the Y-axis, and then the Z-axis.

Rotation X/Y/Z
Sets the orientation of the projection object for each axis, independent from the rotation order.

Tile U/V/W
Defines how often a texture fits into the projected UV space on the applicable axis. Note that the UVW
coordinates are transformed, not a texture. This works best when used in conjunction with the Create
Texture node.

Flip U/V/W
Mirrors the texture coordinates around the applicable axis.

Chapter 29 3D Nodes 735

Flip Faces (Cube Map Mode Only)
Mirrors the texture coordinates on the individual faces of the cube.

NOTE: To utilize the full capabilities of the UV Map 3D node, it helps to have a basic
understanding of how 2D images are mapped onto 3D geometry. When a 2D image is
applied to a 3D surface, it is converted into a texture map that uses UV coordinates to
determine how the image translates to the object. Each vertex on a mesh has a (U, V) texture
coordinate pair that describes the appearance the object takes when it is unwrapped and
flattened. Different mapping modes use different methods for working out how the vertices
transform into a flat 2D texture. When using the UV Map 3D node to modify the texture
coordinates on a mesh, it’s best to do so using the default coordinate system of the mesh or
primitive. So the typical workflow would look like Shape 3D > UV Map 3D > Transform 3D. The
Transformation tab on the Shape node would be left to its default values, and the Transform
3D node following the UV Map 3D does any adjustments needed to place the node in the
scene. Modifying/animating the transform of the Shape node causes the texture to slide
across the shape, which is generally undesirable. The UV Map 3D node modifies texture
coordinates per vertex and not per pixel. If the geometry the UV map is applied to is poorly
tessellated, then undesirable artifacts may appear.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Weld 3D [3WE]

The Weld 3D node

Weld 3D Node Introduction

Sometimes 3D geometry has vertices that should have been joined when the geometry was created,
but for one reason or another they are not joined. This can cause artifacts, especially when the two
vertices have different normals.
For example, you may find:
– The different normals produce hard shading/lighting edges where none were intended.
– If you try to Displace 3D the vertices along their normals, they crack.
– Missing pixels or doubled-up pixels in the rendered image.
– Particles pass through the tiny invisible cracks.

Instead of round tripping back to your 3D modeling application to fix the “duplicated” vertices, the
Weld 3D node allows you to do this in Fusion. Weld 3D welds together vertices with the same or
nearly the same positions. This can be used to fix cracking issues when vertices are displaced by

Chapter 29 3D Nodes 736

welding the geometry before the Displace. There are no user controls to pick vertices. Currently, this
node welds together just position vertices; it does not weld normals, texcoords, or any other vertex
stream. So, although the positions of two vertices have been made the same, their normals still have
their old values. This can lead to hard edges in certain situations.

Inputs
The Weld 3D node has a single input for a 3D scene or 3D object you want to repair.
– Scene Input: The orange scene input is connected to the 3D scene or 3D object
you want to fix.

Basic Node Setup

The Weld 3D node is placed after the geometry that has duplicate vertices problems. Sometimes
problems are exposed when displacing the geometry. In that case, placing the weld after the
geometry but before the Displace 3D can repair the issues.

Weld 3D is placed after the 3D geometry that needs repair

Inspector

Weld 3D controls

Controls Tab
The Controls tab for the Weld 3D node includes a simple Weld Mode menu. You can choose between
welding vertices or fracturing them.

Fracture
Fracturing is the opposite of welding, so all vertices are unwelded. This means that all polygon
adjacency information is lost. For example, an Image Plane 3D normally consists of connected quads
that share vertices. Fracturing the image plane causes it to become a bunch of unconnected quads.

Tolerance
In auto mode, the Tolerance value is automatically detected. This should work in most cases.
It can also be adjusted manually if needed.

Chapter 29 3D Nodes 737

 Usage

Use Weld 3D when issues occur with the geometry. Don’t use it everywhere just because it’s
there, as it influences render time.
Weld 3D is intended to be used as a mesh robustness tool and not as a mesh editing tool to
merge vertices. If you can see the gap between the vertices you want to weld in the 3D view,
you are probably misusing Weld 3D. Unexpected things may happen when you do this; do so
at your own peril.

Limitations

Setting the tolerance too large can cause edges/faces to collapse to points.
If your model has detail distributed over several orders of scale, picking a tolerance value can
be hard or impossible.
For example, suppose you have a model of the International Space Station and there are lots
of big polygons and lots of really tiny polygons. If you set the tolerance too large, small
polygons that shouldn’t merge do; if you set the tolerance too small, some large polygons
won’t be merged.
Vertices that are far from the origin can fail to merge correctly. This is because bignumber +
epsilon can exactly equal bignumber in float math. This is one reason it may be best to merge
in local coordinates and not in world coordinates.
Sometimes Weld 3D-ing a mesh can make things worse. Take Fusion’s cone, for instance. The
top vertex of the cone is currently duplicated for each adjoining face, and they all have
different normals. If you weld the cone, the top vertices merge and only have one normal,
making the lighting look weird.
Weld 3D is not multithreaded.

Warning

Do not misuse Weld 3D to simplify (reduce the polygon count of) meshes. It is designed to
efficiently weld vertices that differ by only very small values, like a 0.001 distance.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 29 3D Nodes 738

Modifier
Coordinate Transform 3D
Because of the hierarchical nature of the Fusion 3D node tree, the original position of an object in the
3D scene often fails to indicate the current position of the object. For example, an image plane might
initially have a position at 1, 2, 1, but then be scaled, offset, and rotated by other nodes further
downstream in the 3D scene, ending up with an absolute location of 10, 20, 5.
This can complicate connecting an object further downstream in the composition directly to the
position of an upstream object. The Coordinate Transform modifier can be added to any set of XYZ
coordinate controls and calculate the current position of a given object at any point in the scene
hierarchy.
To add a Coordinate Transform modifier, simply right-click a number field on any node and select
Modify With/CoordTransform Position from the Controls’ contextual menu.

Inspector

Weld 3D modifier tools

Target Object
This control should be connected to the 3D node that produces the original coordinates to be
transformed. To connect a node, drag and drop a node from the node tree into the Text Edit control, or
right-click the control and select the node from the contextual menu. It is also possible to type the
node’s name directly into the control.

Sub ID
The Sub ID slider can be used to target an individual sub-element of certain types of geometry, such
as an individual character produced by a Text 3D node or a specific copy created by a
Duplicate 3D node.

Scene Input
This control should be connected to the 3D node that outputs the scene containing the object at the
new location. To connect a node, drag and drop a node from the node tree into the Text Edit control,
or right-click the control and select an object from the Connect To submenu.

Chapter 29 3D Nodes 739

The Common Controls
Nodes that handle 3D geometry share several identical controls in the Inspector. This section
describes controls that are common among 3D nodes.

Common Controls Tab

Common Controls 3D tab

These controls are often displayed in the lower half of the Controls tab. They appear in nodes that
create or contain 3D geometry.

Visibility
– Visible: If this option is enabled, the object is visible in the viewers and in final renders. When
disabled, the object is not visible in the viewers nor is it rendered into the output image by the
Renderer 3D node. Also, a non-visible object does not cast shadows.
– Unseen by Cameras: When the Unseen by Cameras checkbox is enabled, the object is visible
in the viewers (unless the Visible checkbox is disabled), except when viewed through a camera.
Also, the object is not rendered into the output image by the Renderer 3D node. However,
shadows cast by an unseen object are still visible when rendered by the software renderer in the
Renderer 3D node, though not by the OpenGL renderer.
– Cull Front Face/Back Face: Use these options to eliminate rendering and display of certain
polygons in the geometry. If Cull Back Face is selected, polygons facing away from the camera are
not rendered and do not cast shadows. If Cull Front Face is selected, polygons facing toward the
camera are not rendered and do not cast shadows. Enabling both options has the same effect as
disabling the Visible checkbox.
– Suppress Aux Channels for Transparent Pixels: In previous versions of Fusion, transparent pixels
were excluded by the software and Open GL render options in the Renderer 3D node. To be
more specific, the software renderer excluded pixels with R,G,B,A set to 0, and the GL renderer
excluded pixels with A set to 0. This is now optional. The reason you might want to do this is to
get aux channels (e.g., Normals, Z, UVs) for the transparent areas. For example, suppose you want
to replace the texture on a 3D element that is transparent in certain areas with a texture that is

Chapter 29 3D Nodes 740

 transparent in different areas. It would then be useful to have transparent areas set aux channels
(particularly UVs). As another example, suppose you are adding depth of field. You probably do
not want the Z-channel to be set on transparent areas, as this gives you a false depth. Also, keep
in mind that the exclusion is based on the final pixel color including lighting, if it is on. So, if you
have a specular highlight on a clear glass material, this checkbox does not affect it.

Lighting
– Affected by Lights: Disabling this checkbox causes lights in the scene to not affect the object.
The object does not receive nor cast shadows, and it is shown at the full brightness of its color,
texture, or material.
– Shadow Caster: Disabling this checkbox causes the object not to cast shadows on other objects
in the scene.
– Shadow Receiver: Disabling this checkbox causes the object not to receive shadows cast by
other objects in the scene.

Matte
Enabling the Is Matte option applies a special texture, causing the object to not only become invisible
to the camera, but also making everything that appears directly behind the camera invisible as well.
This option overrides all textures. For more information on Fog 3D and Soft Clipping, see Chapter 86,
“3D Compositing Basics” in the DaVinci Resolve Reference Manual or Chapter 25 in the Fusion
Reference Manual.
– Is Matte: When activated, objects whose pixels fall behind the matte object’s pixels in Z do not get
rendered. Two additional options are displayed when the Is Matte checkbox is activated.
– Opaque Alpha: When the Is Matte checkbox is enabled, the Opaque Alpha checkbox sets the
Alpha value of the matte object to 1.
– Infinite Z: This option sets the value in the Z-channel to infinite. This checkbox is visible only when
the Is Matte option is enabled.

Blend Mode
A Blend mode specifies which method is used by the renderer when combining this object with the
rest of the scene. The blend modes are essentially identical to those listed in the section for the 2D
Merge node. For a detailed explanation of each mode, see the section for that node.
The blending modes were originally designed for use with 2D images. Using them in a lit 3D
environment can produce undesirable results. For best results, use the Apply modes in unlit 3D
scenes using the software option in the Renderer 3D node.
– OpenGL Blend Mode: Use this menu to select the blending mode that is used when the geometry
is processed by the OpenGL renderer in the Renderer 3D node. This is also the mode used when
viewing the object in the viewers. Currently the OpenGL renderer supports a limited number of
blending modes.
– Software Blend Mode: Use this menu to select the blending mode that is used when the
geometry is processed by the software renderer. Currently, the software renderer supports all the
modes described in the Merge node documentation, except for the Dissolve mode.

Normal/Tangents
Normals are imaginary lines perpendicular to each point on the surface of an object. They are used to
illustrate the exact direction and orientation of every polygon on 3D geometry. Knowing the direction
and orientation determines how the object gets shaded. Tangents are lines that exists along the
surface’s plane. These lines are tangent to a point on the surface. The tangent lines are used to
describe the direction of textures you apply to the surface of 3D geometry.
– Scale: This slider increases or decreases the length of the vectors for both normals and tangents.

Chapter 29 3D Nodes 741

 – Show Normals: Displays blue vectors typically extending outside the surface of the geometry.
These normal vectors help indicate how different areas of the surface are illuminated based on the
angle at which the light hits it.
– Show Tangents: Displays green vectors for Y and red vectors of X. The X and Y vectors represent
the direction of the image or texture you are applying to the geometry.

Object ID
Use this slider to select which ID is used to create a mask from the object of an image. Use the
Sample button in the same way as the Color Picker to grab IDs from the image displayed in the viewer.
The image or sequence must have been rendered from a 3D software package with those
channels included.

Common Materials Tab

Common Materials 3D tab

The controls in the Materials tab are used to determine the appearance of the 3D object when lit. Most
of these controls directly affect how the object interacts with light using a basic shader. For more
advanced control over the objects appearance, you can use tools from the 3D Materials category of
the Effects Library. These tools can be used to assemble a more finely detailed and precise shader.
When a shader is constructed using the 3D Material tools and connected to the 3D Object’s material
input, the controls in this tab are replaced by a label that indicates that an external material is
currently in use.

Chapter 29 3D Nodes 742

Diffuse
Diffuse describes the base surface characteristics without any additional effects like reflections or
specular highlights.

Diffuse Color
The Diffuse Color determines the basic color of an object when the surface of that object is either lit
indirectly or lit by an ambient light. If a valid image is provided to the tools diffuse texture input, then
the RGB values provided here are also multiplied by the color values of the pixels in the diffuse
texture. The Alpha channel of the diffuse material can be used to control the transparency of
the surface.

Alpha
This slider sets the material’s Alpha channel value. This affects diffuse and specular colors equally, and
affects the Alpha value of the material in the rendered output. If the tools diffuse texture input is used,
then the Alpha value provided here is multiplied by the Alpha channel of the pixels in the image.

Opacity
Reducing the material’s Opacity decreases the color and Alpha values of the specular and diffuse
colors equally, making the material transparent and allowing hidden objects to be seen through the
material.

Specular
The Specular section provides controls for determining the characteristics of light that reflects toward
the viewer. These controls affect the appearance of the specular highlight that appears on the surface
of the object.

Specular Color
Specular Color determines the color of light that reflects from a shiny surface. The more specular a
material is, the glossier it appears. Surfaces like plastics and glass tend to have white specular
highlights, whereas metallic surfaces like gold have specular highlights that tend to inherit their color
from the material color. The basic shader material does not provide an input for textures to control the
specularity of the object. Use tools from the 3D Material category when more precise control is
required over the specular appearance.

Specular Intensity
Specular Intensity controls how strong the specular highlight is. If the specular intensity texture input
has a valid connection, then this value is multiplied by the Alpha value of the input.

Specular Exponent
Specular Exponent controls the falloff of the specular highlight. The greater the value, the sharper the
falloff, and the smoother and glossier the material appears. The basic shader material does not
provide an input for textures to control the specular exponent of the object. Use tools from the 3D
Material category when more precise control is required over the specular exponent.

Transmittance
Transmittance controls the way light passes through a material. For example, a solid blue sphere casts
a black shadow, but one made of translucent blue plastic would cast a much lower density
blue shadow.
There is a separate opacity option. Opacity determines how transparent the actual surface is when it is
rendered. Fusion allows adjusting both opacity and transmittance separately. This might be a bit
counter-intuitive to artists who are unfamiliar with 3D software at first. It is possible to have a surface
that is fully opaque but transmits 100% of the light arriving upon it, effectively making it a luminous/
emissive surface.

Chapter 29 3D Nodes 743

Attenuation
Attenuation determines how much color is transmitted through the object. For an object to have
transmissive shadows, set the attenuation to (1, 1, 1), which means 100% of green, blue, red light passes
through the object. Setting this color to RGB (1, 0, 0) means that the material transmits 100% of the red
arriving at the surface but none of the green or blue light. This allows “stained glass” shadows.

Alpha Detail
When the Alpha Detail slider is set to 0, the Alpha channel of the object is ignored and the entire
object casts a shadow. If it is set to 1, the Alpha channel determines what portions of the object cast
a shadow.

Color Detail
The Color Detail slider modulates light passing through the surface by the diffuse color + texture
colors. Use this to throw a shadow that contains color details of the texture applied to the object.
Increasing the slider from 0 to 1 brings in more of diffuse color + texture color into the shadow. Note
that the Alpha and opacity of the object are ignored when transmitting color, allowing an object with a
solid Alpha to still transmit its color to the shadow.

Saturation
The Saturation slider controls the saturation of the color component transmitted to the shadow. Setting
this to 0.0 results in monochrome shadows.

Receives Lighting/Shadows
These checkboxes control whether the material is affected by lighting and shadows in the scene. If
turned off, the object is always fully lit and/or unshadowed.

Two-Sided Lighting
This makes the surface effectively two-sided by adding a second set of normals facing the opposite
direction on the back side of the surface. This is normally off, to increase rendering speed, but can be
turned on for 2D surfaces or for objects that are not fully enclosed, to allow the reverse or interior
surfaces to be visible as well.
Normally, in a 3D application, only the front face of a surface is visible and the back face is culled, so
that if a camera were to revolve around a plane in a 3D application, when it reached the backside, the
plane would become invisible. Making a plane two sided in a 3D application is equivalent to adding
another plane on top of the first but rotated by 180 degrees so the normals are facing the opposite
direction on the backside. Thus, when you revolve around the back, you see the second image plane
that has its normals facing the opposite way.
Fusion does exactly the same thing as 3D applications when you make a surface two sided. The
confusion about what two-sided lighting does arises because Fusion does not cull backfacing
polygons by default. If you revolve around a one-sided plane in Fusion, you still see it from the
backside (but you are seeing the frontside bits duplicated through to the backside as if it were
transparent). Making the plane two sided effectively adds a second set of normals to the backside of
the plane.
Note that this can become rather confusing once you make the surface transparent, as the same rules
still apply and produce a result that is counterintuitive. If you view from the frontside a transparent
two-sided surface illuminated from the backside, it looks unlit.

Material ID
This control is used to set the numeric identifier assigned to this material. The Material ID is an integer
number that is rendered into the MatID auxiliary channel of the rendered image when the Material ID
option is enabled in the Renderer 3D tool. For more information, see Chapter 86, “3D Compositing
Basics” in the DaVinci Resolve Reference Manual or Chapter 25 in the Fusion Reference Manual.

Chapter 29 3D Nodes 744

Common Transform Tab

Common Transform 3D tab

Many tools in the 3D category include a Transform tab used to position, rotate, and scale the object
in 3D space.

Translation
X, Y, Z Offset
These controls can be used to position the 3D element.

Rotation
Rotation Order
Use these buttons to select which order is used to apply rotation along each axis of the object. For
example, XYZ would apply the rotation to the X axis first, followed by the Y axis and then finally
the Z axis.

X, Y, Z Rotation
Use these controls to rotate the object around its pivot point. If the Use Target checkbox is selected,
then the rotation is relative to the position of the target; otherwise, the global axis is used.

Pivot
X, Y, Z Pivot
A Pivot point is the point around which an object rotates. Normally, an object rotates around its own
center, which is considered to be a pivot of 0,0,0. These controls can be used to offset the pivot from
the center.

Scale
X, Y, Z Scale
If the Lock X/Y/Z checkbox is checked, a single Scale slider is shown. This adjusts the overall size of
the object. If the Lock checkbox is unchecked, individual X, Y, and Z sliders are displayed to allow
individual scaling in each dimension. Note: If the Lock checkbox is checked, scaling of individual
dimensions is not possible, even when dragging specific axes of the Transformation Widget in
scale mode.

Chapter 29 3D Nodes 745

Use Target
Selecting the Use Target checkbox enables a set of controls for positioning an XYZ target. When
target is enabled, the object always rotates to face the target. The rotation of the object becomes
relative to the target.

Import Transform
Opens a file browser where you can select a scene file saved or exported by your 3D application. It
supports the following file types:

LightWave Scene .lws

Max Scene .ase

Maya Ascii Scene .ma

dotXSI .xsi

The Import Transform button imports only transformation data. For 3D geometry, lights, and cameras,
consider using the File > FBX Import option.

Onscreen Transformation Controls

Viewer Transform controls

Most of the controls in the Transform tab are represented in the viewer with onscreen controls for
transformation, rotation, and scaling. To change the mode of the onscreen controls, select one of the
three buttons in the toolbar in the upper left of the viewer. The modes can also be toggled using the
keyboard shortcut Q for translation, W for rotation, and E for scaling. In all three modes, individual axes
of the control may be dragged to affect just that axis, or the center of the control may be dragged to
affect all three axes.
The scale sliders for most 3D tools default to locked, which causes uniform scaling of all three axes.
Unlock the Lock X/Y/Z Scale checkbox to scale an object on a single axis only.

Settings Tab

Common Settings 3D controls

Chapter 29 3D Nodes 746

The Common Settings tab can be found on most tools in Fusion. The following controls are specific
settings for 3D nodes.

Hide Incoming Connections

Enabling this checkbox can hide connection lines from incoming nodes, making a node tree appear
cleaner and easier to read. When enabled, empty fields for each input on a node are displayed in the
Inspector. Dragging a connected node from the node tree into the field hides that incoming
connection line as long as the node is not selected in the node tree. When the node is selected in the
node tree, the line reappears.

Comment Tab
The Comment tab contains a single text control that is used to add comments and notes to the tool.
When a note is added to a tool, a small red dot icon appears next to the setting’s tab icon and a text
bubble appears on the node. To see the note in the Node Editor, hold the mouse pointer over the
node for a moment. The contents of the Comments tab can be animated over time, if required.

Scripting Tab
The Scripting tab is present on every tool in Fusion. It contains several edit boxes used to add scripts
that process when the tool is rendering. For more details on the contents of this tab, please consult the
scripting documentation.

Chapter 29 3D Nodes 747

Chapter 30

3D Light Nodes
This chapter details the 3D Light nodes available when creating 3D composites in
Fusion. The abbreviations next to each node name can be used in the Select Tool
dialog when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Ambient Light [3AL]  749
Directional Light [3DL]  750
Point Light [3PL]  752
Spot Light [3SL]  754
The Common Controls  758

Chapter 30 3D Light Nodes 748

Ambient Light [3AL]

The Ambient Light node

Ambient Light Node Introduction

An Ambient Light is a directionless light that globally illuminates a scene. It has no real position or
rotation, although an onscreen control appears in the viewer to indicate that a light is present in the
scene. Position controls for the viewer are provided to move the widget out of the way of other
geometry, if necessary.
Similar to a Camera 3D, you connect lights into a Merge 3D and view them in the scene by viewing the
Merge 3D node. Selecting a light node and loading it into the viewer does not show anything.

Inputs
The Ambient Light node includes a single optional orange input for a 3D scene or 3D geometry.
– SceneInput: The orange input is an optional input that accepts a 3D scene. If a scene is
provided, the Transform controls in this node apply to the entire scene provided.

Basic Node Setup

The Ambient Light node is designed to be part of a larger 3D scene. You connect the light directly into
a Merge 3D. Separating lights into different Merge 3D nodes allows you to control which lights affect
which objects.

Ambient Light node structure

Chapter 30 3D Light Nodes 749

Inspector

Ambient Light controls

Controls Tab
The Controls tab is used to set the color and brightness of the ambient light.

Enabled
When the Enabled checkbox is turned on, the ambient light affects the scene. When the checkbox to
turned off, the light is turned off. This checkbox performs the same function as the red switch to the
left of the node’s name in the Inspector.

Color
Use this standard Color control to set the color of the light.

Intensity
Use this slider to set the Intensity of the ambient light. A value of 0.2 indicates 20% percent light. A
perfectly white texture lit only with a 0.2 ambient light would render at 20% gray (.2, .2, .2).

Common Controls
Transform and Settings Tabs
The options presented in the Transform and Settings tabs are commonly found in other lighting nodes.
For more detail on the controls found in these tabs, see “The Common Controls” section at the end of
this chapter.

Directional Light [3DL]

The Directional Light node

Chapter 30 3D Light Nodes 750

Directional Light Node Introduction
A directional light is a light with a clear direction but without a clear source or distance, similar to
sunlight. This light shows an onscreen control, but the position of the control has no meaning. The
rotation of the control is used to determine from where in the scene the light appears to be coming.
Similar to a Camera 3D, you connect lights into a Merge 3D and view them in the scene by viewing the
Merge 3D node. Selecting a light node and loading it into the viewer does not show anything.

Inputs
The Directional Light node includes a single optional orange input for a 3D scene or 3D geometry.
– SceneInput: The orange input is an optional input that accepts a 3D scene. If a scene is
provided, the Transform controls in this node apply to the entire scene provided.

Basic Node Setup

The Directional Light node is designed to be part of a larger 3D scene. You connect the light directly
into a Merge 3D. Separating lights into different Merge 3D nodes allows you to control which lights
affect which objects.

Directional Light node structure

Inspector

Directional Light controls

Chapter 30 3D Light Nodes 751

Controls Tab
The Controls tab is used to set the color and brightness of the directional light. The direction of the
light source is controlled by the rotation controls in the Transform tab.

Enabled
When the Enabled checkbox is turned on, the directional light affects the scene. When the checkbox
is turned off, the light is turned off. This checkbox performs the same function as the red switch to the
left of the node’s name in the Inspector.

Color
Use this standard Color control to set the color of the light.

Intensity
Use this slider to set the Intensity of the directional light. A value of 0.2 indicates 20% percent light.

Common Controls
Transform and Settings Tabs
The options presented in the Transform and Settings tabs are commonly found in other lighting nodes.
For more detail on the controls found in these tabs, see “The Common Controls” section at the end of
this chapter.

Point Light [3PL]

The Point Light node

Point Light Node Introduction

A point light is a light source with a clear position in space that emits light in all directions. A light bulb
is a good example of a point light.
This light shows an onscreen control, although only the position and distance of the control affect the
light. Since the light is a 360-degree source, rotation has no meaning. Additionally, a point light may
fall off with distance, unlike an ambient or directional light.
Similar to a Camera 3D, you connect lights into a Merge 3D and view them in the scene by viewing the
Merge 3D node. Selecting a light node and loading it into the viewer does not show anything.

Inputs
The Point Light node includes a single optional orange input for a 3D scene or 3D geometry.
– SceneInput: The orange input is an optional input that accepts a 3D scene. If a scene is
provided, the Transform controls in this node apply to the entire scene provided.

Basic Node Setup

The Point Light node is designed to be part of a larger 3D scene. You connect the light directly into a
Merge 3D. Separating lights into different Merge 3D nodes allows you to control which lights affect
which objects.

Chapter 30 3D Light Nodes 752

Point Light node structure

Inspector

Point Light controls

Controls Tab
The Controls tab is used to set the color and brightness of the point light. The position and distance of
the light source are controlled in the Transform tab.

Enabled
When the Enabled checkbox is turned on, the point light affects the scene. When the checkbox is
turned off the light is turned off. This checkbox performs the same function as the red switch to the left
of the node’s name in the Inspector.

Color
Use this standard Color control to set the color of the light.

Intensity
Use this slider to set the Intensity of the point light. A value of 0.2 indicates 20% percent light.

Decay Type
A point light defaults to No Decay, meaning that its light has equal intensity at all points in the scene.
To cause the intensity to fall off with distance, set the Decay Type to either Linear or Quadratic modes.

Chapter 30 3D Light Nodes 753

Common Controls
Transform and Settings Tabs
The options presented in the Transform and Settings tabs are commonly found in other lighting nodes.
For more detail on the controls found in these tabs, see “The Common Controls” section at the end of
this chapter.

Spot Light [3SL]

The Spot Light node

Spot Light Node Introduction

A spotlight is a light that comes from a specific point and that has a clearly defined cone, with falloff of
the light to the edges. Experienced stage and theatre lighting technicians may recognize the spotlight
as being very similar to practical lights used in live productions. This is the only type of light capable of
casting shadows.
Similar to a Camera 3D, you connect lights into a Merge 3D and view them in the scene by viewing the
Merge 3D node. Selecting a light node and loading it into the viewer does not show anything.

Inputs
The Spot Light node includes a single optional orange input for a 3D scene or 3D geometry.
– SceneInput: The orange input is an optional input that accepts a 3D scene. If a scene is
provided, the Transform controls in this node apply to the entire scene provided.

Basic Node Setup

The Spot Light node is designed to be part of a larger 3D scene. You connect the light directly into a
Merge 3D. Separating lights into different Merge 3D nodes allows you to control which lights affect
which objects.

Spot Light node structure

Chapter 30 3D Light Nodes 754

Inspector

Spot Light control

Controls Tab
The Controls tab is used to set the color and brightness of the spotlight. The position, rotation, and
distance of the light source are controlled in the Transform tab.

Enabled
When the Enabled checkbox is turned on, the spotlight affects the scene. When the checkbox is
turned off the light is turned off. This checkbox performs the same function as the red switch to the left
of the node’s name in the Inspector.

Color
Use this standard Color control to set the color of the light.

Intensity
Use this slider to set the Intensity of the spot light. A value of 0.2 indicates 20% percent light.

Decay Type
A spotlight defaults to No Falloff, meaning that its light has equal intensity on geometry despite the
distance from the light to the geometry. To cause the intensity to fall off with distance, set the Decay
type to either Linear or Quadratic modes.

Cone Angle
The Cone Angle of the light refers to the width of the cone where the light emits its full intensity.
The larger the angle, the wider the cone angle, up to a limit of 90 degrees.

Penumbra Angle
The Penumbra Angle determines the area beyond the cone angle where the light’s intensity falls off
toward 0. A larger penumbra angle defines a larger falloff, while a value of 0 generates a hard-
edged light.

Chapter 30 3D Light Nodes 755

Dropoff
The Dropoff controls how quickly the penumbra angle falls off from full intensity to 0.

Shadows
This section provides several controls used to define the shadow map used when this spotlight
creates shadows. For more information, see Chapter 25, “3D Compositing Basics” in the Fusion
Reference Manual or Chapter 86 in the DaVinci Resolve Reference Manual.

Enable Shadows
The Enable Shadows checkbox should be selected if the light is to produce shadows. This defaults
to selected.

Shadow Color
Use this standard Color control to set the color of the shadow. This defaults to black (0, 0, 0).

Density
The shadow density determines the transparency of the shadow. A density of 1.0 produces a
completely opaque shadow, whereas lower values make the shadow more transparent.

Shadow Map Size

The Shadow Map Size control determines the size of the bitmap used to create the shadow map.
Larger values produce more detailed shadow maps at the expense of memory and performance.

Shadow Map Proxy

Shadow Map Proxy determines the size of the shadow map used when the Proxy or Auto Proxy modes
are enabled. A value of 0.5 would produce a shadow map at half the resolution defined in the
Shadow Map Size.

Multiplicative/Additive Bias
Shadows are essentially textures applied to objects in the scene, so there is occasionally Z-fighting,
where the portions of the object that should be receiving the shadows render over the top of the
shadow. Biasing works by adding a small depth offset to move the shadow away from the surface it is
shadowing, eliminating the Z-fighting. Too little bias and the objects can self-shadow themselves. Too
much bias and the shadow can become separated from the surface. Adjust the Multiplicative Bias first,
and then fine tune the result using the Additive Bias control.
For more information, see the Multiplicative and Additive Bias section of Chapter 86, “3D Compositing”
in the DaVinci Resolve Reference Manual, or Chapter 25 in the Fusion Reference Manual.

Force All Materials Non-Transmissive

Normally, an RGBAZ shadow map is used when rendering shadows. By enabling this option, you are
forcing the renderer to use a Z-only shadow map. This can lead to significantly faster shadow
rendering while using a fifth as much memory. The disadvantage is that you can no longer cast
“stained glass”-like shadows.

Shadow Map Sampling

Sets the quality for sampling of the shadow map.

Softness
Soft edges in shadows are produced by filtering the shadow map when it is sampled. Fusion provides
two separate filtering methods for rendering shadows, which produce different effects.

Chapter 30 3D Light Nodes 756

 NOTE: Shadows have a hard edge. No filtering of the shadow map is done at all. The
advantage of this method is that you only have to sample one pixel in the shadow map, so
it is fast.

– Constant: Shadows edges have a constant softness. A filter with a constant width is used when
sampling the shadow map. Adjusting the Constant Softness slider controls the size of the filter.
Note that the larger you make the filter, the longer it takes to render the shadows.
– Variable: The shadow edge softness grows the further the shadow receiver is positioned from the
shadow caster. The variable softness is achieved by changing the size of the filter based on the
distance between the receiver and caster. When this option is selected, the Softness Falloff, Min
Softness, and Max Softness sliders appear.

Constant Softness
If the Softness is set to Constant, then this slider appears. It can be used to set the overall softness of
the shadow.

Softness Falloff
The Softness Falloff slider appears when the Softness is set to variable. This slider controls how fast
the softness of shadow edges grows with distance. More precisely, it controls how fast the shadow
map filter size grows based upon the distance between the shadow caster and receiver. Its effect is
mediated by the values of the Min and Max Softness sliders.

Min Softness
The Min Softness slider appears when the Softness is set to Variable. This slider controls the Minimum
Softness of the shadow. The closer the shadow is to the object casting the shadow, the sharper it is,
up to the limit set by this slider.

Max Softness
The Max Softness slider appears when the Softness is set to Variable. This slider controls the
Maximum Softness of the shadow. The further the shadow is from the object casting the shadow, the
softer it is, up to the limit set by this slider.

Common Controls
Transform and Settings Tabs
The options presented in the Transform and Settings tabs are commonly found in other lighting nodes.
For more detailed information on the controls found in these tabs, see “The Common Controls” section
at the end of this chapter.

Chapter 30 3D Light Nodes 757

The Common Controls
Nodes that handle 3D lighting share several identical controls in the Inspector. This section describes
controls that are common among 3D lighting nodes.

Common Transform Tab

Common Transform 3D tab

Many tools in the 3D category include a Transform tab used to position, rotate, and scale the object
in 3D space.

Translation
X, Y, Z Offset
These controls can be used to position the 3D element.

Rotation
Rotation Order
Use these buttons to select which order is used to apply Rotation along each axis of the object. For
example, XYZ would apply the rotation to the X axis first, followed by the Y axis, and finally the Z axis.

X, Y, Z Rotation
Use these control to rotate the object around its pivot point. If the Use Target checkbox is selected,
then the rotation is relative to the position of the target; otherwise, the global axis is used.

Pivot
X, Y, Z Pivot
A pivot point is the point around which an object rotates. Normally, an object rotates around its own
center, which is considered to be a pivot of 0,0,0. These controls can be used to offset the pivot from
the center.

Chapter 30 3D Light Nodes 758

Scale
X, Y, Z Scale
If the Lock X/Y/Z checkbox is checked, a single Scale slider is shown. This adjusts the overall size of
the object. If the Lock checkbox is unchecked, individual X, Y, and Z sliders are displayed to allow
individual scaling in each dimension. Note: If the Lock checkbox is checked, scaling of individual
dimensions is not possible, even when dragging specific axes of the Transformation Widget in
scale mode.

Use Target
Selecting the Use Target checkbox enables a set of controls for positioning an XYZ target. When
Target is enabled, the object always rotates to face the target. The rotation of the object becomes
relative to the target.

Import Transform
Opens a file browser where you can select a scene file saved or exported by your 3D application. It
supports the following file types:

LightWave Scene .lws

Max Scene .ase

Maya Ascii Scene .ma

dotXSI .xsi

The Import Transform button imports only transformation data. For 3D geometry, lights, and cameras,
consider using the File > FBX Import option.

Onscreen Transformation Controls

Viewer Transform buttons

Viewer Transform buttons.

Most of the controls in the Transform tab are represented in the viewer with onscreen controls for
transformation, rotation, and scaling. To change the mode of the onscreen controls, select one of the
three buttons in the toolbar in the upper left of the viewer. The modes can also be toggled using the
keyboard shortcut Q for translation, W for rotation, and E for scaling. In all three modes, individual axes
of the control may be dragged to affect just that axis, or the center of the control may be dragged to
affect all three axes.
The Scale sliders for most 3D tools default to locked, which causes uniform scaling of all three axes.
Unlock the Lock X/Y/Z Scale checkbox to scale an object on a single axis only.

Chapter 30 3D Light Nodes 759

Settings Tab

Common Settings 3D controls

The Common Settings tab can be found on almost every tool found in Fusion. The following controls
are specific settings for 3D nodes.

Hide Incoming Connections

Enabling this checkbox can hide connection lines from incoming nodes, making a node tree appear
cleaner and easier to read. When enabled, fields for each input on a node are displayed. Dragging a
connected node from the node tree into the field hides that incoming connection line as long as the
node is not selected in the node tree. When the node is selected in the node tree, the line reappears.

Comment Tab
The Comment tab contains a single text control that is used to add comments and notes to the tool.
When a note is added to a tool, a small red dot icon appears next to the setting’s tab icon, and a text
bubble appears on the node. To see the note in the Node Editor, hold the mouse pointer over the
node for a moment. The contents of the Comments tab can be animated over time, if required.

Scripting Tab
The Scripting tab is present on every tool in Fusion. It contains several edit boxes used to add scripts
that process when the tool is rendering. For more details on the contents of this tab, please consult the
scripting documentation.

Chapter 30 3D Light Nodes 760

Chapter 31

3D Material Nodes
This chapter details the 3D Material nodes available when creating 3D composites in
Fusion. The abbreviations next to each node name can be used in the Select Tool
dialog when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Blinn [3BI]  762
Channel Boolean [3BOL]  766
Cook Torrance [3CT]  769
Material Merge 3D [3MM]  773
Phong [3PH]  774
Reflect [3RR]  778
Stereo Mix [3SMM]  781
Ward [3WD]  783
The Common Controls  787

Chapter 31 3D Material Nodes 761

Blinn [3BI]

The Blinn node

Blinn Node Introduction

The Blinn node is a basic illumination material that can be applied to geometry in the 3D scene. It
describes how the object responds to light and provides multiple texture map inputs to allow fine
control over the diffuse, specular, and bump map components of the material.
The standard basic material provided in the Materials tab of most geometry nodes is a simplified
version of the Blinn node. The primary difference is that the Blinn node provides additional texture
map inputs beyond just diffuse.
The Blinn node outputs a 3D Material that can be connected to the material inputs on any
3D geometry node.
The Blinn model in Fusion calculates the highlight as the dot product of the surface normal and the half
angle vector between light source and viewer (dot(N, H)). This may not always match the Blinn model
illumination model used by other 3D applications.

Inputs
There are five inputs on the Blinn node that accept 2D images or 3D materials. These inputs control
the overall color and image used for the 3D object as well as the color and texture used in the
specular highlight. Each of these inputs multiplies the pixels in the texture map by the equivalently
named parameters in the node itself. This provides an effective method for scaling parts of
the material.
– Diffuse Texture: The orange Diffuse Texture input accepts a 2D image or a 3D material to be
used as a main object texture map.
– Specular Color Material: The green Specular Color material input accepts a 2D image or a 3D
material to be used as the color texture map for specula highlight areas.
– Specular Intensity Materials: The magenta Specular Intensity material input accepts a 2D
image or a 3D material to be used to alter the intensity of specular highlights. When the input
is a 2D image, the Alpha channel is used to create the map, while the color channels are
discarded.
– Specular Exponent Material: The teal Specular Exponent material input accepts a 2D image
or a 3D material that is used as a falloff map for the material’s specular highlights. When the
input is a 2D image, the Alpha channel is used to create the map, while the color channels are
discarded.
– Bump Map Material: The white Bump Map material input accepts only a 3D material. Typically,
you connect the texture into a Bump Map node, and then connect the Bump Map node to this
input. This input uses the RGB information as texture-space normals.

Chapter 31 3D Material Nodes 762

When nodes have as many inputs as this one does, it is often difficult to make connections with any
precision. Hold down the Option (macOS) or Alt (Windows) key while dragging the output from another
node over the node tile, and keep holding Option or Alt when releasing the left mouse button. A small
drop-down menu listing all the inputs provided by the node appears. Click on the desired input to
complete the connection. Alternatively, you can drag the output from a node with the right mouse
button to activate the same menu.

Basic Node Setup

The output of a Blinn node output is connected to the material input on a 3D scene or 3D geometry
node to which you want the shader applied. The Blinn inputs can use images as the diffuse color
material (orange) and specular color material (green). This can lead to a smooth, shiny material.

A Blinn shader with diffuse and specular color materials connected

Inspector

Blinn controls

Controls Tab
The Controls tab is the primary tab for the Blinn node. It controls the color and shininess applied to the
surface of the 3D geometry.

Chapter 31 3D Material Nodes 763

Diffuse
Diffuse describes the base surface characteristics without any additional effects like reflections or
specular highlights. Besides defining the base color of an object, the diffuse color also defines the
transparency of the object. The Alpha in a diffuse texture map can be used to make portions of the
surface transparent.

Diffuse Color
A material’s Diffuse Color describes the base color presented by the material when it is lit indirectly or
by ambient light. If a diffuse texture map is provided, then the color value provided here is multiplied
by the color values in the texture.

Alpha
This slider sets the material’s Alpha channel value. This affects diffuse and specular colors equally and
affects the Alpha value of the material in the rendered output. If a diffuse texture map is provided, then
the Alpha value set here is multiplied by the Alpha values in the texture map.

Opacity
Reducing the material’s opacity decreases the color and Alpha values of the specular and diffuse
colors equally, making the material transparent.

Specular
The parameters in the Specular section describe the look of the specular highlight of the surface.
These values are evaluated in a different way for each illumination model.

Specular Color
Specular Color determines the color of light that reflects from a shiny surface. The more specular a
material is, the glossier it appears. Surfaces like plastics and glass tend to have white specular
highlights, whereas metallic surfaces like gold have specular highlights that inherit their color from the
material color. If a specular texture map is provided, then the value provided here is multiplied by the
color values from the texture.

Specular Intensity
Specular Intensity controls how strong the specular highlight is. If the specular intensity texture is
provided, then this value is multiplied by the Alpha value of the texture.

Specular Exponent
Specular Exponent controls the falloff of the specular highlight. The greater the value, the sharper the
falloff, and the smoother and glossier the material appears. If the specular exponent texture is
provided, then this value is multiplied by the Alpha value of the texture map.

Transmittance
Transmittance controls the way light passes through a material. For example, a solid blue sphere casts
a black shadow, but one made of translucent blue plastic would cast a much lower density
blue shadow.
There is a separate Opacity option. Opacity determines how transparent the actual surface is when it
is rendered. Fusion allows adjusting both opacity and transmittance separately. At first, this might be a
bit counterintuitive to those who are unfamiliar with 3D software. It is possible to have a surface that is
fully opaque but transmits 100% of the light arriving upon it, effectively making it a luminous/
emissive surface.

Attenuation
Attenuation determines how much color is passed through the object. For an object to have
transmissive shadows, set the attenuation to (1, 1, 1), which means 100% of green, blue, and red light
passes through the object. Setting this color to RGB (1, 0, 0) means that the material transmits 100% of

Chapter 31 3D Material Nodes 764

the red arriving at the surface but none of the green or blue light. This can be used for “stained
glass”-styled shadows.

Alpha Detail
When the Alpha Detail slider is set to 0, the Alpha channel of the object is ignored and the entire
object casts a shadow. If it is set to 1, the Alpha channel determines what portions of the object
cast a shadow.

Color Detail
The Color Detail slider modulates light passing through the surface by the diffuse color + texture
colors. Use this to throw a shadow that contains color details of the texture applied to the object.
Increasing the slider from 0 to 1 brings in more diffuse color + texture color into the shadow. Note that
the Alpha and opacity of the object are ignored when transmitting color, allowing an object with a solid
Alpha to still transmit its color to the shadow.

Saturation
The Saturation slider controls the saturation of the color component transmitted to the shadow. Setting
this to 0.0 results in monochrome shadows.

Receives Lighting/Shadows
These checkboxes control whether the material is affected by lighting and shadows in the scene. If
turned off, the object is always fully lit and/or unshadowed.

Two-Sided Lighting
This effectively makes the surface two sided by adding a second set of normals facing the opposite
direction on the backside of the surface. This is normally off to increase rendering speed, but it can be
turned on for 2D surfaces or for objects that are not fully enclosed, to allow the reverse or interior
surfaces to be visible as well.
Normally, in a 3D application, only the front face of a surface is visible and the back face is culled, so
that if a camera were to revolve around a plane in a 3D application, when it reached the backside, the
plane would become invisible. Making a plane two sided in a 3D application is equivalent to adding
another plane on top of the first but rotated by 180 degrees so the normals are facing the opposite
direction on the backside. Thus, when you revolve around the back, you see the second image plane,
which has its normals facing the opposite way.
Fusion does exactly the same thing as 3D applications when you make a surface two sided. The
confusion about what two-sided lighting does arises because Fusion does not cull back-facing
polygons by default. If you revolve around a one-sided plane in Fusion, you still see it from the
backside (but you are seeing the frontside duplicated through to the backside as if it were transparent).
Making the plane two sided effectively adds a second set of normals to the backside of the plane.

NOTE: This can become rather confusing once you make the surface transparent, as the
same rules still apply and produce a result that is counterintuitive. If you view from the
frontside a transparent two-sided surface illuminated from the backside, it looks unlit.

Material ID
This slider sets the numeric identifier assigned to this material. This value is rendered into the MatID
auxiliary channel if the corresponding option is enabled in the renderer.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 31 3D Material Nodes 765

Channel Boolean [3BOL]

The Channel Boolean node

Channel Boolean Node Introduction

The Channel Boolean (not to be confused with the 2D Channel Booleans) can be used to remap and
modify channels of 3D materials using mathematical operations. For example, if you want to use the
red channel of a material to control a scalar input of an illumination model that uses the Alpha channel
(e.g., Blinn. SpecularExponent), you can remap the channels here. Furthermore, it allows the use of
geometry-specific information like texture space coordinates and normals.

Inputs
There are two inputs on the Channel Boolean Node: one for the foreground material, and one for the
background material. Both inputs accept either a 2D image or a 3D material like Blinn, Cook-Torrence,
or Phong node.
– BackgroundMaterial: The orange background material input accepts a 2D image or
a 3D material.
– ForegroundMaterial: The green foreground input also accepts a 2D image or a 3D material.

Basic Node Setup

There are many uses for the material 3D Channel Boolean. Most often it is used to combine material
looks or manipulate UV texture coordinates
In the below example, the Channel Boolean node combines the Cook Torrance and Blinn materials. It
uses the math operands in the Channel Boolean to switch, invert, and mix the two inputs, creating a
neon flickering effect.

A Channel Boolean used to combine and operate on Cook Torrance and Blinn nodes

Chapter 31 3D Material Nodes 766

Inspector

Channel Boolean controls

Controls Tab
The Controls tab includes a section for each RGBA channel. Within each channel are two input menus
called Operand A and Operand B. The function performed on these two inputs is selected in the
Operation menu.

Operand A/B
The Operand menus, one for each output RGBA channel, allow you to set the desired input information
for the corresponding channel.
– Red/Green/Blue/Alpha FG
Reads the color information of the foreground material.
– Red/Green/Blue/Alpha BG
Reads the color information of the background material.
– Black/White/Mid Gray
Sets the value of the channel to 0, 0.5, or 1.
– Hue/Lightness/Saturation FG
Reads the color information of the foreground material, converts it into the HLS color space, and
puts the selected information into the corresponding channel.
– Hue/Lightness/Saturation BG
Reads the color information of the background material, converts it into the HLS color space, and
puts the selected information into the corresponding channel.
– Luminance FG
Reads the color information of the foreground material and calculates the luminance value for
the channel.
– Luminance BG

Chapter 31 3D Material Nodes 767

 Reads the color information of the background material and calculates the luminance value for
the channel.
– X/Y/Z Position FG
Sets the value of the channel to the position of the pixel in 3D space. The vector information is
returned in eye space.
– U/V/W Texture FG
Applies the texture space coordinates of the foreground material to the channels.
– U/V/W EnvCoords FG
Applies the environment texture space coordinates to the channels. Use it upstream of nodes
modifying the environment texture coordinates like the Reflect 3D node.
– X/Y/Z Normal
Sets the value of the channel to the selected axis of the normal vector. The vector is returned in
eye space.

Operation
Determines the Operation of how the operands are combined.
– A: Uses Operand A only for the output channel.
– B: Uses Operand B only for the output channel.
– 1-A: Subtracts the value of Operand A from 1.
– 1-B: Subtracts the value of Operand B from 1.
– A+B: Adds the value of Operand A and B.
– A-B: Subtracts the value of Operand B from A.
– A*B: Multiplies the value of both Operands.
– A/B: Divides the value of Operand B from A.
– min(A,B): Compares the values of Operands A and B and returns the smaller one.
– max(A,B): Compares the values of Operands A and B and returns the bigger one.
– avg(A,B): Returns the average value of both Operands.

Material ID
This slider sets the numeric identifier assigned to this material. This value is rendered into the MatID
auxiliary channel if the corresponding option is enabled in the renderer.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 31 3D Material Nodes 768

Cook Torrance [3CT]

The Cook Torrance node

Cook Torrance Node Introduction

The Cook Torrance node is a basic illumination material that can be applied to geometry in the 3D
scene. The diffuse calculation for this node is similar to that used in the basic material and the Blinn
node, but the specular highlights are evaluated using an optimized Fresnel/Beckmann equation. This
illumination model is primarily used for shading metal or other shiny and highly reflective surfaces.
The Cook Torrance node outputs a 3D Material that can be connected to the material inputs on any
3D geometry node.

Inputs
There are six inputs on the Cook Torrance node that accept 2D images or 3D materials. These inputs
control the overall color and image used for the 3D object as well as controlling the color and texture
used in the specular highlight. Each of these inputs multiplies the pixels in the texture map by the
equivalently named parameters in the node itself. This provides an effective method for scaling parts
of the material.
– Diffuse Color Material: The orange Diffuse Color material input accepts a 2D image or a 3D
material to be used as overall color and texture of the object.
– Specular Color Material: The green Specular Color material input accepts a 2D image or a 3D
material to be used as the color and texture of the specular highlight.
– Specular Intensity Material: The magenta Specular Intensity material input accepts a 2D image
or a 3D material to alter the intensity of the specular highlight. When the input is a 2D image,
the Alpha channel is used to create the map, while the color channels are discarded.
– Specular Roughness Material: The white Specular Roughness material input accepts a 2D
image or a 3D material to be used as a map for modifying the roughness of the specular
highlight. The Alpha of the texture map is multiplied by the value of the roughness control.
– Specular Refractive Index Material: The white Specular Refractive Index material input accepts
a 2D image or a 3D material, using the RGB channels as the refraction texture.
– Bump Map Material: The white Bump Map material input accepts only a 3D material. Typically,
you connect the texture into a Bump Map node, and then connect the Bump Map node to this
input. This input uses the RGB information as texture-space normals.

Each of these inputs multiplies the pixels in the texture map by the equivalently named parameters in
the node itself. This provides an effective method for scaling parts of the material.
When nodes have as many inputs as this one does, it is often difficult to make connections with any
precision. Hold down the Option (macOS) or Alt (Windows) key while dragging the output from another
node over the node tile, and keep holding Option or Alt when releasing the left mouse button. A small
drop-down menu listing all the inputs provided by the node appears. Click on the desired input to
complete the connection.

Chapter 31 3D Material Nodes 769

Basic Node Setup
The output of a Cook Torrance node output is connected to the material input on a 3D scene or 3D
geometry node to which you want the shader applied. The Cook Torrance inputs can use images as
the diffuse color material (yellow) and specular color material (green). This can result in a smooth,
shiny material.

A Cook Torrance shader with diffuse and specular color materials connected

Inspector

Cook Torrance controls

Controls Tab
The Controls tab contains parameters for adjusting the main color, highlight, and lighting properties of
the Cook Torrance shader node.

Diffuse
Diffuse describes the base surface characteristics without any additional effects like reflections or
specular highlights. Besides defining the base color of an object, the diffuse color also defines the
transparency of the object. The Alpha in a diffuse texture map can be used to make portions of the
surface transparent.

Chapter 31 3D Material Nodes 770

Diffuse Color
A material’s Diffuse Color describes the base color presented by the material when it is lit indirectly or
by ambient light. If a diffuse texture map is provided, then the color value provided here is multiplied
by the color values in the texture.

Alpha
This slider sets the material’s Alpha channel value. This affects diffuse and specular colors equally, and
affects the Alpha value of the material in the rendered output. If a diffuse texture map is provided, then
the Alpha value set here is multiplied by the Alpha values in the texture map.

Opacity
Reducing the material’s Opacity decreases the color and Alpha values of the specular and diffuse
colors equally, making the material transparent.

Specular
The parameters in the Specular section describe the look of the specular highlight of the surface.
These values are evaluated in a different way for each illumination model.

Specular Color
Specular Color determines the color of light that reflects from a shiny surface. The more specular a
material is, the glossier it appears. Surfaces like plastics and glass tend to have white specular
highlights, whereas metallic surfaces like gold have specular highlights that inherit their color from the
material color. If a specular texture map is provided, then the value provided here is multiplied by the
color values from the texture.

Specular Intensity
Specular Intensity controls how strong the specular highlight is. If the specular intensity texture is
provided, then this value is multiplied by the Alpha value of the texture.

Roughness
The Roughness of the specular highlight describes diffusion of the specular highlight over the surface.
The greater the value, the wider the falloff, and the more brushed and metallic the surface appears. If
the roughness texture map is provided, then this value is multiplied by the Alpha value from
the texture.

Do Fresnel
Selecting this checkbox adds Fresnel calculations to the materials illumination model. This provides
more realistic-looking metal surfaces by taking into account the refractiveness of the material.

Refractive Index
This slider appears when the Do Fresnel checkbox is selected. The Refractive Index applies only to
the calculations for the highlight; it does not perform actual refraction of light through transparent
surfaces. If the refractive index texture map is provided, then this value is multiplied by the Alpha value
of the input.

Transmittance
Transmittance controls the way light passes through a material. For example, a solid blue sphere casts
a black shadow, but one made of translucent blue plastic would cast a much lower density
blue shadow.
There is a separate Opacity option. Opacity determines how transparent the actual surface is when it
is rendered. Fusion allows adjusting both opacity and transmittance separately. At first, this might be a
bit counterintuitive to those who are unfamiliar with 3D software. It is possible to have a surface that is
fully opaque but transmits 100% of the light arriving upon it, effectively making it a luminous/
emissive surface.

Chapter 31 3D Material Nodes 771

Attenuation
Attenuation determines how much color is passed through the object. For an object to have
transmissive shadows, set the attenuation to (1, 1, 1), which means 100% of green, blue, and red light
passes through the object. Setting this color to RGB (1, 0, 0) means that the material transmits 100% of
the red arriving at the surface but none of the green or blue light. This can be used to create “stained
glass”-styled shadows.

Alpha Detail
When the Alpha Detail slider is set to 0, the Alpha channel of the object is ignored and the entire
object casts a shadow. If it is set to 1, the Alpha channel determines what portions of the object
cast a shadow.

Color Detail
The Color Detail slider modulates light passing through the surface by the diffuse color + texture
colors. Use this to throw a shadow that contains color details of the texture applied to the object.
Increasing the slider from 0 to 1 brings in more diffuse color + texture color into the shadow. Note that
the Alpha and opacity of the object are ignored when transmitting color, allowing an object with a solid
Alpha to still transmit its color to the shadow.

Saturation
The Saturation slider controls the saturation of the color component transmitted to the shadow. Setting
this to 0.0 results in monochrome shadows.

Receives Lighting/Shadows
These checkboxes control whether the material is affected by lighting and shadows in the scene. If
turned off, the object is always fully lit and/or unshadowed.

Two-Sided Lighting
This effectively makes the surface two sided by adding a second set of normals facing the opposite
direction on the backside of the surface. This is normally off to increase rendering speed, but it can be
turned on for 2D surfaces or for objects that are not fully enclosed, to allow the reverse or interior
surfaces to be visible as well.
Normally, in a 3D application, only the front face of a surface is visible and the back face is culled, so
that if a camera were to revolve around a plane in a 3D application, when it reached the backside, the
plane would become invisible. Making a plane two sided in a 3D application is equivalent to adding
another plane on top of the first but rotated by 180 degrees so the normals are facing the opposite
direction on the backside. Thus, when you revolve around the back, you see the second image plane,
which has its normals facing the opposite way.

NOTE: This can become rather confusing once you make the surface transparent, as the
same rules still apply and produce a result that is counterintuitive. If you view from the
frontside a transparent two-sided surface illuminated from the backside, it looks unlit.

Material ID
This slider sets the numeric identifier assigned to this material. This value is rendered into the MatID
auxiliary channel if the corresponding option is enabled in the renderer.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 31 3D Material Nodes 772

Material Merge 3D [3MM]

The Material Merge node

Material Merge Node Introduction

The Material Merge node can be used to combine two separate materials together. This node can be
used to composite Material nodes, combining multiple illumination materials (Blinn, Cook Torrance)
with texture nodes (Bump Map, Reflection) to create complex shader networks.
The node also provides a mechanism for assigning a new material identifier to the combined material.

Inputs
The Material Merge node includes two inputs for the two materials you want to combine.
– Background Material: The orange Background material input accepts a 2D image or a 3D
material to be used as the background material.
– Foreground Material: The green Foreground material input accepts a 2D image or a 3D
material to be used as the foreground material. A 2D image is treated as a diffuse texture map
in the basic shading model.

Basic Node Setup

The output of a Material Merge node is connected to the material input on a 3D scene or 3D geometry
node. The Material Merge node below is taking in a background base layer from the Blinn shader and
combining it with a more textured bump map layer.

A Material Merge node combining a Blinn-based shader (teal underlay) and a Ward-based shader (orange underlay)

Chapter 31 3D Material Nodes 773

Inspector

Material Merge controls

Controls Tab
The Controls tab includes a single slider for blending the two materials together.

Blend
The Blend behavior of the Material Merge is similar to the Dissolve (DX) node for images. The two
materials/textures are mixed using the value of the slider to determine the percentage each input
contributes. While the background and foreground inputs can be a 2D image instead of a material, the
output of this node is always a material.
Unlike the 2D Dissolve node, both foreground and background inputs are required.

Material ID
This slider sets the numeric identifier assigned to the resulting material. This value is rendered into the
MatID auxiliary channel if the corresponding option is enabled in the renderer.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Phong [3PH]

The Phong node

Phong Node Introduction

The Phong node is a basic illumination material that can be applied to geometry in the 3D scene. It
describes how the object responds to light and provides multiple texture map inputs to allow fine
control over the diffuse, specular, and bump map components of the material.
While producing a highlight similar to that produced by the Blinn model, it is more commonly used for
shiny/polished plastic surfaces.

Chapter 31 3D Material Nodes 774

Inputs
There are five inputs on the Phong node that accept 2D images or 3D materials. These inputs control
the overall color and image used for the 3D object as well as controlling the color and texture used in
the specular highlight. Each of these inputs multiplies the pixels in the texture map by the equivalently
named parameters in the node itself. This provides an effective method for scaling parts of
the material.
– Diffuse Material: The orange Diffuse material input accepts a 2D image or a 3D material to be
used as a main color and texture of the object.
– Specular Color Material: The green Specular Color material input accepts a 2D image or a 3D
material to be used as a highlight color and texture of the object.
– Specular Intensity Material: The magenta Specular Intensity material input accepts a 2D image
or a 3D material to be used as an intensity map for the material’s highlights. When the input is a
2D image, the Alpha channel is used to create the map, while the color channels are discarded.
– Specular Exponent Material: The teal Specular Exponent material input accepts a 2D image
or a 3D material to be used as a falloff map for the material’s specular highlights. When the
input is a 2D image, the Alpha channel is used to create the map, while the color channels are
discarded.
– Bump Map Material: The white Bump Map texture input accepts only a 3D material. Typically,
you connect the texture into a Bump Map node, and then connect the Bump Map node to this
input. This input uses the RGB information as texture-space normals.

When nodes have as many inputs as this one does, it is often difficult to make connections with any
precision. Hold down the Option or Alt key while dragging the output from another node over the
node tile, and keep holding Option or Alt when releasing the left mouse button. A small drop-down
menu listing all the inputs provided by the node appears. Click on the desired input to
complete the connection.

Basic Node Setup

The output of a Phong node is connected to the material input on a 3D scene or 3D geometry node.
The Phong node below is taking in a base Color Diffuse input from the Fast Noise node and a bump
map texture also generated from a Fast Noise node.

A Phong node with a diffuse color and Bump Map input

Chapter 31 3D Material Nodes 775

Inspector

Phong controls

Controls Tab
The Controls tab contains parameters for adjusting the main color, highlight, and lighting properties of
the Phong shader node.

Diffuse
Diffuse describes the base surface characteristics without any additional effects like reflections or
specular highlights. Besides defining the base color of an object, the diffuse color also defines the
transparency of the object.
The Alpha in a diffuse texture map can be used to make portions of the surface transparent.

Diffuse Color
A material’s Diffuse Color describes the base color presented by the material when it is lit indirectly or
by ambient light. If a diffuse texture map is provided, then the color value provided here is multiplied
by the color values in the texture.

Alpha
This slider sets the material’s Alpha channel value. This affects diffuse and specular colors equally and
affects the Alpha value of the material in the rendered output. If a diffuse texture map is provided, then
the Alpha value set here is multiplied by the Alpha values in the texture map.

Opacity
Reducing the material’s Opacity decreases the color and Alpha values of the specular and diffuse
colors equally, making the material transparent.

Specular
The parameters in the Specular section describe the look of the specular highlight of the surface.
These values are evaluated in a different way for each illumination model.

Chapter 31 3D Material Nodes 776

Specular Color
Specular Color determines the color of light that reflects from a shiny surface. The more specular a
material is, the glossier it appears. Surfaces like plastics and glass tend to have white specular
highlights, whereas metallic surfaces like gold have specular highlights that inherit their color from the
material color. If a specular texture map is provided, then the value provided here is multiplied by the
color values from the texture.

Specular Intensity
Specular Intensity controls how strong the specular highlight is. If the specular intensity texture is
provided, then this value is multiplied by the Alpha value of the texture.

Specular Exponent
Specular Exponent controls the falloff of the specular highlight. The greater the value, the sharper the
falloff, and the smoother and glossier the material appears. If the specular exponent texture is
provided, then this value is multiplied by the Alpha value of the texture map.

Transmittance
Transmittance controls the way light passes through a material. For example, a solid blue sphere casts
a black shadow, but one made of translucent blue plastic would cast a much lower density
blue shadow.
There is a separate Opacity option. Opacity determines how transparent the actual surface is when it
is rendered. Fusion allows adjusting both opacity and transmittance separately. At first, this might be a
bit counterintuitive to those who are unfamiliar with 3D software. It is possible to have a surface that is
fully opaque but transmits 100% of the light arriving upon it, effectively making it a luminous/
emissive surface.

Attenuation
Attenuation determines how much color is passed through the object. For an object to have
transmissive shadows, set the attenuation to (1, 1, 1), which means 100% of green, blue, and red light
passes through the object. Setting this color to RGB (1, 0, 0) means that the material transmits 100% of
the red arriving at the surface but none of the green or blue light. This can be used to create “stained
glass”-styled shadows.

Alpha Detail
When the Alpha Detail slider is set to 0, the Alpha channel of the object is ignored and the entire
object casts a shadow. If it is set to 1, the Alpha channel determines what portions of the object
cast a shadow.

Color Detail
The Color Detail slider modulates light passing through the surface by the diffuse color + texture
colors. Use this to throw a shadow that contains color details of the texture applied to the object.
Increasing the slider from 0 to 1 brings in more diffuse color + texture color into the shadow. Note that
the Alpha and opacity of the object are ignored when transmitting color, allowing an object with a solid
Alpha to still transmit its color to the shadow.

Saturation
The Saturation slider controls the saturation of the color component transmitted to the shadow. Setting
this to 0.0 results in monochrome shadows.

Receives Lighting/Shadows
These checkboxes control whether the material is affected by lighting and shadows in the scene. If
turned off, the object is always fully lit and/or unshadowed.

Chapter 31 3D Material Nodes 777

Two-Sided Lighting
This effectively makes the surface two sided by adding a second set of normals facing the opposite
direction on the backside of the surface. This is normally off to increase rendering speed, but it can be
turned on for 2D surfaces or for objects that are not fully enclosed, to allow the reverse or interior
surfaces to be visible as well.
Normally, in a 3D application, only the front face of a surface is visible and the back face is culled, so
that if a camera were to revolve around a plane in a 3D application, when it reached the backside, the
plane would become invisible. Making a plane two sided in a 3D application is equivalent to adding
another plane on top of the first but rotated by 180 degrees so the normals are facing the opposite
direction on the backside. Thus, when you revolve around the back, you see the second image plane,
which has its normals facing the opposite way.
Fusion does exactly the same thing as 3D applications when you make a surface two sided. The
confusion about what two-sided lighting does arises because Fusion does not cull back-facing
polygons by default. If you revolve around a one-sided plane in Fusion, you still see it from the
backside (but you are seeing the frontside duplicated through to the backside as if it were transparent).
Making the plane two sided effectively adds a second set of normals to the backside of the plane.

NOTE: This can become rather confusing once you make the surface transparent, as the
same rules still apply and produce a result that is counterintuitive. If you view from the
frontside a transparent two-sided surface illuminated from the backside, it looks unlit.

Material ID
This slider sets the numeric identifier assigned to this material. This value is rendered into the MatID
auxiliary channel if the corresponding option is enabled in the renderer.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Reflect [3RR]

The Reflect node

Reflect Node Introduction

The Reflect node is used to add environment map reflections and refractions to materials.
Control is offered over the face on and glancing strength, falloff, per channel refraction indexes, and
tinting. Several texture map inputs can modify the behavior of each parameter.
Environment mapping is an approximation that assumes an object’s environment is infinitely distant
from the object. It’s best to picture this as a cube or sphere with the object at the center. Specifically,
this infinite distance assumption means that objects cannot interact with themselves (e.g., the
reflections on the handle of a teapot do not show the body of the teapot but rather the infinite

Chapter 31 3D Material Nodes 778

environment map). It also means that if you use the same cube map on multiple objects in the scene,
those objects do not inter-reflect each other (e.g., two neighboring objects would not reflect each
other). If you want objects to reflect each other, you need to render a cube map for each object.
For more information, see Chapter 25, “3D Compositing Basics” in the Fusion Reference Manual or
Chapter 86 in the DaVinci Resolve Reference Manual.

Inputs
There are five inputs on the Reflect node that accept 2D images or 3D materials. These inputs control
the overall color and image used for the 3D object as well as controlling the color and texture used in
the reflective highlights.
– Background Material: The orange Background material input accepts a 2D image or a 3D
material. If a 2D image is provided, the node treats it as a diffuse texture map applied to a basic
material.
– Reflection Color Material: The white Reflection Color material input accepts a 2D image or a
3D material. The RGB channels are used as the reflection texture, and the Alpha is ignored.
– Reflection Intensity Material: The white Reflection Intensity material input accepts a 2D image
or a 3D material. The Alpha channel of the texture is multiplied by the intensity of the reflection.
– Refraction Tint Material: The white Refraction Tint material input accepts a 2D image or a 3D
material. The RGB channels are used as the refraction texture.
– Bump Map Texture: The white Bump Map texture input accepts only a 3D material. Typically,
you connect the texture into a Bump Map node, and then connect the Bump Map node to this
input. This input uses the RGB information as texture-space normals.

When nodes have as many inputs and some using the same color as this one does, it is often difficult
to make connections with any precision. Hold down the Option or Alt key while dragging the output
from another node over the node tile, and keep holding Option or Alt when releasing the left mouse
button. A small drop-down menu listing all the inputs provided by the node appears. Click on the
desired input to complete the connection.

Basic Node Setup

The Reflection node can be the main shader for an object as it is in the example below, or it can be
used to feed the diffuse material input of a Ward, Blinn, Phong, or other material node. Usually, a
Sphere Map node is used as the source of the Reflect node’s reflection color input.

A Reflect node used to create a highly reflective surface of a Shape 3D node

Chapter 31 3D Material Nodes 779

Inspector

Reflect controls

Controls Tab
The Controls tab contains parameters for adjusting the reflective strength based on the orientation of
the object, as well as the tint color of the Reflect shader node.

Reflection
Reflection Strength Variability
This multi-button control can be set to Constant or By Angle for varying the reflection intensity,
corresponding to the relative surface orientation to the viewer. The following three controls are visible
only when this control is set to By Angle.

Glancing Strength
[By Angle] Glancing Strength controls the intensity of the reflection for those areas of the geometry
where the reflection faces away from the camera.

Face On Strength
[By Angle] Face On Strength controls the intensity of the reflection for those parts of the geometry that
reflect directly back to the camera.

Falloff
[By Angle] Falloff controls the sharpness of the transition between the Glancing and Face On Strength
regions. It can be considered similar to applying gamma correction to a gradient between the Face On
and Glancing values.

Constant Strength
[Constant Angle] This control is visible only when the reflection strength variability is set to Constant. In
this case, the intensity of the reflection is constant despite the incidence angle of the reflection.

Chapter 31 3D Material Nodes 780

Refraction
If the incoming background material has a lower opacity than 1, then it is possible to use an
environment map as refraction texture, and it is possible to simulate refraction effects in
transparent objects.

Separate RGB Refraction Indices

When this checkbox is enabled, the Refraction Index slider is hidden, and three sliders for adjusting
the refraction index of the Red, Green, and Blue channels appear in its place. This allows simulation of
the spectral refraction effects commonly seen in thick imperfect glass, for example.

Refraction Index
This slider controls how strongly the environment map is deformed when viewed through a surface.
The overall deformation is based on the incidence angle. Since this is an approximation and not a
simulation, the results are not intended to model real refractions accurately.

Refraction Tint
The refraction texture is multiplied by the tint color for simulating color-filtered refractions. It can be
used to simulate the type of coloring found in tinted glass, as seen in many brands of beer bottles,
for example.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Stereo Mix [3SMM]

The Stereo Mix node

Stereo Mix Node Overview

This node is used to swap the left and right material inputs. It is often used to output to the left and
right eye of the 3D Render.

Inputs
This node has two inputs that are both required for this node to work. Both inputs accept either a 2D
image or a 3D material.
– LeftMaterial: The orange left material input accepts a 2D image or a 3D material to be used as
the material for the left eye rendering. If a 2D image is used, it is converted to a diffuse texture
map using the basic material type.
– RightMaterial: The green right material input accepts a 2D image or a 3D material to be used as
the material for the right eye rendering. If a 2D image is used, it is converted to a diffuse texture
map using the basic material type.

While the inputs can be either 2D images or 3D materials, the output is always a material.

Chapter 31 3D Material Nodes 781

Basic Node Setup
The Stereo Mix node can be used with either stereo images or materials. The example below shows
two images combined in the Stereo Mix node causing the output to be a stereo anaglyph material.

A Stereo Mix node used to combine left and right images into a single stereo material

Inspector

Stereo Mix controls

Controls Tab
The Controls tab contains a single switch that swaps the left and right material inputs.

Swap
This option swaps both inputs of the node.

Material ID
This slider sets the numeric identifier assigned to this material. This value is rendered into the MatID
auxiliary channel if the corresponding option is enabled in the renderer.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 31 3D Material Nodes 782

Ward [3WD]

The Ward node

Ward Node Introduction

The Ward node is a basic illumination material that can be applied to geometry in the 3D scene.
It describes how the object responds to light and provides multiple texture map inputs to allow fine
control over the diffuse, specular, and bump map components of the material.
Specifically, the Ward node is ideal for simulating brushed metal surfaces, as the highlight can be
elongated along the U or V directions of the mapping coordinates. This is known as an
anisotropic highlight.
The Ward node outputs a 3D Material that can be connected to the material inputs on any 3D
geometry node.

Inputs
There are six inputs on the Ward node that accept 2D images or 3D materials. These inputs control
the overall color and image used for the 3D object as well as controlling the color and texture used in
the specular highlight. Each of these inputs multiplies the pixels in the texture map by the equivalently
named parameters in the node itself. This provides an effective method for scaling parts of
the material.
– Diffuse Material: The orange Diffuse material input accepts a 2D image or a 3D material to be
used as a main color and texture of the object.
– Specular Color Material: The green Specular Color material input accepts a 2D image or a
3D material to be used as a highlight color and texture of the object.
– Specular Intensity Material: The magenta Specular Intensity material input accepts a 2D image
or a 3D material to be used as an intensity map for the material’s highlights. When the input is a
2D image, the Alpha channel is used to create the map, while the color channels are discarded.
– Spread U Material: The white Spread U material input accepts a 2D image or a 3D material.
The value of the Spread U option in the node’s controls is multiplied against the pixel values in
the material’s Alpha channel.
– Spread V Material: The white Spread V material input accepts a 2D image or a 3D material.
The value of the Spread V option in the node’s controls is multiplied against the pixel values in
the material’s Alpha channel.
– Bump Map Material: The white Bump Map material input accepts only a 3D material. Typically,
you connect the texture into a Bump Map node, and then connect the Bump Map node to this
input. This input uses the RGB information as texture-space normals.

When nodes have as many inputs and some using the same color as this one does, it is often difficult
to make connections with any precision. Hold down the Option or Alt key while dragging the output
from another node over the node tile, and keep holding Option or Alt when releasing the left mouse
button. A small drop-down menu listing all the inputs provided by the node appears. Click on the
desired input to complete the connection.

Chapter 31 3D Material Nodes 783

Basic Node Setup
The Ward node is used to make a shiny glass surface and replace the 3D text material in the example
below. A diffuse color material comes from Reflect node, and the specular color is altered by a
gradient color Fast Noise node.

A Ward node used with a diffuse connection and specular color connection

Inspector

Ward controls

Controls Tab
The Controls tab contains parameters for adjusting the main color, highlight, and lighting properties of
the Ward shader node.

Diffuse
Diffuse describes the base surface characteristics without any additional effects like reflections or
specular highlights. Besides defining the base color of an object, the diffuse color also defines the
transparency of the object. The Alpha in a diffuse texture map can be used to make portions of the
surface transparent.

Chapter 31 3D Material Nodes 784

Diffuse Color
A material’s Diffuse Color describes the base color presented by the material when it is lit indirectly or
by ambient light. If a diffuse texture map is provided, then the color value provided here is multiplied
by the color values in the texture.

Alpha
This slider sets the material’s Alpha channel value. This affects diffuse and specular colors equally and
affects the Alpha value of the material in the rendered output. If a diffuse texture map is provided, then
the Alpha value set here is multiplied by the Alpha values in the texture map.

Opacity
Reducing the material’s Opacity decreases the color and Alpha values of the specular and diffuse
colors equally, making the material transparent.

Specular
The parameters in the Specular section describe the look of the specular highlight of the surface.
These values are evaluated in a different way for each illumination model.

Specular Color
Specular Color determines the color of light that reflects from a shiny surface. The more specular a
material is, the glossier it appears. Surfaces like plastics and glass tend to have white specular
highlights, whereas metallic surfaces like gold have specular highlights that inherit their color from the
material color. If a specular texture map is provided, then the value provided here is multiplied by the
color values from the texture.

Specular Intensity
Specular Intensity controls how strong the specular highlight is. If the specular intensity texture is
provided, then this value is multiplied by the Alpha value of the texture.

Spread U
Spread U controls the falloff of the specular highlight along the U-axis in the UV map of the object. The
smaller the value, the sharper the falloff, and the smoother and glossier the material appears in this
direction. If the Spread U texture is provided, then this value is multiplied by the Alpha value of
the texture.

Spread V
Spread V controls the falloff of the specular highlight along the V-axis in the UV map of the object. The
smaller the value, the sharper the falloff, and the smoother and glossier the material appear in this
direction. If the Spread V texture is provided, then this value is multiplied by the Alpha value of
the texture.

Transmittance
Transmittance controls the way light passes through a material. For example, a solid blue sphere casts
a black shadow, but one made of translucent blue plastic would cast a much lower density
blue shadow.
There is a separate Opacity option. Opacity determines how transparent the actual surface is when it
is rendered. Fusion allows adjusting both opacity and transmittance separately. At first, this might be a
bit counterintuitive to those who are unfamiliar with 3D software. It is possible to have a surface that is
fully opaque but transmits 100% of the light arriving upon it, effectively making it a luminous/
emissive surface.

Attenuation
Attenuation determines how much color is passed through the object. For an object to have
transmissive shadows, set the attenuation to (1, 1, 1), which means 100% of green, blue, and red light
passes through the object. Setting this color to RGB (1, 0, 0) means that the material transmits 100% of

Chapter 31 3D Material Nodes 785

the red arriving at the surface but none of the green or blue light. This can be used to create “stained
glass”-styled shadows.

Alpha Detail
When the Alpha Detail slider is set to 0, the Alpha channel of the object is ignored, and the entire
object casts a shadow. If it is set to 1, the Alpha channel determines what portions of the object
cast a shadow.

Color Detail
The Color Detail slider modulates light passing through the surface by the diffuse color + texture
colors. Use this to throw a shadow that contains color details of the texture applied to the object.
Increasing the slider from 0 to 1 brings in more diffuse color + texture color into the shadow. Note that
the Alpha and opacity of the object are ignored when transmitting color, allowing an object with a solid
Alpha to still transmit its color to the shadow.

Saturation
The Saturation slider controls the saturation of the color component transmitted to the shadow. Setting
this to 0.0 results in monochrome shadows.

Receives Lighting/Shadows
These checkboxes control whether the material is affected by lighting and shadows in the scene. If
turned off, the object is always fully lit and/or unshadowed.

Two-Sided Lighting
This effectively makes the surface two sided by adding a second set of normals facing the opposite
direction on the backside of the surface. This is normally off to increase rendering speed, but it can be
turned on for 2D surfaces or for objects that are not fully enclosed, to allow the reverse or interior
surfaces to be visible as well.
Normally, in a 3D application, only the front face of a surface is visible and the back face is culled, so
that if a camera were to revolve around a plane in a 3D application, when it reached the backside, the
plane would become invisible. Making a plane two sided in a 3D application is equivalent to adding
another plane on top of the first but rotated by 180 degrees so the normals are facing the opposite
direction on the backside. Thus, when you revolve around the back, you see the second image plane,
which has its normals facing the opposite way.
Fusion does exactly the same thing as 3D applications when you make a surface two sided. The
confusion about what two-sided lighting does arises because Fusion does not cull back-facing
polygons by default. If you revolve around a one-sided plane in Fusion you still see it from the
backside (but you are seeing the frontside duplicated through to the backside as if it were transparent).
Making the plane two sided effectively adds a second set of normals to the backside of the plane.

NOTE: This can become rather confusing once you make the surface transparent, as the
same rules still apply and produce a result that is counterintuitive. If you view from the
frontside a transparent two-sided surface illuminated from the backside, it looks unlit.

Material ID
This slider sets the numeric identifier assigned to this material. This value is rendered into the MatID
auxiliary channel if the corresponding option is enabled in the renderer.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail in the following “The Common Controls” section.

Chapter 31 3D Material Nodes 786

The Common Controls
Nodes that handle 3D geometry share a number of identical controls in the Inspector. This section
describes controls that are common among 3D Material nodes.

Settings Tab

Common Settings 3D controls

Common Settings tab can be found on most tools in Fusion The following controls are specific settings
for 3D nodes

Hide Incoming Connections

Enabling this checkbox can hide connection lines from incoming nodes, making a node tree appear
cleaner and easier to read. When enabled, fields for each input on a node are displayed. Dragging a
connected node from the node tree into the field hide that incoming connection line as long as the
node is not selected in the node tree. When the node is selected in the node tree, the line reappears.

Comment Tab
The Comment tab contains a single text control that is used to add comments and notes to the tool.
When a note is added to a tool, a small red dot icon appears next to the setting’s tab icon, and a text
bubble appears on the node. To see the note in the Node Editor, hold the mouse pointer over the
node for a moment. The contents of the Comments tab can be animated over time, if required.

Scripting Tab
The Scripting tab is present on every tool in Fusion. It contains several edit boxes used to add scripts
that process when the tool is rendering. For more details on the contents of this tab, please consult the
scripting documentation.

Chapter 31 3D Material Nodes 787

Chapter 32

3D Texture Nodes
This chapter details the 3D Texture nodes available when creating 3D composites in
Fusion. The abbreviations next to each node name can be used in the Select Tool
dialog when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Bump Map [3Bu]  789
Catcher [3CA]  792
CubeMap [3CU]  794
Falloff [3FA]  797
Fast Noise Texture [3FN]  799
Gradient 3D [3GD]  802
Sphere Map [3SPM]  805
Texture 2D [3Tx]  807
Texture Transform [3TT]  809
The Common Controls  811

Chapter 32 3D Texture Nodes 788

Bump Map [3Bu]

The Bump Map node

Bump Map Node Overview

The Bump Map node is used to convert a grayscale (height map) image into a bump map and takes an
input directly from a bump map created by the Create Bump Map node. The node outputs a material.

Inputs
The Bump Map node includes a single orange input for connecting a 2D image you want to use as the
bump map texture, or it can accept the output of the Create Bump Map node.
– ImageInput: The orange Image input is used to connect a 2D RGBA image for the bump
calculation or an existing bump map from the Create Bump map node.

Basic Node Setup

The Bump Map node is connected to the Bump Map material input on any one of the material shader
nodes. Below, the example uses a Fast Noise node to generate an image that connects to the Bump
Map node. The output of the Bump Map node connects to the Bump Map material input on a
Ward node.

A Bump Map is connected to the Bump Map material input on a material node.

Chapter 32 3D Texture Nodes 789

Inspector

Bump Map controls

Controls Tab
The Controls tab contains all parameters for modifying the input source and the appearance of
the bump map.

Source Image Type

Toggle between Height Map, which creates a bump map similar to the Create Bump Map node, and
Bump Map, which expects a bump map created by the Create Bump Map node.

Filter Size
A custom filter generates the bump information. The drop-down menu sets the filter size.

Height Channel
Sets the channel from which to extract the grayscale information.

Clamp Z Normal
Clips the lower values of the blue channel in the resulting bump texture.

Height Scale
Changes the contrast of the resulting values in the bump map. Increasing this value yields a more
visible bump map.

Texture Depth
Optionally converts the resulting bump map texture into the desired bit depth.

Wrap Mode
Wraps the image at the borders, so the filter produces correct result when using seamless tile textures.

Chapter 32 3D Texture Nodes 790

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Notes on Bump Maps

There is some confusion of terminology with bump mapping, depending on where you get your
information. Here are Fusion conventions:

Height Map

A grayscale image containing a height value per pixel

Bump Map

An image containing normals stored in the RGB channels used for

modifying the existing normals (usually given in tangent space)

Normals Map

An image containing normals stored in the RGB channels used for

replacing the existing normals (usually given in tangent or object space)

Chapter 32 3D Texture Nodes 791

Catcher [3CA]

The Catcher node

Catcher Node Overview

The Catcher material is used to “catch” texture-mode projections cast from Projector 3D and Camera
3D nodes. The intercepted projections are converted into a texture map and applied by the Catcher
material to the geometry to which it is connected.
To understand the Catcher node, it helps to understand the difference between light-based
projections and texture-based projections. Choosing Light from the projection mode menu on the
Projector 3D or Camera 3D nodes simply adds the values of the RGB channels in the projected image
to the diffuse texture of any geometry that lies within the projection cone. This makes it impossible to
clip away geometry based on the Alpha channel of an image when using light mode projections.
Imagine a scenario where you want to project an image of a building onto an image plane as part of a
set extension shot. You first rotoscope the image to mask out the windows. This makes it possible to
see the geometry of the rooms behind the wall in the final composite. When this image is projected as
light, the Alpha channel is ignored, so the masked windows remain opaque.
By connecting the Catcher to the diffuse texture map of the material applied to the image plane, and
then switching the projection mode menu in the Projector 3D or Camera 3D node from Light or
Ambient Light mode to Texture mode, the projected image is applied as a texture map. When using
this technique for the example above, the windows would become transparent, and it would be
possible to see the geometry behind the window.
The main advantages of this approach over light projection are that the Catcher can be used to project
Alpha onto an object, and it doesn‘t require lighting to be enabled. Another advantage is that the
Catcher is not restricted to the diffuse input of a material, making it possible to project specular
intensity maps, or even reflection and refraction maps.

NOTE: The Catcher material requires a Projector 3D or Camera 3D node in the scene, set to
project an image in Texture mode on the object to which the Catcher is connected. Without a
projection, or if the projection is not set to Texture mode, the Catcher simply makes the
object transparent and invisible.

Inputs
The Catcher node has no inputs. The output of the node is connected to the the diffuse color material
input of the Blinn, Cook Torrance, or other material node applied to the 3D geometry.

Chapter 32 3D Texture Nodes 792

Basic Node Setup
The output of a Catcher node should be connected to the material input of your 3D geometry node.
A camera is set up as a proctor with an image connected to the camera’s image input. When the
camera is set to texture projection mode, the Catcher node is used to determine which geometry
receives the texture.

A Catcher node output is connected to the input of the

geometry node that receives the texture projection

Inspector

Catcher controls

Controls Tab
The Options in the Controls tab determine how the Catcher handles the accumulation of multiple
projections.

Enable
Use this checkbox to enable or disable the node. This is not the same as the red switch in the upper-
left corner of the Inspector. The red switch disables the tool altogether and passes the image on
without any modification. The Enable checkbox is limited to the effect part of the tool. Other parts, like
scripts in the Settings tab, still process as normal.

Color Mode
The Color mode menu is used to control how the Catcher combines the light from multiple projectors.
It has no effect on the results when only one projector is in the scene. This control is designed to work
with the software renderer in the Renderer 3D node and has no effect when using the
OpenGL renderer.

Chapter 32 3D Texture Nodes 793

Alpha Mode
The Alpha mode is used to control how the Catcher combines the Alpha channels from multiple
projectors. It has no effect on the results when only one projector is in the scene. This control is
designed to work with the software renderer in the Renderer 3D node and has no effect when using
the OpenGL renderer.

Threshold
The Threshold can be used to exclude certain low values from the accumulation calculation.
For example, when using the Median Accumulation mode, a threshold of 0.01 would exclude any pixel
with a value of less than 0.01 from the median calculation.

Restrict by Projector ID
When active, the Catcher only receives light from projectors with a matching ID. Projectors with a
different ID are ignored.

Material ID
This slider sets the numeric identifier assigned to this material. This value is rendered into the
MatID auxiliary channel if the corresponding option is enabled in the Renderer 3D node.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

CubeMap [3CU]

The Cube Map node

Cube Map Node Overview

The Cube Map node creates texture maps using separate images for each face of the cube. It can also
extract the individual faces of the cube from a single image containing an unfolded cube in the Vertical
or Horizontal Cross layouts.
A cube map is produced by mounting six cameras at 90 degrees angle of views to point up, down, left,
right, front, and back.
The node provides options to set the reference coordinate system and rotation for the resulting
texture map. The Cube Map node is typically used to produce environment maps for distant areas
(such as skies or horizons) or reflection and refraction maps.

Chapter 32 3D Texture Nodes 794

Sample cube map

Inputs
The Inputs on this node change based on the settings of the Layout menu in the Inspector. The single
input uses a 2D image for the entire cube, while six inputs can handle a different 2D image for each
side of a cube.
– CrossImage: The orange Cross Image input is visible by default or when the Layout menu in
the Inspector is set to either Vertical Cross or Horizontal Cross. The input accepts a 2D image.
– CubeMap.[DIRECTION]: These six multi-colored inputs are visible only when the Layout menu
in the Inspector is set to Separate Images. Each input accepts an image aligned to match the
left, right, top, bottom, front, and back faces.

Basic Node Setup

The Cube Map node uses a vertical or horizontal cross image represented by MediaIn2 node
connected into the orange cross image input. The Cube Map node is used similarly to the Sphere Map
node. It creates an environment that surrounds the geometry connected to a Shader node.

A Cube Map node receives a cross image input, creating an environment for the Shape 3D

Chapter 32 3D Texture Nodes 795

Inspector

Cube Map controls

Controls Tab
Layout
The Layout menu determines the type and number of inputs for the cube map texture.
Valid options are:
– Separate Images: This option exposes six inputs on the node, one for each face of the cube. If the
separate images are not square or not of the same size, they are rescaled into the largest 1:1 image
that can contain all of them.
– Vertical Cross: This option exposes a single input on the node. The image should be an
unwrapped texture of a cube containing all the faces organized into a Vertical Cross formation,
where the height is larger than the width. If the image aspect of the cross image is not 3:4, the
CubeMap node crops it down so it matches the applicable aspect ratio.
– Horizontal Cross: This option exposes a single input on the node. The image should be an
unwrapped texture of a cube containing all the faces organized into a Horizontal Cross formation,
where the width is larger than the height. If the image aspect of the cross image is not 4:3, the
CubeMap node crops it down so that matches the applicable aspect ratio.

Coordinate System
The coordinate system menu sets the position values used when converting the image into a texture.
– Model: This option orients the texture along the object local coordinate system.
– World: This option orients the resulting texture using the global or world coordinate system.
– Eye: This option aligns the texture map to the coordinate system of the camera or viewer.

Rotation
The rotation controls are divided into buttons that select the order of rotation along each axis of the
texture. For example, XYZ would apply the rotation to the X axis first, followed by the Y axis, and finally
the Z axis. The other half of the rotation controls are dials that rotate the texture around its pivot point.

Chapter 32 3D Texture Nodes 796

Warn About Bad Dimensions
Selecting this checkbox displays a warning message on the console if the dimensions of the image
provided did not meet the requirements of the selected orientation mode.

Material ID
This slider sets the numeric identifier assigned to this material. This value is rendered into the MatID
auxiliary channel if the corresponding option is enabled in the renderer.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Falloff [3FA]

The Falloff node

Falloff Node Overview

The Falloff node blends two materials or textures together based on the incidence angle between the
object to which the material is applied and the camera. This is useful when you wish to use one
material for portions of the geometry that would reflect light directly back to the camera and a different
material for parts that reflect light back into the scene.

Falloff example

Chapter 32 3D Texture Nodes 797

Inputs
The two Inputs on the Falloff node are used to connect two images or materials. One is used to reflect
back at the camera, while the other reflects away from the camera and into the scene.
– Face On Material: The orange Face On material input accepts a 2D image or a 3D material. If
a 2D image is provided, it is turned into a diffuse texture map using the basic material shader.
This input is used for the material that is reflecting directly back to the camera
– Glancing Material: The green Glancing material input accepts a 2D image or a 3D material. If
a 2D image is provided, it is turned into a diffuse texture map using the basic material shader.
This input is used for the material that is reflecting away from the camera and into the scene.

While the inputs for this node can be images, the output is always a material.

Basic Node Setup

The Falloff node below is used to control the strength of the Blinn material and the Reflect material.
You connect the Face On input of the Falloff node to the material you want shown for the sides of the
object that face the camera and connect the Glance input to the material you want shown for the sides
not directly facing the camera.

The Falloff node uses one input for the material facing the camera and one for the material not directly facing the camera.

Inspector

Falloff controls

Chapter 32 3D Texture Nodes 798

Controls Tab
The parameters in the Controls tabs modify the tint and opacity of the Face On material and the
Glancing material. A Falloff slider controls the blending between the two.

Color Variation
– Two Tone: Two regular Color controls define the colors for Glancing and Face On.
– Gradient: A Gradient control defines the colors for Glancing and Face On. This can be
used for a multitude of effects, like creating Toon Shaders, for example.

Face On Color
The Face On Color defines the color of surface parts facing the camera. If the Face On texture map is
provided, then the color value provided here is multiplied by the color values in the texture.
Reducing the material’s opacity decreases the color and Alpha values of the Face On material, making
the material transparent.

Glancing Color
The Glancing Color defines the color of surface parts more perpendicular to the camera. If the
Glancing material port has a valid input, then this input is multiplied by this color.
Reducing the material’s opacity decreases the color and Alpha values of the Glancing material, making
the material transparent.

Falloff
This value controls the transition between Glancing and Face On strength. It is very similar to a gamma
operation applied to a gradient, blending one value into another.

Material ID
This slider sets the numeric identifier assigned to this material. This value is rendered into the MatID
auxiliary channel if the corresponding option is enabled in the renderer.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Fast Noise Texture [3FN]

The Fast Noise Texture node

Fast Noise Texture Node Overview

The Fast Noise Texture node is the procedural resolution-independent version of the 2D Fast Noise
node. It creates a noise texture directly as a material for usage with 3D nodes. It offers a 3D volumetric
mode for creating seamless textures in conjunction with nodes providing UVW texture coordinates
(Similar to the UV Map 3D node set to XYZ-to-UVW or Camera).

Chapter 32 3D Texture Nodes 799

Inputs
The Fast Noise Texture node includes an optional input that can be used to connect a 2D image
or material.
– SourceMaterial: The Source Materials input accepts a 2D image or a 3D material. The image is
then altered by the noise pattern.

Basic Node Setup

The Fast Noise Texture node below is used to generate a resolution-independent 3D texture for an
FBX imported model.

A Fast Noise Texture node generates a seamless texture, taking advantage of UVW coordinates.

Inspector

Fast Noise Texture controls

Controls Tab
The parameters of the Fast Noise Texture node control the appearance and, for 2D, the animation of
the noise.

Output Mode
– 2D: Calculates the noise texture based on 2D texture coordinates (UV). This setting allows
smoothly varying the noise pattern with animation.
– 3D: Calculates the noise texture based on 3D texture coordinates (UVW). Nodes like Shape 3D
automatically provide a third texture coordinate; otherwise, a 3D texture space can be created
using the UV Map node. The 3D setting does not support animation of the noise pattern.

Chapter 32 3D Texture Nodes 800

Detail
Increase the value of this slider to produce a greater level of detail in the noise result. Larger values
add more layers of increasingly detailed noise without affecting the overall pattern. High values take
longer to render but can produce a more natural result (not all graphics cards support higher detail
levels in hardware).

Brightness
This control adjusts the overall Brightness of the noise map.

Contrast
This control increases or decreases the overall Contrast of the noise map. It can exaggerate the effect
of the noise.

Scale
The scale of the noise map can be adjusted using the Scale slider, changing it from gentle variations
over the entire image to a tighter overall texture effect. This value represents the scale along
the UV axis.

Scale Z
(3D only) The Scale Z value scales the noise texture along the W-axis in texture space. W represents a
direction perpendicular to the UV plane for a 3D texture map.

Seethe
(2D only) The Seethe control smoothly varies the 2D noise pattern.

Seethe Rate
(2D only) As with the Seethe control above, the Seethe Rate also causes the noise map to evolve and
change. The Seethe Rate defines the rate at which the noise changes each frame, causing an
animated drift in the noise automatically, without the need for spline animation.

Discontinuous
Normally, the noise function interpolates between values to create a smooth continuous gradient of
results. You can enable the Discontinuous checkbox to create hard discontinuity lines along some of
the noise contours. The result is a dramatically different effect.

Invert
Enable the Invert checkbox to invert the noise, creating a negative image of the original pattern. This is
most effective when Discontinuous is also enabled.

Material ID
This slider sets the numeric identifier assigned to this material. This value is rendered into the MatID
auxiliary channel if the corresponding option is enabled in the renderer.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 32 3D Texture Nodes 801

Gradient 3D [3GD]

The Gradient node

Gradient Node Overview

The Gradient 3D node is used to texture objects with a variety of gradient types. It offers many of the
same controls as the Background node. While it is not possible to transform the gradient directly in
3D space, it is orientable using the following nodes:
– Texture Transform Node: The Texture Transform node can be used to adjust the
mapping per pixel.
– UV Map Node: The UV Map node can be used to adjust the mapping per vertex (use the
XYZtoUVW mode). This has onscreen controls, so you can see what the gradient is doing.
Using this node is recommended because it is faster to evaluate.

The gradient defaults to a linear gradient that goes from -1 to +1 along the Z-axis. All primitives in the
Shape 3D node can output a third texture coordinate for UVW mapping.

Inputs
The Gradient node has no Inputs. The output of the node is connected to a material input on
3D geometry.

Basic Node Setup

The Gradient 3D node below is used to generate a resolution-independent 3D texture for an FBX
imported model. Positioning in UVW space is easiest to do using a UV Map tool placed after
the geometry.

A Gradient 3D node generates a resolution-independent gradient texture positioned by the UV Map tool

Chapter 32 3D Texture Nodes 802

Inspector

Gradient 3D controls

Controls Tab
The Controls tab for the Gradient node control the pattern and colors used for the gradient texture.

Gradient Type
Determines the type or pattern used for the gradient.
– Linear: A simple linear gradient.
– Reflect: Based on the Linear mode, this gradient is mirrored at the middle of the textured range.
– Square: The gradient is applied using a square pattern.
– Cross: Similar to the Reflect mode, but Cross uses two axes to apply the gradient.
– Radial: The Radial mode uses a circular pattern to apply the gradient.

Gradient 3D modes

Gradient Bar
The Gradient control consists of a bar where it is possible to add, modify, and remove color stops of
the gradient. Each triangular color stop on the Gradient bar represents a color in the gradient. It is
possible to animate the color as well as the position of the point. Furthermore, a From Image modifier
can be applied to the gradient to evaluate it from an image.

Chapter 32 3D Texture Nodes 803

Interpolation Space
The gradient is linearly interpolated from point to point in RGB color space by default. This can
sometimes lead to unwanted colors. Choosing another color space may provide a better result.

Scale
Allows sizing of the gradient.

Offset
Allows panning through the gradient.

Repeat
Defines how the left and right borders of the gradient are treated.

Gradients set to Once, Repeat, and Ping

Pong from top to bottom, respectively,
and shifting the gradient to the left

– Once: When using the Gradient Offset control to shift the gradient, the border colors keep their
values. Shifting the default gradient to the left results in a white border on the left, while shifting it
to the right results in a black border on the right.
– Repeat: When using the Gradient Offset control to shift the gradient, the border colors wrap
around. Shifting the default gradient to the left results in a sharp jump from white to black, while
shifting it to the right results in a sharp jump from black to white.
– Ping Pong: When using the Gradient Offset control to shift the gradient, the border colors ping-
pong back and forth. Shifting the default gradient to the left results in the edge fading from white
back to black, while shifting it to the right results in the edge fading from black back to white.

Sub Pixel
Determines the accuracy with which the gradient is created.

Material ID
This slider sets the numeric identifier assigned to this material. This value is rendered into the MatID
auxiliary channel if the corresponding option is enabled in the renderer.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 32 3D Texture Nodes 804

Sphere Map [3SPM]

The SphereMap node

Sphere Map Node Overview

The Sphere Map node can be used to create simulated environment mapping, also called reflection
mapping. Ray trace rendering a reflective scene can be very time consuming, but sphere map-based
reflection mapping can generate 360-degree reflections faster with little loss of accuracy. For
example, when creating a reflective environment, a sphere map is created, large enough to surround
the 3D object in your scene. The sphere is mapped with the environment you want reflected and
connected to the Reflection Color input on a Reflect node.

Inputs
The single image input on the Sphere Map node accepts a 2D image texture in an equirectangular
format (where the X-axis represents 0–360 degrees longitude, and the Y-axis represents –90 to +90
degrees latitude.)
– ImageInput: The orange Image input accepts a 2D RGBA image. Preferably, this is an
equirectangular image that shows the entire vertical and horizontal angle of view up
to 360 degrees.

Basic Node Setup

The Sphere Map node below is mapped with a spherical image to generate the environment reflected
on the Shape 3D. It is connected to the Reflection Color input on a Reflect node.

A Sphere Map node generates a reflective environment when connected to a Reflect node Reflection Color input.

Chapter 32 3D Texture Nodes 805

Inspector

Sphere Map controls

Controls Tab
The Controls tab in the Inspector modifies the mapping of the image input to the sphere map.

Angular Mapping
Adjusts the texture coordinate mapping so the poles are less squashed and areas in the texture get
mapped to equal areas on the sphere. It turns the mapping of the latitude lines from a hemispherical
fisheye to an angular fisheye. This mapping attempts to preserve area and makes it easier to paint on
or modify a sphere map since the image is not as compressed at the poles.

Rotation
Offers controls to rotate the texture map.

Material ID
This slider sets the numeric identifier assigned to this material. This value is rendered into the MatID
auxiliary channel if the corresponding option is enabled in the renderer.
The node expects an image with an aspect ratio of 2:1. Otherwise, the image is clamped according to
the following rules:
– 2 * width > height: The width is fitted onto the sphere, and the poles display clamped edges.
– 2 * width < height: The height is fitted onto the sphere, and there is clamping about the 0-degree
longitude line.

Common Controls
Settings tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 32 3D Texture Nodes 806

Sphere Map vs. Connecting the Texture to a Sphere Directly
You can connect an equirectangular texture map directly to a sphere instead of piping it through the
Sphere Map node first. This results in a different rendering if you set the start/end angle and latitude to
less than 360°/180°. In the first case, the texture is squashed. When using the Sphere Map node, the
texture is cropped. Compare:

Spherical mapping differences

NOTE: If you pipe the texture directly into the sphere, it is also mirrored horizontally. You can
change this by using a Transform node first.

Texture 2D [3Tx]

The Texture 2D node

Texture Node Overview

The Texture 2D node sets metadata of an image being used for a texture map. By default, an image
will be (0,0) to (1,1) UV, but that can be changed. The Texture node relies on the presence of U and V
Map channels in 3D rendered images. If these channels are not present, this node has no effect.

NOTE: Background pixels may have U and V values of 0.0, which set those pixels to the color
of the texture’s corner pixel. To restrict texturing to specific objects, use an effect mask based
on the Alpha of the object, or its Object or Material ID channel. For more information, see
Chapter 18, “Understanding Image Channels” in the Fusion Reference Manual or Chapter 79
in the DaVinci Resolve Reference Manual.

Chapter 32 3D Texture Nodes 807

Inputs
– Image Input: The orange image input expects a 2D image.

Basic Node Setup

The Texture 2D node below takes a 2D gradient from the Background node and sets the UV metadata
for it. The texture is then applied to the FBX geometry based on that metadata. If you have the option
to use the UV Map tool, it is recommended because it may be faster and has onscreen controls.

A Texture 2D node is used to set the 3D texture metadata for the input image.

Inspector

Texture 2D controls

Controls Tab
The Controls tab of the Inspector includes the following options.

U/V Offset
These sliders can be used to offset the texture along the U and V coordinates.

U/V Scale
These sliders can be used to scale the texture along the U and V coordinates.

Chapter 32 3D Texture Nodes 808

Wrap Mode
If a texture is transformed in the texture space (using the controls below or the UV Map node), then it’s
possible that areas beyond the image borders will be mapped on the object. The Wrap Mode
determines how the image is applied in these areas.
– Wrap: This wraps the edges of the image around the borders of the image.
– Clamp: The color at the edges of the images is used for texturing. This mode is similar to the
Duplicate mode in the Transform node.
– Black: The image is clipped along its edges. A black color with Alpha = 0 is used instead.
– Mirror: The image is mirrored in both X and Y.

Texture Filtering Mode

The texture can be filtered differently depending on whether you are using the Software Renderer or
OpenGL renderer in the Renderer 3D node. Within the two render engines, you can choose between
high-quality anti-aliasing or low quality. The texture filtering mode provides different filtering options
for the two render engines and the two anti-aliasing settings.
– Nearest: The simplest filtering technique is very fast but can cause artifacts when scaling textures.
– Bilinear: A standard isotropic filtering technique for scaling textures into multiple resolutions.
Works well for magnification of textures.
– Trilinear: An extension of Bilinear filtering. Trilinear tends to be a better option when
scaling down textures
– Anisotropic: The highest-quality filtering method that takes the camera orientation and polygon
perspective into account.
– SAT: SAT (Summed Area Table) is a method of performing high-quality filtering, but it can require
more memory than other options. Works very well on smaller bitmaps.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Texture Transform [3TT]

The Texture Transform node

Texture Transform Node Overview

The Texture Transform node can be used to translate, rotate, and scale the UVW texture coordinates
of a 3D object. While the input can also be an image, the output is always a material.

Chapter 32 3D Texture Nodes 809

Inputs
The Texture Transform node includes a single input that is used to connect the image or material you
want to transform.
– Material Input: The orange Material input accepts a 2D image or 3D material whose texture
coordinates are transformed using the controls in the Inspector.

Basic Node Setup

The Texture Transform node below is used to take in a 2D image, transform it, and output a material to
be used on 3D geometry.

A Texture Transform node transforms a texture applied to 3D geometry.

Inspector

Texture Transform controls

NOTE: Not all Wrap modes are supported by all graphics cards.

Chapter 32 3D Texture Nodes 810

Controls Tab
The Controls tab for the Texture Transform node includes many common transform controls that are
used to transform the texture using UVW coordinates.

Translation
The U, V, W translation sliders shift the texture along U, V, and W axes.

Rotation
Rotation Order buttons set the order in which the rotation is applied. In conjunction with the buttons,
the UVW dials define the rotation around the UVW axes.

Scale
U, V, W sliders scale the texture along the UVW axes.

Pivot
U, V, W Pivot sets the reference point for rotation and scaling.

Material ID
This slider sets the numeric identifier assigned to this material. This value is rendered into the MatID
auxiliary channel if the corresponding option is enabled in the renderer.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other 3D nodes. These common controls are
described in the following “The Common Controls” section.

The Common Controls

Nodes that handle 3D geometry share a number of identical controls in the Inspector. This section
describes controls that are common among 3D Texture nodes.

Settings Tab

Common Settings 3D controls

Chapter 32 3D Texture Nodes 811

The Common Settings tab can be found on most tools in Fusion. The following controls are specific
settings for 3D nodes.

Hide Incoming Connections

Enabling this checkbox can hide connection lines from incoming nodes, making a node tree appear
cleaner and easier to read. When enabled, fields for each input on a node are displayed. Dragging a
connected node from the node tree into the field hides that incoming connection line as long as the
node is not selected in the node tree. When the node is selected in the node tree, the line reappears.

Comment Tab
The Comment tab contains a single text control that is used to add comments and notes to the tool.
When a note is added to a tool, a small red dot icon appears next to the setting’s tab icon, and a text
bubble appears on the node. To see the note in the Node Editor, hold the mouse pointer over the
node for a moment. The contents of the Comments tab can be animated over ime, if required.

Scripting Tab
The Scripting tab is present on every tool in Fusion. It contains several edit boxes used to add scripts
that process when the tool is rendering. For more details on the contents of this tab, please consult the
scripting documentation.

Chapter 32 3D Texture Nodes 812

Chapter 33

Blur Nodes
This chapter details the Blur nodes available in Fusion. The abbreviations next to each
node name can be used in the Select Tool dialog when searching for tools and in
scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Blur [Blur]  814
Defocus [DFO]  816
Directional Blur [DRBL]  818
Glow [GLO]  821
Sharpen [SHRP]  824
Soft Glow [SGlo]  826
Unsharp Mask [USM]  829
Vari Blur [VBL]  830
Vector Motion Blur [VBL]  832
The Common Controls  834

Chapter 33 Blur Nodes 813

Blur [Blur]

The Blur node

Blur Node Introduction

The Blur node does exactly what its name implies – it blurs the input image. This is one of the most
commonly used image-processing operations.

Inputs
The two inputs on the Blur node are used to connect a 2D image and an effect mask that can be used
to limit the blurred area.
– Input: The orange input is used for the primary 2D image that is blurred.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the blur to only
those pixels within the mask. An effect mask is applied to the tool after the tool is processed.

Basic Node Setup

The Blur node, like many 2D image-processing nodes, receives a 2D image like the MediaIn1 shown
below. The output continues the node tree by connecting to another 2D image-processing node or a
Merge node.

A Blur node applied to a MediaIn1 node

Inspector

Blur controls

Chapter 33 Blur Nodes 814

 NOTE: Since a perfect Gaussian filter would require examining an infinite number of pixels, all
practical Gaussians are, of necessity, approximations. The algorithm Fusion uses is a highly-
optimized approach that has many strengths, but can create visible ringing around the edges
in certain extreme cases. This ringing appears only when blurring float-depth images and is
normally far below the limits of visibility, especially in final renders or HiQ mode, but may
appear in subsequent processing. If you experience this, selecting the Multi-box filter may be
a good choice.

Controls Tab
The Controls tab contains the primary controls necessary for customizing the blur operation, including
five filter algorithms.

Filter
The Filter menu is where you select the type of filter used to create the blur.
– Box Blur: This option is faster than the Gaussian blur but produces a lower-quality result.
– Bartlett: This option is a more subtle, anti-aliased blur filter.
– Multi-box: Multi-box uses a Box filter layered in multiple passes to approximate a Gaussian shape.
With a moderate number of passes (e.g., four), a high-quality blur can be obtained, often faster
than the Gaussian filter and without any ringing.
– Gaussian: Gaussian applies a smooth, symmetrical blur filter, using a sophisticated constant-time
Gaussian approximation algorithm.
– Fast Gaussian: Gaussian applies a smooth, symmetrical blur filter, using a sophisticated constant-
time Gaussian approximation algorithm. This mode is the default filter method.

Color Channels (RGBA)

The filter defaults to operating on R, G, B, and A channels. Selective channel filtering is possible by
clicking each channel button to make them active or inactive.

NOTE: This is not the same as the RGBA checkboxes found under the common controls. The
node takes these selections into account before it processes the image, so deselecting a
channel causes the node to skip that channel when processing, speeding up the rendering of
the effect. In contrast, the channel controls under the Common Controls tab are applied after
the node has processed.

Lock X/Y
Locks the X and Y Blur sliders together for symmetrical blurring. This is enabled by default.

Blur Size
Sets the amount of blur applied to the image. When the Lock X and Y control is deselected,
independent control over each axis is provided.

Clipping Mode
This option determines how edges are handled when performing domain-of-definition rendering.
This is profoundly important for nodes like Blur, which may require samples from portions of the image
outside the current domain.

Chapter 33 Blur Nodes 815

 – Frame: The default option is Frame, which automatically sets the node’s domain of definition
to use the full frame of the image, effectively ignoring the current domain of definition. If the
upstream DoD is smaller than the frame, the remaining area in the frame is treated as black/
transparent.
– Domain: Setting this option to Domain respects the upstream domain of definition when applying
the node’s effect. This can have adverse clipping effects in situations where the node employs a
large filter.
– None: Setting this option to None does not perform any source image clipping at all. This means
that any data required to process the node’s effect that would normally be outside the upstream
DoD is treated as black/transparent.

Blend
The Blend slider determines the percentage of the affected image that is mixed with original image. It
blends in more of the original image as the value gets closer to 0.
This control is a cloned instance of the Blend slider in the Common Controls tab. Changes made to this
control are simultaneously made to the one in the common controls.

Examples
Following is a comparison of Blur filters visualized as “cross-sections” of a filtered edge. As you can
see, Box creates a linear ramp, while Bartlett creates a somewhat smoother ramp. Multi-box and
Gaussian are indistinguishable unless you zoom in really close on the slopes. They both lead to even
smoother ramps, but as mentioned above, Gaussian overshoots slightly and may lead to negative
values if used on floating-point images.

Blur filters visualized as “cross sections” of a filtered edge

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Blur nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Defocus [DFO]

The Defocus node

Defocus Node Introduction

The Defocus node simulates the effects of an out-of-focus camera lens, including blooming and image
flaring. It provides a fast Gaussian mode, as well as a more realistic but slower Lens mode.

Chapter 33 Blur Nodes 816

Inputs
The two inputs on the Defocus node are for connecting a 2D image and an effect mask that can be
used to limit the simulated defocused area.
– Input: The orange input is used for the primary 2D image for defocusing.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the defocus to
only those pixels within the mask. An effect mask is applied to the tool after it is processed.

Basic Node Setup

The Defocus node receives a 2D image like the MediaIn1 shown below. The output continues the
node tree by connecting to another 2D image-processing node or a Merge node.

A Defocus node applied to a MediaIn1 node

Inspector

Defocus controls

Controls Tab
The Controls tab contains all the primary controls necessary for customizing the defocus operation.

Filter
Use this menu to select the exact method applied to create the defocus. Gaussian applies a simplistic
effect, while Lens mode creates a more realistic defocus. Lens mode takes significantly longer
than Gaussian.

Lock X/Y
When Lock X/Y is selected, this performs the same amount of defocusing to both the X- and Y-axis of
the image. Deselect to obtain individual control.

Defocus Size
The Defocus Size control sets the size of the defocus effect. Higher values blur the image by greater
amounts and produce larger blooms.

Chapter 33 Blur Nodes 817

Bloom Level
The Bloom Level control determines the intensity and size of the blooming applied to pixels that are
above the bloom threshold.

Bloom Threshold
Pixels with values above the set Bloom Threshold are defocused and have a glow applied (blooming).
Pixels below that value are only defocused.
The following four lens options are available only when the Filter is set to Lens.
– Lens Type: The basic shape used to create the “bad bokeh” effect. This can be refined further
with the Angle, Sides, and Shape sliders.
– Lens Angle: Defines the rotation of the shape. Best visible with NGon lens types. Because of the
round nature of a circle, this slider has no visible effect when the Lens Type is set to Circle.
– Lens Sides: Defines how many sides the NGon shapes have. Best visible with NGon lens types.
Because of the round nature of a circle, this slider has no visible effect when the Lens Type is set
to Circle.
– Lens Shape: Defines how pointed the NGons are. Higher values create a more pointed, starry
look. Lower values create smoother NGons. Best visible with NGon lens types and Lens Sides
between 5 and 10. Because of the round nature of a circle, this slider has no visible effect when
the Lens Type is set to Circle.

Clipping Mode
This option determines how edges are handled when performing domain-of-definition rendering. This
is profoundly important for nodes like Blur, which may require samples from portions of the image
outside the current domain.
– Frame: The default option is Frame, which automatically sets the node’s domain of definition
to use the full frame of the image, effectively ignoring the current domain of definition. If the
upstream DoD is smaller than the frame, the remaining area in the frame is treated as black/
transparent.
– Domain: Setting this option to Domain respects the upstream domain of definition when applying
the node’s effect. This can have adverse clipping effects in situations where the node employs a
large filter.
– None: Setting this option to None does not perform any source image clipping at all. This means
that any data required to process the node’s effect that would normally be outside the upstream
DoD is treated as black/transparent.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Blur nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Directional Blur [DRBL]

The Directional Blur node

Chapter 33 Blur Nodes 818

Directional Blur Node Introduction
This node is used to create Directional and Radial blurs. It is useful for creating simulated motion blur
and light ray-type effects. Directional Blur affects all channels (RGBA).

Inputs
The two inputs on the Directional Blur node are used to connect a 2D image and an effect mask which
can be used to limit the blurred area.
– Input: The orange input is used for the primary 2D image that has the directional blur applied.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the directional
blur to only those pixels within the mask. An effect mask is applied to the tool after it is
processed.

Basic Node Setup

The Directional Blur node receives a 2D image like the MediaIn1 shown below. The output continues
the node tree by connecting to another 2D image-processing node or a Merge node.

A Directional Blur node applied to a MediaIn1 node

Inspector

Directional Blur controls

Controls Tab
The Controls tab contains all the primary controls necessary for customizing the directional blur
operation.

Chapter 33 Blur Nodes 819

Type
This menu is used to select the type of directional blur to be applied to the image.
– Linear: Linear distorts the image in a straight line, resembling the scenery that appears in the
window of a speeding train.
– Radial: Radial creates a distortion that originates at some arbitrary center, radiating outward the
way that a view would appear if one were at the head of the train looking forward.
– Centered: The Centered button produces a similar result to linear, but the blur effect is equally
distributed on both sides of the original.
– Zoom: Zoom creates a distortion in the scale of the image smear to simulate the zoom streaking of
a camera filming with a slow shutter speed.

Center X and Y
This coordinate control and related viewer crosshair affects the Radial and Zoom Motion Blur types
only. It is used to position where the blurring effect starts.

Length
Length adjusts the strength and heading of the effect. Values lower than zero cause blurs to head
opposite the angle control. Values greater than the slider maximum may be typed into the
slider’s edit box.

Angle
In both Linear and Center modes, this control modifies the direction of the directional blur. In the Radial
and Zoom modes, the effect is similar to the camera spinning while looking at the same spot. If the
setting of the length slider is other than zero, the effect creates a whirlpool effect.

Glow
This adds a Glow to the directional blur, which can be used to duplicate the effect of increased camera
exposure to light caused by longer shutter speeds.

Clipping Mode
This option determines how edges are handled when performing domain-of-definition rendering.
This is profoundly important for nodes like Blur, which may require samples from portions of the image
outside the current domain.
– Frame: The default option is Frame, which automatically sets the node’s domain of definition
to use the full frame of the image, effectively ignoring the current domain of definition. If the
upstream DoD is smaller than the frame, the remaining area in the frame is treated as black/
transparent.
– Domain: Setting this option to Domain respects the upstream domain of definition when applying
the node’s effect. This can have adverse clipping effects in situations where the node employs a
large filter.
– None: Setting this option to None does not perform any source image clipping at all. This means
that any data required to process the node’s effect that would normally be outside the upstream
DoD is treated as black/transparent.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Blur nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 33 Blur Nodes 820

Glow [GLO]

The Glow node

Glow Node Introduction

A Glow is created by blurring an image, and then brightening the blurred result and mixing it back with
the original. The Glow node provides a variety of variations on this theme. For example, a Bartlett glow
is a high-quality glow with a smoother drop-off; however, it is more processor-intensive at larger sizes.

Inputs
The Glow node has three inputs: an orange one for the primary 2D image input, a blue one for an
effect mask, and a third white input for a Glow mask.
– Input: The orange input is used for the primary 2D image that has the glow applied.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input restricts the source
of the glow to only those pixels within the mask. An effect mask is applied to the tool after it is
processed.
– Glow Mask: The Glow node supports pre-masking using the white glow mask input. A Glow
pre-mask filters the image before applying the glow. The glow is then merged back over the
original image. This is different from a regular effect mask that clips the rendered result.

The Glow mask allows the glow to extend beyond the borders of the mask, while restricting the source
of the glow to only those pixels within the mask.
Glow masks are identical to Effect masks in every other respect.

Basic Node Setup

The Glow node receives a 2D image like the MediaIn1 shown below. The output continues the node
tree by connecting to another 2D image-processing node or a Merge node.

A Glow node applied to a MediaIn1 node in DaVinci Resolve

Chapter 33 Blur Nodes 821

Inspector

Glow controls

Controls Tab
The Controls tab contains all the primary controls necessary for customizing the glow operation. A
Color Scale section at the bottom of the Inspector can be used for tinting the glow.

Filter
Use this menu to select the method of Blur used in the filter. The selections are described below.
– Box: A simple but very fast Box filter.
– Bartlett: Bartlett adds a softer, subtler glow with a smoother drop-off but may take longer to
render than Box.
– Multi-box: Multi-box uses a Box filter layered in multiple passes to approximate a Gaussian shape.
With a moderate number of passes (e.g., four), a high-quality blur can be obtained, often faster
than the Gaussian filter, and without any ringing.
– Gaussian: Gaussian adds a soft glow, blurred by the Gaussian algorithm.
– Fast Gaussian: Fast Gaussian adds a soft glow, blurred by the Gaussian algorithm. This is the
default method.
– Blend: Blend adds a nonlinear glow that is evenly visible in the whites and blacks.
– Hilight: Hilight adds a glow without creating a halo in the surrounding pixels.
– Solarize: Solarize adds a glow and solarizes the image.

Color Channels (RGBA)

This filter defaults to operating on R, G, B, and A channels. Selective channel filtering is possible by
clicking each channel to make them active or inactive.

NOTE: This is not the same as the RGBA checkboxes found under the common controls. The
node takes these selections into account before it processes the image, so deselecting a
channel causes the node to skip that channel when processing, speeding up the rendering of
the effect. In contrast, the channel controls under the Common Controls tab are applied after
the node has processed.

Chapter 33 Blur Nodes 822

Lock X/Y
When Lock X/Y is checked, both the horizontal and vertical glow amounts are locked. Otherwise,
separate amounts of glow may be applied to each axis.

Glow Size
Glow Size determines the size of the glow effect. Larger values expand the size of the glowing
highlights of the image.

Num Passes
Only available in Multi-box mode. Larger values lead to a smoother distribution of the effect, but also
increase render times. It’s good to find the line between desired quality and acceptable render times.

Glow
The Glow slider determines the intensity of the glow effect. Larger values tend to completely blow the
image out to white.

Clipping Mode
This option determines how edges are handled when performing domain-of-definition rendering. This
is profoundly important for nodes like Blur, which may require samples from portions of the image
outside the current domain.
– Frame: The default option is Frame, which automatically sets the node’s domain of definition
to use the full frame of the image, effectively ignoring the current domain of definition. If the
upstream DoD is smaller than the frame, the remaining area in the frame is treated as black/
transparent.
– Domain: Setting this option to Domain respects the upstream domain of definition when applying
the node’s effect. This can have adverse clipping effects in situations where the node employs a
large filter.
– None: Setting this option to None does not perform any source image clipping at all. This means
that any data required to process the node’s effect that would normally be outside the upstream
DoD is treated as black/transparent.

Blend
The Blend slider determines the percentage of the affected image that is mixed with original image. It
blends in more of the original image as the value gets closer to 0.
This control is a cloned instance of the Blend slider in the Common Controls tab. Changes made to this
control are simultaneously made to the one in the common controls.

Apply Mode
Three Apply Modes are available when it comes to applying the glow to the image.
– Normal: Default. This mode simply adds the glow directly over top of the original image.
– Merge Under: Merge Under places the glow beneath the image, based on the Alpha channel.
Threshold mode permits clipping of the threshold values.
– Threshold: This control clips the effect of the glow. A new range slider appears. Pixels in the
glowed areas with values below the low value are pushed to black. Pixels with values greater than
high are pushed to white.
– High-Low Range Control: Available only in Threshold mode. Pixels in the glowed areas with
values below the low value are pushed to black. Pixels with values greater than high are
pushed to white.

Chapter 33 Blur Nodes 823

Color Scale (RGBA)
These Scale sliders can be used to adjust the amount of glow applied to each color channel
individually, by tinting the glow.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Blur nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Sharpen [SHRP]

The Sharpen node

Sharpen Node Introduction

The Sharpen node uses a convolution filter to enhance detail in an image overall or to an
individual channel.

Inputs
The two inputs on the Sharpen node are used to connect a 2D image and an effect mask that can limit
the area affected by the sharpen.
– Input: The orange input is used for the primary 2D image for sharpening.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the sharpen to
only those pixels within the mask. An effect mask is applied to the tool after it is processed.

Basic Node Setup

The Sharpen node receives a 2D image like the MediaIn1 shown below. The output continues the
node tree by connecting to another 2D image-processing node or a Merge node.

A Sharpen node applied to a MediaIn1 node in DaVinci Resolve

Chapter 33 Blur Nodes 824

Inspector

Sharpen controls

Controls Tab
The Controls tab contains all the primary controls necessary for customizing the sharpen operation.

Color Channels (RGBA)

This filter defaults to operating on R, G, B, and A channels. Selective channel filtering is possible by
clicking the channel buttons to make them active or inactive.

NOTE: This is not the same as the RGBA checkboxes found under the common controls. The
node takes these selections into account before it processes the image, so deselecting a
channel causes the node to skip that channel when processing, speeding up the rendering of
the effect. In contrast, the channel controls under the Common Controls tab are applied after
the node has processed.

Lock X/Y
This locks the X and Y Sharpen sliders together for symmetrical sharpening. This is checked
by default.

Amount
This slider sets the amount of sharpening applied to the image. When the Lock X/Y control is
deselected, independent control over each axis is provided.

Clipping Mode
This option determines how edges are handled when performing domain-of-definition rendering. This
is profoundly important for nodes like Blur, which may require samples from portions of the image
outside the current domain.
– Frame: The default option is Frame, which automatically sets the node’s domain of definition
to use the full frame of the image, effectively ignoring the current domain of definition. If the
upstream DoD is smaller than the frame, the remaining area in the frame is treated as black/
transparent.
– Domain: Setting this option to Domain respects the upstream domain of definition when applying
the node’s effect. This can have adverse clipping effects in situations where the node employs a
large filter.
– None: Setting this option to None does not perform any source image clipping at all. This means
that any data required to process the node’s effect that would normally be outside the upstream
DoD is treated as black/transparent.

Chapter 33 Blur Nodes 825

Blend
The Blend slider determines the percentage of the affected image that is mixed with original image. It
blends in more of the original image as the value gets closer to 0.
This control is a cloned instance of the Blend slider in the Common Controls tab. Changes made to this
control are simultaneously made to the one in the common controls.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Blur nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Soft Glow [SGlo]

The Soft Glow node

Soft Glow Node Introduction

The Soft Glow node is similar to the Glow node but performs additional processing of the image to
create a much softer, more natural glow.
This node is perfect for atmospheric haze around planets, skin tones, and simulating dream
like environments.

Inputs
Like the Glow node, Soft Glow also has three inputs: an orange one for the primary image input, a blue
one for an effect mask, and a third white input for a Glow mask.
– Input: The orange input is used for the primary 2D image for the soft glow.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the soft glow to
only those pixels within the mask. An effect mask is applied to the tool after it is processed.
– Glow Mask: The Soft Glow node supports pre-masking using the white glow mask input.
A Glow pre-mask filters the image before applying the soft glow. The soft glow is then
merged back over the original image. This is different from a regular effect mask that clips the
rendered result.

The Glow mask allows the soft glow to extend beyond the borders of the mask, while restricting the
source of the soft glow to only those pixels within the mask.
Glow masks are identical to effect masks in every other respect.

Basic Node Setup

The Soft Glow node receives a 2D image like the MediaIn1 shown below. The output continues the
node tree by connecting to another 2D image-processing node or a Merge node.

Chapter 33 Blur Nodes 826

A Soft Glow node applied to a MediaIn1 node

Inspector

Soft Glow controls

Controls Tab
The Controls tab contains all the primary controls necessary for customizing the soft glow operation. A
color scale section at the bottom of the Inspector can be used for tinting the soft glow.

Filter
Use this menu to select the method of Blur used in the filter. The selections are described below.
– Box: A simple but very fast Box filter.
– Bartlett: Bartlett adds a softer, subtler glow with a smoother drop-off but may take longer to
render than Box.
– Multi-box: Multi-box uses a Box filter layered in multiple passes to approximate a Gaussian shape.
With a moderate number of passes (e.g., four), a high-quality blur can be obtained, often faster
than the Gaussian filter and without any ringing.
– Gaussian: Gaussian adds a soft glow, blurred by the Gaussian algorithm.
This is the default method.

Color Channels (RGBA)

The filter defaults to operating on R, G, B, and A channels. Selective channel filtering is possible by
clicking the channel buttons to make them active or inactive.

NOTE: This is not the same as the RGBA checkboxes found under the common controls. The
node takes these selections into account before it processes the image, so deselecting a
channel causes the node to skip that channel when processing, speeding up the rendering of
the effect. In contrast, the channel controls under the Common Controls tab are applied after
the node has processed.

Chapter 33 Blur Nodes 827

Threshold
This control is used to limit the effect of the soft glow. The higher the threshold, the brighter the pixel
must be before it is affected by the glow.

Gain
The Gain control defines the brightness of the glow.

Lock X/Y
When Lock X/Y is checked, both the horizontal and vertical glow amounts are locked. Otherwise,
separate amounts of glow may be applied to each axis of the image.

Glow Size
This amount determines the size of the glow effect. Larger values expand the size of the glowing
highlights of the image.

Num Passes
Available only in Multi-box mode. Larger values lead to a smoother distribution of the effect, but also
increase render times. It’s good to find the line between desired quality and acceptable render times.

Clipping Mode
This option determines how edges are handled when performing domain-of-definition rendering. This
is profoundly important for nodes like Blur, which may require samples from portions of the image
outside the current domain.
– Frame: The default option is Frame, which automatically sets the node’s domain of definition
to use the full frame of the image, effectively ignoring the current domain of definition. If the
upstream DoD is smaller than the frame, the remaining area in the frame is treated as black/
transparent.
– Domain: Setting this option to Domain respects the upstream domain of definition when applying
the node’s effect. This can have adverse clipping effects in situations where the node employs a
large filter.
– None: Setting this option to None does not perform any source image clipping at all. This means
that any data required to process the node’s effect that would normally be outside the upstream
DoD is treated as black/transparent.

Blend
The Blend slider determines the percentage of the affected image that is mixed with original image. It
blends in more of the original image as the value gets closer to 0.
This control is a cloned instance of the Blend slider in the Common Controls tab. Changes made to this
control are simultaneously made to the one in the common controls.

Color Scale (RGBA)

These Scale sliders are used to adjust the amount of glow applied to each color channel individually,
by tinting the glow.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Blur nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 33 Blur Nodes 828

Unsharp Mask [USM]

The Unsharp Mask node

Unsharp Mask Introduction

Unsharp masking is a technique used to sharpen only the edges within an image. This node is most
often used to correct for blurring and loss of detail in low-contrast images; for example, to extract
useful detail from long exposure shots of faraway galaxies.
This filter extracts a range of frequencies from the image and blurs them to reduce detail. The blurred
result is then compared to the original images. Pixels with a significant difference between the original
and the blurred image are likely to be an edge detail. The pixel is then brightened to enhance it.

Inputs
The two inputs on the Unsharp Mask node are used to connect a 2D image and an effect mask for
limiting the effect.
– Input: The orange input is used for the primary 2D image for the Unsharp Mask.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the Unsharp
Mask to only those pixels within the mask. An effect mask is applied to the tool after it is
processed.

Basic Node Setup

The Unsharp Mask node receives a 2D image like the MediaIn1 shown below. The output continues
the node tree by connecting to another 2D image-processing node or a Merge node.

An Unsharp mask node applied to a MediaIn1 node

Inspector

Unsharp Mask controls

Chapter 33 Blur Nodes 829

Controls Tab
The Controls tab contains all the primary controls necessary for customizing the
Unsharp Mask operation.

Color Channels (RGBA)

The filter defaults to operating on R, G, B, and A channels. Selective channel filtering is possible by
clicking the channel buttons to make them active or inactive.

NOTE: This is not the same as the RGBA checkboxes found under the common controls.
The node takes these selections into account before it processes the image, so deselecting a
channel causes the node to skip that channel when processing, speeding up the rendering of
the effect. In contrast, the channel controls under the Common Controls tab are applied after
the node has processed.

Lock X/Y
When Lock X/Y is checked, both the horizontal and vertical sharpen amounts are locked. Otherwise,
separate amounts of glow may be applied to each axis of the image.

Size
This control adjusts the size of blur filter applied to the extracted image. The higher this value, the
more likely it is that pixels are identified as detail.

Gain
The Gain control adjusts how much gain is applied to pixels identified as detail by the mask. Higher
values create a sharper image.

Threshold
This control determines the frequencies from the source image to be extracted. Raising the value
eliminates lower-contrast areas from having the effect applied.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Blur nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Vari Blur [VBL]

The Vari Blur node

Vari Blur Node Introduction

The Vari Blur node gives a true per-pixel variable blur, using a second image to control the amount of
blur for each pixel. It is somewhat similar in effect to the Depth Blur node but uses a different approach
for frequently cleaner results.

Chapter 33 Blur Nodes 830

Inputs
There are two inputs on the Vari Blur node for the primary image: the blur map image, and an
effect mask.
– Input: The gold image input is a required connection for the primary image you wish to blur.
– Blur Image: The green input is also required, but it can accept a spline shape, text object, still
image, or movie file as the blur map image. Once connected, you can choose red, green, blue,
Alpha, or luminance channel to create the shape of the blur.
– Effect Mask: The optional blue effect mask input expects a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the Vari Blur to only those pixels within the mask. An effect mask is applied to the
tool after it is processed.

Basic Node Setup

The Vari Blur node receives a 2D image like the MediaIn1 shown below. A gradient Background tool
connects to the Blur image input to control the areas affected by the blur. The output continues the
node tree by connecting to another 2D image-processing node or a Merge node.

A Vari Blur node applied to a MediaIn1 node and a

gradient background directing the blurred regions

Inspector

Vari Blur controls

Controls Tab
The Controls tab contains all the primary controls necessary for customizing the Vari Blur operation.

Method
Use this menu to select the method of Blur used in the filter. The selections are described below.

Chapter 33 Blur Nodes 831

 – Soften: This method varies from a simple Box shape to a Bartlett triangle to a decent-looking
Smooth blur as Quality is increased. It is a little better at preserving detail in less-blurred areas
than Multi-box.
– Multi-box: Similar to Soften, this gives a better Gaussian approximation at higher Quality settings.
– Defocus: Produces a flat, circular shape to blurred pixels that can approximate the
look of a defocus.

Quality
Increasing Quality gives smoother blurs, at the expense of speed. Quality set to 1 uses a very fast but
simple Box blur for all Method settings. A Quality of 2 is usually sufficient for low Blur Size values.
A Quality of 4 is generally good enough for most jobs unless Blur Size is particularly high.

Blur Channel
This selects which channel of the Blur Image controls the amount of blurring applied to each pixel.

Lock X/Y
When selected, only a Blur Size control is shown, and changes to the amount of blur are applied to
both axes equally. If the checkbox is cleared, individual controls appear for both X and Y Blur Size.

Blur Size
Increasing this control increases the overall amount of blur applied to each pixel. Those pixels where
the Blur image is black or nonexistent are blurred, despite the Blur Size.

Blur Limit
This slider limits the useable range from the Blur image.  Some Z-depth images can have values that
go to infinity, which skew blur size. The Blur Limit is a way to keep values within range.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Blur nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Vector Motion Blur [VBL]

The Vector Motion Blur node

Vector Motion Blur Introduction

This node is used to create directional blurs based on a Motion Vector map or AOV (Arbitrary Output
Variable) channels exported from 3D-rendering software like Arnold, Renderman, or VRay. You can
also generate motion vectors using the Optical Flow node in Fusion.
The vector map is typically two floating-point images: one channel specifies how far the pixel is
moving in X, and the other specifies how far the pixel is moving in Y. These channels may be
embedded in OpenEXR or RLA/RPF images, or may be provided as separate images using the node’s
Vectors input.

Chapter 33 Blur Nodes 832

The vector channels should use a float16 or float32 color depth, to provide + and – values.
A value of 1 in the X channel would indicate that pixel has moved one pixel to the right, while a value of
–10 indicates ten pixels of movement to the left.

Inputs
The Vector Motion Blur node has three inputs for a 2D image, a motion vector pass, and an
effect mask.
– Input: The required orange input is for a 2D image that receives the motion blur.
– Vectors: The green input is also required. This is where you connect a motion vector AOV
rendered from a 3D application or an EXR file generated from the Optical Flow node in Fusion.
– Vector Mask: The white Vector Mask input is an optional input that masks the image before
processing.
– Effect Mask: The common blue input is used for a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this input
restricts the source of the motion blur to only those pixels within the mask. An effect mask is
applied to the tool after it is processed.

Basic Node Setup

The Vector Motion Blur node receives a 2D image like the IMAGE shown below. A MediaIn or Loader
node containing motion vectors is connected to the Vector’s input. The output continues the node
tree by connecting to another 2D image-processing node or a Merge node.

A Vector Motion Blur node applied to a MediaIn or Loader

node with motion vectors connected to the Vectors input

Inspector

Vector Motion Blur node

Controls Tab
The Controls tab contains all the primary controls necessary for customizing the Vector Motion
Blur operation.

Chapter 33 Blur Nodes 833

X Channel
Use this menu to select which channel of the image provides the vectors for the movement of the
pixels along the X-axis.

Y Channel
Use this menu to select which channel of the image provides the vectors for the movement of the
pixels along the Y-axis.

Flip Channel
These checkboxes can be used to flip, or invert, the X and Y vectors. For instance, a value of 5 for a
pixel in the X-vector channel would become –5 when the X checkbox is enabled.

Lock Scale X/Y

Selecting this checkbox provides access to separate sliders for X and Y Scale. By default, only a single
Scale slider is provided.

Scale
The X and Y vector channel values for a pixel are multiplied by the value of this slider. For example,
given a scale of 2 and a vector value of 10, the result would be 20. This slider splits to show Scale X
and Scale Y if the Lock Scale X/Y checkbox is not enabled.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Blur nodes. These common controls are
described in the following “The Common Controls” section.

The Common Controls

Nodes that handle blur operations share several identical controls in the Inspector. This section
describes controls that are common among Blur nodes.

Inspector

Common Blur settings

Chapter 33 Blur Nodes 834

Settings Tab
The Settings tab in the Inspector can be found on every tool in the Blur category. The Settings controls
are even found on third-party Blur-type plug-in tools. The controls are consistent and work the same
way for each tool.

Blend
The Blend control is used to blend between the tool’s original image input and the tool’s final modified
output image. When the blend value is 0.0, the outgoing image is identical to the incoming image.
Normally, this causes the tool to skip processing entirely, copying the input straight to the output.

Process When Blend Is 0.0

The tool is processed even when the input value is zero. This can be useful if processing of this node
is scripted to trigger another task, but the value of the node is set to 0.0.

Red/Green/Blue/Alpha Channel Selector

These four buttons are used to limit the effect of the tool to specified color channels. This filter is often
applied after the tool has been processed.
For example, if the Red button on a Blur tool is deselected, the blur is first applied to the image, and
then the red channel from the original input is copied back over the red channel of the result.
There are some exceptions, such as tools where deselecting these channels causes the tool to skip
processing that channel entirely. Tools that do this generally possess a set of identical RGBA buttons
on the Controls tab in the tool. In this case, the buttons in the Settings and the Controls tabs
are identical.

Apply Mask Inverted

Enabling the Apply Mask Inverted option inverts the complete mask channel for the tool. The mask
channel is the combined result of all masks connected to or generated in a node.

Multiply by Mask
Selecting this option causes the RGB values of the masked image to be multiplied by the mask
channel’s values. This causes all pixels of the image not in the mask (i.e. set to 0) to become
black/transparent.

Use Object/Use Material (Checkboxes)

Some 3D software can render to file formats that support additional channels. Notably, the EXR file
format supports object and material ID channels, which can be used as a mask for the effect. These
checkboxes determine whether the channels are used, if present. The specific material ID or object ID
affected is chosen using the next set of controls.

Correct Edges
This checkbox appears only when the Use Object or Use Material checkboxes are selected. It toggles
the method used to deal with overlapping edges of objects in a multi-object image. When enabled,
the Coverage and Background Color channels are used to separate and improve the effect around the
edge of the object. If this option disabled (or no Coverage or Background Color channels are
available), aliasing may occur on the edge of the mask.
For more information on Coverage and Background Color channels, see Chapter 18, “Understanding
Image Channels” in the Fusion Reference Manual or Chapter 79 in the DaVinci Resolve
Reference Manual.

Object ID/Material ID (Sliders)

Use these sliders to select which ID is used to create a mask from the object or material channels of an
image. Use the Sample button in the same way as the Color Picker to grab IDs from the image
displayed in the viewer. The image or sequence must have been rendered from a 3D software
package with those channels included.

Chapter 33 Blur Nodes 835

Use GPU
The GPU menu has three settings. Disable turns off GPU hardware accelerated rendering. Enabled
uses the GPU hardware for rendering the node. Auto uses a capable GPU if one is available and falls
back to software rendering when a capable GPU is not available.

Motion Blur
– Motion Blur: This toggles the rendering of Motion Blur on the tool. When this control is toggled
on, the tool’s predicted motion is used to produce the motion blur caused by the virtual camera’s
shutter. When the control is toggled off, no motion blur is created.
– Quality: Quality determines the number of samples used to create the blur. A Quality setting of
2 causes Fusion to create two samples to either side of an object’s actual motion. Larger values
produce smoother results but increase the render time.
– Shutter Angle: Shutter Angle controls the angle of the virtual shutter used to produce the motion
blur effect. Larger angles create more blur but increase the render times. A value of 360 is the
equivalent of having the shutter open for one whole frame exposure. Higher values are possible
and can be used to create interesting effects.
– Center Bias: Center Bias modifies the position of the center of the motion blur. This allows the
creation of motion trail effects.
– Sample Spread: Adjusting this control modifies the weighting given to each sample. This affects
the brightness of the samples.

Comments
The Comments field is used to add notes to a tool. Click in the field and type the text. When a note is
added to a tool, a small red square appears in the lower-left corner of the node when the full tile is
displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the note
in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 33 Blur Nodes 836

Chapter 34

Color Nodes
This chapter details the Color nodes available in Fusion. The abbreviations next to
each node name can be used in the Select Tool dialog when searching for tools and
in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Auto Gain [AG]  838
Brightness Contrast [BC]  839
Channel Booleans [BOL]  842
Color Corrector [CC]  845
Color Curves [CCV]  855
Color Gain [CLR]  858
Color Matrix [CMX]  862
Color Space [CS]  866
Copy Aux [CPA]  868
Gamut [GMT]  872
Hue Curves [HCV]  874
OCIO CDL Transform [OCD]  877
OCIO Color Space [OCC]  880
OCIO File Transform [OCF]  882
Set Canvas Color [SCV]  883
White Balance [WB]  885
The Common Controls  889

Chapter 34 Color Nodes 837

Auto Gain [AG]

The Auto Gain node

Auto Gain Node Introduction

The Auto Gain node automatically adjusts the tonal range of an image, setting the darkest and
brightest pixels to user-selected values. By default, the darkest pixels get pushed to black, the
brightest pixels get pushed to white, and pixels in between get stretched to cover the tonal
range evenly.
This can be useful when compensating for variations in lighting, dealing with low-contrast images, or
visualizing the full color range of float images (although the viewer’s View Normalized Image option is
generally more suitable for this).

Inputs
The two inputs on the Auto Gain node are the input and effect mask.
– Input: The orange input connects the primary 2D image for the auto gain.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the auto gain
adjustment to only those pixels within the mask. An effect mask is applied to the tool after the
tool is processed.

Basic Node Setup

The Auto Gain node, like many 2D image-processing nodes, receives a 2D image like a Loader node
or the MediaIn1 shown below. The output continues the node tree by connecting to another 2D
image-processing node or a Merge node.

An Auto Gain node applied to a MediaIn1 node

Inspector

Auto Gain controls

Chapter 34 Color Nodes 838

Controls Tab
The Controls tab contains the few primary controls necessary for customizing the AutoGain operation.

NOTE: Variations over time in the input image can cause corresponding variations in the
levels of the result. For example, if a bright object moves out of an otherwise dark shot, the
remaining scene gets suddenly brighter, as the remaining darker values get stretched to
white. This also applies to sudden depth changes when Do Z is applied; existing objects may
be pushed forward or backward when a near or far object enters or leaves the scene.

Do Z
Select the Do Z checkbox to apply the Auto Gain effect to the Z or Depth channels. This can be useful
for matching the ranges of one Z-channel to another, or to view a float Z-channel in the RGB values.

Range
This Range control sets the black point and white point in the image. All tonal values in the image
rescale to fit within this range.

Example
Create a horizontal gradient with the Background node. Set one color to dark gray
(RGB Values 0.2). Set the other color to light gray (RGB Values 0.8).
Add an Auto Gain node and set the Low value to 0.0 and the High value to 0.5. This causes
the brightest pixels to be pushed down to 0.5, and the darkest pixels get pushed to black.
The remainder of the pixel values scale between those limits.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Color nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Brightness Contrast [BC]

The Brightness Contrast node

Brightness Contrast Node Introduction

The Brightness Contrast node adjusts the gain, brightness, contrast, gamma, and saturation of an
image. The order of the controls represents the order in which the operations are applied. For
example, gamma gets applied before contrast but after gain. The Brightness Contrast is also reversible
using the Forward and Reverse buttons. So color corrections, once applied, can be reversed further
downstream.
For this to work best, image processing should operate in 32-bit floating point.

Chapter 34 Color Nodes 839

Inputs
The two inputs on the Brightness Contrast node are the input and effect mask.
– Input: The orange input connects the primary 2D image for the brightness contrast.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the Brightness
Contrast adjustment to only those pixels within the mask. An effect mask is applied to the tool
after the tool is processed.

Basic Node Setup

The Brightness Contrast node, like many 2D image-processing nodes, receives a 2D image like a
Loader node or the MediaIn1 shown below. The output continues the node tree by connecting to
another 2D image-processing node or a Merge node.

A Brightness Contrast node applied to a MediaIn1 node

Inspector

Brightness Control controls

Controls Tab
The Controls tab contains all the primary controls necessary for customizing the brightness, contrast
operations.

Color Channels (RGBA)

The filter defaults to operating on R, G, B, and A channels. Selective channel filtering is possible by
clicking each channel button to make them active or inactive.

Chapter 34 Color Nodes 840

 NOTE: This is not the same as the RGBA checkboxes found under the common controls.
The node takes these selections into account before it processes the image, so deselecting a
channel causes the node to skip that channel when processing, speeding up the rendering of
the effect. In contrast, the channel controls under the Common Controls tab get applied after
the node has processed.

Gain
The gain slider is a multiplier of the pixel value. A Gain of 1.2 makes a pixel that is R0.5 G0.5 B0.4 into
R0.6 G0.6, B0.48 (i.e., 0.4 * 1.2 = 0.48) while leaving black pixels unaffected. Gain affects higher values
more than it affects lower values, so the effect is most influential in the midrange and top range of
the image.

Lift
While Gain scales the color values around black, Lift scales the color values around white. The pixel
values get multiplied by the value of this control. A Lift of 0.5 makes a pixel that is R0.0 G0.0 B0.0 into
R0.5 G0.5, B0.5 while leaving white pixels unaffected. Lift affects lower values more than it affects
higher values, so the effect is most influential in the midrange and low range of the image.

Gamma
Values higher than 1.0 raise the Gamma (mid-gray), whereas lower values decrease it. The effect of this
node is not linear, and existing black or white points are not affected at all. Pure gray colors are
affected the most.

Contrast
Contrast is the range of difference between the light to dark areas. Increasing the value of this slider
increases the contrast, pushing color from the midrange toward black and white. Reducing the contrast
causes the colors in the image to move toward midrange, reducing the difference between the darkest
and brightest pixels in the image.

Brightness
The value of the Brightness slider gets added to the value of each pixel in the image. This control’s
effect on an image is linear, so the effect is applied identically to all pixels regardless of value.

Saturation
Use this control to increase or decrease the amount of Saturation in the image. A saturation of 0 has
no color, reducing the image to grayscale.

Low and High

This range control is similar to the Gain control in some respects. If Low gets anchored at 0.0 and the
High value gets reduced from 1.0, the effect is identical to increasing the gain. High values get
multiplied by the inverse of the high value. (e.g., if high is 0.75, each pixel is multiplied by 1/0.75,
or 1.3333).
Leaving the high anchored at 1.0 and increasing the low is the same as inverting the image colors and
increasing the gain and inverting it back again. This pushes more of the image toward black without
affecting the whites at all.

Direction
Forward applies all values normally. Reverse effectively inverts all values.

Clip Black/White
The Clip Black and Clip White checkboxes clip out-of-range color values that can appear in an image
when processing in floating-point color depth. Out-of-range colors are below black (0.0) or above
white (1.0). These checkboxes have no effect on images processed at 8-bit or 16-bit per channel, as
such images cannot have out-of-range values.

Chapter 34 Color Nodes 841

Pre-Divide/Post-Multiply
Selecting the Pre-Divide/Post-Multiply checkbox causes the image pixel values to be divided by the
Alpha values before the color correction, and then re-multiplied by the Alpha value after
the correction.
This helps to prevent the creation of illegally additive images when color correcting images with
premultiplied Alpha channels.

Common Controls
Settings Tab
The Settings tab in the Inspector appears in other Color nodes. These common controls are described
in detail at the end of this chapter in “The Common Controls” section.

Channel Booleans [BOL]

The Channel Booleans node

Channel Booleans Node Introduction

The Channel Booleans node applies a variety of mathematical and logical operations on the channels
in an image. This node works by using one image’s channels to modify another image’s channels. If a
foreground input is not available, selecting options that use color channels from the foreground ends
up using the background input’s color channels instead.

NOTE: Be aware of another similarly named Channel Boolean (3Bol), which is a 3D node used
to remap and modify channels of 3D materials. When modifying 2D channels, use the
Channel Booleans (with an “s”) node (Bol).

Inputs
There are four inputs on the Channel Booleans node in the Node Editor, but only the orange
Background input is required.
– Background: This orange input connects a 2D image that gets adjusted by the foreground
input image.
– Effect Mask: The blue effect mask input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this input
limits the channel booleans adjustment to only those pixels within the mask.
– Foreground: The green foreground input connects a 2D image that is used to adjust the
background input image.
– Matte: The white matte input can be used to combine external mattes with the foreground and
background operations.

Chapter 34 Color Nodes 842

Basic Node Setup
The Channel Booleans node is an extremely flexible tool used in many different ways. The example
below copies the z-depth channel from the foreground input (green) into the background
image (orange).

A Channel Booleans set to copy from foreground to background

Inspector

Channel Booleans controls

Color Channel Tab

On the Color Channels tab, the controls are divided into two columns.
On the left side are target channels for the image connected into the orange background input.
The drop-down menu to the right lets you choose whether you want to modify the BG image with its
channels (suffix BG after list name) or with the channels from an image connected into the green
foreground input on the node (suffix FG in the drop-down list).

Operation
This menu selects the mathematical operation applied to the selected channels. The options are
as follows:

Copy
Copy the value from one color channel to another. For example, copy the foreground red channel into
the background’s Alpha channel to create a matte.
– Add: Add the color values from one color channel to another channel.
– Subtract: Subtract the color values of one color channel from another color channel.
– And: Perform a logical AND on the color values from color channel to color channel. The
foreground image generally removes bits from the color channel of the background image.
– Or: Perform a logical OR on the color values from color channel to color channel. The foreground
image generally adds bits from the color channel of the background image.

Chapter 34 Color Nodes 843

 – Exclusive Or: Perform a logical XOR on the color values from color channel to color channel. The
foreground image generally flips bits from the color channel of the background image.
– Multiply: Multiply the values of a color channel. This gives the appearance of darkening the image
as the values scale from 0 to 1. White has a value of 1, so the result would be the same. Gray has a
value of 0.5, so the result would be a darker image or, in other words, an image half as bright.
– Divide: Divide the values of a color channel. This gives the appearance of lightening the image as
the values scale from 0 to 1.
– Maximum: Compare the two images and take the maximum, or brightest, values from each image.
– Minimum: Compare the two images and take the minimum, or darkest, values from each image.
– Negative: Invert the FG input to make a negative version of the image.
– Solid: Solid sets a channel to a full value of 255. This is useful for setting the Alpha to full value.
– Clear: Clear sets a channel to a value of zero. This is useful for clearing the Alpha.
– Difference: Difference subtracts the greater color values of one color channel from the lesser
values of another color channel.
– Signed Add: Signed Add subtracts areas that are lower than mid-gray and adds areas that are
higher than mid-gray, which is useful for creating effects with embossed gray images.

To Red, To Green, To Blue, To Alpha

These menus represent the four color channels of the output image. Use the drop-down menu to
select which channel from the source images produces the output channel.
The default setting copies the channels from the foreground channel. Select any one of the four color
channels, as well as several auxiliary channels like Z-buffer, saturation, luminance, and hue.

Inspector

Aux Channel Inspector

Chapter 34 Color Nodes 844

Aux Channel Tab
This tab includes a series of menus where you select a source for the auxiliary channels of the
output image.

Enable Extra Channels

When the Enable Extra Channels checkbox is selected, the Channel Booleans node can output
images with channels beyond the usual RGBA. Once enabled, the remaining controls in the
Aux Channels tab can copy data into the auxiliary channels.

Examples
To copy the Alpha channel of one image to its color channels, set the red, green, and blue
channels to Alpha BG. Set the Operation to Copy.
To copy the Alpha channel from another image, set operation type to Alpha FG.
To replace the existing Alpha channel of an image with the Alpha of another image, choose
“Do Nothing” for To Red, To Green, and To Blue and “Alpha FG” for To Alpha. Pipe the image
containing the Alpha into the foreground input on the Channel Booleans node. Set Operation:
“Copy.” The same operation is available in the Matte Control node.
To combine any mask into an Alpha channel of an image, choose “Do Nothing” for To Red,
To Green, and To Blue and “Matte” for To Alpha. Pipe the mask into the foreground input on
the Channel Booleans node. Set Operation: “Copy.”
To subtract the red channel’s pixels of another image from the blue channel, choose “Do
Nothing” for To Red and To Green and “Red FG” for To Blue. Pipe the image containing the red
channel to subtract into the foreground input on the Channel Booleans node. Set
Operation: “Subtract.”

Common Controls
Settings Tab
The Settings tab in the Inspector appears in other Color nodes. These common controls are described
in detail at the end of this chapter in “The Common Controls” section.

Color Corrector [CC]

The Color Corrector node

Color Corrector Node Introduction

The Color Corrector node is a comprehensive color node with histogram matching, and equalization,
hue shifting, tinting, and color suppression.
Controls in the Color Corrector node are separated into four tabs: Correction, Ranges, Options,
and Settings.

Chapter 34 Color Nodes 845

Inputs
The Color Corrector node includes four inputs in the Node Editor.
– Input: This orange input is the only required connection. It connects a 2D image for color
correction.
– Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this input
limits the color corrector adjustment to only those pixels within the mask. An effect mask is
applied to the tool after the tool is processed.
– Match Reference: The green input is used to connect an image that can be a reference for
histogram matching.
– Match Mask: This optional white input accepts any mask much like an effect mask. However,
this mask defines of the area to match during a Histogram Match. It offers more flexibility in
terms of shape than the built-in Match rectangle in the Inspector.

Basic Node Setup

The Color Corrector node, like many 2D image-processing nodes, receives a 2D image like a Loader
node or the MediaIn1 shown below. The output continues the node tree by connecting to another 2D
image-processing node or a Merge node.

A Color Corrector node applied to a MediaIn1 node

Inspector

Color Corrector controls

Chapter 34 Color Nodes 846

Correction Tab Colors Menu
The main Correction tab is further separated into four types of correction methods: colors, levels,
histogram, and suppress. Selecting one from the menu at the top of the Correction tab causes that
method’s controls to appear. The Color method is described in detail below.

Range
This menu determines the tonal range affected by the color correction controls in this tab. The menu
can be set to Shadows, Midtones, Highlights, and Master, where Master is the default affecting the
entire image.
The selected range is maintained throughout the Colors, Levels, and Suppress sections of the Color
Corrector node.
Adjustments made to the image in the Master channel are applied to the image after any changes
made to the Highlight, Midtone, and Shadow ranges.

NOTE: The controls are independent for each color range. For example, adjusting the
Gamma control while in Shadows mode does not change or affect the value of the Gamma
control for the Highlights mode. Each control is independent and applied separately.

Color Wheel
The color wheel provides a visual representation of adjustments made to Hue and Saturation, as well
as any tinting applied to the image. Adjustments can be made directly by dragging the color indicator,
or by entering values in the numeric boxes under the color wheel.
The tinting is represented in the color wheel color indicator that shows the color and strength of the
tint. The Highlight setting uses a black outline for the color indicator. The Midtones and Shadows use
gray color indicators. The Master color indicator is also black, but it has a white M in the center to
distinguish it from the others.
The mouse can position the color indicator for each range only when the applicable range is selected.
For example, the Highlight color indicator cannot be moved when the Master range is selected.
Holding down the Command or Ctrl key while dragging this indicator allows you to make finer
adjustments by reducing the control’s sensitivity to mouse movements. Holding down the Shift key
limits the movement of the color indicator to a single axis, allowing you to restrict the effect to either
tint or strength.

Tint Mode
This menu is used to select the speed and quality of the algorithm used to apply the hue and
saturation adjustments. The default is Better, but for working with larger images, it may be desirable to
use a faster method.

Hue
This slider is a clone of the Hue control located under the color wheel. The slider makes it easier to
make small adjustments to the value with the mouse. The Hue control provides a method of shifting
the hue of the image (or selected color range) through the color spectrum. The control value has an
effective range between -0.1 and 1.0, which represents the angle of rotation in a clockwise direction. A
value of 0.25 would be 90 degrees (90/360) and would have the effect of shifting red toward blue,
green to red, and so on.
Hue shifting can be done by dragging the slider, entering a value directly into the text control, or by
placing the mouse above the outer ring of the color wheel and dragging the mouse up or down. The
outer ring always shows the shifted colors compared to the original colors shown in the center of
the wheel.

Chapter 34 Color Nodes 847

Saturation
This slider is a clone of the Saturation control located under the color wheel. The slider makes it easier
to make small adjustments to the value with the mouse. The Saturation control is used to adjust the
intensity of the color values. A saturation of 0 produces gray pixels without any color component,
whereas a value of 1.0 produces no change in the chroma component of the input image. Higher
values generate oversaturated values with a high color component.
Saturation values can be set by dragging the slider, entering a value directly into the text control, or by
dragging the mouse to the left and right on the outer ring of the color wheel control.

Channel
This menu is set for the Histogram, Color, and Levels sections of the Color Corrector node. When the
red channel is selected, the controls in each mode affect the red channel only, and so on.
The controls are independent, so switching to blue does not remove or eliminate any changes made
to red, green, or Master. The animation and adjustments made to each channel are separate. This
menu simply determines what controls to display.

Contrast
Contrast is the range of difference between the light to dark areas. Increasing the value of this slider
increases the contrast, pushing color from the midrange toward black and white. Reducing the contrast
causes the colors in the image to move toward midrange, reducing the difference between the darkest
and brightest pixels in the image.

Gain
The Gain slider is a multiplier of the pixel value. A gain of 1.2 makes a pixel that is R0.5 G0.5 B0.4 into
R0.6 G0.6, B0.48 (i.e., 0.4 * 1.2 = 0.48), while leaving black pixels totally unaffected. Gain affects higher
values more than it affects lower values, so the effect is strongest in the midrange and top range of
the image.

Lift
While Gain scales the color values around black, Lift scales the color values around white. The pixel
values are multiplied by the value of this control. A Lift of 0.5 makes a pixel that is R0.0 G0.0 B0.0 into
R0.5 G0.5, B0.5, while leaving white pixels totally unaffected. Lift affects lower values more than it
affects higher values, so the effect is strongest in the midrange and low range of the image.

Gamma
Values higher than 1.0 raise the Gamma (mid gray), whereas lower values decrease it. The effect of this
node is not linear, and existing black or white points are not affected at all. Pure gray colors are
affected the most.

Brightness
The value of the Brightness slider is added to the value of each pixel in your image. This control’s
effect on an image is linear, so the effect is applied identically to all pixels despite value.

Reset All Color Changes

Selecting this button returns all color controls in this section to their default values.

Chapter 34 Color Nodes 848

Color Corrector Levels controls

Correction Tab Levels Menu

The main Correction tab is further separated into four types of correction methods: colors, levels,
histogram, and suppress. When Levels is selected from the menu, you can remap the white and black
points of an image, with a Gamma control to adjust midtones. A histogram provides a view of the tonal
distribution in the image to help guide your adjustments. The Level method is described in
detail below.

Range
Identical to the Range menu when Color is selected in the Menu, the Range menu determines the
tonal range affected by the color correction controls in this tab. The menu can be set to Shadows,
Midtones, Highlights, and Master, where Master is the default affecting the entire image.
The selected range is maintained throughout the Colors, Levels, and Suppress sections of the Color
Corrector node.
Adjustments made to the image in the Master channel are applied to the image after any changes
made to the Highlights, Midtones, and Shadows ranges.

NOTE: The controls are independent for each color range. For example, adjusting the
Gamma control while in Shadows mode does not change or affect the value of the Gamma
control for the Highlights mode. Each control is independent and applied separately.

Channel
This menu is used to select and display the histogram for each color channel or for the Master channel.

Histogram Display
A histogram is a chart that represents the distribution of color values in the scene. The chart reads
from left to right, with the leftmost values representing the darkest colors in the scene and the
rightmost values representing the brightest. The more pixels in an image with the same or similar
value, the higher that portion of the chart is.
Luminance is calculated per channel; therefore, the red, green, and blue channels all have their own
histogram, and the combined result of these comprises the Master Histogram.
To scale the histogram vertically, place the mouse pointer inside the control and drag the pointer up to
zoom in or down to zoom out.

Chapter 34 Color Nodes 849

Display Selector Toolbar
The Display Selector toolbar at the top of the histogram provides a method of enabling and disabling
components of the histogram display. Hold the mouse pointer over the button to display a tooltip that
describes the button’s function.
– Input Histogram: This enables or disables the display of the input image’s histogram.
– Reference Histogram: This enables or disables the display of the reference image’s histogram.
– Output Histogram: This enables or disables the display of the histogram from the post-color-
corrected image.
– Corrective Curve: This toggles the display of a spline used to visualize exactly how auto color
corrections applied using a reference image are affecting the image. This can be useful when
equalizing luminance between the input and reference images.

Histogram Controls
These controls along the bottom of the histogram display are used to adjust the input image’s
histogram, compressing or shifting the ranges of the selected color channel.
The controls can be adjusted by dragging the triangles beneath the histogram display to the left
and right.
Shifting the High value toward the left (decreasing the value) causes the histogram to slant toward
white, shifting the image distribution toward white. The Low value has a similar effect in the opposite
direction, pushing the image distribution toward black.

Output Level
The Output Level control can apply clipping to the image, compressing the histogram. Decreasing the
High control reduces the value of pixels in the image, sliding white pixels down toward gray and gray
pixels toward black.
Adjusting the Low control toward High does the opposite, sliding the darkest pixels toward white.
If the low value were set to 0.1, pixels with a value of 0.0 would be set to 0.1 instead, and other values
would increase to accommodate the change. The best way to visualize the effect is to observe the
change to the output histogram displayed above.

Reset All Levels

Clicking this button resets all the controls in the Levels section to their defaults.

Color Correction in Histogram mode

Chapter 34 Color Nodes 850

Correction Tab Histogram Menu
When the menu is set to Histogram, a histogram display is produced of the input image. If a reference
image is also provided, the histogram for the reference image is also displayed. The controls in this tab
are primarily used to match one image to another, using either the Equalize or Match modes of the
Color Corrector.

Channel
This menu is used to select and display the histogram for each color channel or for the Master channel.

Histogram Display
A histogram is a chart that represents the distribution of color values in the scene. The chart reads
from left to right, with the leftmost values representing the darkest colors in the scene and the
rightmost values representing the brightest. The more pixels in an image with the same or similar
value, the higher that portion of the chart is.
Luminance is calculated per channel; therefore, the red, green, and blue channels all have their own
histogram, and the combined result of these comprises the Master Histogram.
To scale the histogram vertically, place the mouse pointer inside the control and drag the pointer up to
zoom in or down to zoom out.

Display Selector Toolbar

The Display Selector toolbar at the top of the histogram provides a method of enabling and disabling
components of the histogram display. Hold the mouse pointer over the button to display a tooltip that
describes the button’s function.
– Input Histogram: This enables or disables the display of the input image’s histogram.
– Reference Histogram: This enables or disables the display of the reference image’s histogram.
– Output Histogram: This enables or disables the display of the histogram from the post-color-
corrected image.
– Corrective Curve: This toggles the display of a spline used to visualize exactly how auto color
corrections applied using a reference image are affecting the image. This can be useful when
equalizing luminance between the input and reference images.

Float Images and Histogram Equalization or Matching

By using the Histogram Match or Equalize methods on a float image, the color depth of the output
image is converted to 16-bit integer. Two-dimensional histograms are not well suited to working with
the extreme dynamic range of float images, so these operations always revert to 16-bit integer
processing.

Histogram Type
Each of these menu options enables a different type of color correction operation.
– Keep: Keep produces no change to the image, and the reference histogram is ignored.
– Equalize: Selecting Equalize adjusts the source image so that all the color values in the image are
equally represented—in essence, flattening the histogram so that the distribution of colors in the
image becomes more even.
– Match: The Match mode modifies the source image based on the histogram from the reference
image. It is used to match two shots with different lighting conditions and exposures so that they
appear similar.
When selected, the Equalize and Match modes reveal the following controls.

Chapter 34 Color Nodes 851

 – Match/Equalize Luminance: This slider affects the degree that the Color Corrector node attempts
to affect the image based on its luminance distribution. When this control is zero (the default),
matching and equalization are applied to each color channel independently, and the luminance, or
combined value of the three color channels, is not affected.
If this control has a positive value when equalizing the image, the input image’s luminance
distribution is flattened before any color equalization is applied.
If this control has a positive value when the correction mode is set to Match, the luminance values
of the input are matched to the reference before any correction is applied to the R, G, and
B channels.
The Luminance and RGB controls can have a cumulative effect, and generally they are not both
set to full (1.0) simultaneously.
– Lock R/G/B: When this checkbox is selected, color matching is applied to all color channels
equally. When the checkbox is not selected, individual controls for each channel appear.

Equalize/Match R/G/B
The name of this control changes depending on whether the Equalize or Match modes have been
selected. The slider can be used to reduce the correction applied to the image to equalize or match it.
A value of 1.0 causes the full effect of the Equalize or Match to be applied, whereas lower values
moderate the result.

Precision
This menu determines the color fidelity used when sampling the image to produce the histogram.
10-bit produces higher fidelity than 8-bit, and 16-bit produces higher fidelity than 10-bit.

Smooth Correction
Often, color equalization and matching operations introduce posterization in an image, which occurs
because gradients in the image have been expanded or compressed so that the dynamic range
between colors is not sufficient to display a smooth transition. This control can be used to smooth the
correction curve, blending some of the original histogram back into the result for a more even
transition.

Snapshot Match Time

Click this button to take a freeze snapshot of the current reference histogram, storing its current state
as a snapshot in memory. If the reference histogram is not snapshot, the reference histogram is
updated from frame to frame. This can cause flickering and phasing of the correction as the node tries
to match a changing source to a changing reference.

Release Match
Click this button to release the current snapshot of the histogram and return to using the live
reference input.

Reset All Histogram Changes

Selecting this button removes all changes made to the histogram, returning the controls to default and
setting the mode back to Keep.

Correction Tab Suppress Menu

Color Suppression provides a mechanism for removing an unwanted color component from the image.
The Color Wheel control is similar to that shown in the Colors section of the node, but this one is
surrounded by six controls, each representing a specific color along the wheel.
To suppress a color in the selected range, drag the control that represents that color toward the center
of the color wheel. The closer the control is to the center, the more that color is suppressed from
the image.

Chapter 34 Color Nodes 852

Color Corrector Suppression controls

Suppression Angle
Use the Suppression Angle control to rotate the controls on the suppression wheel and zero in on a
specific color.

Reset All Suppression

Clicking this control resets the suppression colors to 1.0, the default value.

Color Corrector Luminance controls

Ranges Tab
The Ranges tab contains the controls used to specify which pixels in an image are considered to be
shadows and which are considered to be highlights. The midrange is always calculated as pixels not
already included in the shadows or the highlights.

Range
This menu is used to select the tonal range displayed in the viewers. They help to visualize the pixels
in the range. When the Result menu option is selected, the image displayed by the color corrector in
the viewers is that of the color corrected image. This is the default.
Selecting one of the other menu options switches the display to a grayscale image showing which
pixels are part of the selected range. White pixels represent pixels that are considered to be part of
the range, and black pixels are not in the range. For example, choosing Shadows would show pixels

Chapter 34 Color Nodes 853

considered to be shadows as white and pixels that are not shadows as black. Mid gray pixels are only
partly in the range and do not receive the full effect of any color adjustments to that range.

Channel
The Channel menu in this tab can be used to examine the range of a specific color channel. By default,
Fusion displays the luminance channel when the color ranges are examined.

Spline Display
The ranges are selected by manipulating the spline handles. There are four spline points, each with
one Bézier handle. The two handles at the top represent the start of the shadow and highlight ranges,
whereas the two at the bottom represent the end of the range. The Bézier handles are used to control
the falloff.
The midtones range has no specific controls since its range is understood to be the space between
the shadow and the highlight ranges.
The X and Y text controls below the spline display can be used to enter precise positions for the
selected Bézier point or handle.

Output the Range You See Now as Final Render

Selecting this checkbox causes the monochrome display of the range shown in the viewers to be
output as the final render. Normally, the Color node outputs the full RGBA image, even if the node
were left to display one of the color ranges in the view instead. This control makes it possible to use
the Color Corrector node to generate a range’s matte for use as an effect mask in other nodes.

Preset Simple/Smooth Ranges

These two buttons can be used to return the spline ranges to either Smooth (default) or Simple
(linear) settings.

Color Corrector controls

Options Tab
The Options tab includes a few very important processing operations including a simple solution when
color correcting premultiplied Alpha channels.

Pre-Divide/Post-Multiply
Selecting this option divides the color channels by the value of the Alpha before applying the color
correction. After the color correction, the color values are re-multiplied by the Alpha to produce a
properly additive image. This is crucial when performing an additive merge or when working with CG
images generated with premultiplied Alpha channels.

Histogram Proxy Scale

The Histogram Proxy Scale determines the precision used when creating and calculating histograms.
Lower values represent higher precision, and higher values produce a rougher, generalized histogram.

Chapter 34 Color Nodes 854

Process Order
This menu is used to select whether adjustments to the image’s gamma are applied before or after any
changes made to the images levels.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Color nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Color Curves [CCV]

The Color Curves node

Color Curves Node Introduction

The Color Curves node is a spline-based node for performing Lookup table (LUT) color manipulations.
A separate spline is provided for each color channel. The effect can be animated or dissolved and can
be applied to the image using RGB, YUV, YIQ, CMY, or HLS color spaces.
The LUT view in the Color Corrector can be scaled using the + and - keys on the numeric keypad. The
color curves LUT fully supports out-of-range values—i.e., pixels with color values above 1.0 or
below 0.0.
The splines shown in this LUT view are also available from the Spline Editor, should greater precision
be required when adjusting the controls.

Inputs
The Color Curves node includes three inputs in the Node Editor.
– Input: This orange input is the only required connection. It connects a 2D image that is
adjusted by the color curves.
– Effect Mask: The optional effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the color curves adjustment to only those pixels within the mask. An effect mask is
applied to the tool after it is processed.
– Reference Image: The optional green input is used to connect a second 2D image that can be
used for reference matching.
– Match Mask: This optional white input accepts any mask much like an effect mask. However,
this mask defines of the area to match during a Match. It offers more flexibility in terms of shape
than the built-in Match reference rectangle in the Inspector.

Basic Node Setup

The Color Curves node, like many 2D image-processing nodes, receives a 2D image like a Loader
node or the MediaIn1 shown below. The output continues the node tree by connecting to another 2D
image-processing node or a Merge node.

Chapter 34 Color Nodes 855

A Color Curves node applied to a MediaIn1 node

Inspector

Color Curves controls

Controls Tab
The Controls tab for the color curves is divided into two sections. The top half of the Inspector
includes the curves and LUT controls. The bottom half is dedicated primarily to matching the
reference image.

Mode
The Mode options change between Animated and Dissolve modes. The default mode is No
Animation, where adjustments to the curves are static. Setting the mode provides a change spline for
each channel, allowing the color curve to be animated over time.
Dissolve mode is essentially obsolete and is included for compatibility reasons only.

Color Space
The splines in the LUT view represent color channels from a variety of color spaces. The default is
Red, Green, and Blue. The options in this menu allow an alternate color space to be selected.

Chapter 34 Color Nodes 856

A detailed description of the color spaces available here are below:
– RGB (Red, Green, Blue): Fusion uses the RGB color space, and most nodes and displays interpret
the primary channels of an image as Red, Green, and Blue.
– YUV (Luma, Blue Chroma, and Red Chroma): The YUV color space is used in the analog
broadcast of PAL video. Historically, this format was often used to color correct images, because
of its familiarity to a large percentage of video engineers. Each pixel is described in terms of its
Luminance, Blue Chroma, and Red Chroma components.
– HLS (Hue, Luminance, and Saturation): Each pixel in the HLS color space is described in terms of
its Hue, Luminance, and Saturation components.
– YIQ (Luma, In Phase, and Quadrature): The YIQ color space is used in the analog broadcast of
NTSC video. This format is much rarer than YUV and almost never seen in production. Each pixel
is described in terms of its Luminance, Chroma (in-phase or red-cyan channel) and Quadrature
(magenta-green) components.
– CMY (Cyan, Magenta, and Yellow): Although more common in print, the CMY format is often
found in computer graphics from other software packages. Each pixel is described in terms of its
Cyan, Magenta, and Yellow components. CMY is nonlinear.

Color Channels (RGBA)

Use the Color Channel controls to select which channel’s spline is currently active for editing.
The labels of these controls change to reflect the names of the channels for the current color space.
Normally, they are read as Red, Green, and Blue. If the Color Curves node is operating in YUV color
space, they are read as Y, U, and V instead.
These controls do not restrict the effect of the node to a specific channel. They only select whether
the spline for that channel is editable. These controls are most often used to ensure that adding or
moving points on one channel’s spline do not unintentionally affect a different channel’s spline.

Spline Window
The Spline Window displays a standard curve editor for each RGBA channel. These splines can be
edited individually or as a group, depending on the color channels selected above.
The spline defaults to a linear range, from 0 in/0 out at the bottom left to the 1 in/1 out at the top right.
At the default setting, a color processes to the same value as the output. If a point is added in the
middle at 0.5 in/0.5 out, and the point is moved up, this raises the mid color of the image brighter.
The spline curves allow precise control over color ranges, so specific adjustments can be made
without affecting other color values.

In and Out
Use the In and Out controls to manipulate the precise values of a selected point. To change a value,
select a point and enter the in/out values desired.

Eyedropper (Pick)
Click the Eyedropper icon, also called the Pick button, and select a color from an image in the display
to automatically set control points on the spline for the selected color. The new points are drawn with
a triangular shape and can only be moved vertically (if point is locked, only the Out value can change).
Points are only added to enabled splines. To add points only on a specific channel, disable the other
channels before making the selection.
One use for this technique is white balancing an image. Use the Pick control to select a pixel from the
image that should be pure gray. Adjust the points that appear so that the Out value is 0.5 to change
the pixel colors to gray.
Use the contextual menu’s Locked Pick Points option to unlock points created using the Pick option,
converting them into normal points.

Chapter 34 Color Nodes 857

Reference
The Reference section includes controls that handle matching to sample areas of the connected
reference image.
– Match Reference: The Match Reference button adds points on the curve to match an image
connected to the green reference image input. The number of points used to match the image is
based on the Number of Samples slider below.
– Sample Reference: Clicking the Sample Reference button samples the center scanline of the
background image and creates a LUT of its color values. The number of points used to match the
samples scanline is based on the Number of Samples slider below.
– Number of Samples: This slider determines how many points are used to match the curve to the
range in the reference image.
– Show Match Rectangle: Enabling this checkbox displays a rectangle in the viewer showing the
area on the reference image used during the match process. The match rectangle affects only the
result of the Match Reference operation. The Sample reference is always done from the center
scaling of the image.
– Match Center: The X and Y parameters allow you to reposition the match rectangle to sample a
different area when matching.
– Match Width: Width controls the width of the match rectangle.
– Match Height: Heigh controls the height of the match rectangle.
– Pre-Divide/Post-Multiply: Selecting this checkbox causes the image’s pixel values to be divided
by the Alpha values prior to the color correction, and then re-multiplied by the Alpha value after
the correction. This helps to avoid the creation of illegally additive images, particularly around the
edges of a blue/green key or when working with 3D-rendered objects.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Color nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Color Gain [CLR]

The Color Gain node

Color Gain Node Introduction

The Color Gain node contains options for adjusting the gain, gamma, saturation, and hue of the image.
Many controls provided by the Color Gain node are also found in the Color Corrector node, but this
simpler node may render more quickly. One feature that distinguishes the Color Gain node from the
Color Corrector is its balance tab controls. These can be used to adjust the tinting of the colors in the
highs, mids, and lows.

Chapter 34 Color Nodes 858

Inputs
The Color Gain node includes two inputs: one for the main image and the other for an effect mask.
– Input: This orange input is the only required connection. It connects a 2D image that gets
adjusted by the color gain.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the color gain adjustment to only those pixels within the mask. An effect mask is
applied to the tool after it is processed.

Basic Node Setup

The Color Gain node, like many 2D image-processing nodes, receives a 2D image like a Loader node
or the MediaIn1 shown below. The output continues the node tree by connecting to another 2D-image
processing node or a Merge node.

A Color Gain node applied to a MediaIn1 node

Inspector

Color Gain controls

Gain Tab
The Gain tab provides control of individual RGBA Lift/Gamma/Gain parameters. These controls can
quickly enable you to fix irregular color imbalances in specific channels.

Chapter 34 Color Nodes 859

Lock R/G/B
When selected, the Red, Green, and Blue channel controls for each effect are combined into one
slider. Alpha channel effects remain separate.

Gain RGBA
The Gain RGBA controls multiply the values of the image channel in a linear fashion. All pixels are
multiplied by the same factor, but the effect is larger on bright pixels and smaller on dark pixels. Black
pixels do not change because multiplying any number times 0 is always 0.

Lift RGBA
While Gain scales the color values around black, Lift scales the color values around white. The pixel
values are multiplied by the value of this control. A Lift of 0.5 makes a pixel that is R0.0 G0.0 B0.0 into
R0.5 G0.5, B0.5, while leaving white pixels totally unaffected. Lift affects lower values more than it
affects higher values, so the effect is strongest in the midrange and low range of the image.

Gamma RGBA
The Gamma RGBA controls affect the brightness of the midrange in the image. The effect of this node
is nonlinear. White and black pixels in the image are not affected when gamma is modified, whereas
pure grays are affected most by changes to this parameter. Large changes to this control tend to push
midrange pixels into black or white, depending on the value used.

Pre-Divide/Post-Multiply
Selecting this checkbox causes the image pixel values to be divided by the Alpha values prior to the
color correction, and then re-multiplied by the Alpha value after the correction. This helps when
attempting to color correct images with premultiplied Alpha channels.

Saturation Tab
This Setting tab includes controls for the intensity of the colors in the individual RGB channels.

Color Gain Saturation setting tab

RGB Saturation
When adjusting an individual channel, a value of 0.0 strips out all that channel’s color. Values greater
than one intensify the color in the scene, pushing it toward the primary color.

Balance Tab
This tab in the Color Gain node offers controls for adjusting the overall balance of a color channel.
Independent color and brightness controls are offered for the High, Mid, and Dark ranges of
the image.
Colors are grouped into opposing pairs from the two dominant color spaces. Red values can be
pushed toward Cyan, Green values toward Magenta and Blue toward Yellow. Brightness can be raised
or lowered for each of the channels.

Chapter 34 Color Nodes 860

Color Gain Balance tab

CMY Brightness Highs/Mids/Darks

By default, the Balance sliders can be adjusted by -1 to +1, but values outside this range can be
entered manually to increase the effect. A value of 0.0 for any slider indicates no change to the image
channel. Positive and negative values indicate that the balance of the image channel has been pushed
toward one color or the other in the pair.

Hue Tab
Use the Hue tab of the Color Gain node to shift the overall hue of the image, without affecting the
brightness, or saturation. Independent controls of the High, Mid, and Dark ranges are offered by
three sliders.
The following is the order of the hues in the RGB color space: Red, Yellow, Green, Cyan, Blue,
Magenta and Red.

Color Gain Hue tab

High/Mid/Dark Hue
Values above 0 push the hue of the image toward the right (red turns yellow). Values below 0 push the
hue toward the left (red turns magenta). At -1.0 or 1.0, the hue completes the cycle and returns to its
original value.
The default range of the hue sliders is -1.0 to +1.0. Values outside this range can be entered manually.

Chapter 34 Color Nodes 861

Ranges Tab
The Ranges tab contains the controls used to specify which pixels in an image are considered to be
shadows and which are considered to be highlights. The midrange is always calculated as pixels not
included in either the shadows or the highlights.

Color Gain Ranges tab

Spline Display
The ranges are selected by manipulating the spline handles. There are four spline points, each with
one Bézier handle. The two handles at the top represent the start of the shadow and highlight ranges,
whereas the two at the bottom represent the end of the range. The Bézier handles are used to control
the falloff.
The midtones range has no specific controls since its range is understood to be the space between
the shadow and the highlight ranges. The X and Y text controls below the Spline display can be used
to enter precise positions for the selected Bézier point or handle.

Preset Simple/Smooth Ranges

These two buttons can be used to return the spline ranges to either Smooth (default) or Simple
(linear) settings.

Settings Tab
The Settings tab in the Inspector is also duplicated in other Color nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Color Matrix [CMX]

The Color Matrix node

Color Matrix Node Introduction

The ColorMatrix allows a vast number of operations to modify values individually in the different
color channels.

Chapter 34 Color Nodes 862

Inputs
The Color Matrix node includes two inputs: one for the main image and the other for an effect mask.
– Input: This orange input is the only required connection. It connects a 2D image that is
adjusted by the color matrix.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the color matrix adjustment to only those pixels within the mask. An effect mask is
applied to the tool after it is processed.

Basic Node Setup

The Color Matrix node, like many 2D image-processing nodes, receives a 2D image like a Loader
node or the MediaIn1 shown below. The output continues the node tree by connecting to another 2D
image-processing node or a Merge node.

A Color Matrix node applied to a MediaIn1 node

Inspector

Color Matrix controls

Controls Tab
Color Matrix multiplies the RGBA channels based on the values entered in a 4 x 4 grid. The fifth
column/row is an Add column.

Update Lock
When this control is selected, Fusion does not render the node. This is useful for setting up each value
of the node, and then turning off Update Lock to render it.

Matrix
This defines what type of operation actually takes place. The horizontal rows define the output values
of the node. From left to right, they are R, G, B, A, and Add. The vertical columns define the input
values. From top to bottom, they are R, G, B, A, and Add. The Add column allows simple adding of
values to the individual color channels.

Chapter 34 Color Nodes 863

By default, the output values are identical to the input values.
– 1.0 means 100% of the Red channel input is copied to the Red channel output.
– 1.0 means 100% of the Green channel input is copied to the Green channel output.
– 1.0 means 100% of the Blue channel input is copied to the Blue channel output.
– 1.0 means 100% of the Alpha channel input is copied to the Alpha channel output.
Written as mathematical equations, the default settings of the matrix would appear as follows:
[R out] = 1 * [R in] + 0 * [G in] + 0 * [B in] + 0 * [A in] + 0
[G out] = 0 * [R in] + 1 * [G in] + 0 * [B in] + 0 * [A in] + 0
[B out] = 0 * [R in] + 0 * [G in] + 1 * [B in] + 0 * [A in] + 0
[A out] = 0 * [R in] + 0 * [G in] + 0 * [B in] + 1 * [A in] + 0

Invert
Enabling this option inverts the Matrix. Think of swapping channels around, doing other operations
with different nodes, and then copying and pasting the original ColorMatrix and setting it to Invert to
get your channels back to the original.

Example 1: Invert
If you want to do a simple invert or negative of the color values, but leave the Alpha channel
untouched, the matrix would look like this:

Color Matrix example

Observe the fact that we have to add 1 to each channel to push the inverted values back into the
positive numbers.
Let’s follow this example step by step by viewing the waveform of a 32-bit grayscale gradient.
1 The original grayscale.

Original Grayscale

Chapter 34 Color Nodes 864

2 RGB set to -1. The values get inverted but fall below 0.

RGB set to -1

3 Adding 1 to each channel keeps the inversion but moves the values back into a positive range.

Adding 1 to each channel

Example 2: Brightness per Channel

This example influences the brightness of each channel individually. This subtracts 0.2 from the red
channel, adds 0.314 to the green channel, and adds 0.75 to the blue channel, while keeping
Alpha as it is.

Brightness per channel example

Chapter 34 Color Nodes 865

Example 3: Copying Values
You can also copy color values back and forth between individual channels. In this example, the red
channel contains the luminance values of the image based on thirds, and the green channel contains
the luminance values based on the proper black-and-white conversion method, whereas the blue
channel uses a third method based on getting more information from red and less from blue. The blue
channel’s brightness is also lowered by 0.1, and the Alpha channel is replaced with the original
blue channel.

Copying Values example

Settings Tab
The Settings tab in the Inspector is also duplicated in other Color nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Color Space [CS]

The Color Space node

Color Space Node Introduction

The Color Space node provides the ability to work on an image in a variety of alternate color space
formats. By default, Fusion uses the RGB color space, and most nodes and displays interpret the
primary channels of an image as Red, Green, and Blue.
Changing the color space from RGB causes most images to look odd, as Fusion’s viewers still interpret
the primary channels as Red, Green, and Blue. For example, viewing an image converted to YUV in
one of the viewers shows the Y channel as Red, the U channel as Green, and the V channel as Blue.
Several common elements of the Fusion interface refer to the RGB channels directly. The four buttons
commonly found on the Inspector’s Settings tab to restrict the effect of the node to a single color
channel are one example. When a conversion is applied to an image, the labels of these buttons
remain R, G, and B, but the values they represent are from the current color space. (For example, Red
is Hue, Green is Luminance, and Blue is Saturation for an RGB to HLS conversion. The Alpha value is
never changed by the color space conversion.)

Chapter 34 Color Nodes 866

Inputs
The Color Space node includes two inputs: one for the main image and the other for an effect mask.
– Input: This orange input is the only required connection. It connects a 2D image that is
converted by the color space operation.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the color space adjustment to only those pixels within the mask. An effect mask is
applied to the tool after it is processed.

Basic Node Setup

The Color Space node, like many 2D image-processing nodes, receives a 2D image like a Loader or
the MediaIn1 shown below, processes the image, and then extends the node tree by connecting to
another 2D image-processing node or a Merge node.

A Color Space node applies a conversion to a MediaIn1 node

Inspector

Color Space controls

Controls Tab
The Controls tab in the Color Space node consists of two menus. The top Conversion menu
determines whether you are converting an image to RGB or from RGB. The bottom menu selects the
alternative color space you are either converting to or from.

Conversion
This menu has three options. The None option has no effect on the image. When To Color is selected,
the input image is converted to the color space selected in the Color Type control found below. When
To RGB is selected, the input image is converted back to the RGB color space from the type selected
in the Color Type menu (for example, YUV to RGB).

Color Type
This menu is used to select the color space conversion applied when the To Color conversion is
selected. When the To RGB option is selected in the Conversion menu, the Color Type option should
reflect the input image’s current color space. There are eight color space options to choose from.
– HSV (Hue, Saturation, and Value): Each pixel in the HSV color space is described in terms of its
Hue, Saturation, and Value components. Value is defined as the quality by which we distinguish
a light color from a dark one or brightness. Decreasing saturation roughly corresponds to adding
white to a paint chip on a palette. Increasing Value is roughly similar to adding black.

Chapter 34 Color Nodes 867

 – YUV (Luma, Blue Chroma, and Red Chroma): The YUV color space is used in the analog
broadcast of PAL video. Historically, this format was often used to color correct images because
of its familiarity to a large percentage of video engineers. Each pixel is described in terms of its
Luminance, Blue Chroma, and Red Chroma components.
– YIQ (Luma, In Phase, and Quadrature): The YIQ color space is used in the analog broadcast of
NTSC video. This format is much rarer than YUV and almost never seen in production. Each pixel
is described in terms of its Luminance, Chroma (in-phase or red-cyan channel), and Quadrature
(magenta-green) components.
– CMY (Cyan, Magenta, and Yellow): Although more common in print, the CMY format is often
found in computer graphics from other software packages. Each pixel is described in terms of its
Cyan, Magenta, and Yellow components. CMY is nonlinear.
– HLS (Hue, Luminance, and Saturation): Each pixel in the HLS color space is described in terms
of its Hue, Luminance, and Saturation components. The differences between HLS and HSV color
spaces are minor.
– XYZ (CIE Format): This mode is used to convert a CIE XYZ image to and from RGB color spaces.
CIE XYZ is a weighted space, instead of a nonlinear one, unlike the other available color spaces.
Nonlinear in this context means that equal changes in value at different positions in the color
space may not necessarily produce the same magnitude of change visually to the eye.
Expressed simply, the CIE color space is a perceptual color system, with weighted values obtained
from experiments where subjects were asked to match an existing light source using three primary
light sources.
This color space is most often used to perform gamut conversion and color space matching
between image display formats because it contains the entire gamut of perceivable colors.
– Negative: The color channels are inverted. The color space remains RGBA.
– BW: The image is converted to black and white. The contribution of each channel to the luminance
of the image is adjustable via slider controls that appear when this option is selected. The default
values of these sliders represent the usual perceptual contribution of each channel to an image’s
luminance. The color space of the image remains RGBA.

Settings Tab
The Settings tab in the Inspector is also duplicated in other Color nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Copy Aux [CPA]

The Copy Aux node

Copy Aux Node Introduction

The Copy Aux node is used to shuffle channels between visible channels and auxiliary data channels
in a single 2D image. Typically, these auxiliary channels are rendered from 3D applications. Auxiliary
channels supported in the Copy Aux node include background color, z-depth, texture coordinates,
coverage, object ID, material ID, normals, vectors, back vectors, and world position.

Chapter 34 Color Nodes 868

The Copy Aux node is mostly a convenience node, as the copying can also be accomplished with
more effort using a Channel Booleans node. Where Channel Booleans deals with individual channels,
Copy Aux deals with channel groups. By default, the Copy Aux node automatically promotes the depth
of its output to match the depth of the aux channel.
Copy Aux also supports static normalization ranges. The advantage of static normalization versus the
dynamic normalization that Fusion’s viewers do is that colors remain constant over time. For example,
if you are viewing Z or WorldPos values for a ball, you see a smooth gradient from white to black.
Now imagine that some other 3D object is introduced into the background at a certain time.
Dynamic normalization turns the ball almost completely white while the background object is now the
new black. Dynamic normalization also causes flicker problems while viewing vector/disparity
channels, which can make it difficult to compare the aux channels of two frames at different
times visually.

Inputs
The Copy Aux node includes two inputs: one for the main image and the other for an effect mask.
– Input: This orange input is the only required connection. It connects a 2D image for the Copy
Aux node operation.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the Copy Aux operation to only those pixels within the mask. An effect mask is
applied to the tool after the tool is processed.

Basic Node Setup

The Copy Aux node, like many 2D image-processing nodes, receives a 2D image like a Loader node
or the MediaIn1 shown below. The output continues the node tree by connecting to another 2D
image-processing node or a Merge node.

A Copy Aux node applied to a MediaIn1 node

Inspector

Copy Aux controls

Chapter 34 Color Nodes 869

Controls Tab
The Controls tab is used to copy auxiliary channel groups into RGBA channels. Although Copy Aux has
quite a few options, most of the time you select only the channel to copy and ignore the remaining
functionality.

Mode
The Mode menu determines whether the auxiliary channel is copied into the RGBA color channel (Aux
to Color) or vice versa (Color to Aux). Using this option, you can use one Copy Aux node to bring an
auxiliary channel into color, do some compositing operations on it, and then use another Copy Aux
node to write the color back into the auxiliary channel. When the Mode is set to Color to Aux, all the
options in the Controls tab except the Aux Channel menu are hidden.

Aux Channel
The Aux Channel menu selects the auxiliary channel to be copied from or written to depending on the
current mode. When the aux channel abcd has one valid component, it is copied as aaa1, two valid
components as ab01, three valid components as abc1, and four components as abcd. For example, the
Z-channel is copied as zzz1, texture coordinates as uv01, and normals as nxnynz1.

Out Color Depth

Out Color Depth controls the color depth of the output image. Most aux channels contain float values
or, if they are integer valued, they can contain values beyond 255. When you copy float values into an
int8 or int16 image, this can be a problem since negative values and values over 1.0 can get clipped. In
addition, precision can be lost. This option determines what happens if the depth of RGBA channels of
the input image is insufficient to contain the copied aux channel.
Be careful when copying float channels into integer image formats, as they can get clipped if you do
not set up Copy Aux correctly. For this node, all aux channels are considered to be float32 except
ObjectID or MaterialID, which are considered to be int16.
– Match Aux Channel Depth: The bit depth of the RGBA channels of the output image is increased
to match the depth of the aux channel. Specifically, this means that the RGBA channels of the
output image are either int16 or float32. Be careful when using this option because, for example,
if you normally have int8 color channels, you are now using 2x or 4x more memory for the
color channels. Particularly, the Z, Coverage, TextureCoordinate, Normal, Vector, BackVector,
WorldPosition, and Disparity channels are always output as float, and the Material/ObjectID
channels are output as int16.
– Match Source Color Depth: The bit depth of the RGBA channels of the output image is the same
as the input image. This can have some unexpected consequences. For example, if your input
image is int8, the XYZ components of normals that are floating-point numbers in the [-1, 1] range
are clipped to non-negative numbers [0, 1] range. As a more extreme example, consider what
happens to Z values. Z values are floating-point numbers stored in the [-1e30, 0] range, and they
all get truncated to the [0, 1] range, which means your Z-channel is full of zeroes.
– Force Float32: The bit depth of the RGBA channels of the output image is always float32.

Channel Missing
Channel Missing determines what happens if a channel is not present. For example, this determines
what happens if you chose to copy Disparity to Color and your input image does not have a Disparity
aux channel.
– Fail: The node fails and prints an error message to the console.
– Use Default Value: This fills the RGBA channels with the default value of zero for everything
except Z, which is -1e30.

Chapter 34 Color Nodes 870

Kill Aux Channels
When this is checked, Copy Aux copies the requested channel to RGBA and then outputs a resulting
image that is purely RGBA with other channels being killed. This is useful if you want to increase the
number of frames of Copy Aux that can be cached for playback—for example, to play back a long
sequence of disparity. A handy tip is that you can use the Kill Aux feature also with just Color to Aux >
Color for a longer color playback.

Kill Aux channels

Enable Remapping
When remapping is enabled, the currently selected aux channel is rescaled, linearly mapping the
range according to the From and To slider selections as explained below. The Remapping options are
applied before the conversion operation. This means you could set the From > Min-Max values to -1, 1
to rescale your normals into the [0, 1] range, or set them to [-1000, 0] to rescale your Z values from
[-1000, 0] into the [0, 1] range before the clipping occurs.
Note that the Remapping options are per channel options. That means the default scale for normals
can be set to [-1, +1] > [0, 1] and for Z it can be set [-1000, 0] > [0, 1]. When you flip between normals and
Z, both options are remembered. One way this could be useful is that you can set up the remapping
ranges and save this as a setting that you can reuse. The remapping can be useful to squash the aux
channels into a static [0, 1] range for viewing or, for example, if you wish to compress normals into the
[0, 1] range to store them in an int8 image.
– From > Min: This is the value of the aux channel that corresponds to To > Min.
– From > Max: This is the value of the aux channel that corresponds to To > Max. It is possible to set
the max value less than the min value to achieve a flip/inversion of the values.
– Detect Range: This scans the current image to detect the min/max values and then sets the From
> Min/ From > Max Value controls to these values.
– Update Range: This scans the current image to detect the min/max values and then enlarges the
current [From > Min, From > Max] region so that it contains the min/max values from the scan.
– To > Min: This is the minimum output value, which defaults to 0.
– To > Max: This is the maximum output value, which defaults to 1.
– Invert: After the values have been rescaled into the [To > Min, To > Max] range, this inverts/flips
the range.

Settings Tab
The Settings tab in the Inspector is also duplicated in other Color nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 34 Color Nodes 871

Gamut [GMT]

The Gamut node

Gamut Node Introduction

The Gamut node has controls to transform one color space to another and remove/add gamma curves.
This node, along with the Cineon Log node, is primarily used to linearize incoming images and then
reapply the applicable output gamma curve at the end of a node tree.

Inputs
The Gamut node includes two inputs: one for the main image and the other for an effect mask to limit
the conversion area.
– Input: This orange input is the only required connection. It connects a 2D image output that is
the source of the gamut conversion.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the Gamut operation to only those pixels within the mask. An effect mask is applied
to the tool after the tool is processed.

Basic Node Setup

A Gamut node is most often placed directly after the MediaIn node in DaVinci Resolve or a Loader
node in Fusion Studio. Another Gamut node is usually placed at the end of a node tree before a
MediaOut node in DaVinci Resolve or a Saver node in Fusion Studio.

A Gamut node applied to a MediaIn1 node

Inspector

Chapter 34 Color Nodes 872

Gamut controls

Controls Tab
The Controls tab is where all the conversion operations take place. It has a section for incoming
images and a section for the node’s output. Which section you use depends on whether you are
stripping an image of a gamma curve to make it linear or converting a linear image to a specific color
space and gamma curve for output.

Source Space
Source Space determines the input color space of the image. When placed directly after a Loader
node in Fusion or a MediaIn node in DaVinci Resolve, you would select the applicable color space
based on how the image was created and check the Remove Gamma checkbox. The output of the
node would be a linearized image. You leave this setting at No Change when you are adding gamma
using the Output Space control and placing the node directly before the Saver node in Fusion or a
MediaOut node in DaVinci Resolve.

DCI-P3
The DCI-P3 color space is most commonly used in association with DLP projectors. It is frequently
provided as a color space available with DLP projectors and as an emulation mode for 10-bit LCD
monitors such as the HP Dreamcolor and Apple’s Pro Display XDR. This color space is defined in the
SMPTE-431-2 standard.

Custom
The Custom gamut allows you to describe the color space according to CIE 1931 primaries and white
point, which are expressed as XY coordinates, as well as by gamma, limit, and slope. For example, the
DCI-P3 gamut mentioned above would have the following values if described as a Custom
color space.

Red Primary 0.68 0.32

Green Primary 0.265 0.69

Blue Primary 0.15 0.06

White Point 0.314 0.351

Gamma 2.6 –

Linear Limit 0.0313 –

To understand how these controls work, you could view the node attached to a gradient background
in Waveform mode and observe how different adjustments modify the output.

Chapter 34 Color Nodes 873

Output Space
Output Space converts the gamut to the desired color space. For instance, when working with
linearized images in a composite, you place the Gamut node just before the Saver node and use the
Output Space to convert to the gamut of your final output file. You leave this setting at No Change
when you want to remove gamma using the Source Space control.

NOTE: When outputting to HD specification Rec. 709, Fusion uses the term Scene to refer to
a gamma of 2.4 and the term Display for a gamma of 2.2.

Remove/Add Gamma
Select these checkboxes to do the gamut conversion in a linear or nonlinear gamma, or simply remove
or add the applicable gamma values without changing the color space.

Pre-Divide/Post-Multiply
Selecting this checkbox causes the image’s pixel values to be divided by the Alpha values prior to the
color correction, and then re-multiplied by the Alpha value after the correction. This helps to avoid the
creation of illegally additive images, particularly around the edges of a blue/green key or when
working with 3D-rendered objects.

Settings Tab
The Settings tab in the Inspector is also duplicated in other Color nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Hue Curves [HCV]

The Hue Curves node

Hue Curves Node Introduction

The Hue Curves node allows you to adjust the color in an image using a series of spline curves.
Splines are provided to control the image’s hue, saturation, and luminance as well as each individual
color channel. An additional set of curves allows you to apply suppression to individual color channels.
The advantage of the Hue Curves node over other color correction nodes in Fusion is that the splines
can be manipulated to restrict the node’s effect to a very narrow portion of the image, or expanded to
include a wide-ranging portion of the image. Additionally, these curves can be animated to follow
changes in the image over time. Since the primary axis of the spline is defined by the image’s hue, it is
much easier to isolate a specific color from the image for adjustment.

Chapter 34 Color Nodes 874

Inputs
The Hue Curves node includes two inputs: one for the main image and the other for an effect mask to
limit the color correction area.
– Input: This orange input is the only required connection. It connects a 2D image for the Hue
Curves color correction.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the Hue Curves operation to only those pixels within the mask. An effect mask is
applied to the tool after it is processed.

Basic Node Setup

The Hue Curves node, like many 2D image-processing nodes, receives a 2D image like a Loader node
or the MediaIn1 shown below. The output continues the node tree by connecting to another 2D
image-processing node or a Merge node.

A Hue Curves node applied to a Loader node in Fusion Studio

Inspector

Hue Curves controls

Controls Tab
The Controls tab consists of color attribute checkboxes that determine which splines are displayed in
the Spline window. The spline graph runs horizontally across with control points placed horizontally at
each of the primary colors. You can manipulate these control points to change the selected color
attribute.

Chapter 34 Color Nodes 875

Mode
The Mode options change between No Animation and Animated Points modes. The default mode is
No Animation, where adjustments to the curves are applied consistently over time. Setting the Mode
to Animated Points or Dissolve allows the color curve to be animated over time.
Dissolve mode is essentially obsolete and is included for compatibility reasons only.

Color Channel Checkboxes

These checkboxes define which splines are editable and are included when using the Eyedropper to
pick a color in the image.
Any number of activated splines can be edited simultaneously; however it’s more convenient to have
only the currently modified spline active to avoid unwanted changes to other splines.
When using the Eyedropper icon, a point is created on all active splines, representing the
selected color.

Spline Window
This graph display is the main interface element of the Hue Curves node, which hosts the various
splines. In appearance, the node is very similar to the Color Curves node, but here the horizontal axis
represents the image’s hue, while the vertical axis represents the degree of adjustment. The Spline
window shows the curves for the individual channels. It is a miniature Spline Editor. In fact, the curves
shown in this window can also be found and edited in the Spline Editor.
The spline curves for all components are initially flat, with control points placed horizontally at each of
the primary colors. From left to right, these are: Red, Yellow, Green, Cyan, Blue, and Magenta. Because
of the cyclical design of the hue gradient, the leftmost control point in each curve is connected to the
rightmost control point of the curve.
Right-clicking in the graph displays a contextual menu containing options for resetting the curves,
importing external curves, adjusting the smoothness of the selected control points, and more.

In and Out
Use the In and Out controls to manipulate the precise values of a selected point. To change a value,
select a point and enter the In/Out values desired.

Eyedropper
Left-clicking and dragging from the Eyedropper icon changes the current mouse cursor to an
Eyedropper. While still holding down the mouse button, drag the cursor to a viewer to pick a pixel from
a displayed image. This causes control points, which are locked on the horizontal axis, to appear on
the currently active curves. The control points represent the position of the selected color on the
curve. Use the contextual menu’s Lock Selected Points toggle to unlock points and restore the option
of horizontal movement.
Points are only added to enabled splines. To add points only on a specific channel, disable the other
channels before making the selection.

Pre-Divide/Post-Multiply
Selecting this checkbox causes the image’s pixel values to be divided by the Alpha values prior to the
color correction, and then re-multiplied by the Alpha value after the correction. This helps when color
correcting images that include a premultiplied Alpha channel.

Settings Tab
The Settings tab in the Inspector is also duplicated in other Color nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 34 Color Nodes 876

OCIO CDL Transform [OCD]

The OCIO CDL Transform node

OCIO CDL Transform Node Introduction

Fusion supports the Open Color IO color management workflow by way of three OCIO nodes.
– The OCIO CDL Transform node allows you to create, save, load, and apply a Color Decision
List (CDL) grade.
– The OCIO Color Space allows sophisticated color space conversions, based on an
OCIO config file.
– The OCIO File Transform allows you to load and apply a variety of Lookup tables (LUTs).

Generally, the OCIO color pipeline is composed from a set of color transformations defined by OCIO-
specific config files, commonly named with a “.ocio” extension. These config files allow you to share
color settings within or between facilities. The path to the config file to be used is normally specified
by a user-created environment variable called “OCIO,” although some tools allow overriding this. If no
other *.ocio config files are located, the DefaultConfig.ocio file in Fusion’s LUTs directory is used.
For in-depth documentation of the format’s internals, please refer to the official pages on
opencolorio.org.

Inputs
The OCIO CDL Transform node includes two inputs: one for the main image and the other for an effect
mask to limit the area where the CDL is applied.
– Input: This orange input is the only required connection. It connects a 2D image output for the
CDL grade.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the CDL grade to only those pixels within the mask. An effect mask is applied to the
tool after it is processed.

Basic Node Setup

The OCIO CDL Transform node is often applied after a Gamut node converts the Loader to linear color
in Fusion Studio.

A OCIO CDL Transform node applied to a Loader node

after a Gamut conversion to linear in Fusion Studio

Chapter 34 Color Nodes 877

Inspector

OCIO Transform controls

Controls Tab
The Controls tab for the OCIO CDL Transform contains primary color grading color correction controls
in a format compatible with CDLs. You can make R, G, B adjustments based on the Slope, Offset, and
Power. There is also overall Saturation control. You can also use the Controls tab to import and export
the CDL compatible adjustments.

Operation
This menu switches between File and Controls. In File mode, standard ASC-CDL files can be loaded.
In Controls mode, manual adjustments can be made to Slope, Offset, Power, and Saturation, and the
CDL file can be saved.

NOTE: Using DaVinci Resolve terminology, slope is similar to gain. It controls mids-to-high
contrast. Offset is the overall offset of color balance and exposure. Power is very similar to
contrast with a raised pivot, giving you control over shadow contrast.

Chapter 34 Color Nodes 878

Direction
Toggles between Forward and Reverse. Forward applies the corrections specified in the node, while
Reverse tries to remove those corrections. Keep in mind that not every color correction can
be undone.
Imagine that all slope values have been set to 0.0, resulting in a fully black image. Reversing that
operation is not possible, neither mathematically nor visually.

Slope

Slope multiplies the color values; this is the same as

Gain in the Brightness Contrast node.

Offset

Offset adds to the color values; this is the same as

Brightness in the Brightness Contrast node.

Power
Applies a Gamma Curve. This is an inverse of the Gamma function of the Brightness Contrast node.

Saturation
Enhances or decreases the color saturation. This works the same as Saturation in the Brightness
Contrast node.

Export File
Allows the user to export the settings as a CDL file.

Settings Tab
The Settings tab in the Inspector is also duplicated in other Color nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 34 Color Nodes 879

OCIO Color Space [OCC]

The OCIO Color Space node

OCIO Color Space Node Introduction

Fusion supports the Open Color IO color management workflow by way of three OCIO nodes.
– The OCIO CDL Transform node allows you to create, save, load, and apply a Color Decision
List (CDL) grade.
– The OCIO Color Space allows sophisticated color space conversions, based on an
OCIO config file.
– The OCIO File Transform allows you to load and apply a variety of Lookup tables (LUTs).

Generally, the OCIO color pipeline is composed from a set of color transformations defined by OCIO-
specific config files, commonly named with a “.ocio” extension. These config files allow you to share
color settings within or between facilities. The path to the config file to be used is normally specified
by a user-created environment variable called “OCIO,” though some tools allow overriding this. If no
other *.ocio config files are located, the DefaultConfig.ocio file in Fusion’s LUTs directory is used.
For in-depth documentation of the format’s internals, please refer to the official pages on
opencolorio.org.
Sample configs can be obtained from https://opencolorio.readthedocs.io/en/latest/quick_start/
downloads.html
The functionality of the OCIO Color Space node is also available as a View LUT node from the
View LUT menu.

Inputs
The OCIO Color Space node includes two inputs: one for the main image and the other for an effect
mask to limit the area where the color space conversion is applied.
– Input: This orange input is the only required connection. It connects a 2D image for the color
space conversion.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the color space conversion to only those pixels within the mask. An effect mask is
applied to the tool after it is processed.

Basic Node Setup

The OCIO Color Space node is typically placed directly after a MediaIn node in DaVinci Resolve or a
Loader node in Fusion Studio. Another OCIO Color Space node is placed just before a Media Out
node in DaVinci Resolve or a Saver node in Fusion Studio.

Chapter 34 Color Nodes 880

An OCIO Color Space node applied to a Loader node and a Saver node in Fusion Studio

Inspector

OCIO Color Space controls

Controls Tab
The Controls tab for the OCIO Color Space node allows you to convert an image from one color space
to another based on an OCIO config file. By default, it uses the config file included with Fusion;
however, the Controls tab does allow you to load your own config file as well.

OCIO Config
Displays a File > Open dialog to load the desired config file.

Source Space
Based on the config file, the available source color spaces are listed here.
The content of this list is based solely on the loaded profile and hence can vary immensely. If no other
OCIO config file is loaded, the DefaultConfig.ocio file in Fusion’s LUTs directory is used to populate
this menu.

Output Space
Based on the config file, the available output color spaces are listed here.
The content of this list is based solely on the loaded profile and hence can vary immensely. If no other
OCIO config file is loaded, the DefaultConfig.ocio file in Fusion’s LUTs directory is used to populate
this menu.

Look
Installed OCIO Color Transform Looks appear in this menu. If no looks are installed, this menu has only
None listed as an option.

Settings Tab
The Settings tab in the Inspector is also duplicated in other Color nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 34 Color Nodes 881

OCIO File Transform [OCF]

The OCIO File Transform node

OCIO File Transform Node Introduction

Fusion supports the Open Color IO color management workflow by way of three OCIO nodes.
– The OCIO CDL Transform node allows you to create, save, load,
and apply a Color Decision List (CDL) grade.
– The OCIO Color Space allows sophisticated color space conversions,
based on an OCIO config file.
– The OCIO File Transform allows you to load and apply a variety of Lookup tables (LUTs).

Generally, the OCIO color pipeline is composed from a set of color transformations defined by OCIO-
specific config files, commonly named with a “.ocio” extension. These config files allow you to share
color settings within or between facilities. The path to the config file to be used is normally specified
by a user-created environment variable called “OCIO,” though some tools allow overriding this. If no
other *.ocio config files are located, the DefaultConfig.ocio file in Fusion’s LUTs directory is used.
For in-depth documentation of the format’s internals, please refer to the official pages on
opencolorio.org.
The functionality of the OCIO File Transform node is also available as a View LUT node from the
View LUT menu.

Inputs
The OCIO File Transform node includes two inputs: one for the main image and the other for an effect
mask to limit the area where the color space conversion is applied.
– Input: This orange input is the only required connection. It connects a 2D image for the LUT.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the applied LUT to only those pixels within the mask. An effect mask is applied to
the tool after it is processed.

Basic Node Setup

The OCIO File Transform node is often applied after a Gamut node converts the Loader to linear color
in Fusion Studio.

An OCIO File Transform node applied to a Loader node

after a Gamut node conversion to linear color space

Chapter 34 Color Nodes 882

Inspector

OCIO File Transform controls

Controls Tab
The Controls tab for the OCIO File Transform node includes options to import the LUT, invert the
transform, and select the color interpolation method.

LUT File
Displays a File > Open dialog to load the desired LUT.

CCC ID
This is the ID key used to identify the specific file transform located within the ASC CDL color
correction XML file.

Direction
Toggles between Forward and Reverse. Forward applies the corrections specified in the node, while
Reverse tries to remove those corrections. Keep in mind that not every color correction can be
undone. Imagine that all slope values have been set to 0.0, resulting in a fully black image. Reversing
that operation is not possible, neither mathematically nor visually.

Interpolation
Allows the user to select the color interpolation to achieve the best quality/render time ratio. Nearest is
the fastest interpolation, while Best is the slowest.

Settings Tab
The Settings tab in the Inspector is also duplicated in other Color nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Set Canvas Color [SCV]

The Set Canvas Color node

Set Canvas Color Node Introduction

Set Canvas Color is used to set the color of the area outside the domain of definition (DoD). This is the
workspace area beyond the raster by default, which is invisible since outside the raster is not

Chapter 34 Color Nodes 883

rendered. However, the DoD can be within the raster as well. This can occur when compositing
images smaller than the raster, or with transforms. By default, the canvas color used is black/no Alpha
(transparent). However, since some nodes may change an image’s canvas color—for example,
inverting a mask changes the mask’s canvas from black to white—the Set Canvas Color allows you to
control the color of the canvas to whatever you require.
The Set Canvas Color node sets the color of the workspace outside the domain of definition (DOD).
For example, if you create a circular gradient, the DoD is a square around the circular gradient in the
viewer. Everything outside the DoD is understood to be black and therefore does not have to be
rendered. To change the area outside the DoD, attach the Set Canvas Color node after the
background and change the color.

NOTE: Position the mouse pointer in a black area outside the raster to view the RGB canvas
color in the status bar at the bottom left of the Fusion window.

Inputs
The Set Canvas Color node includes two inputs: one for the main image and a second for a
foreground.
– Input: This orange input is the only required connection. It accepts a 2D image that reveals the
canvas color if the image’s DoD is smaller than the raster.
– Foreground: The optional green foreground input allows the canvas color to be sampled from
an image connected to this input.

Basic Node Setup

The Set Canvas Color node is placed after the image is transformed to reveal part of the raster outside
the domain of definition.

A Set Canvas Color node applied to a transformed MediaIn 1 node

The Set Canvas Color node is often used for adjusting keys. In the example above, the Luma Keyer is
extracting a key, and therefore assigns the area outside the DoD, which is black, as an opaque
foreground. If the element is scaled down and composited, you do not see the background. To correct
this, insert a SetBGColor before the keyed element is placed in the composite. For example, LumaKey
> Set Canvas Color > Transform > Merge.

Chapter 34 Color Nodes 884

Inspector

Set Canvas Color controls

Controls Tab
The Controls tab for the Set Canvas Color is used for simple color selection. When the green
foreground is connected, the tab is empty.

Color Picker
Use these controls to adjust the Color and the Alpha value for the image’s canvas. It defaults to black
with zero Alpha.

Settings Tab
The Settings tab in the Inspector is also duplicated in other Color nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

White Balance [WB]

The White Balance node

White Balance Node Introduction

The White Balance node can be used to automatically remove color casts in the image caused by the
incorrect setup of a camera or bad lighting conditions.
Correction can be done by selecting a color temperature or by choosing a neutral color from the
original image that exhibits the color cast to be corrected.

IMPORTANT
When picking neutral colors using the Custom method, make sure you are picking from the
source image, not the results of the White Balance node. This ensures that the image doesn’t
change while you are still picking, and that the White Balance node gets an accurate idea of
the original colors it needs to correct.

Chapter 34 Color Nodes 885

Inputs
The White Balance node includes two inputs: one for the main image and the other for an effect mask
to limit the area where the white balance is applied.
– Input: This orange input is the only required connection. It connects a 2D image for the white
balance.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the white balance to only those pixels within the mask. An effect mask is applied to
the tool after it is processed.

Basic Node Setup

The White Balance node, like many 2D image-processing nodes, receives a 2D image like the
MediaIn1 shown below. The output continues the node tree by connecting to another 2D image-
processing node or a Merge node.

A White Balance node applied to a MediaIn1 node

Inspector

White Balance controls

Chapter 34 Color Nodes 886

Balance Tab
Space
Use this menu to select the color space of the source image, if it is known. This can make the
correction more accurate since the node can take the natural gamma of the color space into account
as part of the correction. If the color space that the image uses is unknown, leave this menu at its
default value.

Method
The White Balance node can operate using one of two methods: a Custom method or a color
Temperature method.
– Custom: The Custom method requires the selection of a pixel from the scene that should have
been pure gray. The node uses this information to calculate the color correction required to
convert the pixel so that it actually is gray. When the correction is applied without an effect
mask connected and the LockBlack/Mid/White checkbox enabled, the node white balances the
entire shot.
– Temperature: The color Temperature method requires that the actual color temperature of the
shot be specified.

Lock Black/Mid/White
This checkbox locks the Black, Midtones, and White points together so that the entire image is
affected equally. Unchecking the control provides individual controls for white balancing each range
separately. This control affects both methods equally.

Black/Mid/White Reference
These controls appear only if the Custom method is selected. They are used to select a color from a
pixel in the source image. The White Balance node color corrects the image so that the selected color
is transformed to the color set in the Result Color Picker below. Generally, this is gray. A color that is
supposed to be pure gray but is not truly gray for one reason or another should be selected.
If the Lock Black/Mid/White checkbox is deselected, different references can be selected for each
color range.
For example, try to select a pixel for the black and white references that are not clipped in any of the
color channels. In the high end, an example would be a pixel that is light pink with values of 255, 240,
240. The pixel is saturated/clipped in the red, although the color is not white. Similarly, a really dark
blue-gray pixel might be 0, 2, 10. It is clipped in red as well, although it is not black.
Neither example would be a good choice as a reference pixel because there would not be enough
headroom left for the White Balance node.

Black/Mid/White Result
These controls appear only if the Custom method is selected. They are used to select the color that
the node uses to balance the reference color. This generally defaults to pure, midrange gray.
If the Lock Black/Mid/White checkbox is deselected, different results can be selected for each
color range.

Chapter 34 Color Nodes 887

White Balance Temperature controls

Temperature Reference
When the Method menu is set to Temperature, the Temperature reference control is used to set the
color temperature of the source image. If the Lock Black/ Mid/White checkbox is deselected, different
references can be selected for each color range.

Temperature Result
Use this control to set the target color temperature for the image. If the Lock Black/Mid/White
checkbox is deselected, different results can be selected for each color range.

Use Gamma
This checkbox selects whether the node takes the gamma of the image into account when applying
the correction, using the default gamma of the color space selected in the Space menu at the top
of the tab.

White Balance range tab

Ranges Tab
The Ranges tab can be used to customize the range of pixels in the image considered to be shadows,
midtones, and highlights by the node.

Spline Display
The ranges are selected by manipulating the spline handles. There are four spline points, each with
one Bézier handle. The two handles at the top represent the start of the shadow and highlight ranges,
whereas the two at the bottom represent the end of the range. The Bézier handles are used to control
the falloff.
The midtones range has no specific controls since its range is understood to be the space between
the shadow and the highlight ranges.
The X and Y text controls below the Spline display can be used to enter precise positions for the
selected Bézier point or handle.

Chapter 34 Color Nodes 888

Preset Simple/Smooth Ranges
These two buttons can be used to return the spline ranges to either Smooth (default) or Simple
(linear) settings.

Settings Tab
The Settings tab in the Inspector is also duplicated in other Color nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

The Common Controls

Nodes that handle Color adjustment operations share several identical controls in the Inspector. This
section describes controls that are common among color nodes.

Inspector

Color Settings Inspector

Settings Tab
The Settings tab in the Inspector can be found on every tool in the Color category. The Settings
controls are even found on third-party color type plug-in tools. The controls are consistent and work
the same way for each tool, although some tools do include one or two individual options that are also
covered here.

Chapter 34 Color Nodes 889

Blend
The Blend control is used to blend between the tool’s original image input and the tool’s final modified
output image. When the blend value is 0.0, the outgoing image is identical to the incoming image.
Normally, this causes the tool to skip processing entirely, copying the input straight to the output.

Process When Blend Is 0.0

The tool is processed even when the input value is zero. This can be useful if processing of this node
is scripted to trigger another task, but the value of the node is set to 0.0.

Red/Green/Blue/Alpha Channel Selector

These four buttons are used to limit the effect of the tool to specified color channels. This filter is often
applied after the tool has been processed.
For example, if the red button on a blur tool is deselected, the blur is first applied to the image, and
then the red channel from the original input is copied back over the red channel of the result.
There are some exceptions, such as tools for which deselecting these channels causes the tool to skip
processing that channel entirely. Tools that do this generally possess a set of identical RGBA buttons
on the Controls tab in the tool. In this case, the buttons in the Settings and the Controls tabs are
identical.

Apply Mask Inverted

Enabling the Apply Mask Inverted option inverts the complete mask channel for the tool. The mask
channel is the combined result of all masks connected to or generated in a node.

Multiply by Mask
Selecting this option causes the RGB values of the masked image to be multiplied by the mask
channel’s values. This causes all pixels not included in the mask (i.e., set to 0) to become black/
transparent.

Use Object/Use Material (Checkboxes)

Some 3D software can render to file formats that support additional channels. Notably, the EXR file
format supports Object and Material ID channels, which can be used as a mask for the effect. These
checkboxes determine whether the channels are used, if present. The specific Material ID or Object ID
affected is chosen using the next set of controls.

Correct Edges
This checkbox appears only when the Use Object or Use Material checkboxes are selected. It toggles
the method used to deal with overlapping edges of objects in a multi-object image. When enabled,
the Coverage and Background Color channels are used to separate and improve the effect around the
edge of the object. If this option disabled (or no Coverage or Background Color channels are
available), aliasing may occur on the edge of the mask.
For more information on Coverage and Background Color channels, see Chapter 18, “Understanding
Image Channels” in the Fusion Reference Manual or Chapter 79 in the DaVinci Resolve
Reference Manual.

Object ID/Material ID (Sliders)

Use these sliders to select which ID is used to create a mask from the object or material channels of an
image. Use the Sample button in the same way as the Color Picker: to grab IDs from the image
displayed in the viewer. The image or sequence must have been rendered from a 3D software
package with those channels included.

Chapter 34 Color Nodes 890

Clipping Mode
This option determines how edges are handled when performing domain of definition rendering.
This is mostly important for nodes like Blur, which may require samples from portions of the image
outside the current domain.
– Frame: The default option is Frame, which automatically sets the node’s domain of definition
to use the full frame of the image, effectively ignoring the current domain of definition. If
the upstream DoD is smaller than the frame, the remaining area in the frame is treated as
black/transparent.
– Domain: Setting this option to Domain respects the upstream domain of definition when applying
the node’s effect. This can have adverse clipping effects in situations where the node employs a
large filter.
– None: Setting this option to None does not perform any source image clipping at all. This means
that any data required to process the node’s effect that would normally be outside the upstream
DoD is treated as black/transparent.

Use GPU
The Use GPU menu has three settings. Setting the menu to Disable turns off hardware-accelerated
rendering using the graphics card in your computer. Enabled uses the hardware. Auto uses a capable
GPU if one is available and falls back to software rendering when a capable GPU is not available.

Motion Blur
– Motion Blur: This toggles the rendering of Motion Blur on the tool. When this control is toggled
on, the tool’s predicted motion is used to produce the motion blur caused by the virtual camera’s
shutter. When the control is toggled off, no motion blur is created.
– Quality: Quality determines the number of samples used to create the blur. A quality setting of
2 causes Fusion to create two samples to either side of an object’s actual motion. Larger values
produce smoother results but increase the render time.
– Shutter Angle: Shutter Angle controls the angle of the virtual shutter used to produce the motion
blur effect. Larger angles create more blur but increase the render times. A value of 360 is the
equivalent of having the shutter open for one whole frame exposure. Higher values are possible
and can be used to create interesting effects.
– Center Bias: Center Bias modifies the position of the center of the motion blur. This allows the
creation of motion trail effects.
– Sample Spread: Adjusting this control modifies the weighting given to each sample. This affects
the brightness of the samples.

Comments
The Comments field is used to add notes to a tool. Click in the field and type the text. When a note is
added to a tool, a small red square appears in the lower-left corner of the node when the full tile is
displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the note
in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 34 Color Nodes 891

Chapter 35

Composite Nodes
This chapter details the Dissolve and Merge nodes available in Fusion.

The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Dissolve [DX]  893
Merge [MRG]  895
The Common Controls  903

Chapter 35 Composite Nodes 892

Dissolve [DX]

The Dissolve node

Dissolve Node Introduction

The Dissolve node is typically used to mix two images together, providing a gradual transition
between two clips. A Background/Foreground slider controls the mix between the foreground and
background images. Dissolves are commonly used to transition between one clip and another and are
a very common effect in editing. However, you can also use the extreme left and right positions of the
Background/Foreground slider to switch between inputs. Unlike other nodes in Fusion, the Dissolve
node does not require you to connect an image to the background but lets you output either the
background or foreground according to the setting of the Background/Foreground slider.
This quality makes it possible for you use the Dissolve node as an automatic layer switching tool when
connected to background and foreground clips with different durations. Simply connect each clip to
the background and foreground inputs, respectively, and set the Background/Foreground slider to the
input of shorter duration, to determine which is “on top.” After the last frame of that clip has ended, the
Dissolve node automatically switches to the clip that’s connected to the other input.
Besides the default dissolve, the Gradient Wipe setting of the Operation menu allows you to create
arbitrary animated dissolve patterns based on the luminance of an image connected to the optional
Gradient Wipe input. You can use this capability with images of geometric shapes or gradients of
different kinds, movie clips of fire, water ripples, or rain, the Fast Noise node, or even particle systems
you create within the Fusion page to create a variety of unique and creative transitions. Soft-edged
effect masks may also be used to add to the possible effects.
Ultimately, animating the Background/Foreground control allows you to control the transition that’s
being used to switch from the foreground input to the background, or vice versa.

Inputs
The Dissolve node provides three image inputs, all of which are optional:
– Background: The first of two images you want to switch between or mix. Unlike most
other nodes, it is unnecessary to connect the background input before connecting the
foreground input.
– Foreground: The second of two images you want to switch between or mix. The Dissolve node
works best when both foreground and background inputs are connected to images with the
same resolution.
– Gradient Map: (Optional) The Gradient Map is required only when Gradient Wipe is selected.

Basic Node Setup

Dissolve nodes are typically connected in the following way, with two input images connected to the
background and foreground inputs, and the output connected to the next node in the composition.

Chapter 35 Composite Nodes 893

A typical dissolve node structure in Fusion

Resolution Handling
It is recommended to make sure that all images connected to the foreground, background, and
gradient map inputs of the Dissolve node have the same resolution and the same pixel aspect. This is
not required, however. But, the result if you mix resolutions depends on how you set the Background/
Foreground slider.
– If the input images are different sizes, but the Foreground/Background slider is set to full
Foreground (all the way to the right) or full Background (all the way to the left), then the output
resolution will be identical to the image resolution of the corresponding node input.
– If input images of different sizes are mixed by setting the Background/Foreground slider
somewhere between, the output resolution will be set to the larger of the two input resolutions
to make sure there’s enough room to contain both images. In this case, you may experience
undesirable resolution changes when the slider moves from full foreground or background to
somewhere in between.
For example, if you try to dissolve between a 4K image (connected to the background) and an 8K
image (connected to the foreground), the output of the Dissolve node will be 4K when the slider is
set to full Background, but will suddenly jump to 8K when set to full Foreground or when mixed
somewhere between the foreground and background.

Inspector

Dissolve controls

Controls Tab
These are the main controls that govern the Dissolve node’s behavior.
– Operation Pop-Up: The Operation menu contains one of seven different methods for mixing
the Foreground and Background inputs. The two images are mixed using the value of the
Background/Foreground slider to determine the percentage each image contributes.
– Dissolve: The standard Dissolve mode is the equivalent of a cross dissolve: one clip fades out
as another clip fades in.
– Additive Dissolve: Similar in look to a standard film dissolve, an Additive dissolve adds the
second clip and then fades out the first one.
– Erode: The Erode method transitions between the two images by growing the darkest areas of
the background image to reveal the foreground image. The effect appears similar to a filmstrip
burning out.
– Random Dissolve: A randomly generated dot pattern is used to perform the mix of the images.

Chapter 35 Composite Nodes 894

 – Random Noise Dissolve: A moving random dot pattern is used to perform the mix of the images.
– Gradient Wipe: The dissolve is controlled by the luminance values of the image in the Gradient
Map input. The edges of this dissolve can be softened. The density and the color of the border
can be adjusted independently.
– SMPTE Wipe: The SMPTE wipe is similar to the basic effect wipes found on many video
effects switchers. There is a horizontal wipe and a vertical wipe provided. The wipes
can have soft edges and borders added. The density and the color of the border can be
adjusted independently.

– Background/Foreground Slider: Defaults to Foreground. This control determines whether the

output is the background image, the foreground image, or a mix between the two. The type of mix
is determined by the Operation control. If one of the input images is not currently available, the
other one will be output despite the setting of this slider.

Gradient/SMPTE Wipe Controls

The following controls appear only when Gradient Wipe or SMPTE Wipe are selected.
– Wipe Style: (SMPTE Wipe only) This drop-down list allows the selection of two wipe styles:
Horizontal - Left to Right and Vertical - Top to Bottom. The direction of the wipes can be reversed
by using the Invert Wipe checkbox.
– Invert Wipe: (SMPTE Wipe only) When checked, the direction of the wipe will be reversed.
– Softness: Use this control to soften the edge of the transition.
– Border: Select the Border to enable coloring of the transition’s edge and to reveal the associated
controls. The effect is to create a border around the transition edge.
– Border Softness: (Appears only when Border is turned on) The Border Softness slider controls the
width and density of the border. Higher values will create a denser border, and lower values will
create a thinner one.
– Border Color: (Appears only when Border is turned on) Use Border Color to select the color used
in the border.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in both the Dissolve and Merge nodes. These
common controls are described in detail at the end of this chapter in “The Common Controls” section.

Merge [MRG]

The Merge node

Merge Node Introduction

The Merge node combines two images based on the Alpha (opacity) channel associated with the one
in front. This node takes two inputs: a background and a foreground image. The Operation mode
determines which method is used to combine the foreground and background images, supporting the
standard Over, In, Held Out, Atop, and XOr methods for compositing images. Meanwhile, an Apply

Chapter 35 Composite Nodes 895

Mode pop-up lets you use different composite modes, transfer modes, or blend modes (whichever is
your preferred terminology) to combine the foreground against the background in different ways. This
includes such standard modes as screen, dissolve, multiply, overlay, as well as many others.
The Merge node can perform both additive (premultiplied) and subtractive (non-premultiplied)
compositing, depending on how your compositions and media are set up. However, you also have the
flexibility of using the Additive/Subtractive slider to blend between additive and subtractive composite
results, which has the bonus of providing solutions for problem edges in some cases.
Ordinarily, the foreground and background input connections determine the layer order of images
composited with this node. However, you can also enable Z-Depth compositing if Z-channels are
available in the input images. Z-merging compares the depth value of each pixel in each layer to
determine which pixels should be in front and which should be behind.

Inputs
The Merge node provides three image inputs, all of which are optional:
– Background: The orange background input is for the first of two images you want to composite
together. You should connect the background input before connecting the foreground input. If
you connect an image to the background without connecting anything to the foreground input,
the Merge node will output the background image.
– Foreground: The green foreground input is for the second of two images you want to
composite together, which is typically a foreground subject that should be in front of the
background. If you connect an image to the foreground input without connecting anything to
the background input first, the Merge node won’t output anything.
– Effect Mask: (Optional) The effect mask input lets you mask a limited area of the output image
to be merged where the mask is white (where the foreground image shows in front of the
background), letting the background image show through by itself where the mask is black.

Basic Node Setup

Merge nodes are typically connected in the following way, with two input images connected to the
background and foreground inputs, and the output connected to the next node in the composition. In
this example, the effect mask input is not used, as this is not typical.

A typical Merge node structure in DaVinci Resolve

Resolution Handling
While you can connect images of any resolution to the background and foreground inputs of the
Merge node, the image that’s connected to the background input determines the resolution of
the output.

TIP: If you want to change the resolution of the image connected to the background, you can
use the Crop node to change the “canvas” resolution of the image without changing the size
of the original image, or you can use the Resize node to change both the resolution and the
size of the image.

Chapter 35 Composite Nodes 896

Inspector

Merge node controls

Merge Tab
The Merge tab contains most of the controls necessary for customizing most merge operations.

Foreground Sizing Controls

These controls let you adjust the sizing of the image connected to the foreground input, making it
unnecessary to use a separate Transform node to fit the foreground layer to match the background
layer in simple compositions.
– Center X and Y: This control determines the position of the foreground image in the composite.
The default is 0.5, 0.5, which centers the foreground image in the exact center of the background
image. The value shown is always the actual position in normalized coordinates, multiplied by the
reference size. See below for a description of the reference size controls.
– Size: Use this control to increase or decrease the size of the foreground image before it is
composited over the background. The range of values for this slider is 0.0 to 5.0, but any value
greater than 0 can be entered manually. A size of 1.0 gives a pixel-for-pixel composition, where a
single pixel in the foreground is the same size as a single pixel in the background.
– Angle: Use this control to rotate the foreground image before it is combined with the background.

Compositing Mode and Adjustment Controls

The next six parameters control how the background and foreground input images are combined to
create a single output image.
– Apply Modes: The Apply Mode setting determines the math used when blending or combining
the foreground and background pixels.
– Normal: The default Normal merge mode uses the foreground’s Alpha channel as a mask to
determine which pixels are transparent and which are not. When this is active, another menu
shows possible operations, including Over, In, Held Out, Atop, and XOr.

Chapter 35 Composite Nodes 897

– Screen: Screen merges the images based on a multiplication of their color values. The Alpha
channel is ignored, and layer order becomes irrelevant. The resulting color is always lighter.
Screening with black leaves the color unchanged, whereas screening with white will always
produce white. This effect creates a similar look to projecting several film frames onto the same
surface. When this is active, another menu shows possible operations, including Over, In, Held
Out, Atop, and XOr.
– Dissolve: Dissolve mixes two image sequences together. It uses a calculated average of the
two images to perform the mixture.
– Multiply: Multiplies the values of a color channel. This will give the appearance of darkening
the image as the values are scaled from 0 to 1. White has a value of 1, so the result would be the
same. Gray has a value of 0.5, so the result would be a darker image, or an image half as bright.
– Overlay: Overlay multiplies or screens the color values of the foreground image, depending
on the color values of the background image. Patterns or colors overlay the existing pixels
while preserving the highlights and shadows of the color values of the background image. The
background image is not replaced but is mixed with the foreground image to reflect the original
lightness or darkness of the background image.
– Soft Light: Soft Light darkens or lightens the foreground image, depending on the color values
of the background image. The effect is similar to shining a diffused spotlight on the image.
– Hard Light: Hard Light multiplies or screens the color values of the foreground image,
depending on the color values of the background image. The effect is similar to shining a harsh
spotlight on the image.
– Color Dodge: Color Dodge uses the foreground’s color values to brighten the background
image. This is similar to the photographic dark room technique of dodging by reducing the
exposure of an area of a print.
– Color Burn: Color Burn uses the foreground’s color values to darken the background image.
This is similar to the photographic dark room technique of burning by increasing the exposure
of an area of a print.
– Darken: Darken looks at the color information in each channel and selects the background or
foreground image’s color value, whichever is darker, as the result color. Pixels lighter than the
merged colors are replaced, and pixels darker than the merged color do not change.
– Lighten: Lighten looks at the color information in each channel and selects the background or
foreground image’s color values, whichever is lighter, as the result color value. Pixels darker
than the merged color are replaced, and pixels lighter than the merged color do not change.
– Difference: Difference looks at the color information in each channel and subtracts the
foreground color values from the background color values or the background from the
foreground, depending on which has the greater brightness value. Merging with white inverts
the color. Merging with black produces no change.
– Exclusion: Exclusion creates an effect similar to, but lower in contrast than, the Difference
mode. Merging with white inverts the base color values. Merging with black produces no
change.
– Hue: Hue creates a result color with the luminance and saturation of the background color
values and the hue of the foreground color values.
– Saturation: Saturation creates a result color with the luminance and hue of the base color and
the saturation of the blend color.
– Color: Color creates a result color with the luminance of the background color value and the
hue and saturation of the foreground. This preserves the gray levels in the image and is useful
for coloring monochrome images.

Chapter 35 Composite Nodes 898

– Luminosity: Luminosity creates a result color with the hue and saturation of the background color
values and the luminance of the foreground color values. This mode creates an inverse effect from
that of the Color mode.
– Operator Modes: This menu is used to select the Operation mode of the merge. Changing
the Operation mode changes how the foreground and background are combined to produce
a result. This pop-up menu is visible only when the Merge node’s Apply mode is set to either
Normal or Screen.
For an excellent description of the math underlying the Operation modes, read Compositing
Digital Images, Porter, Thomas, and T. Duff, ACM SIGGRAPH Computer Graphics proceedings,
1984, pages 253-259. Essentially, the math is as described below. Note that some modes not
listed in the Operator drop-down menu (Under, In, Held In, Below) are easily obtained by swapping
the foreground and background inputs (with Command-T or Ctrl-T) and choosing a corresponding
mode. The formula used to combine pixels in the merge is always (fg * x) + (bg * y). The different
operations determine exactly what x and y are, as shown in the description for each mode.
The Operator modes are as follows:
– Over: The Over mode adds the foreground layer to the background layer by replacing the
pixels in the background with the pixels from the Z wherever the foreground’s Alpha channel is
greater than 1.

x = 1, y = 1-[foreground Alpha]

– In: The In mode multiplies the Alpha channel of the background input against the pixels in
the foreground. The color channels of the foreground input are ignored. Only pixels from the
foreground are seen in the final output. This essentially clips the foreground using the mask
from the background.

x = [background Alpha], y = 0

– Held Out: Held Out is essentially the opposite of the In operation. The pixels in the foreground
image are multiplied against the inverted Alpha channel of the background image. Accomplish
exactly the same result using the In operation and a Matte Control node to invert the matte
channel of the background image.

x = 1-[background Alpha], y = 0

– Atop: Atop places the foreground over the background only where the background
has a matte.

x = [background Alpha], y = 1-[foreground Alpha]

– XOr: XOr combines the foreground with the background wherever either the foreground or the
background has a matte, but never where both have a matte.

x = 1-[background Alpha], y = 1-[foreground Alpha]

– Subtractive/Additive slider: This slider controls whether Fusion performs an Additive merge, a
Subtractive merge, or a blend of both. This slider defaults to Additive merging for most operations,
assuming the input images are premultiplied (which is usually the case). If you don’t understand
the difference between Additive and Subtractive merging, here’s a quick explanation.
– An Additive merge is necessary when the foreground image is premultiplied, meaning that the
pixels in the color channels have been multiplied by the pixels in the Alpha channel. The result
is that transparent pixels are always black, since any number multiplied by 0 always equals 0.
This obscures the background (by multiplying with the inverse of the foreground Alpha), and
then simply adds the pixels from the foreground.

Chapter 35 Composite Nodes 899

 – A Subtractive merge is necessary if the foreground image is not pre-multiplied. The compositing
method is similar to an additive merge, but the foreground image is first multiplied by its own
Alpha to eliminate any background pixels outside the Alpha area.
While the Additive/Subtractive option could easily have been a checkbox to select one mode or
another, the Merge node lets you blend between the Additive and Subtractive versions of the
merge operation—an operation that is occasionally useful for dealing with problem composites
with edges that are calling attention to themselves as too bright or too dark.
For example, using Subtractive merging on a premultiplied image may lead to darker edges,
whereas using Additive merging with a non-premultiplied image will cause any non-black area
outside the foreground’s Alpha to be added to the result, thereby lightening the edges. By
blending between Additive and Subtractive, you can tweak the edge brightness to be just right for
your situation.

– Alpha Gain slider: Alpha Gain linearly scales the values of the foreground’s Alpha channel. In
Subtractive merges, this controls the density of the composite, similarly to Blend. In Additive
merges, this effectively reduces the amount that the background is obscured, thus brightening the
overall result. In an Additive merge with Alpha Gain set to 0.0, the foreground pixels are simply
added to the background.
– Burn In slider: The Burn In control adjusts the amount of Alpha used to darken the background,
without affecting the amount of foreground added in. At 0.0, the merge behaves like a straight
Alpha blend, whereas at 1.0, the foreground is effectively added onto the background (after Alpha
multiplication if in Subtractive mode). This gives the effect of the foreground image brightening
the background image, as with Alpha Gain. For Additive merges, increasing the Burn In gives an
identical result to decreasing Alpha Gain.
– Blend slider: This is a cloned instance of the Blend slider in the Common Controls tab. Changes
made to this control are simultaneously made to the one in the common controls. The Blend slider
mixes the result of the node with its input, blending back the effect at any value less than 1.0. In
this case, it will blend the background with the merged result.

Additional Controls
The remaining controls let you fine-tune the results of the above settings.
– Filter Method: For input images that are being resized, this setting lets you choose the filter
method used to interpolate image pixels when resizing clips. The default setting is Linear. Different
settings work better for different kinds of resizing. Most of these filters are useful only when
making an image larger. When shrinking images, it is common to use the Linear filter; however,
the Catmull-Rom filter will apply some sharpening to the results and may be useful for preserving
detail when scaling down an image.
– Nearest Neighbor: This skips or duplicates pixels as needed. This produces the fastest but
crudest results.
– Box: This is a simple interpolation resize of the image.
– Linear: This uses a simplistic filter, which produces relatively clean and fast results.
– Quadratic: This filter produces a nominal result. It offers a good compromise between speed
and quality.
– Cubic: This produces better results with continuous-tone images. If the images have fine detail
in them, the results may be blurrier than desired.
– Catmull-Rom: This produces good results with continuous-tone images that are resized down.
Produces sharp results with finely detailed images.
– Gaussian: This is very similar in speed and quality to Bi-Cubic.

Chapter 35 Composite Nodes 900

 – Mitchell: This is similar to Catmull-Rom but produces better results with finely detailed images.
It is slower than Catmull-Rom.
– Lanczos: This is very similar to Mitchell and Catmull-Rom but is a little cleaner and also slower.
– Sinc: This is an advanced filter that produces very sharp, detailed results; however, it may
produce visible “ringing” in some situations.
– Bessel: This is similar to the Sinc filter but may be slightly faster.
– Window Method: Some filters, such as Sinc and Bessel, require an infinite number of pixels to
calculate exactly. To speed up this operation, a windowing function is used to approximate the
filter and limit the number of pixels required. This control appears when a filter that requires
windowing is selected.
– Hanning: This is a simple tapered window.
– Hamming: Hamming is a slightly tweaked version of Hanning.
– Blackman: A window with a more sharply tapered falloff.
– Kaiser: A more complex window, with results between Hamming and Blackman.

Resize Filters from left to right: Nearest Neighbor, Box, Linear, Quadratic, Cubic,
Catmull-Rom, Gaussian, Mitchell, Lanczos, Sinc, and Bessel

– Edges Buttons: Four buttons let you choose how to handle the space around images that are
smaller than the current DoD of the canvas as defined by the resolution of the background image.
– Canvas: The area outside the frame is set to the current color/opacity of the canvas. If you want
to change this value, you can attach a Set Canvas Color node between the image connected to
the foreground input and the foreground input itself, using Set Canvas Color to choose a color
and/or transparency setting with which to fill the canvas.
– Wrap: Creates a “video wall” effect by duplicating the foreground image as a grid.
– Duplicate: Duplicates the outermost pixels along the edge of the foreground image, duplicating
them to stretch up, down, left, and right from each side to reach the end of the DoD.
– Mirror: Similar to duplicate, except every other iteration of the foreground image is flipped and
flopped to create a repeating pattern.

– Invert Transform: Select the Invert Transform control to invert any position, rotation, or scaling
transformation. This option is useful when connecting the merge to the position of a tracker for
match moving.
– Flatten Transform: The Flatten Transform option prevents this node from concatenating its
transformation with subsequent nodes. The node may still concatenate transforms from its input,
but it will not concatenate its transformation with the node at its output.
– Reference Size: The controls under Reference Size do not directly affect the image. Instead, they
allow you to control how Fusion represents the position of the Merge node’s center.
Normally, coordinates are represented as values between 0 and 1, where 1 is a distance equal to
the full width or height of the image. This allows resolution independence, because the size of the
image can be changed without having to change the value of the center.
One disadvantage to this approach is that it complicates making pixel-accurate adjustments to an
image. To demonstrate, imagine an image that is 100 x 100 pixels in size. To move the center of the
foreground element to the right by 5 pixels, we would change the X value of the merge center
from 0.5, 0.5 to 0.55, 0.5. We know the change must be 0.05 because 5/100 = 0.05.

Chapter 35 Composite Nodes 901

 If you specify the dimensions of the background image in the Reference Size controls, this
changes the way the center control values are displayed so that it shows the actual pixel positions
in its X and Y fields.
Extending the example, set the width and height to 100 each and the center will now be shown as
50, 50, and we would move it 5 pixels toward the right by entering 55, 50.
Internally, the Merge node still stores this value as a number between 0 to 1 and, if the center
control’s value was to be queried via scripting or the center control was to be published for use by
other nodes, the original normalized value would be retrieved. The change is only visible in the
value shown for merge center in the node control.
– Use Frame Format Settings: Select this to force the merge to use the composition’s current
frame format settings to set the reference width and reference height values.
– Width and Height: Set these sliders to the width and height of the image to change the way
that Fusion displays the values of the Merge node’s center control.

Channels Tab

Merge node Channels tab

The Channels tab has controls that let the Merge node use Z-channels embedded within each image
to define what’s in front and what’s behind during a Merge operation. The following controls let you
customize the result.
– Perform Depth Merge: Off by default. When turned on, the Z-channel of both images will be used
to determine the composite order. Alpha channels are still used to define transparency, but the
values of the Z-Depth channels will determine the ordering of image elements, front to back. If a
Z-channel is not available for either image, the setting of this checkbox will be ignored, and no
depth compositing will take place. If Z-Depth channels are available, turning this checkbox off
disables their use within this operation.
– Foreground Z-Offset: This slider sets an offset applied to the foreground image’s Z value. Click
the Pick button to pick a value from a displayed image’s Z-channel, or enter a value using the slider
or input boxes. Raising the value causes the foreground image’s Z-channel to be offset further
away along the Z-axis, whereas lowering the value causes the foreground to move closer.
– Subtractive/Additive: When Z-compositing, it is possible for image pixels from the background to
be composited in the foreground of the output because the Z-buffer for that pixel is closer than the
Z of the foreground pixel. This slider controls whether these pixels are merged in an Additive or a
Subtractive mode, in exactly the same way as the comparable slider in the Merge tab.
When merged over a background of a different color, the original background will still be visible in
the semitransparent areas. An Additive merge will maintain the transparencies of the image but
will add their values to the background.

Chapter 35 Composite Nodes 902

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in both the Dissolve and Merge nodes. These
common controls are described in the following “The Common Controls” section.

The Common Controls

The Merge and Dissolve nodes share several identical controls in the Inspector. This section describes
controls that are common among those two nodes.

Inspector

Color Setting Inspector

Settings Tab
The Settings tab in the Inspector can be found on both tools in the Composite category. The Settings
controls are even found on third-party color type plug-in tools. The controls are consistent and work
the same way for each tool, although some tools do include one or two individual options that are also
covered here.

Blend
The Blend control is used to blend between the tool’s original image input and the tool’s final modified
output image. When the blend value is 0.0, the outgoing image is identical to the incoming image.
Normally, this will cause the tool to skip processing entirely, copying the input straight to the output.

Chapter 35 Composite Nodes 903

Process When Blend Is 0.0
The tool is processed even when the input value is zero. This can be useful if processing of this node
is scripted to trigger another task, but the value of the node is set to 0.0.

Red/Green/Blue/Alpha Channel Selector

These four buttons are used to limit the effect of the tool to specified color channels. This filter is often
applied after the tool has been processed.
For example, if the red button on a blur tool is deselected, the blur will first be applied to the image,
then the red channel from the original input will be copied back over the red channel of the result.
There are some exceptions, such as tools for which deselecting these channels causes the tool to skip
processing that channel entirely. Tools that do this will generally possess a set of identical RGBA
buttons on the Controls tab in the tool. In this case, the buttons in the Settings and the Controls tabs
are identical.

Apply Mask Inverted

Enabling the Apply Mask Inverted option inverts the complete mask channel for the tool. The mask
channel is the combined result of all masks connected to or generated in a node.

Multiply by Mask
Selecting this option will cause the RGB values of the masked image to be multiplied by the mask
channel’s values. This will cause all pixels of the image not in the mask (i.e., set to 0) to become black/
transparent.

Use Object/Use Material (Checkboxes)

Some 3D software can render to file formats that support additional channels. Notably, the EXR file
format supports Object and Material ID channels, which can be used as a mask for the effect. These
checkboxes determine whether the channels will be used, if present. The specific Material ID or
Object ID affected is chosen using the next set of controls.

Correct Edges
This checkbox appears only when the Use Object or Use Material checkboxes are selected. It toggles
the method used to deal with overlapping edges of objects in a multi-object image. When enabled,
the Coverage and Background Color channels are used to separate and improve the effect around the
edge of the object. If this option disabled (or no Coverage or Background Color channels are
available), aliasing may occur on the edge of the mask.
For more information on Coverage and Background Color channels, see Chapter 18, “Understanding
Image Channels” in the Fusion Reference Manual or Chapter 79 in the DaVinci Resolve
Reference Manual.

Object ID/Material ID (Sliders)

Use these sliders to select which ID will be used to create a mask from the object or material channels
of an image. Use the Sample button in the same way as the Color Picker: to grab IDs from the image
displayed in the view. The image or sequence must have been rendered from a 3D software package
with those channels included.

Use GPU
The Use GPU menu has three settings. Setting the menu to Disable turns off hardware-accelerated
rendering using the graphics card in your computer. Enabled uses the hardware. Auto uses a capable
GPU if one is available and falls back to software rendering when a capable GPU is not available

Chapter 35 Composite Nodes 904

Motion Blur
– Motion Blur: This toggles the rendering of Motion Blur on the tool. When this control is toggled
on, the tool’s predicted motion is used to produce the motion blur caused by the virtual camera’s
shutter. When the control is toggled off, no motion blur is created.
– Quality: Quality determines the number of samples used to create the blur. A quality setting of 2
will cause Fusion to create two samples to either side of an object’s actual motion. Larger values
produce smoother results but increase the render time.
– Shutter Angle: Shutter Angle controls the angle of the virtual shutter used to produce the motion
blur effect. Larger angles create more blur but increase the render times. A value of 360 is the
equivalent of having the shutter open for one whole frame exposure. Higher values are possible
and can be used to create interesting effects.
– Center Bias: Center Bias modifies the position of the center of the motion blur. This allows the
creation of motion trail effects.
– Sample Spread: Adjusting this control modifies the weighting given to each sample. This affects
the brightness of the samples.

Comments
The Comments field is used to add notes to a tool. Click in the field and type the text. When a note is
added to a tool, a small red square appears in the lower left corner of the node when the full tile is
displayed or a small text bubble icon appears on the right when nodes are collapsed. To see the note
in the node editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 35 Composite Nodes 905

Chapter 36

Deep Pixel Nodes

This chapter details Deep Pixel nodes found in Fusion. Deep Pixel nodes are capable
of handling AOVs (Arbitrary Output Variables) from 3D-rendered files. The
abbreviations next to each node name can be used in the Select Tool dialog when
searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Ambient Occlusion [SSAO]  907
Depth Blur [DBl]  910
Fog [Fog]  912
Shader [Shd]  914
Texture [Txr]  917
The Common Controls  919

Chapter 36 Deep Pixel Nodes 906

Ambient Occlusion [SSAO]

The Ambient Occlusion node

Ambient Occlusion Node Introduction

Ambient Occlusion (AO) is the lighting caused when a scene is surrounded by a uniform diffuse
spherical light source. Think of the scene as being surrounded by a humongous sphere that uniformly
emits light from its surface. AO captures the low frequency lighting. It does not capture sharp shadows
or Diffuse or Specular lighting. So, AO is usually combined with Diffuse and Specular lighting to create
a full lighting solution.
The Ambient Occlusion node generates global lighting effects in 3D-rendered scenes as a post effect.
It quickly approximates computationally expensive ray-traced global illumination. Being a post effect, it
exposes similar aliasing issues like the Shader, Texture, and Volume Fog nodes. Hence, artifacts may
appear in certain situations.

Usage
The AO node rarely works out of the box, and requires some tweaking. The setup process involves
adjusting the Kernel Radius and Number Of Samples to get the desired affect.
The Kernel Radius depends on the natural “scale” of the scene. Initially, there might appear to be no
AO at all. In most cases, the Kernel Radius is too small or too big, and working values must be found.

Inputs
There are three inputs on the AO node. The standard effect mask is used to limit the AO effect. The
Input and Camera connections are required. If either of these is not supplied, the node does not
render an image on output.
– Input: This orange input accepts a 2D RGBA image, Z-Depth, and Normals.
– Camera: The green camera input can take either a 3D Scene or a 3D Camera that rendered the
2D image.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the Ambient Occlusion to only those pixels within the mask. An effects mask is
applied to the tool after the tool is processed.

Basic Node Setup

The Ambient Occlusion node is typically placed after a Renderer 3D node. The Renderer 3D node
must have Z-Depth and Normals enabled in its output channels. The Camera3D that is rendered by the
Renderer3D node is then connected to the camera input on the AO node.

Chapter 36 Deep Pixel Nodes 907

Ambient occlusion as a 2D post process

Inspector

Ambient Occlusion controls

Controls Tab
The controls tab includes all the main controls for compositing with AO. It controls the quality and
appearance of the effect.

Output Mode
– Color: Using the Color menu option combines the incoming image with Ambient Occlusion applied.
– AO: This option outputs the pure Ambient Occlusion as a grayscale image. White corresponds
to regions in the image that should be bright, while black correspond to regions that should be
darker. This allows you to create a lighting equation by combining separate ambient/diffuse/
specular passes. Having the AO as a separate buffer allows creative freedom to combine the
passes in various ways.

Chapter 36 Deep Pixel Nodes 908

Kernal Type
To determine the AO, rays are cast from a point on the surface being shaded, outward to a large
enclosed sphere.
The AO factor is determined by the unoccluded rays that reach the sphere.
– Hemisphere: Rays are cast toward a hemisphere oriented to the surfaces normal. This option
is more realistic than Sphere and should be used unless there is a good reason otherwise. Flat
surfaces receive 100% ambient intensity, while other parts are darkened.
– Sphere: Rays are cast toward a sphere centered about the point being shaded. This option is
provided to produce a stylistic effect. Flat surfaces receive 50% ambient intensity, while other
parts are made darker or brighter.

Number of Samples
Increase the samples until artifacts in the AO pass disappear. Higher values can generate better
results but also increase render time.

Kernel Radius
The Kernel Radius controls the size of the filter kernel in 3D space. For each pixel, it controls how far
one searches in 3D space for occluders. The Filter Kernel should be adjusted manually for each
individual scene.
If made too small, nearby occluders can be missed. If made too large, the quality of the AO decreases
and the samples must be increased dramatically to get the quality back.
This value is dependent on the scene Z-depth. That means with huge Z values in the scene, the kernel
size must be large as well. With tiny Z values, a small kernel size like 0.1 should be sufficient.

Lift/Gamma/Tint
You can use the lift, gamma, and tint controls to adjust the AO for artistic effects.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Deep Pixel nodes. These common controls
are described in detail at the end of this chapter in “The Common Controls” section.

TIP: Combining multiple AO passes with different kernel radii can produce better effects.

AO Tips and Limitations

Transparency/Translucency: AO is designed to work with opaque objects. There are known
limitations with transparent receivers and those with transparent occluders. You can work
around some of these limitations by splitting out the transparent/translucent objects into
separate scenes and only computing AO on the opaque objects.

Chapter 36 Deep Pixel Nodes 909

 Particles: Because of the transparency/translucency limitations, do not use AO on particles,
unless the particles are solid opaque geometry. Anti-aliased edges are another form of
transparency, so they also cause problems with AO.
Supersampling: To render anti-aliasing with Ambient Occlusion, enable HiQ for the Z and
Normals pass in the Renderer 3D.
Viewer Dependence: AO methods work in viewer space, and the results are viewer
dependent. This means the amount of darkening can vary depending on the view location,
when in reality it should be constant. If at a point on an object the AO is 0.5, moving the
camera could change it to 0.4.
Baking of AO: The OpenGL UV renderer can be used to bake AO into the textures on models.

Depth Blur [DBl]

The Depth Blur node

Depth Blur Node Introduction

The Depth Blur node is primarily used to create focal length or depth-of-field effects. It blurs
3D-rendered images based on included Z-channel values, and can also be used for general per-pixel
blurring effects through the Blur Channel controls.

Inputs
The Depth Blur node includes three inputs: one for the main image, one for a blur image, and another
for an effect mask to limit the area where the depth blur is applied.
– Input: This orange input is the only required connection. It accepts a 2D image that includes a Z
channel. The Z channel is used to determine the blur amount in different regions of the image.
– Blur Image: If the Blur Image input is connected, channels from the image are used to control
the blur. This allows general 2D per-pixel blurring effects.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the depth blur to only those pixels within the mask. An effects mask is applied to the
tool after the tool is processed.

Basic Node Setup

The Depth Blur node receives the image containing the Z channel. Below, the Z depth channel is
provided in a separate image file and combined with the RGB (beauty) image using a Channels
Booleans tool. Channel Booleans sets the Z buffer channel to copy into the luminance foreground.

Chapter 36 Deep Pixel Nodes 910

Depth Blur is applied to the beauty image based on a Z depth channel

Inspector

Depth Blur controls

Controls Tab
The Controls tab includes parameters for adjusting the amount of blur applied and the depth of the
blurred area. It also includes options for selecting channels other than the Z channel for the blur map.

Filter
This menu selects the filter used for the blur.
– Box: This applies a basic depth-based box blur effect to the image.
– Soften: This applies a depth-based general softening filter effect.
– Super Soften: This applies a depth-based high-quality softening filter effect.

Blur Channel
Select one of these options to determine the channel used to control the level of blur applied to each
pixel. The channel from the main image input is used, unless an image is connected to the node’s
green Blur Image input.

Lock X/Y
When toggled on, this control locks the X and Y Blur sliders together for symmetrical blurring.

Blur Size
This slider is used to set the strength of the horizontal and vertical blurring.

Focal Point
This control is visible only when the Blur channel menu is set to use the Z channel.
Use this control to select the distance of the simulated point of focus. Lowering the value causes the
Focal Point to be closer to the camera; raising the value causes the Focal Point to be farther away.

Chapter 36 Deep Pixel Nodes 911

Depth of Field
This control is used to determine the depth of the area in focus. The focal point is positioned in the
middle of the region, and all pixels with a Z-value within the region stay in focus. For example, if the
focal point were selected from the image and set to a value of 300, and the depth of field is set to 200,
any pixel with a Z-value between 200 and 400 would remain in focus.

Z Scale
Scales the Z-buffer value by the selected amount. Raising the value causes the distances in the
Z-channel to expand. Lowering the value causes them to contract. This is useful for exaggerating the
depth effect. It can also be used to soften the boundaries of the blur. Some images with small depth
values may require the Z-scale to be set quite low, below 1.0.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Deep Pixel nodes. These common controls
are described in detail at the end of this chapter in “The Common Controls” section.

Fog [Fog]

The Fog node

Fog Node Introduction

The Fog node is used to create simulated fog effects on 3D-rendered images that contain a valid
Z-buffer channel. The fog can be placed in front of or behind various elements of a rendered image
based on the selected Z-channel planes.

Inputs
The Fog node includes three inputs: one for the main image with a Z channel, one for a blur image,
and another for an effect mask to limit the area where the depth blur is applied.
– Input: This orange input is the only required connection. It accepts a 2D image that includes a Z
channel. The Z channel is used to determine the fog amount in different regions of the image.
– Blur Image: The green second image input connects an image that is used as the source of the
fog. If no image is provided, the fog consists of a single color. Generally, a noise map of some
sort is connected here.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the fog to only those pixels within the mask. An effects mask is applied to the tool
after the tool is processed.

Basic Node Setup

The Fog node is typically placed after a Renderer 3D node. The Renderer 3D node must have Z-Depth
enabled in its output channels.

Chapter 36 Deep Pixel Nodes 912

A Fog node added as a 2D post process after the Renderer 3D node

Inspector

Fog controls

Controls Tab
The Controls tab includes parameters for adjusting the density and color of the fog.

Z-Buffer Near Plane and Far Plane

These controls are used to select the extents of the fog within the scene. To pick a value, drag the
Pick button to an area on the image being viewed where the plane is to be located.
The Near Plane is used to select the depth where the fog thins out to nothing. The Far Plane is used to
select the depth at which the fog becomes opaque.

Z Depth Scale
This option scales the Z-buffer values by the selected amount. Raising the value causes the distances
in the Z-channel to expand, whereas lowering the value causes the distances to contract. This is useful
for exaggerating the fog effect.

Chapter 36 Deep Pixel Nodes 913

Fog Color
This option displays and controls the current fog color. Alpha adjusts the fog’s transparency value.

Fog Opacity
Use this control to adjust the opacity on all channels of the fog.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Deep Pixel nodes. These common controls
are described in detail at the end of this chapter in “The Common Controls” section.

Shader [Shd]

The Shader node

Shader Node Introduction

The Shader node can control the lighting, reflection mapping, and 3D shading of elements in a
rendered image. The node relies on the presence of the normal map channel in a rendered image. If
this channel is not present, this node has no effect.

Inputs
The Shader node includes three inputs: one for the main image with normal map channels, one for a
reflection map, and another for an effect mask to limit the area where the depth blur is applied.
– Input: This orange input is the only required connection. It accepts a 2D image that includes a
normals channel.
– Reflection Map Image: The green reflection map image input projects an image onto all
elements in the scene or to elements selected by the Object and Material ID channels in
the Common Controls. Reflection maps work best as 32-bit floating point, equirectangular
formatted images
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the shader to only those pixels within the mask. An effects mask is applied to the
tool after the tool is processed.

Basic Node Setup

The Shader node is inserted after a 2D image that contains a Normals channel. Below, a Renderer 3D
is used to add a Normals channel to an image. The Shader node uses the normals for refining the
surface appearance with a reflection map connected.

Chapter 36 Deep Pixel Nodes 914

The Shader node using normals from a Renderer 3D and a reflection input

Inspector

Shader controls

Controls Tab
The Controls tab for the Shader node includes parameters for adjusting the overall surface reaction to
light sources. You can modify the ambient, diffuse, specular, and reflection properties of the image
connected to the orange image input.

Light Tab
The Controls tab includes parameters for basic lighting brightness and reflections.

Ambient
Ambient controls the Ambient color present in the scene or the selected object. This is a base level of
light added to all pixels, even in completely shadowed areas.

Diffuse
This option controls the Diffuse color present in the scene or for the selected object. This is the normal
color of the object, reflected equally in all directions.

Specular
This option controls the Specular color present in the scene or for the selected object. This is the color
of the glossy highlights reflected toward the eye from a light source.

Reflection
This option controls the Reflection contribution in the scene or for the selected object. High levels
make objects appear mirrored, while low levels overlay subtle reflections giving a polished effect. It
has no effect if no reflection map is connected.

Chapter 36 Deep Pixel Nodes 915

Reflection Type
This menu determines the type of reflection mapping used to project the image in the second input.
– Screen: Screen causes the reflection map to appear as if it were projected on to a screen behind
the point of view.
– Spherical: Spherical causes the reflection map to appear as if it were projected on to a huge
sphere around the whole scene.
– Refraction: Refraction causes the reflection map to appear as if it were refracting or distorting
according to the geometry in the scene.

Equator Angle
Equator Angle controls the left to right angle of the light generated and mapped by the Shader node
for the scene or the selected object.

Polar Height
Polar Height controls the top to bottom angle of the light generated and mapped by the Shader node
for the scene or the selected object.

Shader tab controls

Shader Tab
The Shader tab is used to adjust the falloff of the Diffuse and Specular light and the tint color of the
specular highlight.

Diffuse and Specular

When enabled, these checkboxes allow you to edit the Diffuse and/or Specular Shader curves in the
Shader spline window.

In and Out
These options are used to display and edit point values on the spline.

Chapter 36 Deep Pixel Nodes 916

Specular Color
Use the Diffuse curve to manipulate the diffuse shading and the Specular curve to affect the specular
shading. Drag a box over several points to group-select them. Right-clicking displays a menu with
options for adjusting the spline curves.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Deep Pixel nodes. These common controls
are described in detail at the end of this chapter in “The Common Controls” section.

Texture [Txr]

The Texture node

Texture Node Introduction

The Texture node controls the texture mapping of elements in a rendered image. The Texture node
relies on the presence of U and V Map channels in a 3D-rendered image connected to the main Image
input. If these channels are not present, this node has no effect.

NOTE: Background pixels may have U and V values of 0.0, which set those pixels to the color
of the texture’s corner pixel. To restrict texturing to specific objects, use an effect mask based
on the Alpha of the object or its Object or Material ID channel.

Inputs
The Texture node includes three inputs: one for the main image with UV map channels, one for a
texture map image, and another for an effect mask to limit the area where the replace texture
is applied.
– Input: This orange input accepts a 2D image that includes UV channels. If the UV channels are
not in the images, this node has no effect.
– Texture: The green texture map input provides the texture that is wrapped around objects,
replacing the current texture.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the texture to only those pixels within the mask. An effects mask is applied to the
tool after the tool is processed.

Chapter 36 Deep Pixel Nodes 917

Basic Node Setup
The Texture node is inserted after a 2D image that contains a Texture UV channel. Below, a Renderer
3D is used to add texture coordinates to 3D text. The Texture node can then be used to manipulate
those coordinates using a new texture connected to the green texture input.

A Texture node used to manipulate the texture coordinates and add texture to a Text 3D node

Inspector

Texture controls

Texture Tab
The Texture tab controls allow you to flip, swap, scale, and offset the UV texture image connected to
the texture input.

Flip Horizontal and Vertical

Use these two buttons to flip the texture image horizontally and/or vertically.

Swap UV
When this checkbox is selected, the U and V channels of the source image are swapped.

Rotate 90
The texture map image is rotated 90 degrees when this checkbox is enabled.

U and V Scale
These controls change the scaling of the U and V coordinates used to map the texture. Changing
these values effectively enlarges and shrinks the texture map as it is applied.

U and V Offset
Adjust these controls to offset the U and V coordinates. Changing the values causes the texture to
appear to move along the geometry of the object.

Chapter 36 Deep Pixel Nodes 918

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Deep Pixel nodes. These common controls
are described in the following “The Common Controls” section.

The Common Controls

Nodes that handle Deep Pixel compositing operations share several identical controls in the Inspector.
This section describes controls that are common among Deep Pixel nodes.

Inspector

Common Settings tab in Deep Pixel Nodes

Settings Tab
The Settings tab in the Inspector can be found on every tool in the Deep Pixel category. The Settings
controls are even found on third-party Deep Pixel-type plug-in tools. The controls are consistent and
work the same way for each tool although some tools do include one or two individual options that are
also covered here.

Blend
The Blend control is used to blend between the tool’s original image input and the tool’s final modified
output image. When the blend value is 0.0, the outgoing image is identical to the incoming image.
Normally, this causes the tool to skip processing entirely, copying the input straight to the output.

Process When Blend Is 0.0

The tool is processed even when the input value is zero. This can be useful if processing of this node
is scripted to trigger another task, but the value of the node is set to 0.0.

Chapter 36 Deep Pixel Nodes 919

Red/Green/Blue/Alpha Channel Selector
These four buttons are used to limit the effect of the tool to specified color channels. This filter is often
applied after the tool has been processed.
For example, if the red button on a Blur tool is deselected, the blur is first applied to the image, and
then the red channel from the original input is copied back over the red channel of the result.
There are some exceptions, such as tools for which deselecting these channels causes the tool to skip
processing that channel entirely. Tools that do this generally possess a set of identical RGBA buttons
on the Controls tab in the tool. In this case, the buttons in the Settings and the Controls tabs are
identical.

Apply Mask Inverted

Enabling the Apply Mask Inverted option inverts the complete mask channel for the tool. The mask
channel is the combined result of all masks connected to or generated in a node.

Multiply by Mask
Selecting this option causes the RGB values of the masked image to be multiplied by the mask
channel’s values. This causes all pixels of the image not in the mask (i.e., set to 0) to become black/
transparent.

Use Object/Use Material (Checkboxes)

Some 3D software can render to file formats that support additional channels. Notably, the EXR file
format supports Object and Material ID channels, which can be used as a mask for the effect. These
checkboxes determine whether the channels are used if present. The specific Material ID or Object ID
affected is chosen using the next set of controls.

Correct Edges
This checkbox appears only when the Use Object or Use Material checkboxes are selected. It toggles
the method used to deal with overlapping edges of objects in a multi-object image. When enabled,
the Coverage and Background Color channels are used to separate and improve the effect around the
edge of the object. If this option is disabled (or no Coverage or Background Color channels are
available), aliasing may occur on the edge of the mask.
For more information, see Chapter 18, “Understanding Image Channels” in the Fusion Studio
Reference Manual or Chapter 79 in the DaVinci Resolve Reference Manual.

Object ID/Material ID (Sliders)

Use these sliders to select which ID is used to create a mask from the object or material channels of an
image. Use the Sample button in the same way as the Color Picker: to grab IDs from the image
displayed in the viewer. The image or sequence must have been rendered from a 3D software
package with those channels included.

Use GPU
The Use GPU menu has three settings. Setting the menu to Disable turns off hardware-accelerated
rendering using the graphics card in your computer. Enabled uses the hardware. Auto uses a capable
GPU if one is available, but falls back to software rendering when a capable GPU is not available.

Chapter 36 Deep Pixel Nodes 920

Motion Blur
– Motion Blur: This toggles the rendering of Motion Blur on the tool. When this control is toggled
on, the tool’s predicted motion is used to produce the motion blur caused by the virtual camera’s
shutter. When the control is toggled off, no motion blur is created.
– Quality: Quality determines the number of samples used to create the blur. A quality setting of
2 causes Fusion to create two samples to either side of an object’s actual motion. Larger values
produce smoother results but increase the render time.
– Shutter Angle: Shutter Angle controls the angle of the virtual shutter used to produce the motion
blur effect. Larger angles create more blur but increase the render times. A value of 360 is the
equivalent of having the shutter open for one whole frame exposure. Higher values are possible
and can be used to create interesting effects.
– Center Bias: Center Bias modifies the position of the center of the motion blur. This allows the
creation of motion trail effects.
– Sample Spread: Adjusting this control modifies the weighting given to each sample. This affects
the brightness of the samples.

Comments
The Comments field is used to add notes to a tool. Click in the field and type the text. When a note is
added to a tool, a small red square appears in the lower-left corner of the node when the full tile is
displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the note
in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 36 Deep Pixel Nodes 921

Chapter 37

Effect Nodes
This chapter details the Effect nodes in Fusion.

The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Duplicate [Dup]  923
Highlight [HIL]  929
Hot Spot [HOT]  931
Pseudo Color [PSCL]  937
Rays [CIR]  938
Shadow [SH]  940
Trails [TRLS]  942
TV [TV]  947
The Common Controls  950

Chapter 37 Effect Nodes 922

Duplicate [Dup]

The Duplicate node

Duplicate Node Introduction

Similar to the Duplicate 3D node, the Duplicate node can be used to quickly duplicate any 2D image,
applying a successive transformation to each, and creating repeating patterns and complex arrays of
objects. The options in the Jitter tab allow for non-uniform transformations, such as random
positioning or sizes.

Inputs
The two inputs on the Duplicate node are used to connect a 2D image and an effect mask, which can
be used to limit the area where duplicated objects appear.
– Input: The orange input is used for the primary 2D image that is duplicated.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the duplicated
objects to appear only those pixels within the mask. An effects mask is applied to the tool after
the tool is processed.

Basic Node Setup

The Duplicate node can be used in a variety of different ways and with a variety of different inputs.
Below, to create motion graphics, a masked Background node creates a circular shape that is
duplicated in the Duplicate node.

A Duplicate node used to create a repeating circular object

Chapter 37 Effect Nodes 923

Inspector

Duplicate controls

Controls Tab
The Controls tab includes all the parameters you can use to create, offset, and scale copies of the
object connected to the input on the node.

Copies
Use this slider to set the number of copies made. Each copy is a copy of the last copy. So, when set to
5, the parent is copied, then the copy is copied, then the copy of the copy is copied, and so on. This
allows for some interesting effects when transformations are applied to each copy using the
following controls.

Time Offset
Use the Time Offset slider to offset any animations that are applied to the original image by a set
amount per copy. For example, set the value to -1.0 and use a square set to rotate on the Y-axis as the
source. The first copy shows the animation from a frame earlier. The second copy shows animation
from a frame before that, and so forth. This can be used with great effect on textured planes, for
example, where successive frames of a clip can be shown.

Center
The X and Y Center controls set the offset position applied to each copy. An X offset of 1 would offset
each copy 1 unit along the X-axis from the last copy.

Pivot
The Pivot controls determine the position of the pivot point used when changing the size, position, or
angle of each copy. The pivot does not move with the original object or the duplicated array. To have
the pivot follow the army, you must modify the pivot controls.

Size
The Size control determines how much scaling to apply to each copy.

Angle
The Angle control sets the amount of Z rotation applied to each copy. The angle adjustment is linear
based on the location of the pivot point.

Chapter 37 Effect Nodes 924

Angle adjustment with centered pivot

Angle adjustment with offset pivot

Apply Mode
The Apply Mode setting determines the math used when blending or combining duplicated objects
that overlap.
– Normal: The default mode uses the foreground object’s Alpha channel as a mask to determine
which pixels are transparent and which are not. When this is active, another menu shows possible
operations, including Over, In, Held Out, Atop, and XOr.
– Screen: Screen blends the objects based on a multiplication of their color values. The Alpha
channel is ignored, and layer order becomes irrelevant. The resulting color is always lighter.
Screening with black leaves the color unchanged, whereas screening with white always produces
white. This effect creates a similar look to projecting several film frames onto the same surface.
When this is active, another menu shows possible operations, including Over, In, Held Out, Atop,
and XOr.
– Dissolve: Dissolve mixes overlapping objects. It uses a calculated average of the objects to
perform the mixture.
– Multiply: Multiplies the values of a color channel. This gives the appearance of darkening the
object as the values are scaled from 0 to 1. White has a value of 1, so the result would be the same.
Gray has a value of 0.5, so the result would be a darker object or, in other words, an object half as
bright.
– Overlay: Overlay multiplies or screens the color values of the foreground object, depending
on the color values of the object behind. Patterns or colors overlay the existing pixels while
preserving the highlights and shadows of the color values of the objects behind the foreground
objects. Objects behind other objects are not replaced but mixed with the front objects to reflect
the original lightness or darkness of the objects behind.
– Soft Light: Soft Light darkens or lightens the foreground object, depending on the color values of
the objects behind them. The effect is similar to shining a diffused spotlight on the image.
– Hard Light: Hard Light multiplies or screens the color values of the foreground object, depending
on the color values of the objects behind them. The effect is similar to shining a harsh spotlight on
the image.
– Color Dodge: Color Dodge uses the foreground object’s color values to brighten the objects
behind them. This is similar to the photographic practice of dodging by reducing the exposure of
an area of a print.
– Color Burn: Color Burn uses the foreground object’s color values to darken the objects behind
them. This is similar to the photographic practice of burning by increasing the exposure of an area
of a print.
– Darken: Darken looks at the color information in each channel and selects the object’s foreground
or background’s color value, whichever is darker, as the result color. Pixels lighter than the
blended colors are replaced, and pixels darker than the blended color do not change.

Chapter 37 Effect Nodes 925

 – Lighten: Lighten looks at the color information in each channel and selects the object’s foreground
or background’s color values, whichever is lighter, as the result color value. Pixels darker than the
blended color are replaced, and pixels lighter than the blended color do not change.
– Difference: Difference looks at the color information in each channel and subtracts the foreground
object’s color values from the background object’s color values or the behind object’s values from
the foreground object’s values, depending on which has the higher brightness value. Blending
with white inverts the color. Blending with black produces no change.
– Exclusion: Exclusion creates an effect similar to but lower in contrast than the Difference mode.
Blending with white inverts the base color values. Blending with black produces no change.
– Hue: Hue creates a result color with the luminance and saturation of the background objects color
values and the hue of the foreground object’s color values.
– Saturation: Saturation creates a result color with the luminance and hue of the base color and the
saturation of the blend color.
– Color: Color creates a result color with the luminance of the background object’s color value
and the hue and saturation of the objects in the foreground. This preserves the gray levels in the
image and is useful for colorizing monochrome objects.
– Luminosity: Luminosity creates a color using the hue and saturation of the background object
and the luminance of the foreground object. This mode creates an inverse effect from that of the
Color mode.

Operator
This menu is used to select the Operation mode used when the duplicate objects overlap. Changing
the Operation mode changes how the overlapping objects are combined. This drop-down menu is
visible only when the Apply mode is set to Normal.
The formula used to combine pixels in the Duplicate node is always (fg object * x) + (bg object * y). The
different operations determine what x and y are, as shown in the description for each mode.
The Operator Modes are as follows:
– Over: The Over mode adds the foreground object to the background object by replacing the
pixels in the background with the pixels from the Z wherever the foreground object’s Alpha
channel is greater than 1.
x = 1, y = 1 - [foreground object Alpha]

– In: The In mode multiplies the Alpha channel of the background object against the pixels in the
foreground object. The color channels of the foreground object are ignored. Only pixels from the
foreground object are seen in the final output. This essentially clips the foreground object using
the mask from the background object.
x = [background Alpha], y = 0

– Held Out: Held Out is essentially the opposite of the In operation. The pixels in the foreground
object are multiplied against the inverted Alpha channel of the background object.
x = 1 - [background Alpha], y = 0

– Atop: Atop places the foreground object over the background object only where the background
object has a matte.
x = [background Alpha], y = 1 - [foreground Alpha]

– XOr: XOr combines the foreground object with the background object wherever either the
foreground or the background have a matte, but never where both have a matte.
x = 1 - [background Alpha], y = 1-[foreground Alpha]

Chapter 37 Effect Nodes 926

Subtractive/Additive
This slider controls whether Fusion performs an Additive composite, a Subtractive composite, or a
blend of both when the duplicate objects overlap. This slider defaults to Additive assuming the input
image’s Alpha channel is premultiplied (which is usually the case). If you don’t understand the
difference between Additive and Subtractive compositing, here’s a quick explanation.
An Additive blend operation is necessary when the foreground image is premultiplied, meaning that
the pixels in the color channels have been multiplied by the pixels in the Alpha channel. The result is
that transparent pixels are always black since any number multiplied by 0 always equals 0. This
obscures the background (by multiplying with the inverse of the foreground Alpha), and then adds the
pixels from the foreground.
A Subtractive blend operation is necessary if the foreground image is not premultiplied. The
compositing method is similar to an additive composite, but the foreground image is first multiplied by
its Alpha, to eliminate any background pixels outside the Alpha area.
While the Additive/Subtractive option is often an either/or mode in most other applications, the
Duplicate node lets you blend between the Additive and Subtractive versions of the compositing
operation. This can be useful for dealing with problem composites with bright or dark edges.
For example, using Subtractive merging on a premultiplied image may result in darker edges, whereas
using Additive merging with a non-premultiplied image causes any non-black area outside the
foreground’s Alpha to be added to the result, thereby lightening the edges. By blending between
Additive and Subtractive, you can tweak the edge brightness to be just right for your situation.

Gain
The Gain RGB controls multiply the values of the image channel linearly. All pixels are multiplied by the
same factor, but the effect is larger on bright pixels and smaller on dark pixels. Black pixels are not
changed since multiplying any number times 0 always equals 0.
Alpha Gain linearly scales the Alpha channel values of objects in front. This effectively reduces the
amount that the objects in the background are obscured, thus brightening the overall result. When the
Subtractive/Additive slider is set to Additive with Alpha Gain set to 0.0, the foreground pixels are
simply added to the background.
When Subtractive/Additive slider is set to Subtractive, this controls the density of the composite,
similarly to Blend.

Burn In
The Burn In control adjusts the amount of Alpha used to darken the objects that fall behind other
objects, without affecting the amount of foreground objects added. At 0.0, the blending behaves like a
straight Alpha blend, in contrast to a setting of 1.0 where the objects in the front are effectively added
on to the objects in the back (after Alpha multiplication if in Subtractive mode). This gives the effect of
the foreground objects brightening the objects in the back, as with Alpha Gain. In fact, for Additive
blends, increasing the Burn In gives an identical result to decreasing Alpha Gain.

Blend
This blend control is different from the Blend slider in the Common Settings tab. Changes made to this
control apply the blend between objects. The Blend slider fades the results of the last object first, the
penultimate after that, and so on. The blending is divided between 0 and 1, with 1 being all objects are
fully opaque and 0 being only the original object showing.

Merge Under
This checkbox reverses the layer order of the duplicated elements, making the last copy the
bottommost layer and the first copy the topmost layer.

Chapter 37 Effect Nodes 927

Jitter Tab
The options in the Jitter tab allow you to randomize the position, rotation, size, and color of all the
copies created in the Controls tab.

r
Duplicate Jitter tab

Random Seed
The Random Seed slider and Reseed button are used to generate a random starting point for the
amount of jitter applied to the duplicated objects. Two Duplicate nodes with identical settings but
different random seeds produce two completely different results.

Center X and Y
Use these two controls to adjust the amount of variation in the X and Y position of the
duplicated objects.

Axis X and Y
Use these two controls to adjust the amount of variation in the rotational pivot center of the duplicated
objects. This affects only the additional jitter rotation, not the rotation produced by the Rotation
settings in the Controls tab.

X Size
Use this control to adjust the amount of variation in the Scale of the duplicated objects.

Angle
Use this dial to adjust the amount of variation in the Z rotation of the duplicated objects.

Gain
The Gain RGBA controls randomly multiply the values of the image channel linearly.

Blend
Changes made to this control randomize the blend between objects.

Common Controls
Settings Tab
The Settings tab controls are common to all Effect nodes, so their descriptions can be found in
“The Common Controls” section at the end of this chapter.

Chapter 37 Effect Nodes 928

Highlight [HIL]

The Highlight node

Highlight Node Introduction

The Highlight filter creates star-shaped highlights or glints in bright regions of the image, similar to a
lens star filter effect.

Inputs
There are three Inputs on the Highlight node: one for the image, one for the effects mask, and another
for a highlight mask.
– Input: The orange input is used for the primary 2D image that gets the highlight applied.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input restricts the highlight
to be within the pixels of the mask. An effects mask is applied to the tool after the tool is
processed.
– Highlight Mask: The Highlight node supports pre-masking using the white highlight mask input.
The image is filtered before the highlight is applied. The highlight is then merged back over the
original image. Unlike regular effect masks, it does not crop off highlights from source pixels
when the highlight extends past the edges of the mask.

Highlight masks are identical to effects masks in every other respect.

Basic Node Setup

The Highlight node below is used to create glint-type highlights on an incoming image. The highlight
mask is used to limit the area where the effect is applied.

A Highlight node applied to an image, with a

highlight mask limiting the area of the effect

Chapter 37 Effect Nodes 929

Inspector

Highlight controls

Controls Tab
The Controls tab includes parameters for the highlight style except for color, which is handled in the
Color Scale tab.

Low and High

This range control designates the range of Luminance values in the image that generates highlights.
Values less than the Low value do not receive highlights. Values above the High value receive the full
highlight effect.

Curve
The Curve value changes the drop-off over the length of the highlight. Higher values cause the
brightness of the flares to drop off closer to the center of the highlight, whereas lower values drop off
farther from the center.

Length
This designates the length of the flares from the highlight.

Number of Points
This determines the number of flares emanating from the highlight.

Angle
Use this control to rotate the highlights.

Merge Over
When enabled, the effect is overlaid on the original image. When disabled, the output is the highlights
only. This is useful for downstream color correction of the highlights.

Highlight Color Scale controls

Chapter 37 Effect Nodes 930

Color Scale Tab
The Color Scale tab controls the color of the highlight.

Red, Green, and Blue Scale

Moving the sliders of one or all of these channels down changes the falloff color of the highlight.

Alpha Scale
Moving the Alpha slider down makes highlight falloff more transparent.

Common Controls
Setting Tab
The Settings tab controls are common to all Effect nodes, so their descriptions can be found in
“The Common Controls” section at the end of this chapter.

Hot Spot [HOT]

The Hot Spot node

Hot Spot Node Introduction

The Hot Spot node is used to create lens flare, spotlight, and burn/dodge effects of various types.
In the real world, lens flares occur when extremely bright light sources in the scene by the reflections
are reflected off elements inside the lens of the camera. One might see lens flares in a shot when
viewing a strong light source through a camera lens, like the sun or another bright star.

Inputs
There are three inputs on the Hot Spot node: one for the image, one for the effects mask, and another
for an Occlusion image.
– Input: The required orange input is used for the primary 2D image that gets
the hot spot applied.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input restricts the hot
spot to be within the pixels of the mask. An effects mask is applied to the tool after the
tool is processed.
– Occlusion: The green Occlusion input accepts an image to provide the occlusion matte.
The matte is used to block the hot spot, causing it to “wink.” The white pixels in the image
occlude the hot spot. Gray pixels partially suppress the hot spot.

Basic Node Setup

The Hot Spot node is not a stand-alone generator, so it must have an image input that gets the hot
spot applied.

Chapter 37 Effect Nodes 931

Hot Spot node applied to an image

Inspector

Hot Spot controls

Hot Spot Tab

The Hotspot tab is used to control the primary and secondary hot spots. You can adjust their position,
size, strength, angle, and apply mode.

Primary Center X and Y

This is the position of the primary hot spot within the scene. Secondary lens elements and reflections
are positioned relative to the position of the primary hot spot.

Primary Strength
This control determines the brightness of the primary hot spot.

Hot Spot Size

This control determines the diameter of the primary hot spot. A value of 1.0 represents a circle the full
width of the image.

Aspect
This controls the aspect of the spot. A value of 1.0 produces a perfectly circular hot spot. Values above
1.0 elongate the circle horizontally, and values below 1.0 elongate the circle vertically.

Aspect Angle
This control can be used to rotate the primary hot spot.

Secondary Strength
This control determines the strength, which is to say the brightness, of the secondary hot spot. The
secondary hot spot is a reflection of the primary hot spot. It is always positioned on the opposite side
of the image from the primary hot spot.

Secondary Size
This determines the size of the secondary hot spot.

Chapter 37 Effect Nodes 932

Apply Mode
This control determines how the hot spot affects the underlying image.
– Add (Burn): This causes the spots created to brighten the image.
– Subtract (Dodge): This causes the spots created to dim the image.
– Multiply (Spotlight): This causes the spots created to isolate a portion of the image with light and
to darken the remainder of the image.

Occlude
This menu is used to select which channel of the image connected to the Hot Spot node’s Occlusion
input is used to provide the occlusion matte. Occlusion can be controlled from Alpha or R, G, or B
channels of any image connected to the Occlusion input on the node’s tile.

Lens Aberration
Aberration changes the shape and behavior of the primary and secondary hot spots.
– In and Out Modes: Elongates the shape of the hot spot into a flare. The hot spot stretches toward
the center when set to In mode and stretches toward the corners when set to Out mode.
– Flare In and Flare Out Modes: This option is a lens distortion effect that is controlled by the
movement of the lens effect. Flare In causes the effect to become more severe, the closer the hot
spot gets to the center. Flare Out causes the effect to increase as the hot spot gets closer to the
edges of the image.
– Lens: This mode emulates a round, ringed lens effect.

Aberration
The Aberration slider controls the overall strength of the lens aberration effect.

Hot Spot color controls

Color Tab
The Color tab is used to modify the color of the primary and secondary hot spots.

Color Mode
This menu allows you to choose between animated or static color modifications using the small curves
editor in the Inspector.
– None: The default None setting retains a static curve adjustment for the entire range.
– Animated Points: This setting allows the color curves in the spline area to be animated over time.
Once this option is selected, moving to the desired frame and making a change in the Spline
Editor sets a keyframe.
– Dissolve mode: Dissolve mode is mostly obsolete and is included for compatibility reasons only.

Chapter 37 Effect Nodes 933

Color Channel and Mix
When selected, these checkboxes enable the editing of the chosen splines in the small Inspector
Spline Editor. The Mix checkbox enables the Mix Spline, which is used to determine the influence of
the controls that the Radial tab has along the radius of the hot spot.

Red, Green, Blue, and Alpha Splines

The Spline Window shows the curves for the individual channels. It is a miniature Spline Editor. The
Red, Green, Blue, and Alpha splines are used to adjust the color of the spotlight along the radius of
the hot spot.
The vertical axis represents the intensity or strength of the color channel. The horizontal axis
represents the hot spot position along the radius, from the left outside edge to the inside right edge.
The default curve indicates that the red, green, blue, and Alpha channels all have a linear falloff.

Mix Spline
The Mix spline is used to determine the influence that the Radial controls have along the radius of the
hot spot. The horizontal axis represents the position along the circle’s circumference, with 0 being 0
degrees and 1.0 being 360 degrees. The vertical axis represents the amount of the radial hot spot to
blend with the color hot spot. A value of 0 is all radial hot spot, while a value of 1.0 is all color hot spot.

NOTE: Right-clicking in the LUT displays a contextual menu with options related to modifying
spline curves.

For more information on the LUT Editor, see Chapter 7, “Using Viewers” in the Fusion Studio Reference
Manual or Chapter 68 in the DaVinci Resolve Reference Manual.

Hot Spot Radial tab

Radial Tab
Radial On
This control enables the Radial splines. Otherwise, the radial matte created by the splines is not
applied to the hot spot, and the Mix spline in the color controls does not affect the hot spot.

Radial Mode
Similar to the Color mode menu, this menu allows you to choose between animated or static radial hot
spot modifications using the small curves editor in the Inspector.

Chapter 37 Effect Nodes 934

 – No Animation: The default setting retains a static curve adjustment for the entire range.
– Animated Points: This setting allows the radial curves in the spline area to be animated over time.
Once this option is selected, moving to the desired frame and making a change in the Spline
Editor sets a keyframe.
– The Interpolated Values option is mostly obsolete and is included for compatibility reasons only.

Radial Length and Radial Density Splines

The Spline window shows curves for the Length and Density of the hot spot. It is a miniature Spline
Editor. The key to these splines is realizing that the horizontal axis in Inspector’s Spline Editor
represents a position around the circumference of the hot spot. A value of 0.0 is 0 degrees, and 1.0 is
360 degrees. With that in mind, the length determines the radius of light making up the hot spot along
the circumference. The density represents how bright the light is along the circumference.

Radial Repeat
This control repeats the effect of the radial splines by x number of times. For example, a repeat of 2.0
causes the spline to take effect between 0 and 180 degrees instead of 0 and 360, repeating the spline
between 180 and 360.

Length Angle
This control rotates the effect of the Radial Length spline around the circumference of the hot spot.

Density Angle
This control rotates the effect of the Radial Density spline around the circumference of the hot spot.

NOTE: Right-clicking in the spline area displays a contextual menu containing options related
to modifying spline curves.

A complete description of LUT Editor controls and options can be found in Chapter 45, “LUT Nodes.”

L1, L2, and L3 Tab

L1 tab L2 tab L3 tab

Chapter 37 Effect Nodes 935

Lens Reflect Tabs
The three Lens Reflect tabs are used to enable and design additional lens flare elements beyond the
primary and secondary hot spots.

Lens Reflect 1-3

Each of these three checkboxes enables a pair of lens reflection elements that you can modify using
the controls in this tab. The parameters affect all the enabled Lens reflection elements in this tab.

Element Strength
This determines the brightness of element reflections.

Element Size
This determines the size of element reflections.

Element Position
This determines the distance of element reflections from the axis. The axis is calculated as a line
between the hot spot position and the center of the image.

Element Type
Use this group of buttons to choose the shape and density of the element reflections. The presets
available are described below.
– Circular: This creates slightly soft-edged circular shaped reflections.
– Soft Circular: This creates very soft-edged circular shaped reflections.
– Circle: This creates a hard-edged circle shape.
– NGon Solid: This creates a filled polygon with a variable number of sides.
– NGon Star: This creates a very soft-edged star shape with a variable number of sides.
– NGon Shaded Out: This creates soft-edged circular shapes.
– NGon Shaded In: This creates a polygon with a variable number of sides, which has a very soft
reversed (dark center, bright radius) circle.

NGon Angle:
This control is used to determine the angle of the NGon shapes.

NGon Sides:
This control is used to determine the number of sides used when the Element Type is set to Ngon Star,
Ngon Shaded Out, and Ngon Shaded In.

NGon Starriness:
This control is used to bend polygons into star shapes. The higher the value, the more star-like
the shape.

Lens Color Controls

These controls determine the color of the lens that affects the colors of the reflections. To choose a
lens color, pick one from a displayed image or enter RGBA values using the sliders or input boxes.

Common Controls
Settings Tab
The Settings tab controls are common to all Effect nodes, so their descriptions can be found in
“The Common Controls” section at the end of this chapter.

Chapter 37 Effect Nodes 936

Pseudo Color [PSCL]

The Pseudo Color node

Pseudo Color Node Introduction

The Pseudo Color node provides the ability to produce variations of an image’s color based on
waveforms generated by the node’s controls. Static or animated variances of the original image can
be produced.

Inputs
There are two Inputs on the Pseudo Color node: one for an image and one for an effects mask.
– Input: The orange input is used for the primary 2D image that gets its color modified.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input restricts the
pseudo color to be within the pixels of the mask. An effects mask is applied to the tool after
the tool is processed.

Basic Node Setup

The Pseudo Color node is not a stand-alone generator, so it must have an image input that it uses to
generate variations in colors.

Pseudo Color node applied to an image

Inspector

Pseudo Color RGBA controls

Chapter 37 Effect Nodes 937

Red/Green/Blue/Alpha Tabs
The node’s controls are separated into four identical tabs, one for each of the RGBA color channels.

Color Checkbox
When enabled, the Pseudo Color node affects this color channel.

Wrap
When enabled, waveform values that exceed allowable parameter values are wrapped to the
opposite extreme.

High and Low

High and Low determine the range to be affected by the node in a specific color channel.

Soft Edge
This slider determines the soft edge of color transition.

Waveform
This selects the type of waveform to be created by the generator. Four waveforms are available: Sine,
Triangle, Sawtooth, and Square.

Frequency
This controls the frequency of the waveform selected. Higher values increase the number of
occurrences of the variances.

Phase
This modifies the Phase of the waveform. Animating this control produces color cycling effects.

Mean
This determines the level of the waveform selected. Higher values increase the overall brightness of
the channel until the allowed maximum is reached.

Amplitude
Amplitude increases or decreases the overall power of the waveform.

Common Controls
Settings Tab
The Settings tab controls are common to all Effect nodes, so their descriptions can be found in “The
Common Controls” section at the end of this chapter.

Rays [CIR]

The Rays node

Rays Node Introduction

Rays is a modified zoom blur effect that radiates through an object from a specified point.

Chapter 37 Effect Nodes 938

Inputs
There are two inputs on the Rays node: one for the image and one for the effects mask.
– Input: The orange input is used for the primary 2D image that gets the rays applied to it.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input restricts the
rays to be within the pixels of the mask. An effects mask is applied to the tool after the
tool is processed.

Basic Node Setup

The Rays node works best when the image or graphic connected to the orange input includes an
Alpha channel from which the rays emit.

Rays node set up to emit from a line of text

Inspector

Rays node controls

Controls Tab
The Controls tab contains all the primary controls necessary for customizing the rays.

Center X and Y
This coordinate control and related viewer crosshair set the center point for the light source.

Blend
Sets the percentage of the original image that’s blended with the light rays.

Decay
Sets the length of the light rays.

Weight
Sets the falloff of the light rays.

Exposure
Sets the intensity level of the light rays.

Threshold
Sets the luminance limit at which the light rays are produced.

Chapter 37 Effect Nodes 939

Common Controls
Settings Tab
The Settings tab controls are common to all Effect nodes, so their descriptions can be found in the
“The Common Controls” section at the end of this chapter.

Shadow [SH]

The Shadow node

Shadow Node Introduction

Shadow is a versatile node used in the creation of a drop shadow, based on the Alpha channel in an
image. Optionally, a second image can be used as a depth matte to distort the shadow based on the
varying depth in a background image.

Input
The three inputs on the Shadow node are used to connect a 2D image that causes the shadow.
A depth map input and an effect mask can be used to limit the area where trails appear. Typically, the
output of the shadow is then merged over the actual background in the composite.
– Input: The orange input is used for the primary 2D image with Alpha channel that is the source
of the shadow.
– Depth: The green Depth map input takes a 2D image as its input and extracts a depth matte
from a selected channel. The light Position and Distance controls can then be used to modify
the appearance of the shadow based on depth.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the area where
the shadow appears. An effects mask is applied to the tool after the tool is processed.

NOTE: The Shadow node is designed to create simple 2D drop shadows. Use a Spot Light
node and an Image Plane 3D node for full 3D shadow casting.

Basic Node Setup

Below, the Shadow node uses the output of an image with Alpha and connects to the foreground of a
Merge. The shadow is shown over the background input to the Merge.

Chapter 37 Effect Nodes 940

A Shadow node is generated from the
MediaIn2 and shown over MediaIn1

Inspector

Shadow node controls

Controls Tab
The Controls tab contains all the primary controls necessary for customizing the shadow appearance.

Shadow Offset
This control sets the X and Y position of the shadow. When the Shadow node is selected, you can also
adjust the position of the Shadow Offset using the crosshair in the viewer.

Softness
Softness controls how blurry the shadow’s edges appear.

Shadow Color
Use this control to select the color of the shadow. The most realistic shadows are usually not totally
black and razor sharp.

Light Position
This control sets the position of the light relative to the shadow-casting object. The Light Position is
only taken into consideration when the Light Distance slider is not set to infinity (1.0).

Chapter 37 Effect Nodes 941

Light Distance
This slider varies the apparent distance of the light between infinity (1.0) and zero distance from the
shadow-casting object. The advantage of setting the Light Distance is that the resulting shadow is
more realistic-looking, with the further parts of the shadow being longer than those that are closer.

Minimum Depth Map Light Distance

This control is active when an image is connected to the shadow’s Depth Map input. The slider is used
to control the amount that the depth map contributes to the Light Distance. Dark areas of a depth map
make the shadow deeper. White areas bring it closer to the camera.

Z Map Channel
This menu is used to select which color channel of the image connected to the node’s Depth Map
input is used to create the shadow’s depth map. Selections exist for the RGB and A, Luminance, and
Z-buffer channels.

Output
This menu determines if the output image contains the image with shadow applied or the shadow only.
The shadow only method is useful when color correction, perspective, or other effects need to be
applied to the resulting shadow before it is merged back with the object.

Common Controls
Settings Tab
The Settings tab controls are common to all Effect nodes, so their descriptions can be found in “The
Common Controls” section at the end of this chapter.

Trails [TRLS]

The Trails node

Trails Node Introduction

The Trails node is used to create a ghost-like after-trail of the image. This creates an interesting effect
when applied to moving images with an Alpha channel. Unlike a directional blur, only the preceding
motion of an image is displayed as part of the effect. Since the trail effect is based on an image buffer,
it requires you to play or activate the pre-roll for some number of frames before you see the effect.

Input
The two inputs on the Trails node are used to connect a 2D image and an effect mask that can be
used to limit the area where trails appear.
– Input: The orange input is used for the primary 2D image that receives the trails applied.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the area where
the trails effect appears. An effects mask is applied to the tool after the tool is processed.

Chapter 37 Effect Nodes 942

Basic Node Setup
The output of an animated Text node is connected to the input of the Trails node. Trails are generated
based on the motion of the text. The Reset button must be pressed in the Inspector between each
preview, or the trails will accumulate.

A Trails node generates trails for

the animation in the Text node

Inspector

Trails node controls

Controls Tab
The Controls tab contains all the primary controls necessary for customizing the trails.

Restart
This control clears the image buffer and displays a clean frame, without any of the ghosting effects.

Preroll
This makes the Trails node pre-render the effect by the number of frames on the slider.

Reset/Preroll on Render
When this checkbox is enabled, the Trails node resets itself when a preview or final render is initiated.
It pre-rolls the designated number of frames.

This Time Only

Selecting this checkbox makes the pre-roll use this current frame only and not the previous frames.

Chapter 37 Effect Nodes 943

Preroll Frames
This determines the number of frames to pre-roll.

Lock RGBA
When selected, this checkbox allows the Gain of the color channels to be controlled independently.
This allows for tinting of the Trails effect.

Gain
The Gain control affects the overall intensity and brightness of the image in the buffer. Lower values in
this parameter create a much shorter, fainter trail, whereas higher values create a longer, more
solid trail.

Rotate
The Rotate control rotates the image in the buffer before the current frame is merged into the effect.
The offset is compounded between each element of the trail. This is different than each element of the
trail rotating on its pivot point. The pivot remains over the original object.

Offset X/Y
These controls offset the image in the buffer before the current frame is merged into the effect.
Control is given over each axis independently. The offset is compounded between each element of
the trail.

Lock Scale X/Y

When selected, this checkbox allows the X- and Y-axis scaling of the image buffer to be manipulated
separately for each axis.

Scale
The Scale control resizes the image in the buffer before the current frame is merged into the effect.
The size is compounded between each element of the trail.

Lock Blur X/Y

When selected, this checkbox allows the blurring of the image buffer to be controlled separately for
each axis.

Blur Size
The Blur Size control applies a blur to the trails in the buffer before the current frame is merged into
the effect. The blur is compounded between each element of the trail.

Apply Mode
The Apply Mode setting determines the math used when blending or combining the trailing objects
that overlap.
– Normal: The default mode uses the foreground object’s Alpha channel as a mask to determine
which pixels are transparent and which are not. When this is active, another menu shows possible
operations, including Over, In, Held Out, Atop, and XOr.
– Screen: Screen blends the objects based on a multiplication of their color values. The Alpha
channel is ignored, and layer order becomes irrelevant. The resulting color is always lighter.
Screening with black leaves the color unchanged, whereas screening with white always produces
white. This effect creates a similar look to projecting several film frames onto the same surface.
When this is active, another menu shows possible operations, including Over, In, Held Out, Atop,
and XOr.
– Dissolve: Dissolve mixes overlapping objects. It uses a calculated average of the objects to
perform the mixture.

Chapter 37 Effect Nodes 944

 – Multiply: Multiplies the values of a color channel. This gives the appearance of darkening the
object as the values are scaled from 0 to 1. White has a value of 1, so the result would be the same.
Gray has a value of 0.5, so the result would be a darker object or, in other words, an object half as
bright.
– Overlay: Overlay multiplies or screens the color values of the foreground object, depending on
the color values of the background object. Patterns or colors overlay the existing pixels while
preserving the highlights and shadows of the color values of the objects behind the foreground
objects. The objects behind the foreground objects are not replaced but mixed with the
foreground objects to reflect the original lightness or darkness of the background objects.
– Soft Light: Soft Light darkens or lightens the foreground object, depending on the color values of
the objects behind them. The effect is similar to shining a diffused spotlight on the image.
– Hard Light: Hard Light multiplies or screens the color values of the foreground object, depending
on the color values of the objects behind them. The effect is similar to shining a harsh spotlight on
the image.
– Color Dodge: Color Dodge uses the foreground object’s color values to brighten the objects
behind them. This is similar to the photographic practice of dodging by reducing the exposure of
an area of a print.
– Color Burn: Color Burn uses the foreground object’s color values to darken the objects behind
them. This is similar to the photographic practice of burning by increasing the exposure of an area
of a print.
– Darken: Darken looks at the color information in each channel and selects the color value from the
object in front or behind, whichever is darker. Pixels lighter than the blended colors are replaced,
and pixels darker than the blended color do not change.
– Lighten: Lighten looks at the color information in each channel and selects the color value
from the object in front or behind, whichever is lighter. Pixels darker than the blended color are
replaced, and pixels lighter than the blended color do not change.
– Difference: Difference looks at the color information in each channel and subtracts the foreground
object’s color values from the background object’s color values or vice versa, depending on
which has the higher brightness value. Blending with white inverts the color. Blending with black
produces no change.
– Exclusion: Exclusion creates an effect similar to but lower in contrast than the Difference mode.
Blending with white inverts the base color values. Blending with black produces no change.
– Hue: Hue creates color with the luminance and saturation of the background object’s color and
the hue of the foreground object’s color.
– Saturation: Saturation creates color with the luminance and hue of the base color and the
saturation of the blend color.
– Color: Color creates color with the luminance of the background object’s color and the hue and
saturation of the object in front. This preserves the gray levels in the image and is useful for
colorizing monochrome objects.
– Luminosity: Luminosity creates color with the hue and saturation of the background object’s color
and the luminance of the foreground object’s color. This mode creates an inverse effect from that
of the Color mode.

Operator
This menu is used to select the Operation mode used when the trailing objects overlap. Changing the
Operation mode changes how the overlapping objects are combined to produce a result. This
drop-down menu is visible only when the Apply mode is set to Normal.
The formula used to combine pixels in the trails node is always (fg object * x) + (bg object * y).
The different operations determine what x and y are, as shown in the description for each mode.

Chapter 37 Effect Nodes 945

The Operator Modes are as follows:
– Over: The Over mode adds the foreground object to the background object by replacing the
pixels in the background with the pixels from the Z wherever the foreground object’s Alpha
channel is greater than 1.
x = 1, y = 1 - [foreground object Alpha]

– In: The In mode multiplies the Alpha channel of the background object against the pixels in the
foreground object. The color channels of the foreground object are ignored. Only pixels from the
foreground object are seen in the final output. This essentially clips the foreground object using
the mask from the background object.
x = [background Alpha], y = 0

– Held Out: Held Out is essentially the opposite of the In operation. The pixels in the foreground
object are multiplied against the inverted Alpha channel of the background object.
x = 1 - [background Alpha], y = 0

– Atop: Atop places the foreground object over the background object only where the background
object has a matte.
x = [background Alpha], y = 1 - [foreground Alpha]

– XOr: XOr combines the foreground object with the background object wherever either the
foreground or the background have a matte, but never where both have a matte.
x = 1 - [background Alpha], y = 1 - [foreground Alpha]

Subtractive/Additive
This slider controls whether Fusion performs an Additive composite, a Subtractive composite, or a
blend of both when the trailing objects overlap. This slider defaults to Additive assuming the input
image’s Alpha channel is premultiplied (which is usually the case). If you don’t understand the
difference between Additive and Subtractive compositing, below is a quick explanation.

NOTE: An Additive blend operation is necessary when the foreground image is premultiplied,
meaning that the pixels in the color channels have been multiplied by the pixels in the Alpha
channel. The result is that transparent pixels are always black since any number multiplied by
0 always equals 0. This obscures the background (by multiplying with the inverse of the
foreground Alpha), and then adds the pixels from the foreground.
A Subtractive blend operation is necessary if the foreground image is not premultiplied.
The compositing method is similar to an additive composite, but the foreground image is first
multiplied by its Alpha, to eliminate any background pixels outside the Alpha area.
Although the Additive/Subtractive option is often an either/or checkbox in other software, the
Trails node lets you blend between the Additive and Subtractive versions of the compositing
operation. This can be useful when dealing with problem edges that are too bright
or too dark.
For example, using Subtractive merging on a premultiplied image may result in darker edges,
whereas using Additive merging with a non-premultiplied image causes any non-black area
outside the foreground’s Alpha to be added to the result, thereby lightening the edges.
By blending between Additive and Subtractive, you can tweak the edge brightness to be just
right for your situation.

Chapter 37 Effect Nodes 946

Alpha Gain
Alpha Gain linearly scales the Alpha channel values of the trailing objects in front. This effectively
reduces the amount that the trailing objects in the background are obscured, thus brightening the
overall result. When the Subtractive/Additive slider is set to Additive with Alpha Gain set to 0.0, the
foreground pixels are added to the background.
When the Subtractive/Additive slider is set to Subtractive, this controls the density of the composite,
similar to Blend.

Burn In
The Burn In control adjusts the amount of Alpha used to darken the objects that trail under other
objects, without affecting the amount of foreground objects added. At 0.0, the blending behaves like a
straight Alpha blend. At 1.0, the objects in the front are effectively added onto the objects in the back
(after Alpha multiplication if in Subtractive mode). This gives the effect of the foreground objects
brightening the objects in the back, as with Alpha Gain. In fact, for Additive blends, increasing the
Burn In gives an identical result to decreasing Alpha Gain.

Merge Under
When enabled, the current image is placed under the generated trail, rather than the usual, over top
operation. The layer order of the trailing elements is also reversed, making the last trail the
topmost layer.

Common Controls
Settings Tab
The Settings tab controls are common to all Effect nodes, so their descriptions can be found in “The
Common Controls” section at the end of this chapter.

TV [TV]

The TV node

TV Node Introduction
The TV node is a simple node designed to mimic some of the typical flaws seen in analog television
broadcasts and screens. This Fusion-specific node is mostly obsolete when using DaVinci Resolve
because of the more advanced Analog Damage ResolveFX.

Input
The two inputs on the TV node are used to connect a 2D image and an effect mask, which can be
used to limit the area where the TV effect appears.
– Input: The orange input is used for the primary 2D image that gets the TV distortion applied.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the area where
the the TV effect to appears. An effects mask is applied to the tool after the tool is processed.

Chapter 37 Effect Nodes 947

Basic Node Setup
The output of an image is connected to the input of the TV node. The style of TV interference is then
customized using the controls in the Inspector.

The TV node simulates TV-style flaws in the

image connected to the orange input

Inspector

TV node controls

Controls Tab
The Controls tab is the first of three tabs used to customize the analog TV distortion. The Controls tab
modifies the scan lines and image distortion of the effect.

Scan Lines
This slider is used to emulate the interlaced look by dropping lines out of the image. Setting it to black,
with a transparent Alpha, drops a line. A value of 1 (default) drops every second line. A value of 2
shows one line, and then drops the second and third and repeats. A value of zero turns off the effect.

Horizontal
Use this slider to apply a simple Horizontal offset to the image.

Vertical
Use this slider to apply a simple Vertical offset to the image.

Skew
This slider is used to apply a diagonal offset to the image. Positive values skew the image to the top
left. Negative values skew the image to the top right. Pixels pushed off frame wrap around and
reappear on the other side of the image.

Amplitude
The Amplitude slider can be used to introduce smooth sine wave-type deformation to the edges of the
image. Higher values increase the intensity of the deformation. Use the Frequency control to
determine how often the distortion is repeated.

Chapter 37 Effect Nodes 948

Frequency
The Frequency slider sets the frequency of the sine wave used to produce distortion along the edges
of the image when the amplitude control is greater than 1.

Offset
Use Offset to adjust the position of the sine wave, causing the deformation applied to the image via
the Amplitude and Frequency controls to see across the image.

The TV Noise tab

Noise Tab
The Noise tab is the second of three tabs used to customize the analog TV distortion. The Noise tab
modifies the noise in the image to simulate a weak analog antenna signal.

Power
Increase the value of this slider above 0 to introduce noise into the image. The higher the value, the
stronger the noise.

Size
Use this slider to scale the noise map larger.

Random
If this thumbwheel control is set to 0, the noise map is static. Change the value over time to cause the
static to change from frame to frame.

The TV Roll Bar tab

Roll Bar Tab

The Roll Bar tab is the third of three tabs used to customize the analog TV distortion. The Roll Bar tab
animates the bar.

Bar Strength
At the default value of 0, no bar is drawn. The higher the value, the darker the area covered by the
bar becomes.

Bar Size
Increase the value of this slider to make the bar taller.

Chapter 37 Effect Nodes 949

Bar Offset
Animate this control to scroll the bar across the screen.

Common Controls
Settings Tab
The Settings tab controls are common to all Effect nodes, so their descriptions can be found in the
following “The Common Controls” section.

The Common Controls

Effect nodes share several identical controls in the Inspector. This section describes controls that are
common among Effect nodes.

Inspector

The Common Effects Settings tab

Settings Tab
The Settings tab in the Inspector can be found on every tool in the Effects category. The Settings
controls are even found on third-party Effects-type plug-in tools. The controls are consistent and work
the same way for each tool, although some tools do include one or two individual options, which are
also covered here.

Chapter 37 Effect Nodes 950

Blend
The Blend control is used to blend between the tool’s original image input and the tool’s final modified
output image. When the blend value is 0.0, the outgoing image is identical to the incoming image. This
causes the tool to skip processing entirely, copying the input straight to the output.

Process When Blend Is 0.0

The tool is processed even when the input value is zero. This is useful when this node is scripted to
trigger another task, but the blend is set to 0.0.

Red/Green/Blue/Alpha Channel Selector

These four buttons are used to limit the effect of the tool to specified color channels. This filter is often
applied after the tool has been processed.
For example, if the red button on a Blur tool is deselected, the blur is first applied to the image, and
then the red channel from the original input is copied back over the red channel of the result.
There are some exceptions, such as tools for which deselecting these channels causes the tool to skip
processing that channel entirely. Tools that do this possess a set of like RGBA buttons on the Controls
tab in the tool. In this case, the buttons in the Settings and the Control tabs are identical.

Apply Mask Inverted

Enabling the Apply Mask Inverted option inverts the complete mask channel for the tool. The mask
channel is the combined result of all masks connected to or generated in a node.

Multiply by Mask
Selecting this option causes the RGB values of the masked image to be multiplied by the mask
channel’s values. This causes all pixels of the image not included in the mask (i.e., set to 0) to become
black/transparent.

Use Object/Use Material (Checkboxes)

Some 3D software can render to file formats that support additional channels. Notably, the EXR file
format supports Object and Material ID channels, which can be used as a mask for the effect. These
checkboxes determine whether the channels are used, if present. The specific Material ID or Object ID
affected is chosen using the next set of controls.

Correct Edges
This checkbox appears only when the Use Object or Use Material checkboxes are selected. It toggles
the method used to deal with overlapping edges of objects in a multi-object image. When enabled,
the Coverage and Background Color channels are used to separate and improve the effect around the
edge of the object. If this option is disabled (or no Coverage or Background Color channels are
available), aliasing may occur on the edge of the mask.
For more information on Coverage and Background Color channels, see Chapter 18,
“Understanding Image Channels” in the Fusion Reference Manual or Chapter 79 in the
DaVinci Resolve Reference Manual.

Object ID/Material ID (Sliders)

Use these sliders to select which ID is used to create a mask from the object or material channels of an
image. Use the Sample button in the same way as the Color Picker: to grab IDs from the image
displayed in the view. The image or sequence must have been rendered from a 3D software package
with those channels included.

Chapter 37 Effect Nodes 951

Clipping Mode
This option determines how the domain of definition rendering handles edges. The Clipping mode is
most important when blur or softness is applied, which may require samples from portions of the
image outside the current domain.
– Frame: The default option is Frame, which automatically sets the node’s domain of definition
to use the full frame of the image, effectively ignoring the current domain of definition. If
the upstream DoD is smaller than the frame, the remaining area in the frame is treated as
black/transparent.
– None: Setting this option to None does not perform any source image clipping. Any data required
to process the node’s effect that would usually be outside the upstream DoD is treated as
black/transparent.

Use GPU
The Use GPU menu has three settings. Setting the menu to Disable turns off hardware-accelerated
rendering using the graphics card in your computer. Enabled uses the hardware. Auto uses a capable
GPU if one is available and falls back to software rendering when a capable GPU is not available

Motion Blur
– Motion Blur: This toggles the rendering of Motion Blur on the tool. When this control is toggled
on, the tool’s predicted motion is used to produce the motion blur caused by the virtual camera’s
shutter. When the control is toggled off, no motion blur is created.
– Quality: Quality determines the number of samples used to create the blur. A quality setting of
2 causes Fusion to create two samples to either side of an object’s actual motion. Larger values
produce smoother results but increase the render time.
– Shutter Angle: Shutter Angle controls the angle of the virtual shutter used to produce the motion
blur effect. Larger angles create more blur but increase the render times. A value of 360 is the
equivalent of having the shutter open for one full frame exposure. Higher values are possible and
can be used to create interesting effects.
– Center Bias: Center Bias modifies the position of the center of the motion blur. This allows for the
creation of motion trail effects.
– Sample Spread: Adjusting this control modifies the weighting given to each sample. This affects
the brightness of the samples.

Comments
The Comments field is used to add notes to a tool. Click in the empty field and type the text. When a
note is added to a tool, a small red square appears in the lower-left corner of the node when the full
tile is displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the
note in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 37 Effect Nodes 952

Chapter 38

Film Nodes
This chapter details the Film nodes in Fusion.

The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Cineon Log [LOG]  954
Film Grain [FGR]  957
Grain [GRN]  960
Light Trim [LT]  963
Remove Noise [RN]  965
The Common Controls  967

Chapter 38 Film Nodes 953

Cineon Log [LOG]

The Cineon Log node

Cineon Log Node Introduction

The Cineon Log node is used to convert several different log camera formats to linear gamma and
back again. Although the name implies that it should be used with Cineon files, it handles “log” gamma
from many different digital cinema sources such as Blackmagic Design, Arri, and Red cameras.

Input
There are two Inputs on the Cineon Log node: one for the log image and one for the effects mask.
– Input: The orange input is used for the primary 2D image that gets the highlight applied.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input restricts the log
conversion to be within the pixels of the mask. An effects mask is applied to the tool after the
tool is processed.

Basic Node Setup

The Cineon Log node is placed directly after a MediaIn node in DaVinci Resolve or a Loader node in
Fusion Studio. It is also commonly placed before a MediaOut or Saver node to convert back to a
Log-encoded image.

A Cineon Log node placed after a Loader node in Fusion Studio

Inspector

Cineon Log controls

Chapter 38 Film Nodes 954

Controls Tab
The Controls tab includes settings for converting from log gamma to linear or from linear to log. You
first select the Mode and then the Log Type. For instance, choose Log to Lin from the Mode menu, and
then select BMD Film if you are compositing with a RAW clip from a Blackmagic Design camera. Those
settings output a linear image ready for compositing.

Depth
The Depth menu is used to select the color depth used to process the input image. The default option
is Auto. Auto determines the color depth based on the file format loaded. For example, JPEG files
automatically process at 8 bit because the JPEG file format does not store color depths greater than 8.
Blackmagic RAW files load at Float, etc. If the color depth of the format is undetermined, the default
depth defined in the Frame Format preferences is used.

Mode
The Mode menu offers two options: one for converting log images to linear and one for converting
linear images to logarithmic.

Log Type
The Log Type menu allows you to select the source of the file. Typically, you select the camera used to
create the image, although the Josh Pines option is specific to film scan workflows. This menu contains
the following camera log types:
– Cineon – Canon Log – Panasonic V-Log – Viper Film Stream
– Arri Log C – Nikon N Log – Red Log Film – ACESlog
– BMD Film – Panalog – Sony S-Log

Lock RGB
When enabled, the settings in this tab affect all color channels equally.
Disable this control to convert the red, green, and blue channels of the image using separate settings
for each channel.

Level
Use this range control to set the black level and white level in the log image before converting. The
left handle adjusts the black level, while the right handle adjusts the white level. Pixels with values in
log space below the black level become out-of-range values below 0.0. Pixels with values above the
white level become out-of-range values above 1.0 after conversion.
When processing in floating-point color space, both negative and high out-of-range values are
preserved. When using 16-bit or 8-bit mode, the out-of-range values are clipped.

Soft Clip (Knee)

The Soft Clip control is used to draw values that are out of range back into the image. This is done by
smoothing the conversion curve at the top and bottom of the curve, allowing more values to be
represented.
Applying a soft clip of any value other than 1 causes the node to process at 16-bit integer, eliminating
all out-of-range values that do not fit within the soft clip.

Film Stock Gamma, Conversion Gamma, and Conversion Table

These controls are used to set the response curves of the logarithmic data during conversion. In
addition to the settings above, a custom ASCII file Lookup Table (LUT) can be created with specific
conversion values. The ASCII LUT file can be loaded using the Browse button.

Chapter 38 Film Nodes 955

Common Controls
Settings Tab
The Settings tab controls are common to all Film nodes, so their descriptions can be found in
“The Common Controls” section at the end of this chapter.

Black Rolloff
Since a mathematical log() operation on a value of zero or lower results in invalid values,
Fusion clips values below 1e-38 (0 followed by 38 zeros) to 0 to ensure correct results. This is
almost never an issue, since values that small have no visual impact on an image. To see such
tiny values, you would have to add three Brightness Contrast nodes, each with a gain set to
1,000,000. Even then, the values would hover very close to zero.
We have seen processes where instead of cropping these minimal values, they are instead
scaled. So values between 0.0 and 1e-16 are scaled between 1e-18 and 1e-16. The idea is to
crush the majority of the visual range in a float image into values very near to zero, then
expand them again, forcing a gentle ramp to produce a small ramp in the extreme black
values. Should you find yourself facing a color pipeline using this process, here is how you can
mimic it with the help of a Custom node.
The process involves converting the log image to linear with a very small gamma and a wider
than normal black level to white level (e.g., conversion gamma of 0.6, black of 10,
white of 1010). This crushes most of the image’s range into very small values. This is followed
by a Custom node (described below), and then by a linear to log conversion that reverses the
process but uses a slightly higher black level. The difference between the black levels defines
the falloff range.
Since this lifts the blacks, the image is usually then converted back to linear one more time,
using more traditional values (i.e., 95-685) to reset the black point.
The Custom node should use the following equation in the red, green, and blue expressions:

if (c1< 1e-16, 1e-18 + (c1/1e-16)*(1e-16 - 1e-18), c1)

Falloff Comparison

The black falloff from the native Fusion process

Virtually identical black falloff from the ramped clipping process

Chapter 38 Film Nodes 956

Film Grain [FGR]

The Film Grain node

Film Grain Node Introduction

The Film Grain node adds generated film grain to an image. Typically, when compositing with different
elements shot on film and in digital, the grain is removed for compositing operations and then
reapplied to the final composite. This helps create the appearance that all elements were shot as a
single scene with the same film stock.

NOTE: Although more accurate, the Film Grain node does not replace the older Grain node,
which is still provided to allow older compositions to load and render, but in almost every
case, it is better to use the Film Grain node.

Input
There are two inputs on the Film Grain node: one for the image and one for the effects mask.
– Input: The orange input is used for the primary 2D image that gets the grain applied.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the grain to be
within the pixels of the mask. An effects mask is applied to the tool after the tool is processed.

Basic Node Setup

The Film Grain node is commonly used before a MediaOut node in DaVinci Resolve or before a Saver
node in Fusion Studio. Since grain is often removed from source material to achieve cleaner keys and
composite, film grain is added back before the output to create a more realistic composite.

Film Grain node adding grain back before output

Chapter 38 Film Nodes 957

Inspector

Film Grain controls

Controls Tab
The Controls tab includes all the parameters for modifying the appearance of the film grain.

Complexity
The Complexity setting indicates the number of “layers” of grain applied to the image. With a
complexity of 1, only one grain layer is calculated and applied to the image. When complexity is set to
4, the node calculates four separate grain layers and applies the mean combined result of each pass
to the final image. Higher complexities produce visually more sophisticated results, without the
apparent regularity often perceivable in digitally-produced grain.

Alpha Multiply
When the Alpha Multiply checkbox is enabled, the Film Grain node multiplies its results by the source
image’s Alpha channel. This is necessary when working with post-multiplied images to ensure that the
grain does not affect areas of the image where the Alpha is 0.0 (transparent).

NOTE: Since it is impossible to say what the final value of semitransparent pixels in the image
are until after they are composited with their background, you should avoid applying log-
processed grain to the elements until after they have been composited. This ensures that the
strength of the grain is accurate.

Log Processing
When this checkbox is enabled (default), the grain applied to the image has its intensity applied
nonlinearly to match the grain profile of most film. Roughly speaking, the intensity of the grain
increases exponentially from black to white. When this checkbox is disabled, the grain is applied
uniformly, regardless of the brightness of the affected pixel.
One of the primary features of grain in film is that the appearance of the grain varies radically with the
exposure so that there appears to be minimal grain present in the blacks, with the amount and
deviation of the grain increasing as the pixels exposure increases. In a film negative, the darkest
portions of the developed image appear entirely opaque, and this obscures the grain. As the negative
becomes progressively clearer, more of the grain becomes evident in the result. Chemical differences
in the R, G, B, layer’s response to light also cause each color component of the film to present a
different grain profile, typically with the blue channel presenting the most significant amount of grain.

Chapter 38 Film Nodes 958

As a result, an essential control in the Film Grain node is the Log Processing checkbox, which should
be enabled when matching film, and disabled when working with images that require a more linear
grain response. Having this checkbox enabled closely mimics the results of preceding the old Grain
node with a Linear to Log conversion and following with a Log to Linear conversion immediately after.

Seed
The Seed slider and Reseed button are presented whenever a Fusion node relies on a random result.
Two nodes with the same seed values produce the same random results. Click on the Reseed button
to randomly select a new seed value, or adjust the slider to select a new seed value manually.

Time Lock
Enabling Time Lock stops the random seed from generating new grain on every frame.

Monochrome
When the Monochrome checkbox is enabled (default), the grain is applied to the red, green, and blue
color channels of the image equally. When deselected, individual control over the Size, Strength, and
Roughness of the grain in each channel becomes possible.

Lock Size X/Y

Deselect the Lock Size X/Y checkbox to control the size of the grain along the X- and Y-axis
individually.

Size
The grain size is calculated relative to the size of a pixel. Consequently, changing the resolution of the
image does not impact the relative appearance of the grain. The default grain size of 1.0 produces
grain kernels that cover roughly 2 pixels.

Strength
Grain is expressed as a variation from the original color of a pixel. The stronger the grain’s strength,
the wider the possible variation from the original pixel value. For example, given a pixel with an original
value of p, and a Grain node with complexity = 1 size = 1; roughness = 0; log processing = off; the grain
produces an output value of p +/- strength. In other words, a pixel with a value of 0.5 with a grain
strength of 0.02 could end up with a final value between 0.48 and 0.52.
Once again, that’s a slight oversimplification, especially when the complexity exceeds 1. Enabling the
Log Processing checkbox also causes that variation to be affected such that there is less variation in
the blacks and more variation in the whites of the image.

NOTE: When visualizing the effect of the grain on the image, the more mathematically
inclined may find it helps to picture a sine wave, where each lobe of the sine wave covers 1
pixel when the Grain Size is 1.0. The Grain Size controls the frequency of the sine wave, while
the Grain Strength controls its amplitude. Again, this is something of an oversimplification.

Roughness
The Roughness slider applies low frequency variation to give the impression of clumping in the grain.
Try setting the roughness to 0, and observe that the grain produced has a very even luminance
variation across the whole image. Increase the roughness to 1.0 and observe the presence of “cellular”
differences in the luminance variation.

Offset
The Offset control helps to match the intensity of the grain in the deep blacks by offsetting the values
before the intensity (strength) of the grain is calculated. So an offset of 0.1 would cause a pixel with a
value of 0.1 to receive grain as if its value was 0.2.

Chapter 38 Film Nodes 959

Common Controls
Settings Tab
The Settings tab controls are common to all Film nodes, so their descriptions can be found in
“The Common Controls” section at the end of this chapter.

Processing Examples

Log Processing On

In the default setting, the different amounts of Grain are

applied to the blacks and the whites of the image

Log Processing Off

When Log processing is off, the Grain is applied

evenly to the entire image, as shown here

Grain [GRN]

The Grain node

Grain Node Introduction

The Grain node offers an older film grain emulation than the more modern and accurate Film Grain
node. The Grain node is still provided to allow older compositions to load and render, but in almost
every case, it is better to use the Film Grain node.

Chapter 38 Film Nodes 960

Input
There are two Inputs on the Grain node: one for the 2D image and one for the effects mask.
– Input: The orange input is used for the primary 2D image that gets the grain applied.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the grain to be
within the pixels of the mask. An effects mask is applied to the tool after the tool is processed.

Basic Node Setup

The Grain node is commonly placed before a MediaOut node in DaVinci Resolve, or before a Saver
node in Fusion Studio. It adds back grain previously removed in a composite.

A Grain node used to add grain back for a more realistic composite

Inspector

Grain controls

Controls Tab
The Controls tab includes all the parameters for modifying the appearance of the grain.

Power
This slider determines the strength of the grain. A higher value increases visibility, making the grain
more prevalent.

RGB Difference
Separate Red, Green, and Blue sliders are used to modify the strength of the effect on a per
channel basis.

Chapter 38 Film Nodes 961

Grain Softness
This slider controls the blurriness or fuzziness of the grain. Smaller values cause the grain to be more
sharp or coarse.

Grain Size
This slider determines the size of the grain particles. Higher values increase the grain size.

Grain Spacing
This slider determines the density or amount of grain per area. Higher values cause the grain to
appear more spaced out.

Aspect Ratio
This slider adjusts the aspect of the grain so that it can be matched with anamorphic images.

Alpha-Multiply
When enabled, this checkbox multiplies the image by the Alpha, clearing the black areas of any
grain effect.

Grain Spread controls

Spread Tab
The Spread tab uses curves for the red, green, and blue channels to control the amount of grain over
each channel’s tonal range.

RGB Checkboxes
The red, green, and blue checkboxes enable each channel’s custom curve, allowing you to control
how much grain appears in each channel. To mimic usual film responses, more grain would appear in
the blue channel than the red, and the green channel would receive the least. Right-clicking in the
spline area displays a contextual menu containing options related to modifying spline curves.
For more information on the LUT Editor’s controls see Chapter 106, “LUT Nodes” in the
DaVinci Resolve Reference Manual or Chapter 45 in the Fusion Reference Manual.

In and Out
This control provides direct editing of points on the curve by setting In/Out point values.

Chapter 38 Film Nodes 962

 Examples
Default Spread
In the default setting, the grain is applied evenly to the entire image, as shown here. However,
film often shows a different amount of grain in the blacks, mids, and whites.

Bell-Shaped Spread
Setting a bell shape is often a good starting point to create a more realistic-looking grain.
Here we have a non-uniform distribution with different amounts of grain in the red, green, and
blue channels.
In both examples, the grain’s power has been exaggerated to show the effect a bit better.

Common Controls
Setting Tab
The Settings tab controls are common to all Film nodes, so their descriptions can be found in “The
Common Controls” section at the end of this chapter.

Light Trim [LT]

The Light Trim node

Chapter 38 Film Nodes 963

Light Trim Node Introduction
This node emulates film scanner light trims. By design, this node works best with logarithmic data,
such as the images stored by Cineon, Arri, or Blackmagic RAW files. When logarithmic data is
provided, the Light Trim node can be used to increase or decrease the apparent exposure level of
the image.

Inputs
There are two Inputs on the Light Trim node: one for the 2D image and one for the effects mask.
– Input: The orange input is used for the primary Log 2D image that gets its exposure adjusted.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the exposure
change to be within the pixels of the mask. An effects mask is applied to the tool after the tool
is processed.

Basic Node Setup

The Light Trim node is placed after a LOG clip but before the Log clip is converted by a
Cineon LOG node.

A Light Trim node used to adjust exposure on a LOG clip

Inspector

Light Trim controls

Controls Tab
The Controls tab includes a single slider that adjusts the exposure of the image.

Lock RGBA
When selected, the Lock RGBA control collapses control of all image channels into one slider. This
selection is on by default. To manipulate the various color channels independently, deselect
this checkbox.

Trim
This slider shifts the color in film, optical printing, and lab printing points. 8 points equals one stop
of exposure.

Chapter 38 Film Nodes 964

Common Controls
Settings Tab
The Settings tab controls are common to all Film nodes, so their descriptions can be found in
“The Common Controls” section at the end of this chapter.

Remove Noise [RN]

The Remove Noise node

Remove Noise Node Introduction

The Remove Noise node provides simple noise management. The basic operation is that the node
blurs the image channels, and then compares the blurred image to the original to extract the noise.
A sharpness is then applied to the image, except where noise was detected.
To use this node, view the image and look at the red channel. Then increase the Red Softness until the
grain appears to be gone. Next, increase the sharpness until the detail reappears, but stop before the
grain reappears. Repeat for the green and blue channels.

Inputs
There are two inputs on the Remove Noise node: one for the 2D image and one for the effects mask.
– Input: The orange input is used for the primary 2D image that gets noise removed.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the noise
removal change to be within the pixels of the mask. An effects mask is applied to the tool after
the tool is processed.

Basic Node Setup

The Remove Noise node can be used on any clip with noise. For example, it is used below to remove
noise prior to keying the clip using the DeltaKeyer.

A Remove Noise node used to remove noise prior to keying

Chapter 38 Film Nodes 965

Inspector

Remove Noise controls

Controls Tab
The Controls tab switches the noise removal between two methods: Color and Chroma. When the
Method is set to Color, the Controls tab adjusts the amount of blur and sharpness individually for each
RGB channel. When the Method is set to Chroma, the blur and sharpness is adjusted based on Luma
and Chroma controls.

Method
This menu is used to choose whether the node processes color using the Color or Chroma method.
This also gives you a different set of control sliders.

Lock
This checkbox links the Softness and Detail sliders of each channel together.

Softness Red, Green, and Blue

The Softness sliders determine the amount of blur applied to each channel of the image. In Chroma
mode, you have sliders for the softness in the Luminance and Chrominance channels, respectively.

Detail Red, Green, and Blue

The Sharpness sliders determine how much detail is reintroduced into each channel after each
channel is softened. In Chroma mode, you have sliders for Luminance and Chrominance
channels, respectively.

Chroma Method controls

Common Controls
Settings Tab
The Settings tab controls are common to all Film nodes, so their descriptions can be found in the
following “The Common Controls” section.

Chapter 38 Film Nodes 966

The Common Controls
Film nodes share a number of identical controls in the Inspector. This section describes controls that
are common among Film nodes.

Inspector

The Common Film settings inspector

Settings Tab
The Settings tab in the Inspector can be found on every tool in the Film category. The Settings
controls are even found on third-party Film-type plug-in tools. The controls are consistent and work
the same way for each tool, although some tools do include one or two individual options, which are
also covered here.

Blend
The Blend control is used to blend between the tool’s original image input and the tool’s final modified
output image. When the blend value is 0.0, the outgoing image is identical to the incoming image.
Commonly, this causes the tool to skip processing entirely, copying the input straight to the output.

Process When Blend Is 0.0

The tool is processed even when the input value is zero. This can be useful if processing of this node
is scripted to trigger another task, but the value of the node is set to 0.0.

Red/Green/Blue/Alpha Channel Selector

These four buttons are used to limit the effect of the tool to specified color channels. This filter is often
applied after the tool has been processed.
For example, if the red button on a Blur tool is deselected, the blur is first applied to the image, and
then the red channel from the original input is copied back over the red channel of the result.

Chapter 38 Film Nodes 967

There are some exceptions, such as tools for which deselecting these channels causes the tool to
skip processing that channel entirely. Tools that do this generally possess a set of identical RGBA
buttons on the Controls tab in the tool. In this case, the buttons in the Settings and the Controls tabs
are identical.

Apply Mask Inverted

Enabling the Apply Mask Inverted option inverts the complete mask channel for the tool. The mask
channel is the combined result of all masks connected to or generated in a node.

Multiply by Mask
Selecting this option causes the RGB values of the masked image to be multiplied by the mask
channel’s values. This causes all pixels not included in the mask (i.e., set to 0) to become black/
transparent.

Use Object/Use Material (Checkboxes)

Some 3D software can render to file formats that support additional channels. Notably, the EXR file
format supports Object and Material ID channels, which can be used as a mask for the effect. These
checkboxes determine whether the channels are used, if present. The specific Material ID or Object ID
affected is chosen using the next set of controls.

Correct Edges
This checkbox appears only when the Use Object or Use Material checkboxes are selected. It toggles
the method used to deal with overlapping edges of objects in a multi-object image. When enabled,
the Coverage and Background Color channels are used to separate and improve the effect around the
edge of the object. If this option disabled (or no Coverage or Background Color channels are
available), aliasing may occur on the edge of the mask.
For more information on Coverage and Background Color channels, see Chapter 18, “Understanding
Image Channels” in the Fusion Studio Reference Manual or Chapter 79 in the DaVinci Resolve
Reference Manual.

Object ID/Material ID (Sliders)

Use these sliders to select which ID is used to create a mask from the object or material channels of an
image. Use the Sample button in the same way as the Color Picker: to grab IDs from the image
displayed in the view. The image or sequence must have been rendered from a 3D software package
with those channels included.

Clipping Mode
This option determines how edges are handled when performing domain of definition rendering. This
is mostly important for nodes like Blur, which may require samples from portions of the image outside
the current domain.
– Frame: The default option is Frame, which automatically sets the node’s domain of definition
to use the full frame of the image, effectively ignoring the current domain of definition. If the
upstream DoD is smaller than the frame, the remaining area in the frame is treated as black/
transparent.
– Domain: Setting this option to Domain respects the upstream domain of definition when applying
the node’s effect. This can have adverse clipping effects in situations where the node employs a
large filter.
– None: Setting this option to None does not perform any source image clipping at all. This means
that any data required to process the node’s effect that would normally be outside the upstream
DoD is treated as black/transparent.

Chapter 38 Film Nodes 968

Use GPU
The Use GPU menu has three settings. Setting the menu to Disable turns off hardware-accelerated
rendering using the graphics card in your computer. Enabled uses the hardware. Auto uses a capable
GPU if one is available and falls back to software rendering when a capable GPU is not available

Comments
The Comments field is used to add notes to a tool. Click in the empty field and type the text. When a
note is added to a tool, a small red square appears in the lower-left corner of the node when the full
tile is displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the
note in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 38 Film Nodes 969

Chapter 39

Filter Nodes
This chapter details the Filter nodes available in Fusion.

The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Create Bump Map [CBu]  971
Custom Filter Node [CFlt]  973
Erode Dilate Node [ErDl]  977
Filter Node [Fltr]  979
Rank Filter Node [RFlt]  981
The Common Controls  983

Chapter 39 Filter Nodes 970

Create Bump Map [CBu]

The Create Bump Map node

Create Bump Map Node Introduction

The Create Bump Map node converts a grayscale (height map) image into a bump map. Unlike the
Bump Map node that turns an image into a 3D material, the Create Bump Map node creates bump
vector data and provides the output as an RGB image so other image-processing operations can
be applied. 

Input
The Create Bump Map node includes two inputs: one for the main image and the other for an effect
mask to limit the area where the bump map is created.
– Input: The orange input takes the RGBA channels from an image to calculate the bump map.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the creation of the bump map to only those pixels within the mask. An effects mask
is applied to the tool after the tool is processed.

Basic Node Setup

The Create Bump Map node accepts a 2D grayscale image like a fast noise, which can then go
through various 2D image-processing filters to create the bump map texture.

A Create Bump Map node produces a bump map as an RGB image for further image processing

Chapter 39 Filter Nodes 971

Inspector

Create Bump Map controls

Controls Tab
The Controls tab contains all parameters for creating the bump map.

Filter Size
This menu sets the filter size for creating the bump map. You can set the filter size at 3 x 3 pixels or 5 x
5 pixels, thus determining the radius of the pixels sampled. The larger the size, the more time it takes
to render.

Height Source
The Height Source menu selects the channel for extracting the grayscale information.

Clamp Normal.Z
This slider clips the lower values of the blue channel in the resulting bump texture.

Wrap Mode
This menu determines how the image wraps at the borders, so the filter produces a correct result
when using seamless tiling textures.

Height Scale
The height scale menu modifies the contrast of the resulting values in the bump map. Increasing this
value yields in a more visible bump map.

Bump Map Texture Depth

This menu matches or converts the resulting bump texture into the desired bit depth.

NOTE: The below definitions are provided to clarify some of the terminology used in the
Create Bump Map node and other similar types of nodes.
– Height Map: A grayscale image containing a height value per pixel.
– Bump Map: An image containing normals stored in the RGB channels used for modifying the
existing normals (usually given in tangent space).
– Normal Map: An image containing normals stored in the RGB channels used for replacing
the existing normals (usually given in tangent or object space).

Common Controls
Settings Tab
The Settings tab controls are common to all Filter nodes, so their descriptions can be found in “The
Common Controls” section at the end of this chapter.

Chapter 39 Filter Nodes 972

Custom Filter Node [CFlt]

The Custom Filter node

Custom Filter Node Introduction

The Custom Filter node is used to apply custom convolution filters to images. A custom convolution
filter can give a wide variety of image effects. For example, emboss, relief, sharpen, blurring, and edge
detection are all convolution filters. There are many supplied custom filters in the Filters directory that
can be loaded by right-clicking on the control header and selecting Settings > Load from the
contextual menu.
The Custom filter uses an array (or grid) of either 3 x 3, 5 x 5, or 7 x 7 values. (Note: The array in the
Inspector always shows a 7 x 7 grid; however, setting the Matrix Size to 3 x 3 uses only the center
9 cells.) The center of the array represents the current pixel, and entries nearby represent adjacent
pixels. A value of 1 applies the full value of the pixel to the filter. A value of 0 ignores the pixel’s value.
A value greater than 1 multiplies the pixel’s effect on the result. Negative values can also be entered,
where the value of the pixel is subtracted from the average. Only integer values can be entered;
0.x is not valid.

Input
The Custom Filter node includes two inputs: one for the main image and the other for an effect mask
to limit the area where the custom filter is applied.
– Input: The orange input takes the RGBA channels from an image to calculate the custom filter.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the custom filter to only those pixels within the mask. An effects mask is applied to
the tool after the tool is processed.

Basic Node Setup

The Custom Filter node can be inserted after an image, mask, or any node that needs a custom
convolution filter applied.

A Custom Filter node placed after a

MediaIn node in DaVinci Resolve

Chapter 39 Filter Nodes 973

Inspector

Custom Filter controls

Controls Tab
The Controls tab is used to set the filter size and then use the filter matrix to enter convolution
filter values.

Color Channels (RGBA)

The custom filter defaults to operating on R, G, B, and A channels. Selective channel editing is
possible by enabling or disabling the checkboxes beside each channel.
This is not the same as the RGBA checkboxes found under the Common Controls. The node takes
these controls into account before it processes. Deselecting a channel causes the node to skip that
channel when processing, speeding up the rendering of the effect. In contrast, these controls under
the Common Controls tab are applied after the node has processed.

Matrix Size
This menu is used to set the size of the filter at 3 x 3 pixels, 5 x 5 pixels, or 7 x 7 pixels, thus setting the
radius of the pixels sampled. The larger the size, the more time it takes to render.

Update Lock
When this control is selected, Fusion does not render the filter. This is useful for setting up each value
of the filter, and then turning off Update Lock and rendering the filter.

Filter Matrix
The Filter Matrix control is a 7 x 7 grid of text boxes where a number is entered to represent how much
influence each pixel has on the overall convolution filter. The text box in the center represents the
pixel that is processed by the filter. The text box to the left of the center represents the pixel to the
immediate left, and so forth.
The default Matrix size is 3 x 3. Only the pixels immediately adjacent to the current pixel are analyzed.
If a larger Matrix size is set, more of the text boxes in the grid are enabled for input.

Normalize
This controls the amount of filter normalization that is applied to the result. Zero gives a normalized
image. Positive values brighten or raise the level of the filter result. Negative values darken or lower
the level.

Floor Level
This adds or subtracts a minimum, or Floor Level, to the result of the filtered image. Zero does not add
anything to the image. Positive values add to the filtered image, and negative values subtract from
the image.

Chapter 39 Filter Nodes 974

Examples
Original Image Example
For example, a filter with the values...
0 0 0
0 1 0
0 0 0

...has zero effect from its neighboring pixels, and the resulting image would be unchanged.

Original image

Softening Example
A slight softening effect would be...
1 1 1
1 1 1
1 1 1

...where the neighboring pixels are averaged with the center.

Before and after averaging neighboring pixels to soften the image

Emboss Example
The example below subtracts five times the value from the top left and adds five times
the value from the lower right.
-5 0 0
0 1 0
0 0 5

If parts of the processed image are very smooth in color, the neighboring values are
very similar.

Chapter 39 Filter Nodes 975

In parts of the image where the pixels are different (e.g., an edge), the results are
different and tend to highlight or emboss edges in the image.

A Custom Filter adding and subtracting neighboring pixels to create an embossed image

Exposure Example
Using the values...
1 1 1
1 1 1
1 1 1

...and adjusting Normalize to a positive value makes the image brighter or glow,
simulating film overexposure.

The Custom Filter Normalize slider used to change exposure

Relief Example
Using the values...
-1 0 0
0 0 0
0 0 1

... and adjusting Floor Level to a positive value creates a Relief filter.

Custom Filter Floor Level slider used to create a relief

Chapter 39 Filter Nodes 976

Common Controls
Settings Tab
The Settings tab controls are common to all Filter nodes, so their descriptions can be found in “The
Common Controls” section at the end of this chapter.

Erode Dilate Node [ErDl]

The Erode Dilate node

Erode Dilate Node Introduction

The Erode Dilate node contracts or expands the image, depending on whether the Amount slider is
set to a negative or positive value.

Inputs
The Erode Dilate node includes two inputs: one for the main image and the other for an effect mask to
limit the area where the erode or dilate is applied.
– Input: The orange input takes the RGBA channels from an image to calculate the custom filter.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the erode or dilate to only those pixels within the mask. An effects mask is applied
to the tool after the tool is processed.

Basic Node Setup

The Erode Dilate node is commonly used to contract or expand mattes. Below, a Luma Keyer is
connected to the Erode Dilate and passes the modified key to a Matte Control where it is embedded
into the image.

An Erode Dilate node placed after a Luma

Keyer to operate on a Matte Control

Chapter 39 Filter Nodes 977

Inspector

Erode Dilate controls

Controls Tab
The Controls tab includes the main Amount slider that determines whether you are performing an
erode by entering a negative value or a dilate by entering a positive value.

Color Channels (RGBA)

The Erode Dilate node defaults to operating on R, G, B, and A channels. Selective channel editing is
possible by enabling or disabling the checkboxes beside each channel.
This is not the same as the RGBA checkboxes found under the Common Controls. The node takes
these controls into account before it processes. Deselecting a channel causes the node to skip that
channel when processing, speeding up the rendering of the effect. In contrast, the channel controls
under the Common Controls tab are applied after the node has processed.

Lock X/Y
The Lock X/Y checkbox is used to separate the Amount slider into amount X and amount Y, allowing a
different value for the effect on each axis.

Amount
A negative value for Amount causes the image to erode. Eroding simulates the effect of an
underexposed frame, shrinking the image by growing darker areas of the image so that they eat away
at brighter regions.
A positive value for Amount causes the image to dilate, similar to the effect of overexposing a camera.
Regions of high luminance and brightness grow, eating away at the darker regions of the image. Both
techniques eradicate fine detail in the image and tend to posterize fine gradients.
The Amount slider scale is based on the input image width. An amount value of 1 = image width. So, if
you want to erode or dilate by exactly 1 pixel on an HD image, you would enter 1/1920, or 0.00052083.

Common Controls
Settings Tab
The Settings tab controls are common to all Filter nodes, so their descriptions can be found in “The
Common Controls” section at the end of this chapter.

Chapter 39 Filter Nodes 978

Filter Node [Fltr]

The Filter node

Filter Node Introduction

The Filter node contains several standard convolution filters, easily selectable from a list. This node
enables a variety of effects, from radically changing the look of an image to adding subtle randomly-
generated film grain. The Sobel and Laplacian settings are often used for edge detection.

Inputs
The Filter node includes two inputs: one for the main image and the other for an effect mask to limit
the area where the filter is applied.
– Input: The orange input is used for the primary 2D image that gets the filter applied.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the filter to only those pixels within the mask. An effects mask is applied to the tool
after the tool is processed.

Basic Node Setup

The Filter node can be inserted after an image, mask, or any node that needs a filter applied. Below, it
is used to create an edge matte, which is then used to mask the soft glow around the keyed
foreground.

A Filter node using the Sobel setting to extract an edge matte

Chapter 39 Filter Nodes 979

Inspector

Filter controls

Controls Tab
The Controls tab is used to set the filter type, the channels the filter is applied to, and the amount it
blends with the original image.

Filter Type
The Filter Type menu provides a selection of filter types described below.
– Relief: This appears to press the image into metal, such as an image on a coin. The image appears
to be bumped and overlaid on gray.
– Emboss Over: Embosses the image over the top of itself, with adjustable highlight and shadow
height and direction.
– Noise: Uniformly adds noise to images. This is often useful for 3D computer-generated images
that need to be composited with live action, as it reduces the squeaky-clean look that is inherent
in rendered images. The frame number acts as the random generator seed. Therefore, the effect
is different on each frame and is repeatable.
– Defocus: This filter type blurs the image.
– Sobel: Sobel is an advanced edge detection filter. Used in conjunction with a Glow filter, it creates
impressive neon light effects from live-action or 3D-rendered images.
– Laplacian: Laplacian is a very sensitive edge detection filter that produces a finer edge than the
Sobel filter.
– Grain: Adds noise to images similar to the grain of film (mostly in the midrange). This is useful for
3D computer-generated images that need to be composited with live action as it reduces the
squeaky-clean look that is inherent in rendered images. The frame number acts as the random
generator seed. Therefore, the effect is different on each frame and is repeatable.

Color Channels (RGBA)

The Filter node defaults to operating on R, G, B, and A channels. Selective channel editing is possible
by enabling or disabling the checkboxes beside each channel.

Power
Values range from 1 to 10. Power proportionately increases the amount by which the selected filter
affects the image. This does not apply to the Sobel or Laplacian filter type.

Chapter 39 Filter Nodes 980

Angle
This control has a range from 0 to 315 degrees and changes the effect in increments of 45 degrees.
This applies only to the Relief and Emboss filters.

Median
Depending on which Filter Type is selected, the Median control may appear. It varies the Median
filter’s effect. A value of 0.5 produces the true median result, as it finds the middle values. A value of
0.0 finds the minimums, and 1.0 finds the maximums. This applies to the Median setting only.

Seed
This control is visible only when applying the Grain or Noise filter types. The Seed slider can be used
to ensure that the random elements of the effect are seeded with a consistent value. The randomizer
always produces the same result, given the same seed value.

Animated
This control is visible only when applying the Grain or Noise filter types. Select the checkbox to cause
the noise or grain to change from frame to frame. To produce static noise, deselect this checkbox.

Common Controls
Settings Tab
The Settings tab controls are common to all Filter nodes, so their descriptions can be found in “The
Common Controls” section at the end of this chapter.

Rank Filter Node [RFlt]

The Rank Filter node

Rank Filter Node Introduction

The Rank Filter examines nearby pixels, sorts the pixels by value, and then replaces the color of the
examined pixels with the color of the pixel with the selected rank.

Inputs
The Rank Filter node includes two inputs: one for the main image and the other for an effect mask to
limit the area where the filter is applied.
– Input: The orange input is used for the primary 2D image that gets the Rank filter applied.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the rank filter to only those pixels within the mask. An effects mask is applied to the
tool after the tool is processed.

Chapter 39 Filter Nodes 981

Basic Node Setup
The Rank Filter node can be placed anywhere in a node tree to apply an effect to an image.

A Rank Filter node placed after a MediaIn node in DaVinci Resolve

Inspector

l
Rank Filter controls

Controls Tab
The Controls tab is used to set the size and rank value of the filter.

Size
This control determines the size in pixels of the area sampled by the filter. A value of 1 samples 1 pixel
in each direction, adjacent to the center pixel. This produces a total of 9 pixels, including the center
sampled pixel. Larger values sample from a larger area.
Low Size settings are excellent for removing salt and pepper style noise, while larger Size settings
produce an effect similar to watercolor paintings.

Rank
The Rank slider determines which value from the sampled pixels is chosen. A value of 0 is the lowest
value (darkest pixel), and 1 is the highest value (brightest pixel).

Example
Below is a before and after example of a Rank filter with Size set to 7 and a Rank of 0.7 to
create a watercolor effect.

Before and after of a Rank Filter producing a watercolor-style effect

Chapter 39 Filter Nodes 982

The Common Controls
Filter nodes share a number of identical controls in the Inspector. This section describes controls that
are common among Filter nodes.

Inspector

Common Filter settings Inspector

Settings Tab
The Settings tab in the Inspector can be found on every tool in the Filter category. The Settings
controls are even found on third-party filter-type plug-in tools. The controls are consistent and work
the same way for each tool, although some tools do include one or two individual options, which are
also covered here.

Blend
The Blend control is used to blend between the tool’s original image input and the tool’s final modified
output image. When the blend value is 0.0, the outgoing image is identical to the incoming image.
Normally, this causes the tool to skip processing entirely, copying the input straight to the output.

Process When Blend Is 0.0

The tool is processed even when the input value is zero. This can be useful if processing of this node
is scripted to trigger another task, but the value of the node is set to 0.0.

Chapter 39 Filter Nodes 983

Red/Green/Blue/Alpha Channel Selector
These four buttons are used to limit the effect of the tool to specified color channels. This filter is often
applied after the tool has been processed.
For example, if the red button on a Blur tool is deselected, the blur is first applied to the image, and
then the red channel from the original input is copied back over the red channel of the result.
There are some exceptions, such as tools for which deselecting these channels causes the tool to skip
processing that channel entirely. Tools that do this generally possess a set of identical RGBA buttons
on the Controls tab in the tool. In this case, the buttons in the Settings and the Controls tabs are
identical.

Apply Mask Inverted

Enabling the Apply Mask Inverted option inverts the complete mask channel for the tool. The mask
channel is the combined result of all masks connected to or generated in a node.

Multiply by Mask
Selecting this option causes the RGB values of the masked image to be multiplied by the mask
channel’s values. This causes all pixels of the image not included in the mask (i.e., set to 0) to become
black/transparent.

Use Object/Use Material (Checkboxes)

Some 3D software can render to file formats that support additional channels. Notably, the EXR file
format supports Object ID and Material ID channels, which can be used as a mask for the effect. These
checkboxes determine whether the channels are used, if present. The specific Material ID or Object ID
affected is chosen using the next set of controls.

Correct Edges
This checkbox appears only when the Use Object or Use Material checkboxes are selected. It toggles
the method used to deal with overlapping edges of objects in a multi-object image. When enabled,
the Coverage and Background Color channels are used to separate and improve the effect around the
edge of the object. If this option disabled (or no Coverage or Background Color channels are
available), aliasing may occur on the edge of the mask.
For more information on Coverage and Background Color channels, see Chapter 18, “Understanding
Image Channels” in the Fusion Reference Manual or Chapter 79 in the DaVinci Resolve
Reference Manual.

Object ID/Material ID (Sliders)

Use these sliders to select which ID is used to create a mask from the object or material channels of an
image. Use the Sample button in the same way as the Color Picker: to grab IDs from the image
displayed in the view. The image or sequence must have been rendered from a 3D software package
with those channels included.

Use GPU
The Use GPU menu has three settings. Setting the menu to Disable turns off hardware-accelerated
rendering using the graphics card in your computer. Enabled uses the hardware. Auto uses a capable
GPU if one is available and falls back to software rendering when a capable GPU is not available

Chapter 39 Filter Nodes 984

Motion Blur
– Motion Blur: This toggles the rendering of Motion Blur on the tool. When this control is toggled
on, the tool’s predicted motion is used to produce the motion blur caused by the virtual camera’s
shutter. When the control is toggled off, no motion blur is created.
– Quality: Quality determines the number of samples used to create the blur. A quality setting of
2 causes Fusion to create two samples to either side of an object’s actual motion. Larger values
produce smoother results but increase the render time.
– Shutter Angle: Shutter Angle controls the angle of the virtual shutter used to produce the motion
blur effect. Larger angles create more blur but increase the render times. A value of 360 is the
equivalent of having the shutter open for one whole frame exposure. Higher values are possible
and can be used to create interesting effects.
– Center Bias: Center Bias modifies the position of the center of the motion blur. This allows for the
creation of motion trail effects.
– Sample Spread: Adjusting this control modifies the weighting given to each sample. This affects
the brightness of the samples.

Comments
The Comments field is used to add notes to a tool. Click in the empty field and type the text. When a
note is added to a tool, a small red square appears in the lower-left corner of the node when the full
tile is displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the
note in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 39 Filter Nodes 985

Chapter 40

Flow Nodes
This chapter details the Sticky Note and Underlay features available in Fusion.

The abbreviations next to each feature name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Sticky Note [NTE]  987
Sticky Note Introduction  987
Usage  987
Underlay [UND]  988
Underlay Introduction  988
Usage  988

Chapter 40 Flow Nodes 986

Sticky Note [NTE]

The Sticky Note

Sticky Note Introduction

A Sticky Note is not a node at all. It is a useful way of attaching notes, comments, and history to a
specific area of a comp. By changing their size and color, they can provide unobtrusive comments or
important notices, as required. Sticky Notes make an excellent complement to the Comments tab in
the Inspector.

Usage

A Sticky Note added to a node tree can provide a

description for other people or as a reminder to yourself

To create a Sticky Note, click in an empty area of the Node Editor where you want a Sticky Note to
appear. Then, from the Effects Library, click the Sticky Note effect located in the Tools > Flow category
or press Shift-Spacebar and search for the Sticky Note in the Select Tool window.
Like Groups, Sticky Notes are created in a smaller, collapsed form. They can be expanded by double-
clicking on them. Once expanded, they can be resized using any side or corner of the note or moved
by dragging on the name header. To collapse the Sticky Note again, click the icon in the top-
left corner.
To rename, delete, copy, or change the color of the note, right-click over the note and choose from the
contextual menu. Using this menu, you can also lock the note to prevent editing.
To edit the text in a Sticky Note, first expand it by double-clicking anywhere on the note, and then click
below its title bar. If the note is not locked, you can edit the text.

Chapter 40 Flow Nodes 987

Underlay [UND]

The Underlay

Underlay Introduction
Underlays are a convenient method of visually organizing areas of a composition. As with Groups,
Underlays can improve the readability of a comp by separating it into labeled functional blocks. While
Groups are designed to streamline the look of a comp by collapsing complex layers down to single
nodes, Underlays highlight, rather than hide, and do not restrict outside connections.

Usage

Underlay node structure

As with Sticky Notes , an Underlay can be added to a comp by selecting it from the Flow category in
the Effects Library or searching for it in the Select Tool window. The Underlay to the Node Editor with
its title bar is centered on the last-clicked position.
Underlays can be resized using any side or corner. This will not affect any nodes.
Underlays can also be used as simple selection groups. Activating an Underlay, by clicking its title, will
select all the tools contained wholly within it as well, allowing the entire set to be moved, duplicated,
passed through, and so on.
To rename an Underlay, first ensure that nodes contained within the Underlay are not selected. Then,
Option-click on the Underlay title to select the Underlay without selecting the nodes it contains. Once
selected, right-click over the title and choose Rename. Underlays can be assigned a color using the
same right-click contextual menu.

Chapter 40 Flow Nodes 988

Chapter 41

Flow Organizational
Nodes
This chapter details the Groups, Macro, and Pipe Router nodes, which are designed to
help organize your compositions, making the node tree easier to see and understand.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Groups  990
Groups Introduction  990
Usage  990
Macro  991
Macro Introduction  991
Usage  991
Macro Editor  991
The Final Macro  992
Pipe Router  992
Router Introduction  992
Usage  993
Router  993

Chapter 41 Flow Organizational Nodes 989

Groups

The Group node

Groups Introduction
Groups are used to keep complex node trees organized. You can select any number of nodes in the
node tree and then group them to create a single node icon in the Node Editor. Groups are non-
destructive and can be opened at any time.

Usage
– To group nodes, select them in the Node Editor, and then right-click over any of the selected
nodes and choose Group from the contextual menu.

Selecting nodes to group

– To edit the individual nodes in a group, right-click and choose Expand Group from the contextual
menu. All individual nodes contained in the group are displayed in a floating node tree window.
When opened, groups hover over existing elements, allowing editing of the enclosed nodes.

An opened node group

– To remove or decompose a group and retain the individual nodes, right-click the group and
choose Ungroup.

Chapter 41 Flow Organizational Nodes 990

Macro
Macros are not technically a node. Instead, they are a group of nodes that act as a single node.

Macro Introduction
Macros can be used to combine multiple nodes and expose a user-definable set of controls.
They are meant as a fast and convenient way of building custom nodes.

Usage
To create a Macro, select the nodes intended for the macro. The order in which the nodes are
selected becomes the order in which they are displayed in the Macro Editor. Right-click on any of the
selected nodes and choose Macro > Create Macro from the contextual menu.

Select the nodes to include in the macro

Macro Editor
The Macro Editor allows you to specify and rename the controls that are exposed in the final
macro tool.
In the example below, the tool is named Light_Wrap at the top. The Blur slider for Matte Control 1 is
enabled and renamed to Softness, as it will appear in the Inspector.

The Macro Editor

Chapter 41 Flow Organizational Nodes 991

After setting up the macro to your liking, click the Close button in the lower-right corner of the dialog.
Then, in the Save dialog, click Yes to save the macro, click No to leave Macro Editor without saving the
changes, or click Cancel to return to the Macro Editor.

The Macro Editor save dialog

To add the macro to your node tree, right-click anywhere on the node tree and select Macro >
[NameOfYourMacro] from the contextual menu.

Saving a Macro as a Title Template in the Edit Page (DaVinci Resolve)

When using DaVinci Resolve, macros are available only in the Fusion page. However, if the macro is a
title animation, you can save it to the Titles Templates folder and have it appear in the Edit page
Effects Library.
To save a title macro so it appears in the Edit page Effects Library, save the macro to:

macOS:
Users > UserName > Library > Application Support > Blackmagic Design > DaVinci Resolve > Fusion >
Templates > Edit > Titles

Windows:
C Drive > Users > UserName > AppData > Roaming > Blackmagic Design > DaVinci Resolve > Support
> Fusion > Templates > Edit > Titles

The Final Macro

The final macro looks and behaves just like any other node in Fusion.
As another example, you could take a single Channel Boolean, set it to Add mode, and make it into a
macro exposing no controls at all, thus creating the equivalent of an Add Mix node like the one that
can be found in programs like Nuke.

Pipe Router
Pipe Routers are another type of organizational tool you use to improve the layout and appearance of
the node tree.

Router Introduction
Routers can be used to neatly organize your comps by creating “elbows” in your node tree, so the
connection lines do not overlap nodes, making them easier to understand. Routers do not have any
influence on render times.

Chapter 41 Flow Organizational Nodes 992

Usage

A Pipe Router usage example

Router
To insert a router along a connection line, Option- or Alt-click on the line. The router can then be
repositioned to arrange the connections as needed.
Although routers have no actual controls, they still can be used to add comments to a comp.

An example comment in a Router node

Chapter 41 Flow Organizational Nodes 993

Chapter 42

Fuses
This chapter introduces Fuses, which are scriptable plug-ins that can
be used within Fusion.

Contents
Fuses [FUS]  995
Fuses Introduction  995
Installing Fuses  995
Working with Fuses in a Composition  995

Chapter 42 Fuses 994

Fuses [FUS]

A Fuse node

Fuses Introduction
Fuses are plug-ins. The difference between a Fuse and an Open FX plug-in is that a Fuse is created
using a Lua script. Fuses can be edited within Fusion or DaVinci Resolve, and the changes you make
compile on-the-fly.
Using a Lua script makes it easy for even non-programmers to prototype and develop custom nodes.
A new Fuse can be added to a composition, edited and reloaded, all without having to close the
current composition. They can also be used as modifiers to manipulate parameters, curves, and text
very quickly. ViewShader Fuses can make use of the GPU for faster performance. This makes Fuses
much more convenient than an Open FX plug-in that uses Fusion’s OFX SDK. However, this flexibility
comes at a cost. Since a Fuse is compiled on-the-fly, it can be significantly slower than the identical
node created using the Open FX SDK.
As an example, Fuses could generate a mask from the over-exposed areas of an image, or create
initial particle positions based on the XYZ position stored within a text file.
Please contact Blackmagic Design for access to the SDK (Software Developer Kit) documentation.

Installing Fuses
Fuses are installed in the Fusion:\Fuses path map. By default this folder is located at
Users/ User_Name/Library Application Support/Blackmagic Design/Fusion (or DaVinci Resolve)/Fuses
on macOS or C:\Users\User_Name\AppData\Roaming\Blackmagic Design\Fusion (or DaVinci Resolve)\
Fuses, on Windows. Files must use the extension .fuse, or they will be ignored by Fusion.

Working with Fuses in a Composition

Fuses can be designed to appear in any category in the Effects Library. Once installed, they are added
to a composition exactly as any native or third-party plug-in node. However, since a Fuse is just a text
document, it can be edited by clicking the Edit button that appears at the top of the Inspector when
the Fuse node is selected. This will open the Fuse in the default script editor specified in the Global
Preferences/Scripting panel.

NOTE: Any changes made to a Fuse’s script do not immediately affect other copies of the
same Fuse node already added to a composition. To use the updated Fuse script on all
similar Fuses in the composition, either close and reopen the composition, or click on the
Reload button in each Fuse’s Inspector.

When a composition containing a Fuse node is opened, the currently saved version of the Fuse script
is used. The easiest way to ensure that a composition is running the current version of a Fuse is to
close and reopen the composition.

Chapter 42 Fuses 995

Chapter 43

Generator Nodes
This chapter details the Generator nodes available in Fusion. The abbreviations next
to each node name can be used in the Select Tool dialog when searching for tools
and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Background [BG]  997
Day Sky [DS]  1001
Fast Noise [FN]  1003
Mandelbrot [MAN]  1006
Plasma [PLAS]  1009
Text+ [TXT+]  1011
Text+ Modifiers  1023
Character Level Styling  1023
Comp Name  1024
Follower  1024
Text Scramble  1026
Text Timer  1027
Time Code  1027
The Common Controls  1028

Chapter 43 Generator Nodes 996

Background [BG]

The Background node

Background Node Introduction

The Background node can be used to produce anything from a solid color background to complex
looped gradients.

Inputs
There is one input on the Background node for an effect mask input.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the background color to only those pixels within the mask.

Basic Node Setup

The Background node generates a color and passes the color off to another node like the Paint
node below.

A Background node generating a background for the Paint tool

Inspector

The Background node controls

Chapter 43 Generator Nodes 997

Color Tab
The Color tab is used to control the style and color(s) of the generated background.

Type
This control is used to select the style of background generated by the node. Four selections are
available:
– Solid Color: This default creates a single-color image.
– Horizontal: This creates a two-color horizontal gradation.
– Vertical: This creates a two-color vertical gradation.
– Four Corner: This creates a four-color corner gradation.

Horizontal gradient controls

Horizontal/Vertical/Four Point
When the Type menu is set to Horizontal, Vertical, or Four Corner, two- or four-color swatches are
displayed where the left/right, top/bottom, or four corners of the gradient colors can be set.

Chapter 43 Generator Nodes 998

Gradient controls

Gradient
When the Type menu is set to Gradient, additional controls are displayed where the gradient colors’
direction can be customized.

Gradient Type
This menu selects the form used to draw the gradient. There are six choices:
– Linear: Draws the gradient along a straight line from the starting color stop to the
ending color stop.
– Reflect: Draws the gradient by mirroring the linear gradient on either side of the starting point.
– Square: Draws the gradient by using a square pattern when the starting point is at the
center of the image.
– Cross: Draws the gradient using a cross pattern when the starting point is at the center
of the image.
– Radial: Draws the gradient in a circular pattern when the starting point is at the center
of the image.
– Angle: Draws the gradient in a counterclockwise sweep when the starting point is at the
center of the image.

Viewer Start and End Position

The Start and End positions in the viewer are represented by two red control points connected by a
green line. They determine where the gradient begins and ends.

Gradient Colors
This gradient color bar is used to select the colors for the gradient. The default two color stops set the
start and end colors. You can change the colors used in the gradient by selecting the color stop, and
then using the Eyedropper or color swatch to set a new color.
You can add, move, copy, and delete color from the gradient using the gradient bar.

Chapter 43 Generator Nodes 999

To modify one of the colors, select the triangle below the color on the bar.

To add a color stop to the gradient bar:

1 Click anywhere along the bottom of the gradient bar.
2 Use the Eyedropper or color swatch to select the color for the color stop.

To move a color stop on the gradient bar:

– Drag a color stop left or right along the bar.

To copy a color stop:

– Hold Command (macOS) or Ctrl (Windows) while you drag a color stop.

To delete a color stop:

– Drag the color stop up past the gradient bar.

Interpolation Space
This menu determines what color space is used to calculate the colors between color stops.

Offset
The Offset control is used to offset the position of the gradient relative to the start and end markers.
This control is most useful when used in conjunction with the repeat and ping-pong modes
described below.

Repeat
This menu includes three options used to set the behavior of the gradient when the Offset control
scrolls the gradient past its start and end positions. Selecting Once keeps the color continuous for
offset. Selecting Repeat loops around to the start color when the offset goes beyond the end color.
Selecting Ping-pong repeats the color pattern in reverse.

Sub-Pixel
The Sub-Pixel menu controls the sub-pixel precision used when the edges of the gradient become
visible in repeat mode, or when the gradient is animated. Higher settings will take significantly longer
to render but are more precise.

Gradient Contextual Menu

Gradients have their own contextual menu that you can bring up by right-clicking on the gradient bar.
The Gradient contextual menu includes options for animating, publishing, and connecting one gradient
to another. There is also a gradient-specific modifier that builds a custom gradient by sampling colors
from the output of a node.

Common Controls
Image and Settings Tabs
The Image and Settings tabs in the Inspector are duplicated in many Generator nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Chapter 43 Generator Nodes 1000

Day Sky [DS]

The Day Sky node

Day Sky Node Introduction

The Day Sky node aims to produce a simulation of the daylight produced at a specific time and
location on the earth, and generates a high dynamic range image that represents a map of that light. It
is not a sky generator, although it could be combined with a cloud generator or Noise node to
produce one.

NOTE: This generator is a practical implementation of the research paper, A Practical

Analytical Model for Daylight, by Preetham, Shirley, and Smits. A copy of the original paper
can be found at the website for the Visual Simulation Group at the University of Utah
[https://www.cs.utah.edu/~shirley/papers/sunsky/sunsky.pdf].

Inputs
There is a single input on the Day Sky node for an effect mask to limit the area where the day sky
simulation occurs is applied.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the Day Sky to only those pixels within the mask.

Basic Node Setup

The Day Sky node is a generator, so it typically starts the branch of a node tree and connects to some
other node, like a Merge node.

A Day Sky connected as the background to a Merge node

Chapter 43 Generator Nodes 1001

Inspector

Day Sky node controls

Controls Tab
The Controls tab is used to set the location and time of the daylight simulation. This will determine the
overall look that is generated.

Location
The Latitude and Longitude sliders are used to specify the location used to create the Day Sky
simulation.

Date and Time

The Day, Month, and Time controls are used to select the specific time for the Day Sky simulation.

Turbidity
Turbidity causes light to be scattered and absorbed instead of transmitted in straight lines through the
simulation. Increasing the turbidity will give the sky simulation a murky feeling, as if smoke or
atmospheric haze were present.

Do Tone Mapping
Since the simulation is calculated in 32-bit floating-point color space, it generates color values well
above 1.0 and well below 0.0. Tone mapping is a process that takes the full dynamic range of the
resulting simulation and compresses the data into the desired exposure range while attempting to
preserve as much detail from the highlights and shadows as possible. Deselect this checkbox to
disable any tone mapping applied to the simulation.
Generally, this option should be deselected only if the resulting image will later be color corrected as
part of a floating-point color pipeline.

Exposure
Use this control to select the exposure used for tone mapping.

Chapter 43 Generator Nodes 1002

Day sky node controls

Advanced Tab
The Advanced tad provides more specific controls over the brightness and width of the different
ranges in the generated sky.

Horizon Brightness
Use this control to adjust the brightness of the horizon relative to the sky.

Luminance Gradient
Use this control to adjust the width of the gradient separating the horizon from the sky.

Circumsolar Region Intensity

Use this control to adjust the intensity or brightness of the sky nearest to the sun.

Circumsolar Region Width

Use this control to adjust the width or size of the area in the sky affected by the sun.

Backscattered Light
Use this control to increase or decrease the backscatter light in the simulation.

Common Controls
Image and Settings Tabs
The Image and Settings tabs in the Inspector are duplicated in many Generator nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Fast Noise [FN]

The Fast Noise node

Fast Noise Node Introduction

The Fast Noise node is a very fast and flexible Perlin Noise generator. It can be useful for a wide range
of effects, from clouds, to swirling fog, waves, water caustics, stylized fire, and smoke, and other

Chapter 43 Generator Nodes 1003

organic textures. It is also invaluable as a noise source for other effects, such as heat shimmers,
particle systems, and dirtiness maps.

Inputs
The two map inputs on the Fast Noise node allow you to use masks to control the value of the noise
detail and brightness controls for each pixel. These two optional inputs can allow some interesting and
creative effects. There is also a standard effect mask input for limiting the Fast Noise size.
– Noise Detail Map: A soft-edged mask connected to the gray Noise Detail Map input will give
a flat noise map (zero detail) where the mask is black, and full detail where it is white, with
intermediate values smoothly reducing in detail. It is applied before any gradient color mapping.
This can be very helpful for applying maximum noise detail in a specific area, while smoothly
falling off elsewhere.
– Noise Brightness Map: A mask connected to this white input can be used to control the noise
map completely, such as boosting it in certain areas, combining it with other textures, or if Detail
is set to 0, replacing the Perlin Noise map altogether.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the Fast Noise to only those pixels within the mask.

Basic Node Setup

The Fast Noise node is used to generate images for other nodes to take advantage of. For example,
below the Fast Noise node is used as a bitmap source in the Particle Emitter.

A Fast Noise node used as a bitmap source for a Particle Emitter

Inspector

Fast Noise controls

Chapter 43 Generator Nodes 1004

Noise Tab
The Noise tab controls the shape and pattern of the noise for the Fast Noise node.

Discontinuous
Normally, the noise function interpolates between values to create a smooth continuous gradient of
results. Enable this checkbox to create hard discontinuity lines along some of the noise contours. The
result will be a dramatically different effect.

Inverted
Select this checkbox to invert the noise, creating a negative image of the original pattern. This is most
effective when Discontinuous is also enabled.

Center
Use the Center coordinate control to pan and move the noise pattern.

Detail
Increase the value of this slider to produce a greater level of detail in the noise result. Larger values
add more layers of increasingly detailed noise without affecting the overall pattern. High values take
longer to render but can produce a more natural result.

Brightness
This control adjusts the overall brightness of the noise map, before any gradient color mapping is
applied. In Gradient mode, this has a similar effect to the Offset control.

Contrast
This control increases or decreases the overall contrast of the noise map, prior to any gradient color
mapping. It can exaggerate the effect of the noise and widen the range of colors applied in
Gradient mode.

Lock and Scale X/Y

The size of the noise map can be adjusted using the Scale slider, changing it from gentle variations
over the whole image to a tighter overall texture effect. The Scale slider can be separated into
independent X- and Y-axis scale sliders by clicking on the Lock X/Y checkbox immediately above,
which can be useful for a brushed-metal effect.

Angle
Use the Angle control to rotate the noise pattern.

Seethe
Adjust this thumbwheel control to interpolate the noise map against a different noise map.
This will cause a crawling shift in the noise, as if it was drifting or flowing. This control must be
animated to affect the gradient over time, or you can use the Seethe Rate control below.

Seethe Rate
As with the Seethe control above, the Seethe Rate also causes the noise map to evolve and change.
The Seethe Rate defines the rate at which the noise changes each frame, causing an animated drift in
the noise automatically, without the need for spline animation.

Chapter 43 Generator Nodes 1005

Fast Noise node Color tab

Color Tab
The Color tab allows you to adjust the gradient colors used in the generated noise pattern.

Two Color
A simple two-color gradient is used to color the noise map. The noise function will smoothly transition
from the first color into the second.

Gradient
The Advanced Gradient control in Fusion is used to provide more control over the color gradient used
with the noise map.

Common Controls
Image and Settings Tabs
The Image and Settings tabs in the Inspector are duplicated in many Generator nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Mandelbrot [MAN]

The Mandelbrot node

Chapter 43 Generator Nodes 1006

Mandelbrot Node Introduction
This node creates an image pattern based on the Mandelbrot fractal theory set.

Inputs
The one input on the Mandelbrot node is for an effect mask to limit the area where the fractal noise
is applied.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the fractals to only those pixels within the mask.

Basic Node Setup

The Mandelbrot node generates images that can be used for motion graphics and sci-fi effects. Below,
it is used to generate motion graphics for the background of a title.

A Mandelbrot node applied as a background to the Merge node

Inspector

Mandelbrot node controls

Noise Tab
The Noise tab controls the shape and pattern of the noise for the Mandelbrot node.

Position X and Y
This chooses the image’s horizontal and vertical position or seed point.

Zoom
Zoom magnifies the pattern in or out. Every magnification is recalculated so that there is no practical
limit to the zoom.

Escape Limit
Defines a point where the calculation of the iteration is aborted. Low values lead to blurry halos.

Chapter 43 Generator Nodes 1007

Iterations
This determines the repetitiveness of the set. When animated, it simulates a growing of the set.

Rotation
This rotates the pattern. Every new angle requires recalculation of the image.

The Mandelbrot node Color tab

Color Tab
The Color tab allows you to adjust the gradient and repetition of the gradient colors for the
generated pattern.

Grad Method
Use this control to determine the type of gradation applied at the borders of the pattern.

Continuous Potential
This causes the edges of the pattern to blend to the background color.

Iterations
This causes the edges of the pattern to be solid.

Gradient Curve
This affects the width of the gradation from the pattern to the background color.

R/G/B/A Phase and Repetitions

These controls set the color values of the pattern generators.

Common Controls
Image and Settings Tabs
The Image and Settings tabs in the Inspector are duplicated in other generator nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Chapter 43 Generator Nodes 1008

Plasma [PLAS]

The Plasma node

Plasma Node Introduction

The Plasma node is a background generation node that uses four circular patterns to generate images
similar to plasma. It is useful as a deforming pattern for the Shadow and Deform nodes and to create a
variety of other useful shapes and patterns. It is similar to the Fast Noise node.

Inputs
The one input on the Plasma node is for an effect mask to limit the area where the plasma pattern
is applied.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the plasma to only those pixels within the mask.

Basic Node Setup

Like the Mandelbrot node, the Plasma node generates images that can be used for motion graphics
and sci-fi effects. Below, it is used to generate motion graphics for the background of a title.

A Plasma node applied as a background to a Merge node

Inspector

Plasma node Circles tab

Chapter 43 Generator Nodes 1009

Circles Tab
The Circles tab controls the shape and pattern generated by the Plasma node.

Scale
The Scale control is used to adjust the size of the pattern created.

Operation
The options in this menu determine the mathematical relationship among the four circles whenever
they intersect.

Circle Type
Select the type of circle to be used.

Circle Center
Report and change the position of the circle center.

Circle Scale
Determines the size of the circle to be used for the pattern.

Color Tab
The Color tab allows you to adjust the colors and location within the pattern of the colors for the
generated plasma.

Plasma node Color tab

Phase
Phase changes the color phase of the entire image. When animated, this creates psychedelic
color cycles.

R/G/B/A Phases
Changes the phase of the individual color channels and the Alpha. When animated, this creates color
cycling effects.

Common Controls
Image and Settings Tabs
The Image and Settings tabs in the Inspector are duplicated in many Generator nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Chapter 43 Generator Nodes 1010

Text+ [TXT+]

The Text+ node

Text+ Node Introduction

Fusion’s Text+ node is an advanced character generator capable of multiple styles, 3D
transformations, and several layers of shading. Text can be laid out to a user-defined frame, circle, or
along a path.
Any TrueType or PostScript 1 font installed on the computer can be used to create text. Support for
multibyte and Unicode characters allows text generation in any language, including right to left and
vertically oriented text.
This node generates a 2D image. To produce extruded 3D text with optional beveling, see the
Text 3D node.

Inputs
The one input on the Text+ node is for an effect mask to crop the text.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the text to only those pixels within the mask.

Basic Node Setup

The Text+ node can be combined with many other nodes to create motion graphics. Below it is
connected as the foreground of a Merge node.

Text+ node connected as the foreground to a Merge node

Inspector

Chapter 43 Generator Nodes 1011

Text+ Text tab controls

Text Tab
The Text tab in the Inspector is divided into three sections: Text, Advanced Controls, and Tab Spacing.
The Text section includes parameters that will be familiar to anyone who has used a word processor.
It includes commonly used text formatting options. The Advanced controls are used for kerning
options, and the Tab Spacing is used to define the location and alignment of tabs in the layout.

Styled Text
The edit box in this tab is where the text to be created is entered. Any common character can be
typed into this box. The common OS clipboard shortcuts (Command-C or Ctrl-C to copy, Command-X
or Ctrl-X to cut, Command-V or Ctrl-V to paste) will also work; however, right-clicking in the edit box
displays a custom contextual menu. More information on these modifiers can be found at the end of
this section.
The Styled Text contextual menu includes the following options:
– Animate: Used to animate the text over time.
– Character Level Styling: Used to change the font, color, size and
transformations of individual characters or words through the Modifiers tab.
– Comp Name: Places the name of the composition in the Styled text box for creating slates.
– Follower: A text modifier used to ripple animation across each character of the text.
– Publish: Publishes the text for connection to other text nodes.
– Text Scramble: A text modifier used to randomize the characters in the text.
– Text Timer: A text modifier used to display a countdown or the current date and time.
– Time Code: A text modifier used to display Time Code for the current frame.
– Connect To: Used to connect the text to the published output of another node.

Font
Two Font menus are used to select the font family and typeface, such as Regular, Bold, and Italic.

Chapter 43 Generator Nodes 1012

Color
Sets the basic fill color of the text. This is the same control displayed in the Shading tab color swatch.

Size
This control is used to increase or decrease the size of the text. This is not like selecting a point size in
a word processor. The size is relative to the width of the image.

Tracking
The Tracking parameter adjusts the uniform spacing between each character of text.

Line Spacing
Line Spacing adjusts the distance between each line of text. This is sometimes called leading in
word-processing applications.

V Anchor
The vertical anchor controls consist of three buttons and a slider. The three buttons are used to align
the text vertically to the top of the text, middle of the text, or bottom baseline. The slider can be used
to customize the alignment. Setting the vertical anchor will affect how the text is rotated as well as the
location for line spacing adjustments. This control is most often used when the Layout type is set to
Frame in the Layout tab.

V Justify
The vertical justify slider allows you to customize the vertical alignment of the text from the V Anchor
setting to full justification so it is aligned evenly along the top and bottom edges. This control is most
often used when the Layout type is set to Frame in the Layout tab.

H Anchor
The horizontal anchor controls consist of three buttons and a slider. The three buttons justify the text
alignment to the left edge, middle, or right edge of the text. The slider can be used to customize the
justification. Setting the horizontal anchor will affect how the text is rotated as well as the location for
tracking (leading) spacing adjustments. This control is most often used when the Layout type is set to
Frame in the Layout tab.

H Justify
The horizontal justify slider allows you to customize the justification of the text from the H Anchor
setting to full justification so it is aligned evenly along the left and right edges. This control is most
often used when the Layout type is set to Frame in the Layout tab.

Direction
This menu provides options for determining the Direction in which the text is to be written.

Line Direction
These menu options are used to determine the text flow from top to bottom, bottom to top, left to right,
or right to left.

Underline and Strikeout

These buttons apply the addition of emphasis styles to the text.

Write On
This range control is used to quickly apply simple Write On and Write Off effects to the text. To create
a Write On effect, animate the End portion of the control from 1 to 0 over the length of time required.
To create a Write Off effect, animate the Start portion of the range control from 0 to 1.

Chapter 43 Generator Nodes 1013

Advanced font controls

Force Monospaced
This slider control can be used to override the kerning (spacing between characters) defined in the
font. Setting this slider to zero (the default value) will cause Fusion to rely entirely on the kerning
defined with each character. A value of one will cause the spacing between characters to be
completely even, or monospaced.

Use Font Defined Kerning

This enables kerning as specified in the TrueType font and is on by default.

Manual Font Kerning/Placement

Right-clicking on this label will display a contextual menu that can be used to animate the kerning of
the text. See the “Toolbar” section of this node later in this chapter for details on manual kerning.

Tab Spacing

Tab Spacing
The controls in the Tabs section are used to configure the horizontal screen positions of eight separate
tab stops. Any tab characters in the text will conform to these positions.
You can add tabs directly in the Styled Text input as you type. You can also add tabs by copying from
another document, such as Text on macOS or Notepad on Windows, and paste it into the text box.

Position
This control is used to set the horizontal position of the tab in the frame. The values range from -0.5 to
0.5, where 0 is the center. The position of the tab will be indicated in the viewer by a thin vertical white
line when the Text node is selected. At the top of each tab line in the viewer is a handle. The handle
can be used to position the tab manually.

Alignment
Each tab can be set either left aligned, right aligned, or centered. This slider ranges from -1.0 to 1.0,
where -1.0 is a left-aligned tab, 0.0 is a centered tab and 1.0 is a right-aligned tab. Clicking the tab
handles in the viewer will toggle the alignment of the tab among the three states.

Chapter 43 Generator Nodes 1014

Text+ Layout tab controls

Layout Tab
The controls used to position the text are located in the Layout tab. One of four layout types can be
selected using the Type drop-down menu.
– Point: Point layout is the simplest of the layout modes. Text is arranged around an adjustable
center point.
– Frame: Frame layout allows you to define a rectangular frame used to align the text. The alignment
controls are used for justifying the text vertically and horizontally within the boundaries of the
frame.
– Circle: Circle layout places the text around the curve of a circle or oval. Control is offered over
the diameter and width of the circular shape. When the layout is set to this mode, the Alignment
controls determine whether the text is positioned along the inside or outside of the circle’s edge,
and how multiple lines of text are justified.
– Path: Path layout allows you to shape your text along the edges of a path. The path can be
used simply to add style to the text, or it can be animated using the Position on Path control that
appears when this mode is selected.

Center X, Y, and Z
These controls are used to position the center of the layout element in space. X and Y are onscreen
controls, and Center Z is a slider in the node controls.

Size
This slider is used to control the scale of the layout element.

Chapter 43 Generator Nodes 1015

Perspective
This slider control is used to add or remove perspective from the rotations applied by the Angle X, Y,
and Z controls.

Rotation
Rotation consists of a series of buttons allowing you to select the order in which 3D rotations are
applied to the text. Angle dials can be used to adjust the angle of the Layout element along any axis.

Width and Height

The Width control is visible when the Layout mode is set to Circle or Frame. The Height control is
visible only when the Layout mode is set to Frame. Width and Height controls are used to adjust the
dimensions and aspect of the Layout element.

Fit Characters
This menu control is visible only when the Layout type is set to Circle. This menu is used to select how
the characters are spaced to fit along the circumference.

Position on Path
The Position on Path control is used to control the position of the text along the path. Values less than
0 or greater than 1 will cause the text to move beyond the path in the same direction as the vector of
the path between the last two keyframes.

Background Color
The text generated by this node is normally rendered with a transparent background. This Color Picker
control can be used to set a background color.

Right-Click Here for Shape Animation

This label appears only when the Layout type is set to Path. It is used to provide access to a contextual
menu that provides options for connecting the path to other paths in the node tree, and animating the
shape of the path over time.
For more information, see Chapter 11, “Animating with Motion Paths” in the Fusion Reference Manual or
Chapter 72 in the DaVinci Resolve Reference Manual.

Text+ Transform tab controls

Chapter 43 Generator Nodes 1016

Transform Tab
The Transform tab is used to move, rotate, shear and scale text based on a character, word, or line.

Transform
The Transform menu is used to determine the portion of the text affected by the transformations
applied in this tab. Transformations can be applied to line, word, and character levels simultaneously.
This menu is only used to keep the visible controls to a reasonable number.
– Characters: Each character of text is transformed along its own center axis.
– Words: Each word is transformed separately on the word’s center axis.
– Lines: Each line of the text is transformed separately on that line’s center axis.

Spacing
The Spacing slider is used to adjust the space between each line, word, or character. Values less than
1 will usually cause the characters to begin overlapping.

Pivot X, Y, and Z
This provides control over the exact position of the axis. By default, the axis is positioned at the
calculated center of the line, word, or character. The Axis control works as an offset, such that a value
of 0.1, 0.1 in this control would cause the axis to be shifted downward and to the right for each of the
text elements. Positive values in the Z-axis slider will move the axis of rotation further along the axis
(away from the viewer). Negative values will bring the axis of rotation closer.

Rotation
These buttons are used to determine the order in which transforms are applied. X, Y, and Z would
mean that the rotation is applied to X, then Y, and then Z.

X, Y, and Z
These controls can be used to adjust the angle of the text elements in any of the three dimensions.

Shear X and Y
Adjust these sliders to modify the slanting of the text elements along the X- and Y-axis.

Size X and Y
Adjust these sliders to modify the size of the text elements along the X- and Y-axis.

Text+ Shading tab controls

Chapter 43 Generator Nodes 1017

Shading Tab
The Shading tab provides controls to adjust the shading, texture, and softness of the text.
Transformations can be controlled from this tab as well, applying additional transformations to as many
as eight separate text shading elements independently.

Shading Element
The eight number values in the menu are used to select the element affected by adjustments
in this tab.

Enabled
Select this checkbox to enable or disable each layer of shading elements. Element 1, which is the fill
color, is enabled by default. The controls for a shading element will not be displayed unless this
checkbox is selected.

Sort By
This menu allows you to sort the shading elements by number priority, with 1 being the topmost
element and 8 being the bottommost element, or Z depth, based on the Z Position parameter.

Name
This text label can be used to assign a more descriptive name to each shading element you create.

Appearance
The four Appearance buttons determine how the shading element is applied to the text. Different
controls will appear below depending on the appearance type selected.
– Text Fill: The shading element is applied to the entire text. This is the default mode.
– Text Outline: The shading element is drawn as an outline around the edges of the text.
– Border Fill: The shading element fills a border surrounding the text. Five additional controls are
provided with this shading mode.
– Border Outline: The Border Outline mode draws an outline around the border that surrounds the
text. It offers several additional controls.

Opacity
The Opacity slider controls the overall transparency of the shading element. It is usually better to
assign opacity to a shading element than to adjust the Alpha of the color applied to that element.

Blending
This menu is used to select how the renderer deals with an overlap between two characters in
the text.
– Composite: Merges the shading over the top of itself.
– Solid: Sets the pixels in the overlap region to opaque.
– Transparent: Sets the pixels in the overlap region to transparent.

Thickness
(Outline only) Thickness adjusts the thickness of the outline. Higher values equal thicker outlines.

Adapt Thickness to Perspective

(Outline only) Selecting this checkbox will cause your outline to become thinner where the text is
farther away from the camera, and thicker where it is closer. This will create a more realistic outline for
text transformed in 3D but takes significantly longer to render.

Chapter 43 Generator Nodes 1018

Outside Only
(Outline only) Selecting this checkbox will cause the outline to be drawn only on the outside edge of
the text. By default, the outline is centered on the edge and partially overlaps the text.

Join Style
(Outline only) These buttons provide options for how the corners of the outline are drawn. Options
include Sharp, Rounded, and Beveled.

Line Style
(Outline only) This menu offers additional options for the style of the line. Besides the default solid line,
a variety of dash and dot patterns are available.

Level
(Border Fill only) This is used to control the portion of the text border filled.
– Text: This draws a border around the entire text.
– Line: This draws a border around each line of text.
– Word: This draws a border around each word.
– Character: This draws a border around each character.

Extend Horizontal and Extend Vertical

(Border only) Use this slider to change the dimensions of each border.

Round
(Border Fill and Border Outline only) This slider is used to round off the edges of the border.

Color Types
Besides solid shading, it is also possible to use a gradient fill or map an external image onto the text.
This menu is used to determine if the color of the shading element is derived from a user-selected
color or gradient, or if it comes from an external image source. Different controls will be displayed
below depending on the Color Type selected.
– Solid: When the Type menu is set to Solid mode, color selector controls are provided to select the
color of the text.
– Image: The output of a node in the node tree will be used to texture the text. The node used is
chosen using the Color Image control that is revealed when this option is selected.
– Gradient: When the Type menu is set to Gradient, additional controls are displayed where the
gradient colors can direction can be customized.

Using the Gradient Color Bar

The gradient color bar is used to select the colors for the gradient. The default two color stops set the
start and end colors. You can change the colors used in the gradient by selecting the color stop, and
then using the Eyedropper or color swatch to set a new color.
You can add, move, copy and delete color using the gradient bar.

Chapter 43 Generator Nodes 1019

To modify one of the colors, select the triangle below the color on the bar.

To add a color stop to the gradient bar:

1 Click anywhere along the bottom of the gradient bar.
2 Use the Eyedropper or color swatch to select the color for the color stop.

To move a color stop on the gradient bar:

– Drag a color stop left or right along the bar.

To copy a color stop:

– Hold Command (macOS) or Ctrl (Windows) while you drag a color stop.

To delete a color stop:

– Drag the color stop up past the gradient bar.

Image Source
(Image Mode only) The Image Source menu includes three options for acquiring the image used to
fill the text.
– Tool: Displays a Color image text field where you can add a tool from the node tree as
the fill for text.
– Clip: Provides a Browse button to select a media file from your hard drive as the fill for text.
– Brush: Displays a Color Brush menu where you can select one of Fusion’s paint brush bitmaps as
the fill for text.

Color Image/Color Brush

(Image Mode only) The Color Image text box and Color Brush menu are used to select the tool or paint
brush that will fill the text. In the Color Image text box, enter the name of the node in the node tree that
will provide the image. You can type the name in with the keyboard, drag the node from the node tree
into the text box, or right-click and select Connect To from the contextual menu to select the image to
be used. For the Color Brush, select the Brush from the menu.

Image Sampling
(Image Mode only) This menu is used to select the sampling type for shading rendering and
transformations. The default of Pixel shading is sufficient for 90% of tasks. To reduce detectable
aliasing in the text, set the sampling type to Area. This is slower but may produce better-quality results.
A setting of None will render faster, but with no additional sampling applied so the quality will be lower.

Image Edges
(Image Mode only) This menu is used to choose how transformations applied to image shading
elements are handled when they wrap off the text’s edges.

Shading Mapping
(Image Mode only) This menu is used to select whether the entire image is stretched to fill the text or
scaled to fit, maintaining the aspect ratio but cropping part of the image as needed.

Mapping Angle
(Image and Gradient Modes only) This control rotates the image or gradient on the Z-axis.

Mapping Size
(Image and Gradient Modes only) This control scales the image or gradient.

Chapter 43 Generator Nodes 1020

Mapping Aspect
(Image and Gradient Modes only) This control vertically stretches or shrinks the image or gradient.

Mapping Level
(Image and Gradient Modes only) The Mapping Level menu is used to select how the image is mapped
to the text.
– Full Image: Applies the entire image to the text.
– Text: Applies the image to fit the entire set of text.
– Line: Applies the image per line of text.
– Word: Applies the image per each word of text.
– Character: Applies the image per individual character.

Softness X and Y
These sliders control the softness of the text outline used to create the shading element. Control is
provided for the X- and Y-axis independently.

Apply Softness to Fill Color

Selecting this checkbox will cause blur (softness) to be applied to the shading element. The effect is
best seen when applied to a shading element using an external image.

Softness Glow
This slider will apply a glow to the softened portion of the shading element.

Softness Blend
This slider controls the amount that the result of the softness control is blended back with the original.
It can be used to tone down the result of the soften operation.

Priority Back/Front
Only enabled when the Sort By menu is set to Priority, this slider overrides the priority setting and
determines the layer’s order for the shading elements. Slide the control to the right to bring an element
closer to the front. Move it to the left to tuck one shading element behind another.

Offset X, Y, and Z
These controls are used to apply offset from the text’s global center (as set in the Layout tab) for the
shading elements. A value of X0.0, Y0.1 in the coordinate controls would place the shading element
centered with 10 percent of the image further down the screen along the Y-axis. Positive values in the
Z-Offset slider control will push the center further away from the camera, while positive values will
bring it closer to the camera.

Pivot X, Y, and Z
These controls are used to set the exact position of the axis for the currently selected shading
element. By default, the axis is positioned at the calculated center of the line, word, or character. The
axis control works as an offset, such that a value of 0.1, 0.1 in this control would cause the axis to be
shifted downward and to the right for the shading element. Positive values in the Z-axis slider will move
the axis of rotation further along the axis (away from the viewer). Negative values will bring the axis of
rotation closer.

Rotation X, Y, and Z
These controls are used to adjust the angle of the currently selected shading element in any of the
three dimensions.

Chapter 43 Generator Nodes 1021

Shear X and Y
Adjust these sliders to modify the slanting of the currently selected shading element along the X
and Y axis.

Size X and Y
Adjust these sliders to modify the size of the currently selected shading element along the
X and Y axis.

Common Controls
Image and Settings Tabs
The Image and Settings tabs in the Inspector are duplicated in many Generator nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Text+ Toolbar
When the Text node is selected, a toolbar will appear in the viewer. Each button is described below
from left to right.

Text+ toolbar

Allow Typing in Viewer

When selected, you can type and edit text directly in the viewer. Clicking on the text in the viewer will
move the insert cursor within the text. Left and Right Arrow keys will move the insert cursor between
characters. Using the Up and Down Arrows will move between text lines.

Allow Manual Kerning

The Manual Kerning button overrides the automatic kerning normally applied to text. Clicking the small
red handle at the bottom of any character selects the character for kerning. You can also draw a
selection rectangle around the characters you want to kern. Once any characters are selected, hold
down the Option (macOS) or Alt key (Windows) while pressing the Left or Right Arrow key to make
small adjustments to the kerning of the selected characters. Hold the Alt + Shift (Windows) or Option +
Shift (macOS) keys down while pressing arrow keys to move the character in larger increments.
To animate the position of each character, right-click on the control label Manual Font Kerning in the
Inspector’s Advanced Controls and select Animate from the contextual menu. A new key will be set on
the animation spline each time a character is moved. All characters are animated with the same spline,
as with polyline mask animation.

No Text Outline
When this button is selected, it disables the drawing of any outline around the edges of the text. The
outline is not a part of the text; it is an onscreen control used to help identify the position of the text.
This is a three-way toggle with the Text Outline Outside Frame Only, and Show Always Text
Outline buttons.

Text Outline Outside Frame Only

This button draws an outline around the edges of text, which is outside the visible frame. This is useful
for locating text that has moved offscreen and is no longer rendering a visible result. This is a three-
way toggle with the No Text Outline, and Show Always Text Outline buttons.

Chapter 43 Generator Nodes 1022

Show Always Text Outline
This button draws an outline around the edges of text, whether or not the text is visible within the
frame. This is a three-way toggle with the No Text Outline, and Text Outline Outside Frame
Only buttons.

Common Controls
Image and Settings Tabs
The Image and Settings tabs in the Inspector are duplicated in many Generator nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Text+ Modifiers

Text+ modifiers

Text modifiers can be assigned by right-clicking in the Styled Text box and selecting a modifier from
the contextual menu. Once a modifier is selected, its controls are found in the Modifiers tab at the top
of the Inspector.

Character Level Styling

The Character Level Styling modifier works only on Text+ nodes. Once applied, individual characters
can be selected directly in the viewer, and different text attributes can be applied to them using the
controls in the Modifiers tab.

NOTE: Character Level Styling can only be directly applied to Text+ nodes, not to Text 3D
nodes. However, styled text from a Text+ node can be applied to a Text 3D node by copying
the Text+, right-clicking on the Text 3D, and choosing Paste Settings.

Chapter 43 Generator Nodes 1023

Inspector

Character Level Styling Modifiers tab

Text Tab
The Styled Text box in the Modifiers tab displays the same text in the Tools tab of the Text+ Inspector.
However, individual characters you want to modify cannot be selected in the Styled Text box; they
must be selected in the viewer. Once text is selected in the viewer, the Text tab includes familiar text
formatting options that will apply only to the selected characters.

Clear Character Styling on Selection

All changes made to the currently selected characters will be reset.

Clear All Character Styling

All character attributes will be reset to their original values.

Transform Tab and Shading Tab

For details on these Text+ tabs, see the “Text+” section above.

Comp Name
The Comp Name sets the styled text to become the current Composition Name. This is quite useful to
automate burn-ins for daily renderings. See also the TimeCode modifier. It can be applied by right-
clicking in the Styled Text field of a Text+ node and selecting Comp Name.

Controls
This modifier has no controls.

Follower
The Follower modifier allows sequencing text animations. The modifier is applied by right-clicking in
the Styled Text field of a Text+ node and selecting Follower. In the Modifiers tab, you start by
animating the parameters of the text (note that changing any parameter in the Modifiers tab will not be
visible unless a keyframe is added.) Then, in the Timing tab you set the animation’s delay between
characters.

Chapter 43 Generator Nodes 1024

Inspector

Follower Timing tab

Timing Tab
Once the text is animated using the controls in the Modifiers tab, the Timing tab is used to choose the
direction and offset of the animation.

Range
The Range menu is used to select whether all characters should be influenced or only a selected
range. To set the range, you can drag-select over the characters directly in the viewer.

Order
The Order menu determines in which direction the characters are influenced. Notice that empty
spaces are counted as characters as well. Available options include:
– Left to right: The animation ripples from left to right through all characters.
– Right to left: The animation ripples from right to left through all characters.
– Inside out: The animation ripples symmetrically from the center point of the
characters toward the margin.
– Outside in: The animation ripples symmetrically from the margin toward the center point of the
characters.
– Random but one by one: The animation is applied to randomly selected characters but only
influences one character at a time.
– Completely random: The animation is applied to randomly selected characters, influencing
multiple characters at a time.
– Manual curve: The affected characters can be specified by sliders.

Delay Type
Determines what sort of delay is applied to the animation. Available options include:
– Between Each Character: The more characters there are in your text, the longer the animation
will take to the end. A setting of 1 means the first character starts the animation, and the second
character starts 1 frame later, the third character starts 1 frame after the second, and so on.
– Between First and Last Character: No matter how many characters are in your text, the animation
will always be completed in the selected amount of time.

Clear All Character Styling

All character attributes will be reset to their original values.

Text Controls, Alignment, Transform, and Shading Tabs

In these tabs, the actual animation for the characters is done. Observe that simply changing a value in
these tabs will have no influence at all. The value must be animated for the effect to show.
For a detailed description on the various parameters, see the Text+ node documentation.

Chapter 43 Generator Nodes 1025

Text Scramble
The Text Scramble randomly replaces the characters with others from a user definable set. It can be
applied by right-clicking into the Styled Text field of a Text+ node and selecting Text Scramble.
The Controls for the Text Scramble are then adjusted in the Modifiers tab.

Inspector

Text Scramble modifier Controls tab

Controls Tab
The Controls tab in the Text Scramble modifier is used to enter text and scramble it using the
Randomness control. The scrambled characters are taken from the Substitute Chars field at the bottom
of the Inspector.

Randomness
Defines how many characters are exchanged randomly. A value of 0 will change no characters at all.
A value of 1 will change all characters in the text. Animating this thumbwheel to go from 0 to 1 will
gradually exchange all characters.

Input Text
This reflects the original text in the Text+ Styled Text field. Text can be entered either here or in the
Text+ node.

Animate on Time
When enabled, the characters are scrambled randomly on every new frame. This switch has no effect
when Randomness is set to 0.

Animate on Randomness
When enabled, the characters are scrambled randomly on every new frame, when the Randomness
thumbwheel is animated.
This switch has no effect when Randomness is set to 0.

Don’t Change Spaces

When enabled, spaces are not scrambled, allowing the length of the single words to stay the same,
although their characters get scrambled around.

Substitute Chars
This field contains the characters used to scramble the text.

Chapter 43 Generator Nodes 1026

Text Timer
The Text Timer makes the Text+ node either a Countdown, a Timer, or a Clock. This is useful for
onscreen real-time displays or to burn in the creation time of a frame into the picture. It can be applied
by right-clicking in the Styled Text field of a Text+ node and selecting Text Timer.

Inspector

Text Timer modifier Controls tab

Controls Tab
The Controls tab for the Text Timer modifier is used to set up the type of time display that is generated
by this modifier.

Mode
This menu sets the mode the timer is working in. The choices are CountDown, Timer, and Clock. In
Clock mode, the current system time will be displayed.

Hrs, Mins, Secs (Checkboxes)

Defines which parts of the clock should be shown onscreen.

Hrs, Mins, Secs (Sliders)

Sets the start time for the CountDown and Timer mode.

Start
Starts the Counter or Timer. Toggles to Stop once the timer is running.

Reset
Resets the Counter and Timer to the values set by the sliders.

Time Code
The Time Code only works on Text+ nodes. It sets the Styled text to become a counter based on the
current frame. This is quite useful for automating burn-ins for daily renderings.
It can be applied by right-clicking in the Styled Text field of a Text+ node and selecting Time Code.

Inspector

Time Code modifier Controls tab

Chapter 43 Generator Nodes 1027
Controls Tab
The Controls tab for the Time Code modifier is used to set up the time code display that is generated
by this modifier.

Hrs, Mins, Secs, Frms, Flds

Activate or deactivate these options to customize the time code display to show hours, minutes,
seconds, frames, and fields, respectively. Activating frames will only give you a plain frame counter.

Start Offset
Introduce a positive or negative offset to Fusion’s current time to match up with existing time codes.

Frames per Second

This should match with your composition’s FPS setting to provide accurate time measurement.

Drop Frame
Activate this checkbox to match the time code with footage that has drop frames—for example, certain
NTSC formats.

The Common Controls

Nodes that generate images share a number of identical controls in the Inspector. This section
describes controls that are common among Generator nodes.

Inspector

Background node Image tab

Image Tab
The controls in this tab are used to set the resolution, color depth, and pixel aspect of the image
produced by the node.

Process Mode
Use this menu control to select the Fields Processing mode used by Fusion to render changes to the
image. The default option is determined by the Has Fields checkbox control in the Frame
Format preferences.

Chapter 43 Generator Nodes 1028

Global In and Out
Use this control to specify the position of this node within the project. Use Global In to specify which
frame that starts the clip and Global Out to specify which frame ends the clip (inclusive) within the
project’s Global Range.
The node will not produce an image on frames outside this range.

Use Frame Format Settings

When this checkbox is selected, the width, height, and pixel aspect of the image created by the node
will be locked to values defined in the composition’s Frame Format preferences. If the Frame Format
preferences change, the resolution of the image produced by the node will change to match. Disabling
this option can be useful to build a composition at a different resolution than the eventual target
resolution for the final render.

Width/Height
This pair of controls is used to set the Width and Height dimensions of the image to be created
by the node.

Pixel Aspect
This control is used to specify the Pixel Aspect ratio of the created images. An aspect ratio of 1:1 would
generate a square pixel with the same dimensions on either side (like a computer display monitor), and
an aspect of 0.9:1 would create a slightly rectangular pixel (like an NTSC monitor).

NOTE: Right-click on the Width, Height, or Pixel Aspect controls to display a menu listing the
file formats defined in the preferences Frame Format tab. Selecting any of the listed options
will set the width, height, and pixel aspect to the values for that format, accordingly.

Depth
The Depth drop-down menu is used to set the pixel color depth of the image created by the Creator
node. 32-bit pixels require 4X the memory of 8-bit pixels but have far greater color accuracy. Float
pixels allow high dynamic range values outside the normal 0…1 range, for representing colors that are
brighter than white or darker than black.

Source Color Space

You can use the Source Color Space menu to set the Color Space of the footage to help achieve a
linear workflow. Unlike the Gamut tool, this doesn‘t perform any actual color space conversion, but
rather adds the source space data into the metadata, if that metadata doesn‘t exist. The metadata can
then be used downstream by a Gamut tool with the From Image option, or in a Saver, if explicit output
spaces are defined there. There are two options to choose from:
– Auto: Automatically reads and passes on the metadata that may be in the image.
– Space: Displays a Color Space Type menu where you can choose the correct color
space of the image.

Source Gamma Space

Using the Curve type menu, you can set the Gamma Space of the footage and choose to remove it by
way of the Remove Curve checkbox when working in a linear workflow. There are three choices in the
Curve type menu:
– Auto: Automatically reads and passes on the metadata that may be in the image.
– Space: Displays a Gamma Space Type menu where you can choose the correct gamma
curve of the image.
– Log: Brings up the Log/Lin settings, similar to the Cineon tool. For more information, see
Chapter 38, “Film Nodes” in the Fusion Reference Manual or Chapter 99 in the DaVinci Resolve
Reference Manual.

Chapter 43 Generator Nodes 1029

Remove Curve
Depending on the selected Gamma Space or on the Gamma Space found in Auto mode, the Gamma
Curve is removed from, or a log-lin conversion is performed on, the material, effectively converting it
to a linear output space.

Fast Noise Image Tab Options

The following controls are specific the the Image tab in the Fast Noise Node.

Mask Map Inputs

These external connections allow you to use masks to control the value of the Noise Detail and
Brightness controls individually for each pixel. This can allow some interesting and creative effects.

Noise Detail Map

A soft-edged mask connected to the Noise Detail Map will give a flat noise map (zero detail) where the
mask is black, and full detail where it is white, with intermediate values smoothly reducing in detail. It is
applied before any gradient color mapping. This can be very helpful for applying maximum noise detail
in a specific area, while smoothly falling off elsewhere.

Noise Brightness Map

A mask connected to this input can be used to control the noise map completely, such as boosting it in
certain areas, combining it with other textures, or if Detail is set to 0, replacing the Perlin Noise map
altogether.

Settings Tab
The Settings Tab in the Inspector can be found on every tool in the Color category. The Settings
controls are even found on third-party Color-type plug-in tools. The controls are consistent and work
the same way for each tool, although some tools do include one or two individual options, which are
also covered here.

Common Generator settings

Chapter 43 Generator Nodes 1030

Blend
The Blend control is used to blend between the tool’s original image input and the tool’s final modified
output image. When the blend value is 0.0, the outgoing image is identical to the incoming. Normally,
this will cause the tool to skip processing entirely, copying the input straight to the output.

Process When Blend Is 0.0

The tool is processed even when the input value is zero. This can be useful if processing of this node
is scripted to trigger another task, but the value of the node is set to 0.0.

Red/Green/Blue/Alpha Channel Selector

These four buttons are used to limit the effect of the tool to specified color channels. This filter is often
applied after the tool has been processed.
For example, if the red button on a Blur tool is deselected, the blur will first be applied to the image,
and then the red channel from the original input will be copied back over the red channel of the result.
There are some exceptions, such as tools for which deselecting these channels causes the tool to skip
processing that channel entirely. Tools that do this will generally possess a set of identical RGBA
buttons on the Controls tab in the tool. In this case, the buttons in the Settings and the Controls tabs
are identical.

Apply Mask Inverted

Enabling the Apply Mask Inverted option inverts the complete mask channel for the tool. The mask
channel is the combined result of all masks connected to or generated in a node.

Multiply by Mask
Selecting this option will cause the RGB values of the masked image to be multiplied by the mask
channel’s values. This will cause all pixels of the image not in the mask (i.e., set to 0) to become black/
transparent.

Use Object/Use Material (Checkboxes)

Some 3D software can render to file formats that support additional channels. Notably, the EXR file
format supports Object ID and Material ID channels, which can be used as a mask for the effect. These
checkboxes determine whether the channels will be used if present. The specific Material ID or Object
ID affected is chosen using the next set of controls.

Correct Edges
This checkbox appears only when the Use Object or Use Material checkboxes are selected. It toggles
the method used to deal with overlapping edges of objects in a multi-object image. When enabled,
the Coverage and Background Color channels are used to separate and improve the effect around the
edge of the object. If this option disabled (or no Coverage or Background Color channels are
available), aliasing may occur on the edge of the mask.
For more information on the Coverage and Background Color channels, see Chapter 18
“Understanding Image Channels” in the Fusion Reference Manual or Chapter 79 in the
DaVinci Resolve Reference Manual.

Object ID/Material ID (Sliders)

Use these sliders to select which ID will be used to create a mask from the object or material channels
of an image. Use the Sample button in the same way as the Color Picker: to grab IDs from the image
displayed in the view. The image or sequence must have been rendered from a 3D software package
with those channels included.

Chapter 43 Generator Nodes 1031

Motion Blur
– Motion Blur: This toggles the rendering of Motion Blur on the tool. When this control is toggled
on, the tool’s predicted motion is used to produce the motion blur caused by the virtual camera’s
shutter. When the control is toggled off, no motion blur is created.
– Quality: Quality determines the number of samples used to create the blur. A quality setting of 2
will cause Fusion to create two samples to either side of an object’s actual motion. Larger values
produce smoother results but increase the render time.
– Shutter Angle: Shutter Angle controls the angle of the virtual shutter used to produce the motion
blur effect. Larger angles create more blur but increase the render times. A value of 360 is the
equivalent of having the shutter open for one whole frame exposure. Higher values are possible
and can be used to create interesting effects.
– Center Bias: Center Bias modifies the position of the center of the motion blur. This allows the
creation of motion trail effects.
– Sample Spread: Adjusting this control modifies the weighting given to each sample. This affects
the brightness of the samples.

Use GPU
The user GPU menu has three settings. Setting the menu to Disable turns off hardware-accelerated
rendering using the graphics card in your computer. Enabled uses the hardware, and Auto uses a
capable GPU if one is available and falls back to software rendering when a capable GPU is
not available

Comments
The Comments field is used to add notes to a tool. Click in the field and type the text. When a note is
added to a tool, a small red square appears in the lower-left corner of the node when the full tile is
displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the note
in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 43 Generator Nodes 1032

Chapter 44

I/O Nodes
This chapter details the input and output of media using Loader and Saver nodes
within Fusion Studio as well as the MediaIn and MediaOut nodes in DaVinci Resolve.

The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

Contents
Loader Node [LD]  1034
MediaIn Node [MI]  1042
MediaOut Node [MO]  1046
Saver Node [SV]  1047
The Common Controls  1054

Chapter 44 I/O Nodes 1033

Loader Node [LD]

The Loader node

Loader Node Introduction

NOTE: The Loader node in DaVinci Resolve is only used for importing EXR files.

When using Fusion Studio, the Loader node is the node you use to select and load footage from your
hard drive into the Node Editor. There are three ways to add a Loader node, and consequently a clip,
to the Node Editor.
– Add the Loader from the Effects Library or toolbar (Fusion Studio only), and then use Loader’s
file browser to bring a clip into the Node Editor
– Drag clips from an OS window directly into the Node Editor, creating a Loader node in the Node
Editor.
– Choose File > Import > Footage (Fusion Studio only), although this method creates a new
composition as well as adds the Loader node to the Node Editor.

When a Loader is added to the Node editor, a File dialog is displayed automatically to allow the
selection of a clip from your hard drives.

NOTE: You can disable the automatic display of the file browser by disabling Auto Clip
Browse in the Global > General Preferences.

Once clips are brought in using the Loader node, the Loader is used for trimming, looping, and
extending the footage, as well as setting the field order, pixel aspect, and color depth. The Loader is
arguably the most important tool in Fusion Studio.

Inputs
The single input on the Loader node is for an effect mask to crop the image brought in by the Loader.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the loaded
image to appear only within the mask. An effects mask is applied to the tool after the tool is
processed.

Basic Node Setup

The Loader node is a 2D image of any format supported in Fusion Studio. It is limited to an EXR format
in DaVinci Resolve. Below, the LOADER imports an image, which is then masked using an Ellipse
matte. The output of the masked LOADER is passed onto 2D image-processing nodes.

Chapter 44 I/O Nodes 1034

A Loader node used for importing images

Inspector

Loader node File tab

File Tab
The File tab for the Loader includes controls for trimming, creating a freeze frame, looping, and
reversing the clip. You can also reselect the clip that the Loader links to on your hard drive.

Global In and Out

The Global In and Out handles are used to specify the position of this node within the project. Use
Global In to specify the frame on which that the clip starts and use Global Out to specify the frame on
which the clip ends within the project’s global range. The node does not produce an image on frames
outside of this range.
If the Global In and Out values are decreased to the point where the range between the In and Out
values is smaller than the number of available frames in the clip, Fusion automatically trims the clip by
adjusting the Trim range control. If the Global In/Out values are increased to the point where the range
between the In and Out values is larger than the number of available frames in the clip, Fusion
automatically lengthens the clip by adjusting the Hold First/Last Frame controls. Extended frames are
visually represented in the range control by changing the color of the held frames to green in
the control.
To slide the clip in time or move it through the project without changing its length, place the mouse
pointer in the middle of the range control and drag it to the new location, or enter the value manually in
the Global In value box.

Filename
The Filename field shows the file path of the clip imported to the Node Editor by the Loader node.
Clicking on the Browse button opens a standard file browser. The path to the footage can also be
typed directly using the field provided. The text box supports filename completion. As the name of a
directory or file is typed in the text box, Fusion displays a pop-up that lists possible matches. Use the
arrow keys to select the correct match and complete the path.

Chapter 44 I/O Nodes 1035

 NOTE: Loading image sequences is common practice for compositing, whether the image
sequence comes from a 3D renderer or a digital cinema camera. If the last part of a file’s
name is a number (not counting the file extension), Fusion automatically scans the directory
looking for files that match the sequence. For example, the following filenames would be valid
sequences:
image.0001.braw , image.0002.braw, image.0003.braw …
or
image151.exr , image152.exr, image153.exr …

The following would not be considered a sequence since the last characters are not numeric.
shot.1.fg.jpg, shot.2.fg.jpg, shot.3.fg.jpg

It is not necessary to select the first file in the sequence. Fusion searches the entire folder for
files matching the sequence in the selected filename. Also, Fusion determines the length of
the sequence based on the first and last numeric value in the filenames. Missing frames are
ignored. For example, if the folder contains two files with the following names:
image.0001.exr, image.0100.exr

Fusion sees this as a file sequence with 100 frames, not an image sequence containing two
frames. The Missing Frames drop-down menu is used to choose how Fusion handles
missing frames.
The Trim In/Trim Out control’s context menu can also be used to force a specific clip length or
to rescan the folder. Both controls are described in greater detail below.
Occasionally, you want to load only a single frame out of a sequence—e.g., a photograph from
a folder containing many other files as well. By default, Fusion detects those as a sequence,
but if you hold Shift while dragging the file from the OS window to the Node Editor, Fusion
takes only that specific file and disregards any sequencing.

Proxy Filename
The Proxy Filename control only appears once the filename control points to a valid clip. The Proxy
Filename can specify a clip that is loaded when the Proxy mode is enabled. This allows smaller
versions of the image to be loaded to speed up file I/O from disk, and processing. For example, create
a 1/4-scale version of an 8K EXR sequence to use as EXR proxy files. Whenever the Proxy mode of the
Composition is enabled, the smaller resolution proxy clip is loaded from disk, and all processing is
performed at the lower resolution, significantly improving render times. This is particularly useful when
working with large RAW plates stored on a remote file server. Lower-resolution versions of the plates
can be stored locally, reducing network bandwidth, interactive render times, and memory usage.The
proxy clip must have the same number of frames as the source clip, and the sequence numbers for the
clip must start and end on the same frame numbers. It is strongly suggested that the proxies are the
same format as the main files. In the case of formats with options, such as Cineon, DPX, and OpenEXR,
the proxies use the same format options as the primary.

Trim
The Trim range control is used to trim frames from the start or end of a clip. Adjust the Trim In to
remove frames from the start and adjust Trim Out to specify the last frame of the clip. The values used
here are offsets. A value of 5 in Trim In would use the fifth frame in the sequence as the start, ignoring
the first four frames. A value of 95 would stop loading frames after the 95th.

Chapter 44 I/O Nodes 1036

Hold First Frame/Hold Last Frame:
The Hold First Frame and Hold Last Frame controls hold the first or last frame of the clip for the
specified amount of frames. Held frames are included in a loop if the footage is looped.

Reverse
Select this checkbox to reverse the footage so that the last frame is played first, and the first frame is
played last.

Loop
Select this checkbox to loop the footage until the end of the project. Any lengthening of the clip using
Hold First/Last Frame or shortening using Trim In/Out is included in the looped clip.

Missing Frames
The Missing Frames menu determines the Loader behavior when a frame is missing or is unable to
load for any reason.
– Fail: The Loader does not output any image unless a frame becomes available. Rendering is
aborted.
– Hold Previous Output: The last valid frame is held until a frame becomes available again. This fails
if no valid frame has been seen—for example, if the first frame is missing.
– Output Black: Outputs a black frame until a valid frame becomes available again.
– Wait: Fusion waits for the frame to become available, checking every few seconds. This is useful
for rendering a composition simultaneously with a 3D render. All rendering ceases until the
frame appears.

The Magic Comp Variable

Loaders and Savers use the absolute file paths for the location of media. However, if you are
using Fusion Studio, you can use a file path that is relative to the saved composition location.
The Comp variable works for Loaders and Savers and helps you to keep your work organized.
Entering Comp:\ in place of the full file path name is a shortcut for the folder where your Fusion
composition document is saved.
You can either enter the Comp:\ manually into the filename field of a Loader, or turn on the
Enable Reverse Mapping of Paths Preferences checkbox in the Path Map preferences.
Enabling the Path Map preference check box will use the Comp:\ automatically.
So as long as all your source footage is stored in subfolders of your Comp folder, Fusion finds
that footage regardless of the actual hard drive or network share name.
You could, for example, copy an entire shot from the network to your local drive, set up your
Loaders and Savers to use the Comp variable, work all your magic locally (i.e., set up your
composition), and then copy just the composition back to the server and issue a net-render.
All render slaves automatically find the source footage.

Some examples:
Your composition is stored in:
X:\Project\Shot0815\Fusion\Shot0815.comp

Your source footage sits in:

X:\Project\Shot0815\Fusion\Greenscreen\0815Green_0000.dpx

Chapter 44 I/O Nodes 1037

 The relative path in the Loader node would then be:
Comp:\Greenscreen\0815Green_0000.dpx

If your source footage is stored in:

X:\Project\Shot0815\Footage\Greenscreen\0815Green_0000.dpx

The relative path in the Loader node would then be:

Comp:\..\Footage\ Greenscreen\0815Green_0000.dpx

Observe how the two dots .. set the directory to go up one folder, much like CD .. in a
Command Shell window.

Loader Import tab

Import Tab
The Import tab includes settings for the frame format and how to deal with fields, pixel aspect, 3:2 pull
down/pull up conversion, and removing gamma curve types for achieving a linear workflow.

Process Mode
Use this menu to select the Fields Processing mode used by Fusion when loading the image. The Has
Fields checkbox control in the Frame Format preferences determines the default option, and the
default height as well. Available options include:
– Full frames
– NTSC fields
– PAL/HD fields
– PAL/HD fields (reversed)
– NTSC fields (reversed).

The two reversed options load fields in the opposite order and thus result in the fields being spatially
swapped both in time order and in vertical order as well.
Use the Swap Field Dominance checkbox (described below) to swap fields in time only.

Chapter 44 I/O Nodes 1038

Depth
The Depth menu is used to select the color depth used to process footage from this Loader. The
default option is Format.
– Format: The color depth is determined by the color depth supported in the file format loaded.
For example, JPEG files automatically process at 8 bit because the JPEG file format does not
store color depths greater than 8. EXR files load at Float. If the color depth of the format is
undetermined, the default depth defined in the Frame Format preferences is used. Formats that
support multiple color depths are set to the appropriate color depth automatically.
– Default: The color depth is determined by the settings in the composition’s Frame Format
Preferences panel.
– Int 8 Bit/Int 16 Bit/Float 16/Float 32: These options set the color depth for processing the image.

Pixel Aspect
This menu is used to determine the image’s pixel aspect ratio.
– From File: The loader conforms to the image aspect detected in the saved file. There are a few
formats that can store aspect information. TIFF, JPEG, and OpenEXR are examples of image
formats that may have the pixel aspect embedded in the file’s header. When no aspect ratio
information is stored in the file, the default frame format method is used.
– Default: Any pixel aspect ratio information stored in the header of the image file is ignored. The
pixel aspect set in the composition’s frame format preferences is used instead.
– Custom: Select this option to override the preferences and set a pixel aspect for the clip manually.
Selecting this button causes the X/Y Pixel Aspect control to appear.

Custom Pixel Aspect

This control is visible only when Custom is selected from the Pixel Aspect menu. Enter the desired X
and Y aspect or right-click on the control to display a menu of common frame formats and
their aspects.

Import Mode
This menu provides options for removing pull-up from an image sequence. Pull-up is a reversible
method of combining frames used to convert 24 fps footage into 30 fps. It is commonly used to
broadcast NTSC versions of films.
– Normal: This passes the image without applying pull-up or pull-down 2:3.
– Pull Up: This removes existing 3:2 pull-down applied to the image sequence, converting from 30
fps to 24 fps 2:3.
– Pull Down: The footage has pull-down applied, converting 24 fps footage to 30 fps by creating
five frames out of every four. The process mode of a Loader set to Pull Down should always be
Full Frames.

First Frame
This menu appears when the Import Mode is set to either Pull Up or Pull Down. It is used to determine
which frame of the 3:2 sequence is used as the first frame of the loaded clip.

Detect Pull-Down Sequence

This button is used to detect and set the pull-up sequence of the footage automatically. It only works if
Pull-Up or Pull-Down is first selected from the Import Mode menu. If it succeeds in detecting the order,
the First Frame control automatically sets to the correct value.

Chapter 44 I/O Nodes 1039

Make Alpha Solid
When enabled, the original Alpha channel of the clip is cleared and set to solid white
(completely opaque).

Invert Alpha
When enabled, the original Alpha channel of the clip is inverted. This may also be used in conjunction
with Make Alpha Solid to set the Alpha to pure black (completely transparent).

Post-Multiply by Alpha
Enabling this option causes the color value of each pixel to be multiplied by the Alpha channel for that
pixel. This option can be used to convert subtractive (non-premultiplied) images to additive
(premultiplied) images.

Swap Field Dominance

When enabled, the field order (dominance) of the image is swapped, so the order in time the fields
appear is reversed. Unlike the Process Mode control, this is done without spatially swapping the
scanlines in the image

Color Space Type

This menu is used to set the Color Space of the footage to help achieve a linear color space workflow.
Unlike the Gamut tool, this doesn’t perform any actual color space conversion, but instead adds the
source color space data into the metadata, if that metadata doesn‘t already exist. The metadata can
then be used downstream by a Gamut tool with the From Image option, or in a Saver if explicit output
spaces are defined there.
– Auto: Passes along any metadata that might be in the incoming image.
– Space: Allows the user to set the color space based on the recording device used to capture
content or software settings used when rendering the content in another application.

Curve Type
This menu is used to determine the gamma curve of the footage. Once the Gamma Curve Type is set,
you can choose to remove the curve to help achieve a linear workflow.
– Auto: Passes along any metadata that might be in the incoming image.
– Space: Allows the user to set the gamma curve based on the recording device used to capture
content or software settings used when rendering the content in another application.
– Log: Displays the Log/Lin settings, similar to the Cineon Log node. For more information on the
Log settings, refer to Chapter 38, “Film Nodes” in the Fusion Reference Manual or Chapter 99 in
the DaVinci Resolve Reference Manual.

Remove Curve
Depending on the selected Curve Type or on the Gamma Space found in Auto mode, the associated
Gamma Curve is removed from, or a log-lin conversion is performed on, the material, effectively
converting it to a linear output space.

Chapter 44 I/O Nodes 1040

Format tab

Format Tab
The Format tab contains file format-specific controls that dynamically change based on the selected
Loader and the file it links to. Some formats contain a single control or no controls at all. Others like
Camera RAW formats contain RAW-specific debayering controls. A partial format list is provided below
for reference.
– OpenEXR: EXR provides a compact and flexible format to support high dynamic range images
(float). The format also supports a variety of extra channels and metadata.The Format tab for
OpenEXR files provides a mechanism for mapping any non-RGBA channels to the channels
supported natively in Fusion. Using the Format tab, you can enter the name of a channel contained
in the OpenEXR file into any of the edit boxes next to the Fusion channel name. A command line
utility for dumping the names of the channels can be found at https://www.openexr.com.
– QuickTime: QuickTime files can potentially contain multiple tracks. Use the format options to
select one of the tracks.
– Cinema DNG: CinemaDNG is an open format capable of high-resolution raw image data with a
wide dynamic range. It was one of the formats recorded by Blackmagic Design cameras before
switching over to the BRAW format.
– Photoshop PSD Format: Fusion can load any one of the individual layers stored in the PSD file, or
the completed image with all layers. Transformation and adjustment layers are not supported. To
load all layers in a PSD file with appropriate blend modes, use File > Import > PSD.

Common Controls
Settings Tab
The Settings tab controls are common to both Loader and Saver nodes, so their descriptions can be
found in “The Common Controls” section at the end of this chapter.

Chapter 44 I/O Nodes 1041

MediaIn Node [MI]

The MediaIn node

MediaIn Node Introduction

NOTE: The MediaIn node is only available in DaVinci Resolve.

The MediaIn node is the foundation of every composition you create in DaVinci Resolve’s Fusion page.
In most cases, it replaces the Loader node used in Fusion Studio for importing clips. There are four
ways to add a MediaIn node to the Node Editor.
– In the Edit or Cut page, position the playhead over a clip in the Timeline, and then click the Fusion
page button. The clip from the Edit or Cut page Timeline is represented as a MediaIn node in the
Node Editor.
– Drag clips from the Media Pool into the Node Editor, creating a MediaIn node in the Node Editor.
– Drag clips from an OS window directly into the Node Editor, creating a MediaIn node
in the Node Editor.
– Choose Fusion > Import> PSD when importing PSD files into the Node Editor. Each PSD layer is
imported as a separate MediaIn node.

NOTE: Although a MediaIn tool is located in the I/O section of the Effects Library, it is not
used as a method to import clips.

When clips are brought in from the Media Pool, dragged from the OS window, or via the Import
PSD menu option, you can use the MediaIn node’s Inspector for trimming, looping, and extending the
footage, as well as setting the source’s color and gamma space.

Inputs
The single input on the MediaIn node is for an effect mask to crop the image brought in by
the MediaIn.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive
shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the
source image to appear only within the mask. An effects mask is applied to the tool after the
tool is processed.

Basic Node Setup

The MediaIn node is typically the starting point for all composites done in the Fusion page of
DaVinci Resolve. It contains the clip from the Edit page or Cut page. Any clip brought in from the Media
Pool is also added as a MediaIn node.

Chapter 44 I/O Nodes 1042

Two MediaIn nodes: one from the Edit page Timeline and one from the Media Pool

Inspector

MediaIn node Image tab

Image Tab
When brought in from the Media Pool or dragged from an OS window, the MediaIn node’s Image tab
includes controls for trimming, creating a freeze frame, looping, and reversing the clip. You can also
reselect the clip the MediaIn links to on your hard drive. A subset of these controls is available when
the MediaIn node is brought in from the Edit or Cut page Timeline.

Global In and Out

Only used when a clip is brought in through the Media Pool or an OS window, the Global In and Out
handles are used to specify the start and end of this node within the Fusion effect. Use Global In to
specify the frame on which the clip starts and use Global Out to specify the frame on which the clip
ends within the comp’s global range. The node does not produce an image on frames outside of
this range.
If the Global In and Out values are decreased to the point where the range between the In and Out
values is smaller than the number of available frames in the clip. Fusion automatically trims the clip by
adjusting the Trim range control. If the Global In/Out values are increased to the point where the range
between the In and Out values is larger than the number of available frames in the clip, Fusion
automatically lengthens the clip by adjusting the Hold First/Last Frame controls. Extended frames are
visually represented in the range control by changing the color of the held frames to green in
the control.
To slide the clip in time or move it through the project without changing its length, place the mouse
pointer in the middle of the range control and drag it to the new location, or enter the value manually in
the Global In value box.

Chapter 44 I/O Nodes 1043

Process Mode
Use this menu to select the Fields Processing mode used by Fusion when loading the image. The Has
Fields checkbox control in the Frame Format preferences determines the default option, and the
default height as well. Available options include:
– Full frames
– NTSC fields
– PAL/HD fields
– PAL/HD fields (reversed)
– NTSC fields (reversed).

The two reversed options load fields in the opposite order and thus result in the fields being spatially
swapped both in time order and in vertical order as well.

MediaID
An ID assigned by DaVinci Resolve for that clip.

Layer
Used to identify the layer in a PSD file or compound clip. When a PSD file is brought in from the Media
Pool, the drop-down menu allows you to select an individual layer for output instead of the entire PSD
composite.

Trim
The Trim range control is used to trim frames from the start or end of a clip. Adjust the Trim In to
remove frames from the start and adjust Trim Out to specify the last frame of the clip. The values used
here are offsets. A value of 5 in Trim In would use the fifth frame in the sequence as the start, ignoring
the first four frames. A value of 95 would stop loading frames after the 95th frame.

Hold First Frame/Hold Last Frame:

The Hold First Frame and Hold Last Frame controls hold the first or last frame of the clip for the
specified amount of frames. Held frames are included in a loop if the footage is looped.

Reverse
Select this checkbox to reverse the footage so that the last frame is played first, and the first frame is
played last.

Loop
Select this checkbox to loop the footage until the end of the project. Any lengthening of the clip using
Hold First/Last Frame or shortening using Trim In/Out is included in the looped clip.

Source Color Space

Lets you choose a color space for the image data output by this MediaIn node.
– Auto: uses the Timeline color space, or whichever color space is assigned by Resolve Color
Management (RCM) if it’s enabled.
– Space: Space lets you choose a specific setting from a Color Space drop-down menu, while a
visual “horseshoe” graph lets you see a representation of the color space you’ve selected.

Source Gamma Space

Lets you choose a gamma setting for the image data output by this MediaIn node. Once the gamma
curve type is set, you can choose to remove the curve to help achieve a linear workflow.
– Auto: Uses the Timeline gamma, or whichever gamma is assigned by Resolve Color Management
(RCM) if it’s enabled.

Chapter 44 I/O Nodes 1044

 – Space: Lets you choose a specific setting from a Gamma Space drop-down menu, while a visual
graph lets you see a representation of the gamma setting you’ve selected.
– Log: Displays the Log Type drop-down menu where you can choose a specific log encoding
profile. A visual graph shows a representation of the log setting chosen from the menu. Additional
Lock RGB, Level, Soft Clip, Film Stock Gamma, Conversion Gamma, and Conversion table options
are presented to finesse the gamma output.
– Remove Curve: The associated gamma curve is removed from, or a log-lin conversion is
performed on, the material, effectively converting it to a linear output space.
– Pre-Divide/Post-Multiply: Lets you convert “straight” Alpha channels into premultiplied Alpha
channels, when necessary.

The Audio tab in the MediaIn node is used

to select the track for playback, slip the
audio timing and reset the audio cache.

Audio Tab
The Inspector for the MediaIn node contains an Audio tab, where you can choose to solo the audio
from the clip or hear all the audio tracks in the Timeline.
If the audio is out of sync when playing back in Fusion, the Audio tab’s Sound Offset wheel allows you
to slip the audio in subframe frame increments. The slipped audio is only modified in the Fusion page.
All other pages retain the original audio placement.

Audio with Media Pool Clips

Audio from a clip brought in through the Media Pool is muted by default. Hearing the audio from a
Media Pool clip is a two step process.

To hear audio from a clip brought in through the Media Pool, do the following:
1 Select the clip in the Node Editor.
2 In the Inspector, click the Audio tab and select the clip name from the Audio Track
drop-down menu.
If more than one MediaIn node exists in the comp, the audio last selected in the Inspector is
heard. You can use the Speaker icon in the toolbar to switch between the MediaIn node
audio files.
3 Right-click the Speaker icon in the toolbar, then choose the MediaIn for the clip you want to hear.

Purging the Audio Cache

The audio and its settings are cached for faster performance. If you change which audio tracks you
want to play back in Fusion, or you use the Sound Offset wheel to slip the audio tracks, you need to
purge the audio cache. Also, if you return to the Edit, Cut, or Fairlight page and modify the audio
levels, you need to purge the audio cache.

Chapter 44 I/O Nodes 1045

To purge the audio cache after any change to the audio playback:
– Click the Purge Audio Cache button in the Inspector.

The audio will be updated when you next playback the composition.

MediaOut Node [MO]

The MediaOut node

MediaOut Node Introduction

NOTE: The MediaOut node is only available in DaVinci Resolve.

Every composition you create in DaVinci Resolve’s Fusion page must include a MediaOut node. The
MediaOut node sends the final output back to your Timeline on DaVinci Resolve’s Edit or Cut page. In
most cases, it replaces the Saver node used in Fusion Studio for exporting clips.
The composition output by the Fusion page’s MediaOut node is propagated via the Color page’s
source inputs, with the sole exception that if you’ve performed transforms or added plug-ins to that
clip in the Edit or Cut page, then the handoff from the Fusion page to the Color page is as follows:

Fusion Edit Color

Effects Page Effects
Plugs-ins

When using Resolve Color Management or ACES, each MediaOut node converts the output image
back to the Timeline color space for handoff to the Color page.

NOTE: Additional MediaOut nodes can be added to the Node Editor from the Effects Library.
Additional MediaOut nodes are used to pass mattes to the Color page.

Inputs
The single input on the MediaOut node is where you connect the final composite image you want
rendered back into the Edit page.
– Input: The orange input is a required input. It accepts any 2D image that you want rendered
back into the Edit page.

Chapter 44 I/O Nodes 1046

Basic Node Setup
Multiple MediaOut nodes can exist in a single comp. The first MediaOut node always renders the
image back to the Edit page Timeline. Additional MediaOut nodes can be used to pass mattes to the
Color page.

MediaOut1 node rendering to the Edit page, and MediaOut2 sending mattes to the Color page

Saver Node [SV]

The Saver node

Saver Node Introduction

NOTE: The Saver node in DaVinci Resolve is only used for exporting EXR files.

The Saver node represents the final composition output from Fusion Studio. It is used to render out
movie files or sequential images but can be inserted into a composition at any point to render out
intermediate stages of a composition. A composition can contain any number of Saver nodes for
rendering different branches of a comp as well as different formats.
The Saver node can also be used to add scratch track audio to your composition, which can be heard
during interactive playback.

Inputs
The single input on the Saver node is for the final composition you want to render.
– Image Input: The orange input is used to connect the resulting image you want rendered.

Chapter 44 I/O Nodes 1047

Basic Node Setup
The Saver node is placed at the end of the composition. Multiple Savers can be placed in a comp to
render different formats or to render different parts of a composition.

Saver node added to the end of a node tree to render the composition

Inspector

Saver File tab

File Tab
The Saver File tab is used to set the location and output format for the rendered file.

Filename
The Filename dialog is used to select the name and path of the rendered image output. Click on the
Browse button to open a file browser and select a location for the output.
Sequence numbering is automatically added to the filename when rendering a sequential image file
format. For example, if c\renders\image.exr is entered as the filename and 30 frames of output are
rendered, the files are automatically numbered as image0000.tga, image0001.exr, image0003.exr...
and so on. Four-digit padding is automatically used for numbers lower than 10000.
You can specify the number of digits to use for padding by explicitly entering the digits into
the filename.
For example, image000000.exr would apply 6-digit padding to the numeric sequence, image.001.exr
would use 3-digit padding, and image1.exr would use none.

Output Format
This menu is used to select the image format to be saved. Be aware that selecting a new format from
this menu does not change the extension used in the filename to match. Modify the filename manually
to match the expected extension for that format to avoid a mismatch between name and image format.

Chapter 44 I/O Nodes 1048

Save Frames
This control selects between two modes of rendering: Full Renders Only or High Quality Interactive.
– Full Renders Only: This is the common setting for most situations. Images are saved to disk when
a final render is started using the Start Render button in the Time Ruler.
– High Quality Interactive: This render mode is designed for real-time rendering when painting and
rotoscoping. Fusion saves each frame to disk as it is processed interactively. When used correctly,
this feature can eliminate the need to perform a final render after rotoscoping.

NOTE: The High Quality Interactive setting can easily cause confusion when used in
conjunction with a node tree that contains spline-animated parameters. If these splines are
modified in such a way that frames already saved interactively are changed, the frames
already on the disk do not automatically re-render. Either step through each frame again or
perform a final render to make certain that the result is correct

Frame Offset
This thumbwheel control can be used to set an explicit start frame for the number sequence applied to
the rendered filenames. For example, if Global Start is set to 1 and frames 1-30 are rendered, files are
normally numbered 0001-0030. If the Sequence Start Frame is set to 100, the rendered output would
be numbered from 100-131.

Saver Export tab

Export Tab
Process Mode
Use this menu to select the Fields Processing mode used by Fusion when saving the images or movie
file to disk. The Has Fields checkbox control in the Frame Format preferences determines the default
option, and the default height as well. Available options include: 
– Full frames
– NTSC fields
– PAL/HD fields
– PAL/HD fields (reversed)
– NTSC fields (reversed).

The two reversed options save fields in the opposite order and thus result in the fields being spatially
swapped both in time order and in vertical order as well.

Chapter 44 I/O Nodes 1049

Export Mode
This menu is used to render out the file normally or apply a SMPTE standard 3:2 pulldown to the
footage, converting the footage from 24 fps to 30 fps.

Clipping Mode
This menu, sometimes considered source image clipping, defines how the edges of the image should
be treated.
– Frame: The default Frame setting clips to the parts of the image visible within its visible
dimensions. It breaks any infinite-workspace behavior. If the upstream DoD is smaller than the
frame, the remaining area in the frame is treated as black/transparent.
– None: This setting does not perform any source image clipping at all. This means that any data
that would normally be needed outside the upstream DoD is treated as black/transparent. Be
aware that this might create humongous images that can consume a considerate amount of disk
space. So you should use this option only if really needed.

For more information about ROI, DoD, and Infinite Workspace, see Chapter 7, “Using Viewers” in the
Fusion Reference Manual or Chapter 68 in the DaVinci Resolve Reference Manual.

Save Alpha to Color

When enabled, this control causes the Alpha channel to be saved into the color channels as a
grayscale image. This completely overwrites any existing color information.

Color Space Type

This menu is used to set the Color Space of the output file so you could, for example, color space
convert images from linear to Rec709, etc., thereby delivering linear EXRs, 709 Quicktimes, etc. from a
single source. The images are not being converted in the Comp, only in the images saved to disk are
converted.
– Auto: Passes along any metadata that might be in the rendered image.
– Space: Allows the user to set the color space based on the output format.

Curve Type
This menu is used to select a Gamma curve of the rendered file. Once the gamma curve type is set,
you can choose to apply the curve for output.
– Auto: Passes along any metadata that might be in the incoming image.
– Space: Allows the user to set the gamma curve based on the selected file format.
– Log: Displays the Log/Lin settings, similar to the Cineon Log node. For more detail on the Log
settings, see Chapter 38, “Film Nodes” in the Fusion Reference Manual or Chapter 99 in the
DaVinci Resolve Reference Manual.

Apply Curve
Depending on the selected Curve Type or on the Gamma Space found in Auto mode, the associated
Gamma Curve is applied, effectively converting from a linear working space.

Chapter 44 I/O Nodes 1050

Audio Tab

Saver Audio tab

NOTE: This tab is only available in Fusion Studio’s Saver node.

The audio functionality is included in Fusion Studio for scratch track (aligning effects to audio and clip
timing) purposes only. Final renders should almost always be performed without audio. The smallest
possible audio files should be used, as Fusion loads the entire audio file into memory for efficient
display of the waveform in the Timeline. The audio track is included in the saved image if a Quicktime
file format is selected. Fusion currently supports playback of WAV audio.

Source Filename
You can enter the file path and name of the audio clip you want to use in the Source Filename field.
You can also click the Browse button to open a file browser window and locate the audio scratch track.
Select the WAV file of choice, and then in the keyframes panel expand the Saver bar to view the audio
waveform. Drag the pointer over the audio wave in the Timeline layout to hear the track.

Sound Offset
Drag the control left or right to slide the Timeline position of the audio clip, relative to other nodes in
the Node Editor.

Legal Tab
The Legal tab includes settings for creating “broadcast safe” saturation and video range files
for output.

Saver Legal tab

Video Type
Use this menu to select the standard to be used for broadcast legal color correction. NTSC, NHK, or
PAL/SECAM can be chosen.

Action
Use this menu to choose how Fusion treats illegal colors in the image.
– Adjust to Legal: This causes the images to be saved with legal colors relevant to the
Video Type selected.

Chapter 44 I/O Nodes 1051

 – Indicate as Black: This causes the illegal colors to be displayed as black in the views.
– Indicate as White: This causes the illegal colors to be displayed as white in the views.
– No Changes: This causes the images to be saved unaffected.

Adjust Based On
This menu is used to choose whether Fusion makes legal the image to 75% or 100% amplitude. Very
few broadcast markets permit 100% amplitude, but for the most part this should be left to 75%.

Soft Clip
The Soft Clip control is used to draw values that are out of range back into the image. This is done by
smoothing the conversion curve at the top and bottom of the curve, allowing more values to be
represented.

Format Tab

Saver Format tab

The Format tab contains information, options, and settings specific to the image format being saved.
The controls for an EXR sequence is entirely different from the ones displayed when a MOV
file is saved.
EXR is displayed above for reference.

When the Saver node is set to DPX, it’s important to understand the reason for the Bypass
Conversion > Data is Linear option. When saving log data into a DPX, and not using the
Saver’s node’s own lin-log conversion (that is, Bypass Conversion is checked), the Data Is
Linear option should be off. This indicates whether the reason for checking Bypass
Conversion is because the data is linear, or whether it’s already log.
If Data Is Linear is enabled, then the DPX is marked in its header as containing linear data. In
turn, that means that when the DPX is loaded back into Fusion, or into other apps that evaluate
the header, those apps think the data is linear and do not perform any log-lin conversion.

Chapter 44 I/O Nodes 1052

Common Controls
Settings Tab
The Settings tab controls are common to both Loader and Saver nodes, so their descriptions can be
found in “The Common Controls” section at the end of this chapter.

Clipping Mode
This menu, sometimes considered source image clipping, defines how the edges of the image should
be treated.
– Frame: The default Frame setting clips to the parts of the image visible within its visible
dimensions. It breaks any infinite-workspace behavior. If the upstream DoD is smaller than the
frame, the remaining areas in the frame are treated as black/transparent.
– None: This setting does not perform any source image clipping at all. This means that any data
that would normally be needed outside the upstream DoD is treated as black/transparent. Be
aware that this might create humongous images which can consume a considerable amount of
disk space. So you should use this option only if really needed.

For more information about ROI, DoD, and Infinite Workspace, see Chapter 7, “Using Viewers” in the
Fusion Reference Manual or Chapter 68 in the DaVinci Resolve Reference Manual.

Save Alpha to Color

When enabled, this control causes the Alpha channel to be saved into the color channels as a
grayscale image. This completely overwrites any existing color information.

Color Space Type

This menu is used to set the Color Space of the output file so you could, for example, color space
convert images from linear to Rec709, etc., thereby delivering linear EXRs, 709 Quicktimes, etc. from a
single source. The images are not being converted in the Comp, only in the images saved to disk are
converted.
– Auto: Passes along any metadata that might be in the rendered image.
– Space: Allows the user to set the color space based on the output format.

Curve Type
This menu is used to select a Gamma curve of the rendered file. Once the gamma curve type is set,
you can choose to apply the curve for output.
– Auto: Passes along any metadata that might be in the incoming image.
– Space: Allows the user to set the gamma curve based on the selected file format.
– Log: Displays the Log/Lin settings, similar to the Cineon Log node. For more detail on the Log
settings, see Chapter 38, “Film Nodes” in the Fusion Reference Manual or Chapter 99 in the
DaVinci Resolve Reference Manual.

Apply Curve
Depending on the selected Curve Type or on the Gamma Space found in Auto mode, the associated
Gamma Curve is applied, effectively converting from a linear working space.

Chapter 44 I/O Nodes 1053

The Common Controls
I/O nodes share a number of identical controls in the Inspector. This section describes controls that are
common among I/O nodes.

Inspector

Common Saver settings inspector

Settings Tab
The Settings tab in the Inspector can be found on the Loader, Saver, MediaIn, and MediaOut nodes.
The controls are consistent and work the same way for each tool, although some parameters are only
available on individual nodes but are covered here.

Blend
The Blend control is used to blend between the tool’s original image input and the tool’s final modified
output image. When the blend value is 0.0, the outgoing image is identical to the incoming image.
Normally, this causes the tool to skip processing entirely, copying the input straight to the output.

Process When Blend Is 0.0

The tool is processed even when the input value is zero. This can be useful if processing of this node
is scripted to trigger another task, but the value of the node is set to 0.0.

Red/Green/Blue/Alpha Channel Selector

These four buttons are used to limit the effect of the tool to specified color channels. This filter is often
applied after the tool has been processed.
For example, if the red button on a Blur tool is deselected, the blur is first applied to the image, and
then the red channel from the original input is copied back over the red channel of the result.

Chapter 44 I/O Nodes 1054

There are some exceptions, such as tools for which deselecting these channels causes the tool to skip
processing that channel entirely. Tools that do this generally possess a set of identical RGBA buttons
on the Controls tab in the tool. In this case, the buttons in the Settings and the Controls tabs are
identical.

Apply Mask Inverted

Enabling the Apply Mask Inverted option inverts the complete mask channel for the tool. The mask
channel is the combined result of all masks connected to or generated in a node.

Multiply by Mask
Selecting this option causes the RGB values of the masked image to be multiplied by the mask
channel’s values. This causes all pixels of the image not included in the mask (i.e., set to 0) to become
black/transparent.

Use Object/Use Material (Checkboxes)

Some 3D software can render to file formats that support additional channels. Notably, the EXR file
format supports Object and Material ID channels, which can be used as a mask for the effect. These
checkboxes determine whether the channels are used, if present. The specific Material ID or Object ID
affected is chosen using the next set of controls.

Correct Edges
This checkbox appears only when the Use Object or Use Material checkboxes are selected. It toggles
the method used to deal with overlapping edges of objects in a multi-object image. When enabled,
the Coverage and Background Color channels are used to separate and improve the effect around the
edge of the object. If this option disabled (or no Coverage or Background Color channels are
available), aliasing may occur on the edge of the mask.
For more information on Coverage and Background Color channels, see Chapter 18, “Understanding
Image Channels” in the Fusion Reference Manual or Chapter 79 in the DaVinci Resolve
Reference Manual.

Object ID/Material ID (Sliders)

Use these sliders to select which ID is used to create a mask from the object or material channels of an
image. Use the Sample button in the same way as the Color Picker: to grab IDs from the image
displayed in the view. The image or sequence must have been rendered from a 3D software package
with those channels included.

Hide Incoming Connections

Enabling this checkbox can hide connection lines from incoming nodes, making a node tree appear
cleaner and easier to read. When enabled, empty fields for each input on a node are displayed in the
Inspector. Dragging a connected node from the node tree into the field hides that incoming
connection line as long as the node is not selected in the node tree. When the node is selected in the
node tree, the line reappears.

Comments
The Comments field is used to add notes to a tool. Click in the empty field and type the text. When a
note is added to a tool, a small red square appears in the lower-left corner of the node when the full
tile is displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the
note in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 44 I/O Nodes 1055

Chapter 45

LUT Nodes
This chapter details the LUT nodes available in Fusion.

The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn/MediaOut nodes in

DaVinci Resolve are interchangeable with Loader/Saver nodes in Fusion Studio,
unless otherwise noted.

Contents
File LUT [FLU]  1057
LUT Cube Analyzer [LCA]  1059
LUT Cube Apply [LCP]  1060
LUT Cube Creator [LCC]  1061
The Common Controls  1063

Chapter 45 LUT Nodes 1056

File LUT [FLU]

The File LUT node

File LUT Node Introduction

The File LUT node applies a Lookup table (LUT) to the image: either a simple 1D LUT or a supported
3D LUT. Unlike the Color Curves node, it does not use a spline-based LUT. Instead, it loads the LUT
from a file stored on your computer or server.
This approach has two advantages. The first is that the only part of the LUT stored in the composition
is the path to the file. Since LUT files can be large, this can dramatically reduce the file size of a
composition when several LUTs are present. The second advantage is that it becomes possible to
adjust all File LUT nodes using the same file at the same time, just by changing the contents of the
LUT. This can be useful when the same LUT-based color correction is applied in many different
compositions.

Inputs
The File LUT node includes two inputs: one for the main image and the other for an effect mask to limit
the area where the LUT is applied.
– Input: This orange input is the only required connection. It accepts a 2D image output that gets
the LUT applied.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the applied LUT to only those pixels within the mask. An effects mask is applied to
the tool after the tool is processed.

Basic Node Setup

The File LUT node can be placed after a MediaIn node in DaVinci Resolve or a Loader node in Fusion
Studio. Sometimes this setup is used to convert the camera-original image to linear color space for
compositing. Other times, as in the example below, the File LUT is applied at the end as a grading LUT
to apply a look from the colorist.

A File LUT node applied at the end of a node tree as a colorist’s look

Chapter 45 LUT Nodes 1057

Inspector

File LUT controls

Controls Tab
The Controls tab includes options for loading a LUT and making adjustments to the gain, color space,
and Alpha channel, if one exists.

LUT File
This field is used to enter the path to the LUT file. Clicking the Browse button opens a file browser
window to locate the LUT file instead of entering it manually into the LUT File field. Currently, this node
supports LUTs exported from Fusion in .LUT and .ALUT formats, DaVinci Resolve’s .CUBE format, and
several 3D LUT formats. The node fails with an error message on the Console if it is unable to find or
load the specified file.

Pre-Gain:
This slider is a gain adjustment before the LUT being applied. This can be useful for pulling in
highlights before the LUT clips them.

Post-Gain
This slider is a gain adjustment after the LUT is applied.

Color Space
This menu is used to change the color space the LUT is applied in. The default is to apply the curves
described in the LUT to the RGB color space, but options for YUV, HLS, HSV, and others are also
available.

Pre-Divide/Post-Multiply
Selecting the Pre-Divide/Post-Multiply checkbox causes the image pixel values to be divided by the
Alpha values before applying the LUT, and then re-multiplied by the Alpha value after the correction.
This helps to prevent the creation of illegally additive images, particularly around the edges of a blue/
green key or when working with 3D-rendered objects.

Settings Tab
The Settings tab in the Inspector is also duplicated in other LUT nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 45 LUT Nodes 1058

LUT Cube Analyzer [LCA]

The LUT Cube Analyzer node

LUT Cube Analyzer Node Introduction

The LUT Cube Analyzer takes an image originated by the LUT Cube Creator as an input and allows the
user to create a 3D LUT file in ALUT3, ITX, or 3DL format.
Feeding the original LUT Cube Creator image into the node results in an unaltered, or 1:1, LUT file, and
nothing is displayed in the viewer.
You can, however, modify, grade, and color correct the original cube image with as many nodes as you
like and feed the result into the LUT Cube Analyzer. This creates a LUT that exactly resembles your
color pipeline.

Inputs
The LUT Cube Analyzer includes a single orange input.
– Input: The orange input is used to take the output of any node modifying an image that
originated with the LUT Cube Creator.

Basic Node Setup

The example below shows a node tree starting with a LUT Cube Creator node and going through two
color adjustments. The adjusted image is then connected to a LUT Cube Analyzer, which generates
the LUT file.

Generating a LUT starts with the LUT Cube Creator and ends with a LUT Cube Analyzer.

Inspector

LUT Cube Analyzer controls

Chapter 45 LUT Nodes 1059

Controls Tab
The Controls tab for the LUT Cube Analyzer node is used to select the desired LUT output format,
specify a filename, and write the 3D LUT to disk.

Type
Select the desired output format of the 3D LUT.

Filename
Enter the path where you want the file saved and enter the name of the LUT file. Alternatively, you can
click the Browse button to open a file browser to select the location and filename.

Write File
Press this button to generate the 3D LUT file based on the settings above.

Settings Tab
The Settings tab in the Inspector is also duplicated in other LUT nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

LUT Cube Apply [LCP]

The LUT Cube Apply node

LUT Cube Apply Node Introduction

The LUT Cube Apply node takes an image created by the LUT Cube Creator as the foreground input
and applies that LUT to the image connected to the background input.
Feeding the original image into the node would result in an unaltered, or 1:1, output.
You can, however, modify, grade, and color correct the original cube image with as many nodes as you
like and feed the result into the LUT Cube Apply. Or, take a LUT image that has been graded
beforehand to apply the LUT without having to write an actual 3D LUT using the LUT Cube Analyzer.

Inputs
The LUT Cube Apply has three inputs: a green input where the output of the LUT Cube Creator is
connected, an orange input for the image to have the LUT applied, and a blue effect mask input
– Input: This orange input accepts a 2D image that gets the LUT applied.
– Reference Image: The green input is used to connect the output of the LUT Cube Creator or a
node that is modifying the image originating in the LUT Cube Creator.
– Effect Mask: The optional effect mask input accepts a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this input
limits the LUT Cube Apply to only those pixels within the mask. An effects mask is applied to
the tool after the tool is processed.

Chapter 45 LUT Nodes 1060

Basic Node Setup
The example below shows a node tree starting with a LUT Cube Creator node and going through two
color adjustments. The adjusted image is then connected to the green Reference input of the LUT
Cube Apply. The image you want to apply the LUT to is connected to the orange Input.

The LUT generated by the LUT Cube Creator is applied to an image using the LUT Cube Apply node.

Inspector
There are no controls for the LUT Cube Apply node. The LUT connected to the green foreground input
is applied to the image connected to the orange background input without having to write an actual
3D LUT using the LUT Cube Analyzer.

Settings Tab
The Settings tab in the Inspector is also duplicated in other LUT nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

LUT Cube Creator [LCC]

The LUT Cube Creator node

LUT Cube Creator Node Introduction

The LUT Cube Creator generates an image for further use with the LUT Cube Analyzer or LUT Cube
Apply nodes. The output can be graded, color corrected, or modified with any tool inside and outside
of Fusion. If working outside Fusion, make sure to keep the image in 32-bit floating point to preserve
color accuracy.
Feeding the original LUT Cube Creator image into the LUT Cube Analyzer node results in an
unaltered, or 1:1, LUT file, and nothing is displayed in the viewer.

Inputs
There are no inputs on the LUT Cube Creator. The purpose of the node is to generate an image that
can be used to create a LUT.

Chapter 45 LUT Nodes 1061

Basic Node Setup
The example below shows a node tree starting with a LUT Cube Creator node and going through two
color adjustments. The adjusted image is then connected into a LUT Cube Analyzer that generates
the LUT file.

Generating a LUT starts with the LUT Cube Creator and ends with a LUT Cube Analyzer.

Inspector

LUT Cube Creator controls

Controls Tab
The Controls tab creates a test pattern of sorts used to create a 3D LUT. The controls here determine
the complexity of the pattern used to create a LUT using the LUT Cube Analyzer.

Type:
The Type menu is used to create a pattern of color cubes.
– Horizontal: Creates a long, horizontal strip representing a color cube.
– Vertical: Creates a long, vertical strip representing a color cube.
– Rect: Creates a rectangular image, as depicted below, representing a color cube.

A Cube image created with the Rect type The resulting color cube

Chapter 45 LUT Nodes 1062

Size
Determines the resolution of the color cube.
Typical Size settings for color cubes are 33 (33 x 33 x 33) or 65 (65 x 65 x 65). These numbers are the
samples on each side of the cube. A 33 x 33 x 33 cube has around 35,937 color samples.

NOTE: Higher resolutions yield more accurate results but are also more memory and
computationally expensive.

Settings Tab
The Settings tab in the Inspector is also duplicated in other LUT nodes. These common controls are
described in the following “The Common Controls” section.

The Common Controls

LUT nodes share a number of identical controls in the Inspector. This section describes controls that
are common among LUT nodes.

Inspector

Common LUT Settings Inspector

Settings Tab
The Settings tab in the Inspector can be found on every tool in the LUT category. The controls are
consistent and work the same way for each tool, although some tools do include one or two individual
options, which are also covered here.

Chapter 45 LUT Nodes 1063

Use GPU
The Use GPU menu has three settings. Setting the menu to Disable turns off hardware-accelerated
rendering using the graphics card in your computer. Enabled uses the hardware. Auto uses a capable
GPU if one is available and falls back to software rendering when a capable GPU is not available.

Hide Incoming Connections

Enabling this checkbox can hide connection lines from incoming nodes, making a node tree appear
cleaner and easier to read. When enabled, empty fields for each input on a node are displayed in the
Inspector. Dragging a connected node from the node tree into the field hides that incoming
connection line as long as the node is not selected in the node tree. When the node is selected in the
node tree, the line reappears.

Comments
The Comments field is used to add notes to a tool. Click in the empty field and type the text. When a
note is added to a tool, a small red square appears in the lower-left corner of the node when the full
tile is displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the
note in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 45 LUT Nodes 1064

Chapter 46

Mask Nodes
This chapter details the Mask nodes available in Fusion.

The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Bitmap Mask [BMP]  1066
B-Spline Mask [BSP]  1070
Ellipse Mask [ELP]  1074
Mask Paint [PNM]  1077
Polygon Mask [PLY]  1080
Ranges Mask [RNG]  1085
Rectangle Mask [REC]  1090
Triangle Mask [TRI]  1093
Wand Mask [WND]  1096
The Common Controls  1099

Chapter 46 Mask Nodes 1065

Bitmap Mask [BMP]

The Bitmap node

Bitmap Mask Node Introduction

The Bitmap mask node allows images from the node tree to act as masks for nodes and effects.
Bitmap masks can be based on values from any of the color, Alpha, hue, saturation, luminance, and
auxiliary coverage channels of the image. Nodes can also be masked based on the Object ID or
Material ID of a 3D-rendered image (provided those channels were included when the file was
rendered).
The Bitmap mask node is not required for effect masks. For effects masks, the Common Settings tab
for the masked node displays controls to select which channel of the mask image is used to
create the mask.
However, Bitmap mask nodes may still be required to connect to other mask inputs on some nodes,
such as Garbage Mattes and Pre-Masks. Also, using a Bitmap mask node between the mask source
and the target node provides additional options that would not be available when connecting directly,
such as combining masks, blurring the mask, or clipping its threshold.

Inputs
The Bitmap mask node includes two inputs in the Node Editor.
– Input: The orange input accepts a 2D image from which the mask will be created.
– Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input combines
the masks. How masks are combined is handled in the Paint mode menu in the Inspector.

Basic Node Setup

The Bitmap mask node is not required for connecting an image into the effect mask input, but it does
provide options that are otherwise unavailable. It allows for selecting channels other than RGBA for the
mask, as well as softness and clipping. In the node tree below, the two Bitmap masks are combined
using a Paint menu located in the second Bitmap mask node, which allows you to add, subtract,
multiply, and perform other operations on the combined mask.

Bitmap nodes can be chained together for more advanced matte operations.

Chapter 46 Mask Nodes 1066

Inspector

Bitmap Mask controls

Controls Tab
The Controls tab is used to refine how the image connected to the orange input converts into the
Bitmap mask.

Show View Controls

The Show View Controls checkbox is used to enable/disable the display of the mask’s onscreen
controls in the viewer. Onscreen controls, including center position, polylines, angles, and others, do
not appear when this checkbox is disabled, even when the node is selected.

Level
The Level control sets the transparency level of the pixels in the mask channel. When the value is 1.0,
the mask is completely opaque (unless it has a soft edge). Lower values cause the mask to be partially
transparent. The result is identical to lowering the Blend control of an effect.

NOTE: Lowering the level of a mask lowers the values of all pixels covered by the mask in the
mask channel. For example, if a Circle mask is placed over a Rectangle mask, lowering the
level of the Circle mask lowers the values of all the pixels in the mask channel, even though
the Rectangle mask beneath it is still opaque.

Filter
This control selects the filtering algorithm used when applying Soft Edge to the mask.
– Box: This is the fastest method but at reduced quality. Box is best suited for minimal
amounts of blur.
– Bartlett: Otherwise known as a Pyramid filter, Bartlett makes a good compromise between speed
and quality.
– Multi-box: When selecting this filter, the Num Passes slider appears and lets you control the
quality. At 1 and 2 passes, results are identical to Box and Bartlett, respectively. At 4 passes and
above, results are usually as good as Gaussian, in less time and with no edge “ringing.”

Chapter 46 Mask Nodes 1067

 – Gaussian: The Gaussian filter uses a true Gaussian approximation and gives excellent results, but
it is a little slower than the other filters. In some cases, it can produce an extremely slight edge
“ringing” on floating-point pixels.

Soft Edge
Use the Soft Edge slider to blur (feather) the mask, using the selected filter. Higher values cause the
edge to fade off well beyond the boundaries of the mask. A value of 0.0 creates a crisp, well-
defined edge.

Paint Mode
Connecting a mask to the effect mask input displays the Paint mode menu. The Paint mode is used to
determine how the incoming mask for the effect mask input and the mask created in the node
are combined.
– Merge: Merge is the default for all masks. The new mask is merged with the input mask.
– Add: The mask’s values add to the input mask’s values.
– Subtract: In the intersecting areas, the new mask values subtract from the input mask’s values.
– Minimum: Comparing the input mask’s values and the new mask, this displays the lowest
(minimum) value.
– Maximum: Comparing the input mask’s values and the new mask, this displays the highest
(maximum) value.
– Average: This calculates the average (half the sum) of the new mask and the input mask.
– Multiply: This multiplies the values of the input mask by the new mask’s values.
– Replace: The new mask completely replaces the input mask wherever they intersect. Areas that
are zero (completely black) in the new mask do not affect the input mask.
– Invert: Areas of the input mask that are covered by the new mask are inverted; white becomes
black and vice versa. Gray areas in the new mask are partially inverted.
– Copy: This mode completely discards the input mask and uses the new mask for all values.
– Ignore: This mode completely discards the new mask and uses the input mask for all values.

Invert
Selecting this checkbox inverts the entire mask. Unlike the Invert Paint mode, this checkbox affects all
pixels, regardless of whether the new mask covers them.

Fit Input
This menu is used to select how the image source is treated if it does not fit the dimensions of the
generated mask.
In the example below, a 720 x 576 image source (yellow) is used to generate a 1920 x 1080
mask (gray).
– Crop: If the image source is smaller than the generated mask, it will be placed according to
the X/Y controls, masking off only a portion of the mask. If the image source is larger than the
generated mask, it will be placed according to the X/Y controls and cropped off at the borders of
the mask.

Chapter 46 Mask Nodes 1068

– Stretch: The image source will be stretched in X and Y to accommodate the full dimensions of the
generated mask. This might lead to visible distortions of the image source.

– Inside: The image source will be scaled uniformly until one of its dimensions (X or Y) fits the
inside dimensions of the mask. Depending on the relative dimensions of the image source and
mask background, either the image source’s width or height may be cropped to fit the respective
dimensions of the mask.

– Width: The image source will be scaled uniformly until its width (X) fits the width of the mask.
Depending on the relative dimensions of the image source and mask, the image source’s Y
dimension might not fit the mask’s Y dimension, resulting in either cropping of the image source in
Y or the image source not covering the mask’s height entirely.

– Height: The image source will be scaled uniformly until its height (Y) fits the height of the mask.
Depending on the relative dimensions of the image source and mask, the image source’s
X-dimension might not fit the mask’s X-dimension, resulting in either cropping of the image source
in X or the image source not covering the mask’s width entirely.

Chapter 46 Mask Nodes 1069

 – Outside: The image source will be scaled uniformly until one of its dimensions (X or Y) fits the
outside dimensions of the mask. Depending on the relative dimensions of the image source
and mask, either the image source’s width or height may be cropped or not fit the respective
dimension of the mask.

Center X and Y
These controls adjust the position of the Bitmap mask.

Channel
The Channel menu determines the Channel of the input image used to create the mask. Choices
include the red, green, blue, and alpha channels, the hue, luminance, or saturation values, or the
auxiliary coverage channel of the input image (if one is provided).

Threshold Low/High
The Threshold range control can be used to clip the bitmap image. Increasing the low range control
will clip pixels below the specified value to black (0.0). Decreasing the high range control will force
pixels higher than the specified value to white (1.0).

Use Object/Use Material

This control has no effect unless the input image contains a Material or Object ID channel. When
toggled on, the Object ID and Material ID are used to create a mask based on the selected object or
material. When toggled off, the regular color channels will generate the mask.

Common Controls
Image and Settings Tabs
The Image and Settings tabs in the Inspector are also duplicated in other Mask nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

B-Spline Mask [BSP]

The B-Spline node

A B-Spline mask is identical to a Polygon mask in all respects except one. Where Polygon masks use
Bézier splines, this mask node uses B-Splines. Where Bézier splines employ a central point and two
handles to manage the smoothing of the spline segment, a B-Spline requires only a single point. This
means that a B-Spline shape requires far fewer control points to create a nicely smoothed shape.
When first added to a node, the B-Spline mask consists of only Center control, which is visible
onscreen. Points are added to the B-Spline by clicking in the viewer. Each new point is connected to

Chapter 46 Mask Nodes 1070

the last one created, but instead of the spline going directly through each control point, B-Spline
control points only influence the spline shape. The control point pulls the spline in its direction to
create a smooth curve.
Like the Polygon mask tool, the B-Spline mask auto-animates. Adding this node to the Node Editor
adds a keyframe to the current frame. Moving to a new frame and changing the shape creates a new
keyframe and interpolate between the two defined shapes.

Inputs
The B-Spline mask node includes a single effect mask input.
– Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input combines
the masks. How masks are combined is handled in the Paint mode menu in the Inspector.

Basic Node Setup

The B-Spline node can be used to generate a single smooth spline shape or combined with other
masks for more complex shapes. In the node tree below, the B-Spline mask is used to generate a
smooth, curved shape as a garbage matte on the Delta Keyer.

A B-Spline node generates a smooth, curved shape as a garbage matte

Inspector

B-Spline Mask controls

Chapter 46 Mask Nodes 1071

Controls Tab
The Controls tab is used to refine how the B-Spline appears after drawing it in the viewer.

Show View Controls

The Show View Controls checkbox is used to enable/disable the display of the mask’s onscreen
controls in the viewer. Onscreen controls, including center position, polylines, angles, and others, do
not appear when this checkbox is disabled, even when the node is selected.

Level
The Level control sets the transparency level of the pixels in the mask channel. When the value is 1.0,
the mask is completely opaque (unless it has a soft edge). Lower values cause the mask to be partially
transparent. The result is identical to lowering the blend control of an effect.

NOTE: Lowering the level of a mask lowers the values of all pixels covered by the mask in the
mask channel. For example, if a Circle mask is placed over a Rectangle mask, lowering the
level of the Circle mask lowers the values of all of the pixels in the mask channel, even though
the Rectangle mask beneath it is still opaque.

Filter
This control selects the filtering algorithm used when applying Soft Edge to the mask.
– Box: This is the fastest method but at reduced quality. Box is best suited for minimal
amounts of blur.
– Bartlett: Otherwise known as a Pyramid filter, Bartlett makes a good compromise between speed
and quality.
– Multi-box: When selecting this filter, the Num Passes slider appears and lets you control the
quality. At 1 and 2 passes, results are identical to Box and Bartlett, respectively. At 4 passes and
above, results are usually as good as Gaussian, in less time and with no edge “ringing.”
– Gaussian: The Gaussian filter uses a true Gaussian approximation and gives excellent results, but
it is a little slower than the other filters. In some cases, it can produce an extremely slight edge
“ringing” on floating-point pixels.

Soft Edge
Use the Soft Edge slider to blur (feather) the mask, using the selected filter. Higher values cause the
edge to fade off well beyond the boundaries of the mask. A value of 0.0 creates a crisp, well-
defined edge.

Border Width
The Border Width control adjusts the thickness of the mask’s edge. When the solid checkbox is
toggled on, the border thickens or narrows the mask. When the mask is not solid, the mask shape
draws as an outline, and the width uses the Border Width setting.

Paint Mode
Connecting a mask to the effect mask input displays the Paint mode menu. The Paint mode is used to
determine how the incoming mask for the effect mask input and the mask created in the node
are combined.
– Merge: Merge is the default for all masks. The new mask is merged with the input mask.
– Add: The mask’s values add to the input mask’s values.
– Subtract: In the intersecting areas, the new mask values subtract from the input mask’s values.

Chapter 46 Mask Nodes 1072

 – Minimum: Comparing the input mask’s values and the new mask, this displays the
lowest (minimum) value.
– Maximum: Comparing the input mask’s values and the new mask, this displays the
highest (maximum) value.
– Average: This calculates the average (half the sum) of the new mask and the input mask.
– Multiply: This multiplies the values of the input mask by the new mask’s values.
– Replace: The new mask completely replaces the input mask wherever they intersect. Areas that
are zero (completely black) in the new mask do not affect the input mask.
– Invert: Areas of the input mask that are covered by the new mask are inverted; white becomes
black and vice versa. Gray areas in the new mask are partially inverted.
– Copy: This mode completely discards the input mask and uses the new mask for all values.
– Ignore: This mode completely discards the new mask and uses the input mask for all values.

Invert
Selecting this checkbox inverts the entire mask. Unlike the Invert Paint mode, the checkbox affects all
pixels, regardless of whether the new mask covers them or not.

Solid
When the Solid checkbox is enabled, the mask is filled to be transparent (white) unless inverted. When
disabled, the spline is drawn as just an outline whose thickness is determined by the Border
Width slider.

Center X and Y
These controls adjust the position of the B-Spline mask.

Size
Use the Size control to adjust the scale of the B-Spline effect mask, without affecting the relative
behavior of the points that compose the mask or setting a keyframe in the mask animation.

X, Y, and Z Rotation
Use these three controls to adjust the rotation angle of the mask along any axis.

Fill Method
The Fill Method menu offers two different techniques for dealing with overlapping regions of a
polyline. If overlapping segments in a mask are causing undesirable holes to appear, try switching the
setting of this control from Alternate to Non Zero Winding.

Right-Click Here for Shape Animation

By default, all B-Spline masks are animated when they are created. The initial keyframe is set to the
current time, and any changes to the shape at different times will create new keys.
Right-clicking on this label will display a contextual menu that offers options for removing or re-adding
animation to the mask, or publishing and connecting the masks.

Adding Points
Adding Points to a B-Spline effect mask is relatively simple. Immediately after adding the node to the
Node Editor, there are no points, but the tool will be in Click Append mode. Click once in the viewer
wherever a point is required for the mask. Continue clicking to draw the shape of the mask.
When the shape is complete, click on the initial point again to close the mask.
When the shape is closed, the mode of the polyline changes to Insert and Modify. This allows you to
add and adjust additional points on the mask by clicking the spline segments. To lock down the mask’s

Chapter 46 Mask Nodes 1073

shape and prevent accidental changes, switch the Polyline mode to Done using the Polyline toolbar or
contextual menu.

Adjusting Tension on a B-Spline

The tension of the control point determines the smoothness of a B-Spline. To adjust the tension of a
B-Spline’s control points, select the point in the viewer, hold down the W key and drag the mouse
pointer to the left and right to increase or decrease the tension of the curve through that point.

Adjusting Tension on a B-Spline

B-Spline Toolbar
When a B-Spline mask is selected in the Node Editor, a toolbar appears above the viewer with buttons
for easy access to the modes. Position the pointer over any button in the toolbar to display a tooltip
that describes that button’s function.

B-Spline Mask Polygon toolbar

You can change the way the toolbar is displayed by right-clicking on the toolbar and selecting from the
options displayed in the toolbar’s contextual menu.
The functions of the buttons in this toolbar are explained in depth in the Polylines section.

Common Controls
Image and Settings Tabs
The Image and Settings tabs in the Inspector are also duplicated in other mask nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Ellipse Mask [ELP]

The Ellipse node

Chapter 46 Mask Nodes 1074

Ellipse Mask Node Introduction
The Ellipse mask is most useful for masking round objects. It is a circle by default, but independent
control is offered over the width, height, and angle, providing for a wide variety of ellipsoidal shapes.

Inputs
The Ellipse mask node includes a single effect mask input.
– Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input combines
the masks. How masks are combined is handled in the Paint mode menu in the Inspector.

Basic Node Setup

The Ellipse mask node is useful for generating circular masks. Below, the Ellipse mask is used to
generate a circular colored shape by cutting a circular shape from a background node.

Ellipse mask node creating a colored

circle by masking a background node

Inspector

Ellipse Mask controls

Controls Tab
The Controls tab is used to refine how the ellipse appears after drawing it in the viewer.

Show View Controls

The Show View Controls checkbox is used to enable/disable the display of the mask’s onscreen
controls in the viewer. Onscreen controls, including center position, polylines, angles, and others, do
not appear when this checkbox is disabled, even when the node is selected.

Chapter 46 Mask Nodes 1075

Level
The Level control sets the transparency level of the pixels in the mask channel. When the value is 1.0,
the mask is completely opaque (unless it has a soft edge). Lower values cause the mask to be partially
transparent. The result is identical to lowering the blend control of an effect.

NOTE: Lowering the level of a mask lowers the values of all pixels covered by the mask in the
mask channel. For example, if a Circle mask is placed over a Rectangle mask, lowering the
level of the Circle mask lowers the values of all of the pixels in the mask channel, even though
the Rectangle mask beneath it is still opaque.

Filter
This control selects the filtering algorithm used when applying Soft Edge to the mask.
– Box: This is the fastest method but at reduced quality. Box is best suited for
minimal amounts of blur.
– Bartlett: Otherwise known as a Pyramid filter, Bartlett makes a good compromise between speed
and quality.
– Multi-box: When selecting this filter, the Num Passes slider appears and lets you control the
quality. At 1 and 2 passes, results are identical to Box and Bartlett, respectively. At 4 passes and
above, results are usually as good as Gaussian, in less time and with no edge “ringing.”
– Gaussian: The Gaussian filter uses a true Gaussian approximation and gives excellent results, but
it is a little slower than the other filters. In some cases, it can produce an extremely slight edge
“ringing” on floating-point pixels.

Soft Edge
Use the Soft Edge slider to blur (feather) the mask, using the selected filter. Higher values cause the
edge to fade off well beyond the boundaries of the mask. A value of 0.0 creates a crisp, well-
defined edge.

Border Width
The Border Width control adjusts the thickness of the mask’s edge. When the solid checkbox is
toggled on, the border thickens or narrows the mask. When the mask is not solid, the mask shape
draws as an outline, and the width uses the Border Width setting.

Paint Mode
Connecting a mask to the effect mask input displays the Paint mode menu. The Paint mode is used to
determine how the incoming mask for the effect mask input and the mask created in the node
are combined.
– Merge: Merge is the default for all masks. The new mask is merged with the input mask.
– Add: The mask’s values add to the input mask’s values.
– Subtract: In the intersecting areas, the new mask values subtract from the input mask’s values.
– Minimum: Comparing the input mask’s values and the new mask, this displays the lowest
(minimum) value.
– Maximum: Comparing the input mask’s values and the new mask, this displays the highest
(maximum) value.
– Average: This calculates the average (half the sum) of the new mask and the input mask.
– Multiply: This multiplies the values of the input mask by the new mask’s values.

Chapter 46 Mask Nodes 1076

 – Replace: The new mask completely replaces the input mask wherever they intersect. Areas that
are zero (completely black) in the new mask do not affect the input mask.
– Invert: Areas of the input mask that are covered by the new mask are inverted; white becomes
black and vice versa. Gray areas in the new mask are partially inverted.
– Copy: This mode completely discards the input mask and uses the new mask for all values.
– Ignore: This mode completely discards the new mask and uses the input mask for all values.

Invert
Selecting this checkbox inverts the entire mask. Unlike the Invert Paint mode, the checkbox affects all
pixels, regardless of whether the new mask covers them or not.

Solid
When the Solid checkbox is enabled, the mask is filled to be transparent (white) unless inverted. When
disabled, the spline is drawn as just an outline whose thickness is determined by the Border
Width slider.

Center X and Y
These controls adjust the position of the Ellipse mask.

Width
This control allows independent control of the ellipse mask’s Width. In addition to the slider in the
mask’s controls, interactively drag the width (left or right edge) of the mask on the viewer using the
pointer. Any changes will be reflected in this control.

Height
Height allows independent control of the ellipse mask’s height. In addition to the slider in the mask’s
controls, interactively drag the height (top or bottom edge) of the mask on the view using the pointer.
Any changes will be reflected in this control.
To change the mask’s size without affecting the aspect ratio, drag the onscreen control between the
edges (diagonal). This will modify both the width and height proportionately.

Angle
Change the rotational angle of the mask by moving the Angle control left or right. Values can be
entered into the number fields provided. Alternately, use the onscreen controls by dragging the little
circle at the end of the dashed angle line to interactively adjust the rotation of the ellipse.

Common Controls
Image and Settings Tabs
The Image and Settings tabs in the Inspector are also duplicated in other mask nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Mask Paint [PNM]

The Mask Paint node

Chapter 46 Mask Nodes 1077

Mask Paint Node Introduction
The Mask Paint node allows direct painting of mask images, using the pointer as if it was a paintbrush.
In addition to regular paint strokes, it is possible to apply basic primitive shapes and polyline
style strokes.
Each stroke can have a duration that lasts for the entire project, a single frame. or an arbitrary number
of fields. The strokes can have independent durations in the Keyframes Editor for easy manipulation of
time. Alternatively, Multistrokes is a faster but non-editable way for doing many mask clean up
paint tasks.

Inputs
The Paint mask node includes a single effect mask input.
– Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input combines
the masks. How masks are combined is handled in the Paint mode menu in the Inspector.

Basic Node Setup

The Mask Paint node is useful for painting masks using a more free hand, pressure sensitive style. In
the node tree below, the Mask Paint node is used to patch up holes in a Bitmap mask.

A Mask Paint node can be used to repair problematic areas of a matte.

Inspector

Mask Paint controls

Chapter 46 Mask Nodes 1078

As the Controls tab in the Mask Paint node is fundamentally identical to the Paint node, for more detail,
see Chapter 51 “Paint Node” in the Fusion Reference Manual or Chapter 112 in the DaVinci Resolve
Reference Manual. The only difference between the two nodes is that, as Mask Paint operates on
single-channel mask images, there is no Channel Selector control, and all color controls have only a
single Alpha value. The Mask tab, however, includes several parameters that are different from the
Paint tool, so they are covered below.

The Mask Paint tab

Mask Tab
The Mask tab is used to refine the basic mask parameters that do not fall into the category of
“panting.” These include how multiple masks are combined, overall softness control, and level control.

Show View Controls

The Show View Controls checkbox is used to enable/disable the display of the masks onscreen
controls in the viewer. Onscreen controls including, center position, polylines, angles, and others, do
not appear when this checkbox is disabled, even when the node is selected.

Level
The Level control sets the transparency level of the pixels in the mask channel. When the value is 1.0,
the mask is completely opaque (unless it has a soft edge). Lower values cause the mask to be partially
transparent. The result is identical to lowering the blend control of an effect.

NOTE: Lowering the level of a mask lowers the values of all pixels covered by the mask in the
mask channel. For example, if a Circle mask is placed over a Rectangle mask, lowering the
level of the Circle mask lowers the values of all of the pixels in the mask channel, even though
the Rectangle mask beneath it is still opaque.

Filter
This control selects the filtering algorithm used when applying Soft Edge to the mask.
– Box: This is the fastest method but at reduced quality. Box is best suited for minimal
amounts of blur.
– Bartlett: Otherwise known as a Pyramid filter, Bartlett makes a good compromise between speed
and quality.
– Multi-box: When selecting this filter, the Num Passes slider appears and lets you control the
quality. At 1 and 2 passes, results are identical to Box and Bartlett, respectively. At 4 passes and
above, results are usually as good as Gaussian, in less time and with no edge “ringing.”
– Gaussian: The Gaussian filter uses a true Gaussian approximation and gives excellent results, but
it is a little slower than the other filters. In some cases, it can produce an extremely slight edge
“ringing” on floating-point pixels.

Chapter 46 Mask Nodes 1079

Soft Edge
Use the Soft Edge slider to blur (feather) the mask, using the selected filter. Higher values cause the
edge to fade off well beyond the boundaries of the mask. A value of 0.0 creates a crisp, well-
defined edge.

Paint Mode
Connecting a mask to the effect mask input displays the Paint mode menu. The Paint mode is used to
determine how the incoming mask for the effect mask input and the mask created in the node
are combined.
– Merge: Merge is the default for all masks. The new mask is merged with the input mask.
– Add: The mask’s values add to the input mask’s values.
– Subtract: In the intersecting areas, the new mask values subtract from the input mask’s values.
– Minimum: Comparing the input mask’s values and the new mask, this displays the lowest
(minimum) value.
– Maximum: Comparing the input mask’s values and the new mask, this displays the highest
(maximum) value.
– Average: This calculates the average (half the sum) of the new mask and the input mask.
– Multiply: This multiplies the values of the input mask by the new mask’s values.
– Replace: The new mask completely replaces the input mask wherever they intersect. Areas that
are zero (completely black) in the new mask do not affect the input mask.
– Invert: Areas of the input mask that are covered by the new mask are inverted; white becomes
black and vice versa. Gray areas in the new mask are partially inverted.
– Copy: This mode completely discards the input mask and uses the new mask for all values.
– Ignore: This mode completely discards the new mask and uses the input mask for all values.

Invert
Selecting this checkbox inverts the entire mask. Unlike the Invert Paint mode, the checkbox affects all
pixels, regardless of whether the new mask covers them or not.

Common Controls
Image and Settings Tabs
The Image and Settings tabs in the Inspector are also duplicated in other mask nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Polygon Mask [PLY]

The Polygon node

Polygon Mask Node Introduction

The Polygon mask is most useful for masking objects that do not have a regular shape. When first
added to a node, the Polygon mask consists of only Center and Angle controls, which are visible

Chapter 46 Mask Nodes 1080

onscreen. Points are added to the polyline by clicking in the viewer. Each new point is connected to
the last one created.
Like the B-Spline mask tool, the Polygon mask auto-animates. Adding this node to the Node Editor
adds a keyframe to the current frame. Moving to a new frame and changing the shape creates a new
keyframe and interpolate between the two defined shapes.

Inputs
The Polygon mask node includes a single effect mask input.
– Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input combines
the masks. How masks are combined is handled in the Paint mode menu in the Inspector.

Basic Node Setup

The Polygon node can be used to generate a detailed spline shape or combined with other masks for
even more complex shapes. In the node tree below, the Polygon mask is used to generate detailed
shape as a solid matte on the Delta Keyer.

A Polygon node generates a detailed shape as a Solid matte.

Inspector

Polygon Mask controls

Controls Tab
The Controls tab is used to refine how the polyline appears after drawing it in the viewer.

Chapter 46 Mask Nodes 1081

Show View Controls
The Show View Controls checkbox is used to enable/disable the display of the mask’s onscreen
controls in the viewer. Onscreen controls, including center position, polylines, angles, and others, do
not appear when this checkbox is disabled, even when the node is selected.

Level
The Level control sets the transparency level of the pixels in the mask channel. When the value is 1.0,
the mask is completely opaque (unless it has a soft edge). Lower values cause the mask to be partially
transparent. The result is identical to lowering the blend control of an effect.

NOTE: Lowering the level of a mask lowers the values of all pixels covered by the mask in the
mask channel. For example, if a Circle mask is placed over a Rectangle mask, lowering the
level of the Circle mask lowers the values of all of the pixels in the mask channel, even though
the Rectangle mask beneath it is still opaque.

Filter
This control selects the filtering algorithm used when applying Soft Edge to the mask.
– Box: This is the fastest method but at reduced quality. Box is best suited for minimal
amounts of blur.
– Bartlett: Otherwise known as a Pyramid filter, Bartlett makes a good compromise between speed
and quality.
– Multi-box: When selecting this filter, the Num Passes slider appears and lets you control the
quality. At 1 and 2 passes, results are identical to Box and Bartlett, respectively. At 4 passes and
above, results are usually as good as Gaussian, in less time and with no edge “ringing.”
– Gaussian: The Gaussian filter uses a true Gaussian approximation and gives excellent results, but
it is a little slower than the other filters. In some cases, it can produce an extremely slight edge
“ringing” on floating-point pixels.

Soft Edge
Use the Soft Edge slider to blur (feather) the mask, using the selected filter. Higher values cause the
edge to fade off well beyond the boundaries of the mask. A value of 0.0 creates a crisp, well-
defined edge.

Border Width
The Border Width control adjusts the thickness of the mask’s edge. When the solid checkbox is
toggled on, the border thickens or narrows the mask. When the mask is not solid, the mask shape
draws as an outline, and the width uses the Border Width setting.

Paint Mode
Connecting a mask to the effect mask input displays the Paint mode menu. The Paint mode is used to
determine how the incoming mask for the effect mask input and the mask created in the node
are combined.
– Merge: Merge is the default for all masks. The new mask is merged with the input mask.
– Add: The mask’s values add to the input mask’s values.
– Subtract: In the intersecting areas, the new mask values subtract from the input mask’s values.
– Minimum: Comparing the input mask’s values and the new mask, this displays the lowest
(minimum) value.

Chapter 46 Mask Nodes 1082

 – Maximum: Comparing the input mask’s values and the new mask, this displays the highest
(maximum) value.
– Average: This calculates the average (half the sum) of the new mask and the input mask.
– Multiply: This multiplies the values of the input mask by the new mask’s values.
– Replace: The new mask completely replaces the input mask wherever they intersect. Areas that
are zero (completely black) in the new mask do not affect the input mask.
– Invert: Areas of the input mask that are covered by the new mask are inverted; white becomes
black and vice versa. Gray areas in the new mask are partially inverted.
– Copy: This mode completely discards the input mask and uses the new mask for all values.
– Ignore: This mode completely discards the new mask and uses the input mask for all values.

Invert
Selecting this checkbox inverts the entire mask. Unlike the Invert Paint mode, the checkbox affects all
pixels, regardless of whether the new mask covers them or not.

Solid
When the Solid checkbox is enabled, the mask is filled to be transparent (white) unless inverted. When
disabled, the spline is drawn as just an outline whose thickness is determined by the Border
Width slider.

Center X and Y
These controls adjust the position of the polygon spline mask.

Size
Use the Size control to adjust the scale of the polygon spline effect mask, without affecting the relative
behavior of the points that compose the mask or setting a keyframe in the mask animation.

X, Y, and Z Rotation
Use these three controls to adjust the rotation angle of the mask along any axis.

Fill Method
The Fill Method menu offers two different techniques for dealing with overlapping regions of a
polyline. If overlapping segments in a mask are causing undesirable holes to appear, try switching the
setting of this control from Alternate to Non Zero Winding.

Right-Click Here for Shape Animation

By default, all polygon spline masks are animated when they are created. The initial keyframe is set to
the current time, and any changes to the shape at different times create new keys.
Right-clicking on this label displays a contextual menu that offers options for removing or re-adding
animation to the mask, or publishing and connecting the masks together.

Common Controls
Image and Settings Tabs
The Image and Settings tabs in the Inspector are also duplicated in other mask nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Chapter 46 Mask Nodes 1083

Adding Points
Adding Points to a polygonal effect mask is relatively simple. Immediately after adding the node to the
Node Editor, there are no points, but the tool will be in Click Append mode. Click once in the viewer
wherever a point is required for the mask. Continue clicking to draw the shape of the mask. When the
shape is complete, click on the initial point again to close the mask.
When the shape is closed, the mode of the polyline will change to Insert and Modify. This allows for
the adjusting and adding of additional points to the mask by clicking on segments of the polyline. To
lock down the mask’s shape and prevent accidental changes, switch the Polyline mode to Done using
the Polyline toolbar or contextual menu.

B-Spline Mask Polygon toolbar

When a Polygon (or B-Spline) mask is added to a node, a toolbar appears above the viewer, offering
easy access to modes. Hold the pointer over any button in the toolbar to display a tooltip that
describes that button’s function.
– Click: Click is the default option when creating a polyline (or B-Spline) mask. It is a Bézier style
drawing tool. Clicking sets a control point and appends the next control point when you click again
in a different location.
– Draw: Draw is a freehand drawing tool. It creates a mask similar to drawing with a pencil on paper.
You can create a new mask using the Draw tool, or you can extend an existing open spline by
clicking the Draw tool and starting to draw from the last control point.
– Insert: Insert adds a new control point along the spline.
– Modify: Modify allows you to safely move or smooth any exiting point along a spline without
worrying about adding new points accidentally.
– Done: Prevents any point along the spline from being moved or modified. Also, new points cannot
be added. You can, however, move and rotate the entire spline.
– Closed: Closes an open spline.
– Smooth: Changes the selected control point from a linear to a smooth curve.
– Linear: Changes the selected control point from a smooth curve to linear.
– Select All: Selects all the control points on the spline.
– Keys: Shows or hides the control points along the spline.
– Handles: Shows or hides the Bézier handles along the polyline.
– Shape: Places a reshape rectangle around the selected spline shape. Using the reshape
rectangle, you can deform groups of control points or entire shapes much easier than modifying
each point.
– Delete: Deletes the selected control point(s).
– Reduce: Opens a Freehand precision window that can be used to reduce the number of controls
points on a spline. This can make the paint stroke easier to modify, especially if it has been
created using the Draw tool.
– Publish menu: You can use the publish menu to select between publishing the control points or
the path. Publishing is a form of parameter linking, it makes the selected item available for use by
other controls. It also allows you to attach a control point to a tracker.
– Follow Points: Allows a selected point to follow the path of a published point. The point follows
the published point using an offset position.

Chapter 46 Mask Nodes 1084

 – Double Poly: Allows softening part of the spline curve while keeping other portions of the curve
sharp. The double polyline is composed of two shapes, an inner and outer shape. The inner shape
is the original shape from the single polyline, whereas the outer shape is used to determine the
spread of the softness. The further the outer shape gets from the inner shape, the softer that
segment of the shape becomes. Both polylines start with exactly the same shape as the original
single polyline, keeping the mask sharp to start. Any animation already applied to the shape
remains. To select the outer shape, press the Tab key to cycle between the onscreen controls
until the dashed outline is visible, or you can select the outer polyline using the contextual menu’s
Controls > Outer Polygon menu.
– Multiframe: Multiframe is a method of adjusting control points across multiple keyframes. The
default setting of none only adjusts the control point of a spline on the current keyframe. Setting
the menu to All adjusts the controls point for all keyframes. Prev settings adjust the current and
previous keyframe while Next adjusts the current and next keyframe.
– Onion Skinning: Enabling onion skinning displays a mix in the viewer of the spline animation. It is
useful when aligning spline animation and motion. Selecting Onion Skin Settings from the drop
down menu allows you to set the number of overlapping frames.
– Roto Assist: Enable the Roto Assist button when you begin painting with the Polyline Stroke
tool. The polyline points snap to the closest edge as you click to add points to the shape. A cyan
outline indicates the points that have snapped to an edge. There are three main Roto Assist
options selectable through the drop down menu:
– Multiple Points: When enabled, a single click on a high contrast edge adds multiple points to
define the entire edge, instead of having to add each point individually. This is a one time only
click. The second click reverts to single point edge detection.
– Distance 8: Opens a dialog where you can set the pixel range within which searching for an
edge takes place.
– Reset: Used for resetting the snap attribute of all snapped points. After resetting, the points
become unavailable for tracking.

Change the way the toolbar is displayed by right-clicking on the toolbar and selecting from the options
displayed in the toolbar’s contextual menu. The functions of the buttons in this toolbar are explained in
depth in the Polylines chapter.

Ranges Mask [RNG]

The Ranges node

Ranges Mask Node Introduction

Similar to Bitmap mask, the Ranges mask allows images from the node tree to act as masks for nodes
and effects. Instead of creating a simple luminance-based mask from a given channel, Ranges allows
spline-based selection of low, mid and high ranges, akin to Color Corrector.

Chapter 46 Mask Nodes 1085

Inputs
The Ranges mask node includes two inputs in the Node Editor.
– Input: The orange input accepts a 2D image from which the mask will be created.
– Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input combines
the masks. How masks are combined is handled in the Paint mode menu in the Inspector.

Basic Node Setup

The Ranges node is not required for connecting an image into the effect mask input, but like the
Bitmap node, it does provide options that are otherwise unavailable. It allows for selecting channels
other than RGBA for the mask, as well as softness and clipping. In the node tree below, the Ranges
node takes the composite out of the merge, creating a mask for the color correction.

A Ranges node selects a specific range in the image to create a mask.

Inspector

Rangers Mask controls

Chapter 46 Mask Nodes 1086

Controls Tab
The Controls tab is used to refine how the image connected to the orange input converts into the
ranges mask.

Show View Controls

The Show View Controls checkbox is used to enable/disable the display of the mask’s onscreen
controls in the viewer. Onscreen controls, including center position, polylines, angles, and others, do
not appear when this checkbox is disabled, even when the node is selected.

Level
The Level control sets the transparency level of the pixels in the mask channel. When the value is 1.0,
the mask is completely opaque (unless it has a soft edge). Lower values cause the mask to be partially
transparent. The result is identical to lowering the blend control of an effect.

NOTE: Lowering the level of a mask lowers the values of all pixels covered by the mask in the
mask channel. For example, if a Circle mask is placed over a Rectangle mask, lowering the
level of the Circle mask lowers the values of all of the pixels in the mask channel, even though
the Rectangle mask beneath it is still opaque.

Filter
This control selects the filtering algorithm used when applying Soft Edge to the mask.
– Box: This is the fastest method but at reduced quality. Box is best suited for minimal
amounts of blur.
– Bartlett: Otherwise known as a Pyramid filter, Bartlett makes a good compromise between speed
and quality.
– Multi-box: When selecting this filter, the Num Passes slider appears and lets you control the
quality. At 1 and 2 passes, results are identical to Box and Bartlett, respectively. At 4 passes and
above, results are usually as good as Gaussian, in less time and with no edge “ringing.”
– Gaussian: The Gaussian filter uses a true Gaussian approximation and gives excellent results, but
it is a little slower than the other filters. In some cases, it can produce an extremely slight edge
“ringing” on floating-point pixels.

Soft Edge
Use the Soft Edge slider to blur (feather) the mask, using the selected filter. Higher values cause the
edge to fade off well beyond the boundaries of the mask. A value of 0.0 creates a crisp, well-
defined edge.

Paint Mode
Connecting a mask to the effect mask input displays the Paint mode menu. The Paint mode is used to
determine how the incoming mask for the effect mask input and the mask created in the node
are combined.
– Merge: Merge is the default for all masks. The new mask is merged with the input mask.
– Add: The mask’s values add to the input mask’s values.
– Subtract: In the intersecting areas, the new mask values subtract from the input mask’s values.
– Minimum: Comparing the input mask’s values and the new mask, this displays the lowest
(minimum) value.
– Maximum: Comparing the input mask’s values and the new mask, this displays the highest
(maximum) value.
– Average: This calculates the average (half the sum) of the new mask and the input mask.

Chapter 46 Mask Nodes 1087

 – Multiply: This multiplies the values of the input mask by the new mask’s values.
– Replace: The new mask completely replaces the input mask wherever they intersect. Areas that
are zero (completely black) in the new mask do not affect the input mask.
– Invert: Areas of the input mask that are covered by the new mask are inverted; white becomes
black and vice versa. Gray areas in the new mask are partially inverted.
– Copy: This mode completely discards the input mask and uses the new mask for all values.
– Ignore: This mode completely discards the new mask and uses the input mask for all values.

Invert
Selecting this checkbox inverts the entire mask. Unlike the Invert Paint mode, the checkbox affects all
pixels, regardless of whether the new mask covers them or not.

Center X and Y
These controls adjust the position of the ranges mask.

Fit Input
This menu is used to select how the image source is treated if it does not fit the dimensions of the
generated mask.
For example, below, a 720 x 576 image source (yellow) is used to generate a 1920 x 1080 mask (gray).
– Crop: If the image source is smaller than the generated mask, it is placed according to the X/Y
controls, masking off only a portion of the mask. If the image source is larger than the generated
mask it is placed according to the X/Y controls and cropped off at the borders of the mask.

– Stretch: The image source is stretched in X and Y to accommodate the full dimensions of the
generated mask. This might lead to visible distortions of the image source.

– Inside: The image source is scaled uniformly until one of its dimensions (X or Y) fits the inside
dimensions of the mask. Depending on the relative dimensions of the image source and mask
background, either the image source’s width or height may be cropped to fit the respective
dimension of the mask.

Chapter 46 Mask Nodes 1088

 – Width: The image source is scaled uniformly until its width (X) fits the width of the mask.
Depending on the relative dimensions of the image source and mask, the image source’s Y
dimension might not fit the mask’s Y dimension, resulting in either cropping of the image source in
Y or the image source not covering the mask’s height entirely.

– Height: The image source is scaled uniformly until its height (Y) fits the height of the mask.
Depending on the relative dimensions of the image source and mask, the image source’s X
dimension might not fit the mask’s X dimension, resulting in either cropping of the image source in
X or the image source not covering the mask’s width entirely.

– Outside: The image source is scaled uniformly until one of its dimensions (X or Y) fits the outside
dimensions of the mask. Depending on the relative dimensions of the image source and mask,
either the image source’s width or height may be cropped or not fit the respective dimension of
the mask.

Channel
The Channel menu determines the Channel of the input image used to create the mask. Choices
include the red, green, blue, and alpha channels; the hue, luminance, or saturation values; or the
auxiliary coverage channel of the input image (if one is provided).

Shadows/Midtones/Highlights
These buttons are used to select which range is output by the node as a mask. White pixels represent
pixels that are considered to be part of the range, and black pixels are not included in the range. For
example, choosing Shadows would show pixels considered to be shadows as white, and pixels that
are not shadows as black. Mid gray pixels are only partly in the range and do not receive the full effect
of any color adjustments to that range.

Chapter 46 Mask Nodes 1089

Channel
The Channel selection buttons shown in this tab can be used to extract a mask from the range of a
specific color channel. By default, Fusion uses the luminance channel when the color ranges
are examined.

Mini Spline Editor

The extent of the ranges is selected by manipulating the spline handles. There are four spline points,
each with one Bézier handle. The two handles at the top represent the start of the shadow and
highlight ranges; the two handles at the bottom represent the end of the range. The Bézier handles
are used to control the falloff.
The midtones range has no specific control, since its range is understood to be the space between
the shadow and highlight ranges. In other words, after low and high masks have been applied,
midtones are everything else.
The X and Y text controls below the Mini Spline Editor can be used to enter precise positions for the
selected Bézier point or handle.

Presets
This sets the splines to two commonly-used configurations. The Simple button gives a straightforward
linear-weighted selection, while the Smooth button uses a more natural falloff.

Common Controls
Image and Settings Tabs
The Image and Settings tabs in the Inspector are also duplicated in other mask nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Rectangle Mask [REC]

The Rectangle node

Rectangle Mask Node Introduction

The Rectangle mask creates simple square or rectangular masks. By default, it creates a rectangle in
the same aspect ratio as the comp, but independent control is offered over the width, height, and
angle, providing for a wide variety of rectangular shapes.

Inputs
The Rectangle mask node includes a single effect mask input.
– Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input combines
the masks. How masks are combined is handled in the Paint mode menu in the Inspector.

Chapter 46 Mask Nodes 1090

Basic Node Setup
The Rectangle mask node is useful for generating rectangular shapes. Below, the Rectangle mask is
used to generate a colored square by cutting a square shape from a background node.

A Rectangle mask node creating a colored

square by masking a Background node

Inspector

Rectangle Mask controls

Controls Tab
The Controls tab is used to refine how the rectangle appears after drawing it in the viewer.

Show View Controls

The Show View Controls checkbox is used to enable/disable the display of the mask’s onscreen
controls in the viewer. Onscreen controls, including center position, polylines, angles, and others, do
not appear when this checkbox is disabled, even when the node is selected.

Level
The Level control sets the transparency level of the pixels in the mask channel. When the value is 1.0,
the mask is completely opaque (unless it has a soft edge). Lower values cause the mask to be partially
transparent. The result is identical to lowering the Blend control of an effect.

NOTE: Lowering the level of a mask lowers the values of all pixels covered by the mask in the
mask channel. For example, if a Circle mask is placed over a Rectangle mask, lowering the
level of the Circle mask lowers the values of all the pixels in the mask channel, even though
the Rectangle mask beneath it is still opaque.

Chapter 46 Mask Nodes 1091

Filter
This control selects the filtering algorithm used when applying Soft Edge to the mask.
– Box: This is the fastest method but at reduced quality. Box is best suited for minimal
amounts of blur.
– Bartlett: Otherwise known as a Pyramid filter, Bartlett makes a good compromise between speed
and quality.
– Multi-box: When selecting this filter, the Num Passes slider appears and lets you control the
quality. At 1 and 2 passes, results are identical to Box and Bartlett, respectively. At 4 passes and
above, results are usually as good as Gaussian, in less time and with no edge “ringing.”
– Gaussian: The Gaussian filter uses a true Gaussian approximation and gives excellent results, but
it is a little slower than the other filters. In some cases, it can produce an extremely slight edge
“ringing” on floating-point pixels.

Soft Edge
Use the Soft Edge slider to blur (feather) the mask, using the selected filter. Higher values cause the
edge to fade off well beyond the boundaries of the mask. A value of 0.0 creates a crisp, well-
defined edge.

Border Width
The Border Width control adjusts the thickness of the mask’s edge. When the solid checkbox is
toggled on, the border thickens or narrows the mask. When the mask is not solid, the mask shape
draws as an outline, and the width uses the Border Width setting.

Paint Mode
Connecting a mask to the effect mask input displays the Paint mode menu. The Paint mode is used to
determine how the incoming mask for the effect mask input and the mask created in the node
are combined.
– Merge: Merge is the default for all masks. The new mask is merged with the input mask.
– Add: The mask’s values add to the input mask’s values.
– Subtract: In the intersecting areas, the new mask values subtract from the input mask’s values.
– Minimum: Comparing the input mask’s values and the new mask, this displays the lowest
(minimum) value.
– Maximum: Comparing the input mask’s values and the new mask, this displays the highest
(maximum) value.
– Average: This calculates the average (half the sum) of the new mask and the input mask.
– Multiply: This multiplies the values of the input mask by the new mask’s values.
– Replace: The new mask completely replaces the input mask wherever they intersect. Areas that
are zero (completely black) in the new mask do not affect the input mask.
– Invert: Areas of the input mask that are covered by the new mask are inverted: white becomes
black and vice versa. Gray areas in the new mask are partially inverted.
– Copy: This mode completely discards the input mask and uses the new mask for all values.
– Ignore: This mode completely discards the new mask and uses the input mask for all values.

Invert
Selecting this checkbox inverts the entire mask. Unlike the Invert Paint mode, this checkbox affects all
pixels, regardless of whether the new mask covers them.

Chapter 46 Mask Nodes 1092

Solid
When the Solid checkbox is enabled, the mask is filled to be transparent (white) unless inverted. When
disabled, the spline is drawn as just an outline whose thickness is determined by the Border
Width slider.

Center X and Y
These controls adjust the position of the Rectangle mask.

Width and Height

Use these controls to change the X or Y scale of the rectangular effect mask independently of each
other. Alternatively, drag the edges of the rectangle in the viewer to interactively adjust its size.

Corner Radius
Corner Radius allows the corners of the Rectangle mask to be rounded. A value of 0.0 is not rounding
at all, which means that the rectangle has sharp corners. A value of 1.0 applies the maximum amount of
rounding to the corners.

Angle
Change the rotation angle of an effect mask by moving the Angle control left or right. Values can be
entered in the provided input boxes. Alternatively, use the onscreen controls by dragging the little
circle at the end of the dashed angle line to interactively adjust the rotation of the ellipse.

Common Controls
Image and Settings Tabs
The Image and Settings tabs in the Inspector are also duplicated in other mask nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Triangle Mask [TRI]

The Triangle node

Triangle Mask Node Introduction

The Triangle mask creates simple triangular masks. It is unique in that it has no Center, Size, or Angle
control. Unlike most other types of masks, all three points of the triangle can attach to a tracker or
motion path.

Inputs
The Triangle mask node includes a single effect mask input.
– Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input combines
the masks. How masks are combined is handled in the Paint mode menu in the Inspector.

Chapter 46 Mask Nodes 1093

Basic Node Setup
The Triangle mask node is useful for generating triangular shapes. Below, the Triangle mask is used to
generate a colored triangle by cutting a triangular shape from a background node.

A Triangle mask node creating

a colored Triangle shape by
masking a Background node

Inspector

Triangle Mask controls

Controls Tab
The Controls tab is used to refine how the triangle appears after drawing it in the viewer.

Show View Controls

The Show View Controls checkbox is used to enable/disable the display of the mask’s onscreen
controls in the viewer. Onscreen controls, including center position, polylines, angles, and others, do
not appear when this checkbox is disabled, even when the node is selected.

Level
The Level control sets the transparency level of the pixels in the mask channel. When the value is 1.0,
the mask is completely opaque (unless it has a soft edge). Lower values cause the mask to be partially
transparent. The result is identical to lowering the Blend control of an effect.

NOTE: Lowering the level of a mask lowers the values of all pixels covered by the mask in the
mask channel. For example, if a Circle mask is placed over a Rectangle mask, lowering the
level of the Circle mask lowers the values of all the pixels in the mask channel, even though
the Rectangle mask beneath it is still opaque.

Chapter 46 Mask Nodes 1094

Filter
This control selects the filtering algorithm used when applying Soft Edge to the mask.
– Box: This is the fastest method but at reduced quality. Box is best suited for minimal
amounts of blur.
– Bartlett: Otherwise known as a Pyramid filter, Bartlett makes a good compromise between speed
and quality.
– Multi-box: When selecting this filter, the Num Passes slider appears and lets you control the
quality. At 1 and 2 passes, results are identical to Box and Bartlett, respectively. At 4 passes and
above, results are usually as good as Gaussian, in less time and with no edge “ringing.”
– Gaussian: The Gaussian filter uses a true Gaussian approximation and gives excellent results, but
it is a little slower than the other filters. In some cases, it can produce an extremely slight edge
“ringing” on floating-point pixels.

Soft Edge
Use the Soft Edge slider to blur (feather) the mask, using the selected filter. Higher values cause the
edge to fade off well beyond the boundaries of the mask. A value of 0.0 creates a crisp, well-
defined edge.

Border Width
The Border Width control adjusts the thickness of the mask’s edge. When the solid checkbox is
toggled on, the border thickens or narrows the mask. When the mask is not solid, the mask shape
draws as an outline, and the width uses the Border Width setting.

Paint Mode
Connecting a mask to the effect mask input displays the Paint mode menu. The Paint mode is used to
determine how the incoming mask for the effect mask input and the mask created in the node
are combined.
– Merge: Merge is the default for all masks. The new mask is merged with the input mask.
– Add: The mask’s values add to the input mask’s values.
– Subtract: In the intersecting areas, the new mask values subtract from the input mask’s values.
– Minimum: Comparing the input mask’s values and the new mask, this displays the lowest
(minimum) value.
– Maximum: Comparing the input mask’s values and the new mask, this displays the highest
(maximum) value.
– Average: This calculates the average (half the sum) of the new mask and the input mask.
– Multiply: This multiplies the values of the input mask by the new mask’s values.
– Replace: The new mask completely replaces the input mask wherever they intersect. Areas that
are zero (completely black) in the new mask do not affect the input mask.
– Invert: Areas of the input mask that are covered by the new mask are inverted: white becomes
black and vice versa. Gray areas in the new mask are partially inverted.
– Copy: This mode completely discards the input mask and uses the new mask for all values.
– Ignore: This mode completely discards the new mask and uses the input mask for all values.

Invert
Selecting this checkbox inverts the entire mask. Unlike the Invert Paint mode, this checkbox affects all
pixels, regardless of whether the new mask covers them.

Chapter 46 Mask Nodes 1095

Solid
When the Solid checkbox is enabled, the mask is filled to be transparent (white) unless inverted. When
disabled, the spline is drawn as just an outline whose thickness is determined by the Border
Width slider.

Point 1, Point 2, Point 3

These controls show the position coordinates of the three corners of the triangle. Each point can be
published, connected to other controls, animated with a path, or attached to trackers. These tasks are
performed by right-clicking the Position control or directly on the point in the viewer.

Common Controls
Image and Settings Tabs
The Image and Settings tabs in the Inspector are also duplicated in other mask nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Wand Mask [WND]

The Wand node

Wand Mask Node Introduction

The Wand mask masks an image based on a wand-style selection, similar to the Magic Wand tool
found in Adobe Photoshop. As with a Bitmap mask, any image in the composition can be the source of
the mask. Generally, the default is most useful, where the source image is the input of the node to
which the mask is applied.
When adding a Wand mask to a node, a crosshair appears in the viewers. This crosshair should be
positioned in the image to select the color used to create the Wand mask. The mask is created by
examining the pixel color beneath the selection point and adding that color to the mask. The mask
then expands to examine the pixels surrounding the selection point. Surrounding pixels are added to
the mask if they are the same color. The mask stops expanding when no connecting pixels fall within
the color range of the mask.

Inputs
The Wand mask node includes two inputs in the Node Editor.
– Input: The orange input accepts a 2D image from which the mask is created.
– Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input combines
the masks. How masks are combined is handled in the Paint mode menu in the Inspector.

Basic Node Setup

The Wand mask node is not required for connecting an image into the effect mask input, but like the
Bitmap node, it does provide options that are otherwise unavailable. It allows for selecting channels
other than RGBA for the mask, as well as softness and clipping. In the node tree below, the Wand
node takes the composite out of the merge, creating a mask for the color correction.

Chapter 46 Mask Nodes 1096

A Wand node selects a specific area in the image to create a mask.

Inspector

Wand Mask controls

Controls Tab
The Controls tab is used to refine how the mask appears after the Wand makes a selection in
the viewer.

Show View Controls

The Show View Controls checkbox is used to enable/disable the display of the mask’s onscreen
controls in the viewer. Onscreen controls, including center position, polylines, angles, and others,
do not appear when this checkbox is disabled, even when the node is selected.

Level
The Level control sets the transparency level of the pixels in the mask channel. When the value is 1.0,
the mask is completely opaque (unless it has a soft edge). Lower values cause the mask to be partially
transparent. The result is identical to lowering the Blend control of an effect.

NOTE: Lowering the level of a mask lowers the values of all pixels covered by the mask in the
mask channel. For example, if a Circle mask is placed over a Rectangle mask, lowering the
level of the Circle mask lowers the values of all the pixels in the mask channel, even though
the Rectangle mask beneath it is still opaque.

Chapter 46 Mask Nodes 1097

Filter
This control selects the filtering algorithm used when applying Soft Edge to the mask.
– Box: This is the fastest method but at reduced quality. Box is best suited for minimal
amounts of blur.
– Bartlett: Otherwise known as a Pyramid filter, Bartlett makes a good compromise between speed
and quality.
– Multi-box: When selecting this filter, the Num Passes slider appears and lets you control the
quality. At 1 and 2 passes, results are identical to Box and Bartlett, respectively. At 4 passes and
above, results are usually as good as Gaussian, in less time and with no edge “ringing.”
– Gaussian: The Gaussian filter uses a true Gaussian approximation and gives excellent results, but
it is a little slower than the other filters. In some cases, it can produce an extremely slight edge
“ringing” on floating-point pixels.

Soft Edge
Use the Soft Edge slider to blur (feather) the mask, using the selected filter. Higher values cause the
edge to fade off well beyond the boundaries of the mask. A value of 0.0 creates a crisp, well-
defined edge.

Paint Mode
Connecting a mask to the effect mask input displays the Paint mode menu. The Paint mode is used to
determine how the incoming mask for the effect mask input and the mask created in the node
are combined.
– Merge: Merge is the default for all masks. The new mask is merged with the input mask.
– Add: The mask’s values add to the input mask’s values.
– Subtract: In the intersecting areas, the new mask values subtract from the input mask’s values.
– Minimum:  Comparing the input mask’s values and the new mask, this displays the lowest
(minimum) value.
– Maximum:  Comparing the input mask’s values and the new mask, this displays the highest
(maximum) value.
– Average: This calculates the average (half the sum) of the new mask and the input mask.
– Multiply: This multiplies the values of the input mask by the new mask’s values.
– Replace: The new mask completely replaces the input mask wherever they intersect. Areas that
are zero (completely black) in the new mask do not affect the input mask.
– Invert: Areas of the input mask that are covered by the new mask are inverted; white becomes
black and vice versa. Gray areas in the new mask are partially inverted.
– Copy: This mode completely discards the input mask and uses the new mask for all values.
– Ignore: This mode completely discards the new mask and uses the input mask for all values.

Invert
Selecting this checkbox inverts the entire mask. Unlike the Invert Paint mode, this checkbox affects all
pixels, regardless of whether the new mask covers them.

Selection Point
The Selection Point is a pair of X and Y coordinates that determines where in the source image the
Wand mask derives its initial color sample. This control is also seen as a crosshair in the viewers. The
selection point can be positioned manually, connected to a tracker, path, or other expressions.

Chapter 46 Mask Nodes 1098

Color Space
The Color Space button group determines the color space used when selecting the source color for
the mask. The Wand mask can operate in RGB, YUV, HLS, or LAB color spaces.

Channel
The Channel button group is used to select whether the color that is masked comes from all three
color channels of the image, the alpha channel, or an individual channel only.
The exact labels of the buttons depend on the color space selected for the Wand mask operation.
If the color space is RGB, the options are R, G, or B. If YUV is the color space, the options are Y, U, or V.

Range
The Range slider controls the range of colors around the source color that are included in the mask.
If the value is left at 0.0, only pixels of the same color as the source are considered part of the mask.
The higher the value, the more that similar colors in the source are considered to be wholly part
of the mask.

Range Soft Edge

The Range Soft Edge determines the falloff range of the colors selected. Any pixel within the range
defined above are treated as 100% within the mask. If the soft range is set to 0.0, no other pixels are
considered for the mask. Increasing the soft range increases the number of colors close to, but not
quite within, the range included in the mask. These pixels are semitransparent in the mask.

The Common Controls

Nodes that create masks share several identical controls in the Inspector. This section describes
controls that are common among mask nodes.

Inspector

Mask node Image tab

Image Tab
The controls in this tab set the resolution and clipping method used by the generated mask.

Output Size
The Output size menu sets the resolution of the mask node’s output. The three options include the
default resolution of the comp, the source input’s resolution on nodes that have an input, or a custom
resolution.

Custom
When selecting Custom from the Output Size menu, the width, height, and pixel aspect of the mask
created are locked to values defined in the composition’s Frame Format preferences. If the Frame
Format preferences change, the resolution of the mask produced is changed to match. Disabling this
option can be useful for building a composition at a different resolution than the eventual target
resolution for the final render.

Chapter 46 Mask Nodes 1099

 – Width and Height: This pair of controls is used to set the Width and Height dimensions of the
mask to be created.
– Pixel Aspect: This control is used to specify the Pixel Aspect ratio of the created mask. An aspect
ratio of 1:1 would generate a square pixel with the same dimensions on either side (like a computer
monitor), and an aspect of 0.91 would create a slightly rectangular pixel (like an NTSC monitor).
– Depth: The Depth drop-down menu is used to set the pixel color depth of the image created by
the mask. 32-bit pixels require four times the memory of 8-bit pixels but have far greater accuracy.
Float pixels allow high dynamic range values outside the normal 0..1 range, for representing colors
that are brighter than white or darker than black.

NOTE: Right-click on the Width, Height, or Pixel Aspect controls to display a menu listing the
file formats defined in the preferences Frame Format tab. Selecting any of the listed options
sets the width, height, and pixel aspect to the values for that format.

Clipping Mode
This option determines how the domain of definition rendering handles edges. The Clipping mode is
most important when blur or softness is applied, which may require samples from portions of the
image outside the current domain.
– Frame: The default option is Frame, which automatically sets the node’s domain of definition
to use the full frame of the image, effectively ignoring the current domain of definition. If
the upstream DoD is smaller than the frame, the remaining area in the frame is treated as
black/transparent.
– None: Setting this option to None does not perform any source image clipping. Any data required
to process the node’s effect that would usually be outside the upstream DoD is treated as
black/transparent.

Common Mask Settings controls

Chapter 46 Mask Nodes 1100

Settings Tab
The Settings tab in the Inspector can be found on every tool in the Mask category. The Settings
controls are even found on third-party plug-in tools. The controls are consistent and work the same
way for each tool, although some tools do include one or two individual options, which are also
covered here.

Motion Blur
– Motion Blur: This toggles the rendering of Motion Blur on the tool. When this control is toggled
on, the tool’s predicted motion is used to produce the motion blur caused by the virtual camera’s
shutter. When the control is toggled off, no motion blur is created.
– Quality: Quality determines the number of samples used to create the blur. A quality setting of
2 causes Fusion to create two samples to either side of an object’s actual motion. Larger values
produce smoother results but increase the render time.
– Shutter Angle: Shutter Angle controls the angle of the virtual shutter used to produce the motion
blur effect. Larger angles create more blur but increase the render times. A value of 360 is the
equivalent of having the shutter open for one full frame exposure. Higher values are possible and
can be used to create interesting effects.
– Center Bias: Center Bias modifies the position of the center of the motion blur. This allows for the
creation of motion trail effects.
– Sample Spread: Adjusting this control modifies the weighting given to each sample. This affects
the brightness of the samples.

Use GPU
The Use GPU menu has three settings. Setting the menu to Disable turns off hardware-accelerated
rendering using the graphics card in your computer. Enabled uses the hardware, and Auto uses a
capable GPU if one is available and falls back to software rendering when a capable GPU is
not available

Comments
The Comments field is used to add notes to a tool. Click in the empty field and type the text. When a
note is added to a tool, a small red square appears in the lower-left corner of the node when the full
tile is displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the
note in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 46 Mask Nodes 1101

Chapter 47

Matte Nodes
This chapter details the Matte nodes available in Fusion.

The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Alpha Divide [ADV]  1103
Alpha Multiply [AML]  1104
Chroma Keyer [CKY]  1105
Clean Plate  1110
Delta Keyer  1113
Difference Keyer [DfK]  1121
Luma Keyer [LKY]  1125
Matte Control [MAT]  1128
Primatte [Pri]  1133
How to Key with Primatte  1144
Ultra Keyer [UKY]  1146
The Common Controls  1152

Chapter 47 Matte Nodes 1102

Alpha Divide [ADV]

The Alpha Divide node

Alpha Divide Node Introduction

As the name gives away, the Alpha Divide’s sole purpose is to divide an incoming image’s color
channels by its alpha channel. When you color correct an image that contains a premultiplied alpha
channel, first apply an Alpha Divide node before any color correction node to create a non-
premultiplied image. Then you can perform the color correction. After the color correction, add an
Alpha Multiply node to return the image to its premultiplied state.

Inputs
The Alpha Divide node includes two inputs in the Node Editor.
– Input: The orange input accepts a 2D image with a premultiplied Alpha.
– Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input limits
the pixels where the Alpha divide occurs. An effects mask is applied to the tool after
the tool is processed.

Basic Node Setup

The Alpha Divide node is placed before any color correction is done to an image with a premultiplied
Alpha channel. Here the MediaIn node is assumed to have a premultiplied Alpha channel. The Alpha
Divide node is inserted, and then color correction nodes operate on the “straight” Alpha. An Alpha
Multiply node is placed at the end of the chain to premultiply the Alpha channel again. If only a single
color correction node is used, then the Pre-Divide/Post-Multiply checkbox on the Options tab can be
used in place of the Alpha Divide/Alpha Multiple nodes.

An Alpha Divide node is inserted before color correcting an image with premultiplied alpha.

Inspector
This node has no controls.

Chapter 47 Matte Nodes 1103

Alpha Multiply [AML]

The Alpha Multiply node

Alpha Multiply Node Introduction

As the name gives away, the Alpha Multiply’s sole purpose is to multiply an image’s color channels by
its alpha channel. When you color correct an image that contains a premultiplied alpha channel, first
apply an Alpha Divide node before any color correction node to create a non-premultiplied image.
Then you can perform the color correction. After the color correction, add an Alpha Multiply node to
return the image to its premultiplied state.

Inputs
The Alpha Multiply node includes two inputs in the Node Editor.
– Input: The orange input accepts a 2D image with a “straight” or non-premultiplied alpha.
– Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input limits
the pixels where the Alpha multiply occurs. An effects mask is applied to the tool after the
tool is processed.

Basic Node Setup

The Alpha Multiply node is placed after color correction is done to an image with a premultiplied Alpha
channel. Here the MediaIn node is assumed to have a premultiplied Alpha channel. The Alpha Divide
node is inserted, and then color correction nodes operate on the “straight” Alpha. An Alpha Multiply
node is placed at the end of the chain to premultiply the Alpha channel again. If only a single color
correction node is used, then the Pre-Divide/Post-Multiply checkbox on the Options tab can be used
in place of the Alpha Divide/Alpha Multiple nodes.

An Alpha Multiply node is inserted after color correcting an image with premultiplied alpha.

Inspector
This node has no controls.

Chapter 47 Matte Nodes 1104

Chroma Keyer [CKY]

The Chroma Keyer node

Chroma Keyer Node Introduction

The Chroma Keyer node creates an alpha channel (matte) for an image by removing selected colors
from the scene. Unlike the Delta Keyer or Primatte, which use specific optimizations for keying from
blue and green colors, the Chroma Keyer works equally well with any color.

NOTE: When working with blue- or green-screen shots, it is best to use the Delta Keyer or
Primatte node, rather than the more general purpose Chroma Keyer node.

Inputs
The Chroma Keyer node includes four inputs in the Node Editor.
– Input: The orange input accepts a 2D image that contains the color you want to be
keyed for transparency.
– Garbage Matte: The gray garbage matte input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input causes
areas of the image that fall within the matte to be made transparent. The garbage matte is
applied directly to the alpha channel of the image.
– Solid Matte: The white solid matte input accepts a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input causes areas
of the image that fall within the matte to be fully opaque.
– Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input limits
the pixels where the alpha multiply occurs. An effects mask is applied to the tool after the
tool is processed.

Basic Node Setup

The Chroma Keyer node takes in a 2D image like the MediaIn node below and removes a color that
you identify by dragging over it in the viewer. The result is that the selected color is replaced with
transparency, allowing you to composite the image as the foreground in a Merge node.

A Chomra Keyer node creating transparency on the MediaIn node.

Chapter 47 Matte Nodes 1105

Inspector

The Chroma Key tab

Chroma Key Tab

The Chroma Key tab is used to make the initial selection of color for keying.

Key Type
The Key Type menu determines the selection method used for the matte creation.
– Chroma: The Chroma method creates a matte based on the RGB values of the
selected color range.
– Color: The Color method creates a matte based on the hue of the selected color range.

Color Range
Colors are made transparent by selecting the Chroma Keyer node in the node tree, and then dragging
a selection around the colors in the viewer. The range controls update automatically to represent the
current color selection. You can tweak the range sliders slightly, although most often selecting colors
in the displays is all that is required.

Lock Color Picking

When this checkbox is activated, selecting colors from the viewer is disabled to prevent accidental
addition to the range. It is a good idea to activate this checkbox once you make the color selection for
the matte. All other controls in the node remain editable.

Soft Range
This control softens the selected color range, adding additional colors into the matte.

Reset Color Ranges

Clicking this button resets the Chroma Keyer’s range controls, discarding all color selections. All other
sliders and controls maintain their values.

The Chroma Key Image tab

Chapter 47 Matte Nodes 1106

Image Tab
The Image tab primarily handles removing spill color on the foreground subject. Color spill occurs
when light containing the color you are removing is reflected onto the foreground subject.

Spill Color
This menu selects the color used as the base for all spill suppression techniques.

Spill Suppression
This slider sets the amount of spill suppression applied to the foreground subject.
When this slider is set to 0, no spill suppression is applied.

Spill Method
This menu selects the strength of the algorithm used to apply spill suppression to the image.
– None: None is selected when no spill suppression is required.
– Rare: This removes very little of the spill color and is the lightest of all methods.
– Medium: This works best for green screens.
– Well Done: This works best for blue screens.
– Burnt: This works best for blue. Use this mode only for very troublesome shots. Most likely you will
have to add strong color correction after the key to get, for example, your skin tones back.

Fringe Gamma
This control is used to adjust the brightness of the fringe or halo that surrounds the keyed image.

Fringe Size
This expands and contracts the size of the fringe or halo surrounding the keyed image.

Fringe Shape
Fringe Shape forces the fringe toward the external edge of the image or toward the inner edge of the
fringe. Its effect is most noticeable while the Fringe Size slider’s value is large.

Cyan/Red, Magenta/Green, and Yellow/Blue

Use these three controls to color correct the fringe of the image. This is useful for correcting
semitransparent pixels that still contain color from the original background to match the new
background.

The Chroma Key Matte tab

Chapter 47 Matte Nodes 1107

Matte Tab
The Matte tab refines the softness, density, and overall fit of the resulting matte.

Filter
This control selects the filtering algorithm used when applying blur to the matte.
– Box: This is the fastest method but at reduced quality. Box is best suited for minimal
amounts of blur.
– Bartlett: Otherwise known as a Pyramid filter, Bartlett makes a good compromise between speed
and quality.
– Multi-box: When selecting this filter, the Num Passes slider appears and lets you control the
quality. At 1 and 2 passes, results are identical to Box and Bartlett, respectively. At 4 passes and
above, results are usually as good as Gaussian, in less time and with no edge “ringing.”
– Gaussian: The Gaussian filter uses a true Gaussian approximation and gives excellent results, but
it is a little slower than the other filters. In some cases, it can produce an extremely slight edge
“ringing” on floating-point pixels.

Blur
Matte Blur blurs the edge of the matte based on the Filter menu setting. A value of zero results in a
sharp, cutout-like hard edge. The higher the value, the more blur applied to the matte.

Clipping Mode
This option determines how edges are handled when performing domain of definition rendering.
This is profoundly important when blurring the matte, which may require samples from portions of the
image outside the current domain.
– Frame: The default option is Frame, which automatically sets the node’s domain of definition
to use the full frame of the image, effectively ignoring the current domain of definition. If the
upstream DoD is smaller than the frame, the remaining area in the frame will be treated as
black/transparent.
– Domain: Setting this option to Domain will respect the upstream domain of definition when
applying the node’s effect. This can have adverse clipping effects in situations where the node
employs a large filter.
– None: Setting this option to None will not perform any source image clipping at all. This means
that any data required to process the node’s effect that would normally be outside the upstream
DoD will be treated as black/transparent.

Contract/Expand
This slider shrinks or grows the semitransparent areas of the matte. Values above 0.0 expand the
matte, while values below 0.0 contract it.
This control is usually used in conjunction with the Matte Blur to take the hard edge of a matte and
reduce fringing. Since this control affects only semitransparent areas, it will have no effect on a matte’s
hard edge.

Gamma
Matte Gamma raises or lowers the values of the matte in the semitransparent areas. Higher values
cause the gray areas to become more opaque, and lower values cause the gray areas to become
more transparent. Completely black or white regions of the matte remain unaffected.
Since this control affects only semitransparent areas, it will have no effect on a matte’s hard edge.

Chapter 47 Matte Nodes 1108

Threshold
This range slider sets the lower threshold using the handle on the left and sets the upper threshold
using the handle on the right.
Any value below the lower threshold setting becomes black or transparent in the matte.
Any value above the upper threshold setting becomes white or opaque in the matte. All values within
the range maintain their relative transparency values.
This control is often used to reject salt and pepper noise in the matte.

Restore Fringe
This restores the edge of the matte around the keyed subject. Often when keying, the edge of the
subject where you have hair is clipped out. Restore Fringe brings back that edge while keeping the
matte solid.

Invert Matte
When this checkbox is selected, the alpha channel created by the keyer is inverted, causing all
transparent areas to be opaque and all opaque areas to be transparent.

Solid Matte
Solid Mattes are mask nodes or images connected to the solid matte input on the node. The solid
matte is applied directly to the alpha channel of the image. Generally, solid mattes are used to hold out
keying in areas you want to remain opaque, such as someone with blue eyes against a blue screen.
Enabling Invert will invert the solid matte, before it is combined with the source alpha.

Garbage Matte
Garbage mattes are mask nodes or images connected to the garbage matte input on the node. The
garbage matte is applied directly to the alpha channel of the image. Generally, garbage mattes are
used to remove unwanted elements that cannot be keyed, such as microphones and booms. They are
also used to fill in areas that contain the color being keyed but that you wish to maintain.
Garbage mattes of different modes cannot be mixed within a single tool. A Matte Control node is often
used after a Keyer node to add a garbage matte with the opposite effect of the matte applied to
the keyer.
Enabling Invert will invert the garbage matte, before it is combined with the source alpha.

Post-Multiply Image
Select this option to cause the keyer to multiply the color channels of the image against the alpha
channel it creates for the image. This option is usually enabled and is on by default.
Deselect this checkbox and the image can no longer be considered premultiplied for purposes of
merging it with other images. Use the Subtractive option of the Merge node instead of the
Additive option.
For more information on these Merge node settings, see Chapter 35, “Composite Nodes” in the Fusion
Reference Manual or Chapter 96 in the DaVinci Resolve Reference Manual.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other matte nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 47 Matte Nodes 1109

Clean Plate

The Clean Plate node

Clean Plate Node Introduction

The Clean Plate tool is a pre-keying node used to generate an image of the green or blue color screen
to smooth out the lighting differences. The output of the Clean Plate is later connected to the Clean
Plate input on the Delta Keyer so it can key fine detail without choking or clipping the matte.

How to Create a Clean Plate

Creating a clean plate is the opposite of creating a key. When keying, you try to remove the green or
blue color. When creating a clean plate, you try to keep as much of the blue- or green-screen as
possible. By box selecting areas of the screen color in the viewer, you end up with an image of the
green/blue screen. A transparent cutout represents everything that is not part of the blue or
green screen.

Once you have the selection, the Erode control expands the pre-matte, removing any small pixels of
non-green/blue screen around the edges. Then, growing the pre-matte fills in the holes until you have
a solid blue or green image.

Inputs
The Clean Plate node includes three inputs in the Node Editor.
– Input: The orange input accepts a 2D image that contains the green or blue screen.
– Garbage Matte: The white garbage matte input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input causes
areas of the image that fall within the matte to be excluded from the clean plate. For a clean
plate, garbage mattes should contain areas that are not part of the blue or green screen.
– Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input limits the
pixels where the clean plate is generated. An effects mask is applied to the tool after the
tool is processed.

Chapter 47 Matte Nodes 1110

Basic Node Setup
The Clean Plate node and the Delta Keyer are two separate branches stemming from the main image
you want to key. The green-screen or blue-screen clip is breached to connect to both the orange
image input on the Clean Plate and the orange image input on the Delta Keyer. The output of the clean
plate is then connected to the magenta clean plate input on the Delta Keyer. The output of the Delta
Keyer is then used as the foreground to a Merge.

A Clean Plate connected to a Delta Keyer

Inspector

The Clean Plate tab

Plate Tab
The Plate tab contains the primary tools for creating a clean plate. Using this tab, you drag over the
areas in the viewer, and then use the Erode and Grow Edges sliders to create the clean plate.

Method
The Method menu selects the type of color selection you use when sampling colors in the viewer.
– Color: Color uses a difference method to separate the background color. This works well on
screen colors that are even.
– Ranges: Ranges uses a chroma range method to separate the background color. This is a better
option for shadowed screen or screens that have different colors.

Chapter 47 Matte Nodes 1111

Matte Threshold
This range slider sets the lower threshold using the handle on the left and sets the upper threshold
using the handle on the right.
Any value below the lower threshold becomes black or transparent in the matte.
Any value above the upper threshold becomes white or opaque in the matte. All values within the
range maintain their relative transparency values. This control is often used to reject salt and pepper
noise in the matte.

Erode
The Erode slider decreases the size of the screen area. It is used to eat away at small non-screen color
pixels that may interfere with creating a smooth green- or blue-screen clean plate.

Crop
Crop trims in from the edges of the image.

Grow Edges
The Grow Edges slider expands the color of the edges to fill in holes until fully green or blue screen
is created.

Fill
The Fill checkbox fills in remaining holes with color from the surrounding screen color.

Time Mode
– Sequence: Generates a new clean plate every frame.
– Hold Frame: Holds the clean plate at a single frame.

The Clean Plate Mask tab

Mask Tab
The Mask tab is used to invert the mask connected to the garbage mask input on the node. The
garbage mask can be applied to clear areas before growing edges or filling remaining holes.

Invert
Invert uses the transparent parts of the mask to clear the image.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other matte nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 47 Matte Nodes 1112

Delta Keyer

The Delta Keyer node

Delta Keyer Node Introduction

The Delta Keyer is an advanced color difference keyer, with many features and controls for tuning the
matte and separating the subject from a blue or green screen.
It contains several keying systems: the Key tab is the master difference keyer and Pre-Matte is a
built-in clean plate to smooth out screen color. Tuning, Fringe, and Matte finish the keying process. The
tabs are laid out in the rough order you tend to use them.

Inputs
The Delta Keyer node includes five inputs in the Node Editor.
– Input: The orange input accepts a 2D image that contains the color you want to be keyed for
transparency.
– Garbage Matte: The gray garbage matte input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input causes
areas of the image that fall within the matte to be made transparent. The garbage matte is
applied directly to the alpha channel of the image.
– Solid Matte: The white solid matte input accepts a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input causes areas
of the image that fall within the matte to be fully opaque.
– Clean Plate: Accepts the resulting image from the Clean Plate node.
– Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input
limits the pixels where the keying occurs. An effects mask is applied to the tool after
the tool is processed.

Basic Node Setup

A single Delta keyer rarely gets perfect results because most green- or blue-screen shots have
problems that the keyer is not made to handle. Keyers often need the help of garbage mattes or solid
mattes created with a Polygon or B-Spline node. Shots can also require more than just one keyer to
achieve perfect results. That is shown in one of the examples below, where one Delta Keyer is used to
create a hard inner white matte, while a second Delta Keyer is used to capture the soft edges. The
core matte below is then connected into the solid matte input on the second Delta Keyer. Color
correction, including spill suppression, is best handled as a separate branch in your node tree.
Separating your matte from your color correction using underlays makes it clear what branch of nodes
is performing what operation.

Chapter 47 Matte Nodes 1113

Starting point for a Delta Keyer setup

An expanded key setup with two Delta Keyers, a Polygon and

B-Spline nodes for mattes, and a branch for color correction

Inspector

The Delta Keyer View Mode menu

Key Tab View Mode

At the top of the Inspector is the View Mode menu. The default selection shows the final result. You
can change the view to see various intermediate stages of the keying process.
– Pre Matte: Displays the output of the Pre Matte key.
– Matte: Displays the alpha of the key before being combined with solid and garbage masks.
When displaying the matte, set the viewer to show the alpha channel.
– Tuning Ranges: Displays a false color for Shadow, Midtone, and Highlight ranges of the image.
Shadows are in the red channel, Midtones are in the green channel, and Highlights are in
the blue channel.

Chapter 47 Matte Nodes 1114

 – Status: Displays information to indicate areas that are solid, transparent, or in between. It also
displays areas that have been affected by matte adjustments, such as thresholding or erode/dilate
and areas affected by the solid mask.
– Intermediate Result: The untouched source image color channels combined with the final matte.
Optionally, combine this output with additional Delta Keyer nodes.
– Final Result: The final keyed image with spill suppression, ready to merge onto a scene.

The Delta Keyer Key tab

Key Tab
The Key tab is where most keying begins. It is used to select the screen color.

Background Color
This is the color of the blue or green screen, sometimes called the screen color. To create the key with
the Delta Keyer, use the background color Eyedropper to select the screen color from the image.

Pre-Blur
Applies a blur before generating the alpha. This can help with certain types of noise, edge
enhancements, and artifacts in the source image.

Gain
Gain increases the influence of the screen color, causing those areas to become more transparent.

Chapter 47 Matte Nodes 1115

Balance
A color difference keyer, like the Delta Keyer, compares the differences between the dominant
channel determined by the selected background color and the other two channels. Adjusting balance
determines the proportions of the other two channels. A value of 0 uses the minimum of the other two
channels, where a value of 1 uses the maximum. A value of 0.5 uses half of each.

Lock Alpha/Spill Removal Color Balance Reference

Unlocking this allows you to use different color references when generating the alpha and when
determining how much of the background color to subtract from the image. When enabled, spill and
color are combined.

Color Balance Reference

The Color balance reference controls adjust for lighting or white balance that might be reducing
background color purity and saturation. A correction is applied based on the reference of a neutral-
colored object when generating the key and determining the amount of background color subtraction,
without altering the background color that is subtracted.

The Delta Keyer Pre Matte tab

Pre Matte Tab

The Pre Matte tab is in some ways a garbage matte keyer that occurs before the delta keyer process.
It is used to even out the screen color before a more delicate key is done. After selecting the initial
screen color, you can switch to the Pre Matte tab and view the PreMatte Out from view mode menu at
the top of the Inspector. The Pre Matte works on ranges of color, so you start by dragging a bounding
box around areas where lighting is different from the initial screen color pick. Use the Soft range to
expand, and Erode to cut away the edges.
Once you have a more even screen selection, you can move to the Matte tab.

Soft Range
The Soft Range extends the range of selected color and rolloff of the screen color.

Erode
Erode contracts the edge of the pre matte, so the edge detail does not clip.

Chapter 47 Matte Nodes 1116

Blur
This softens the edges of the pre matte.

Pre Matte Range

These controls update automatically to represent the current color selection. The controls tweak the
selection slightly, although selecting colors in the viewer is all that is required.

Lock Color Picking

When this checkbox is activated, it prevents the accidental growth of the selected range by selecting
more colors from the view. It is a good idea to activate this checkbox after selecting the color for the
matte. All other controls in the node remain editable.

Reset Pre Matte Ranges

This button discards all color selection by resetting the ranges but maintains all other slider and
control values.

The Delta Keyer Matte tab

Matte Tab
The Matte tab refines the alpha of the key, combined with any solid and garbage masks connected to
the node. When using the matte tab, set the viewer to display the alpha channel of the Delta Keyer’s
final output.

Threshold
This range slider sets the lower threshold using the handle on the left and sets the upper threshold
using the handle on the right.
Any value below the lower threshold setting becomes black or transparent in the matte.
Any value above the upper threshold setting becomes white or opaque in the matte. All values within
the range maintain their relative transparency values.

Chapter 47 Matte Nodes 1117

Restore Fringe
This restores the edge of the matte around the keyed subject. Often when keying, clipping occurs
around the edge of the subject where there is hair. Restore Fringe brings back that edge while
keeping the matte solid.

Erode/Dilate
Expands or contracts the matte.

Blur
Softens the matte.

Clean Foreground
Fills slightly transparent (light gray) areas of the matte.

Clean Background
Clips the bottom dark range of the matte.

Replace Mode
Determines how matte adjustments restore color to the image.
– None: No color replacement. Matte processing does not affect the color.
– Source: The color from the original image.
– Hard Color: A solid color.
– Soft Color: A solid color weighted by how much background color was originally removed.

Replace Color
The color used with the Hard Color and Soft Color replace modes.

The Delta Keyer Fringe tab

Fringe Tab
The Fringe tab handles the majority of spill suppression in the Delta Keyer. Spill suppression is a form
of color correction that attempts to remove the screen color from the fringe of the matte.
Spill is the transmission of the screen color through the semitransparent areas of the alpha channel. In
the case of blue- or green-screen keying, this usually causes the color of the background to become
apparent in the edges of the foreground subject.

Chapter 47 Matte Nodes 1118

Spill Method
This selects the strength of the algorithm used to apply spill suppression to the image.
– None: None is selected when no spill suppression is required.
– Rare: This removes very little of the spill color and is the lightest of all methods.
– Medium: This works best for green screens.
– Well Done: This works best for blue screens.
– Burnt: This works best for blue screens. Use this mode only for very troublesome shots.

Spill Suppression
When this slider is set to 0, no spill suppression is applied to the image. Increasing the slider increases
the strength of the spill method.

Fringe Gamma
This control can be used to adjust the brightness of the fringe or halo that surrounds the keyed image.

Fringe Size
This expands and contracts the size of the fringe or halo surrounding the keyed image.

Fringe Shape
Fringe Shape presses the fringe toward the external edge of the image or pulls it toward the inner
edge of the fringe. Its effect is most noticeable while the Fringe Size value is large.

Cyan/Red, Magenta/Green, and Yellow/Blue

Use these three controls to color correct the fringe of the image.
This is useful for correcting semitransparent pixels that still contain color from the original background
to match the new background.

The Delta Keyer Tuning tab

Tuning Tab
The Tuning tab is an advanced tab that allows you to determine the size of the shadow, midtone, and
highlight ranges. By modifying the ranges, you can select the strength of the matte and spill
suppression based on tonal values.

Chapter 47 Matte Nodes 1119

Range Controls
The range controls define how much color is considered to fall into the shadows, midtones, and
highlights areas of the image. The spline controls allow for easy adjusting of the tonal ranges of each
Shadow and Highlight tonal map.

Simple/Smooth
The Simple button sets the range to be linear. The Smooth button sets a smooth tonal gradient for
the ranges.

Lock Alpha/Spill Removal Tuning

When this checkbox is disabled, separate tuning controls are used when generating the alpha and
when determining how much of the background color to subtract from the image.
– Shadows: Adjusts the strength of the key in darker areas of the background.
– Midtones: Adjusts the strength of the key in midtone areas of the background.
– Highlights: Adjusts the strength of the key in brighter areas of the background.

The Delta Keyer Mask tab

Mask Tab
The Mask tab determines how the solid and garbage mattes are applied to the key.

Solid Source Alpha

Used to combine the existing alpha from the source image with the solid mask.
– Ignore: Does not combine the alpha from the source image.
– Add: Solid areas of the source image alpha are made solid in the solid mask.
– Subtract: Transparent areas of the source image alpha are made transparent in the solid mask.

Chapter 47 Matte Nodes 1120

Solid Replace Mode
This determines how the solid mask restores color to the image.
– None: No color replacement. The solid mask does not affect the color.
– Source: The color from the original image.
– Hard Color: A solid color.
– Soft Color: A solid color weighted by how much background color was originally removed.

Solid Replace Color

The color used with the Hard Color and Soft Color replace modes.
– Invert: Inverts the solid mask, before it is combined with the source alpha.

Garbage Mask
– Invert: Normally, solid areas of a garbage mask remove the image. When inverted, the transparent
areas of the mask remove the image.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other matte nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Difference Keyer [DfK]

The Difference Keyer node

Difference Keyer Node Introduction

Difference keying is a process that produces a matte based on the differences between two images.
A difference key uses two input images: one containing the subject with the background and another
containing the background without the subject.
Although the process sounds reasonable at first glance, subtle variations in the camera position from
shot to shot usually make it difficult to get clean results. Think of the futile attempt of trying to key
smoke in front of a brick wall and using a clean plate of the brick wall as your difference input. Part of
the wall’s structure is always visible in this keying method. Instead, a Difference Keyer is often used to
produce a rough matte that is combined with other nodes to produce a more detailed matte.

Inputs
The Difference Keyer node includes four inputs in the Node Editor.
– Background: The orange background input accepts a 2D image that contains just the set
without your subject.
– Foreground: The green foreground input accepts a 2D image that contains the shot with your
subject in the frame.

Chapter 47 Matte Nodes 1121

 – Garbage Matte: The gray garbage matte input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input causes
areas of the image that fall within the matte to be made transparent.
– Solid Matte: The white solid matte input accepts a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input causes areas
of the image that fall within the matte to be fully opaque.
– Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input limits the
pixels where the difference matte occurs. An effects mask is applied to the tool after the
tool is processed.

Basic Node Setup

When you do not have content shot on a blue or green screen, the Difference Keyer can be one node
in a chain of many used to extract an object from the background. The example below has the
MediaIn1 as the main subject and a clean background shot without the subject (Background). A
B-Spline is used to limit the area the Difference Keyer must deal with for extraction. The result is a
matte that can be used to help but not solve the key.

A Difference Keyer with two inputs: one of the subject on

a background and the other of just the background

Inspector

The Difference Keyer Controls tab

Chapter 47 Matte Nodes 1122

Controls Tab
The Controls tab in the Difference Keyer contains all the parameters for adjusting the quality of
the matte.

Threshold
This range slider sets the lower threshold using the handle on the left and sets the upper threshold
using the handle on the right. Adjusting them defines a range of difference values between the
images to create a matte.
A difference below the lower threshold setting becomes black or transparent in the matte.
Any difference above the upper threshold setting becomes white or opaque in the matte.
The difference values in the range in between create a grayscale matte.

Filter
This control selects the filtering algorithm used when applying a blur to the matte.
– Box: This is the fastest method but at reduced quality. Box is best suited for minimal
amounts of blur.
– Bartlett: Otherwise known as a Pyramid filter, Bartlett makes a good compromise between speed
and quality.
– Multi-box: When selecting this filter, the Num Passes slider appears and lets you control the
quality. At 1 and 2 passes, results are identical to Box and Bartlett, respectively. At 4 passes and
above, results are usually as good as Gaussian, in less time and with no edge “ringing.”
– Gaussian: The Gaussian filter uses a true Gaussian approximation and gives excellent results, but
it is a little slower than the other filters. In some cases, it can produce an extremely slight edge
“ringing” on floating-point pixels.

Blur
This blurs the edge of the matte using the method selected in the Filter menu. A value of zero results
in a sharp, cutout-like hard edge. The higher the value, the more blur.

Clipping Mode
This option determines how edges are handled when performing domain of definition rendering. This
is profoundly important when blurring the matte, which may require samples from portions of the
image outside the current domain.
– Frame: The default option is Frame, which automatically sets the node’s domain of definition
to use the full frame of the image, effectively ignoring the current domain of definition. If the
upstream DoD is smaller than the frame, the remaining area in the frame is treated as black/
transparent.
– Domain: Setting this option to Domain respects the upstream domain of definition when applying
the node’s effect. This can have adverse clipping effects in situations where the node employs a
large filter.
– None: Setting this option to None does not perform any source image clipping at all. This means
that any data required to process the node’s effect that would usually be outside the upstream
DoD is treated as black/transparent.

Chapter 47 Matte Nodes 1123

Contract/Expand
This slider shrinks or grows the semitransparent areas of the matte. Values above 0.0 expand the
matte, while values below 0.0 contract it.
This control is usually used in conjunction with the blur to take the hard edge of a matte and reduce
fringing. Since this control affects only semitransparent areas, it has no effect on a matte’s hard edge.

Gamma
Matte Gamma raises or lowers the values of the matte in the semitransparent areas. Higher values
cause the gray areas to be more opaque, and lower values cause the gray areas to be more
transparent. Wholly black or white regions of the matte remain unaffected.

Invert
Selecting this checkbox inverts the matte, causing all transparent areas to be opaque and all opaque
areas to be transparent.

Solid Matte
Solid Mattes are mask nodes or images connected to the solid matte input on the node. The solid
matte is applied directly to the alpha channel of the image. Generally, solid mattes are used to hold out
keying in areas you want to remain opaque, such as someone with blue eyes against a blue screen.
Enabling Invert, inverts the solid matte before it is combined with the source alpha.

Garbage Matte
Garbage mattes are mask nodes or images connected to the garbage matte input on the node. The
garbage matte is applied directly to the alpha channel of the image. Generally, garbage mattes are
used to remove unwanted elements that cannot be keyed, such as microphones and booms. They are
also used to fill in areas that contain the color being keyed but that you wish to maintain.
Garbage mattes of different modes cannot be mixed within a single tool. A Matte Control node is often
used after a Keyer node to add a garbage matte with the opposite effect of the matte applied to
the keyer.
Enabling Invert inverts the garbage matte before it is combined with the source alpha.

Post-Multiply Image
Select this option to cause the keyer to multiply the color channels of the image against the alpha
channel it creates for the image. This option is usually enabled and is on by default.
Deselect this checkbox, and the image can no longer be considered premultiplied for purposes of
merging it with other images. Use the Subtractive option of the Merge node instead of the
Additive option.
For more information on these Merge node settings, see Chapter 35, “Composite Nodes” in the Fusion
Reference Manual or Chapter 96 in the DaVinci Resolve Reference Manual.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Matte nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 47 Matte Nodes 1124

Luma Keyer [LKY]

The Luma Keyer node

Luma Keyer Node Introduction

The Luma Keyer node uses the overall luminance of an image to create an Alpha channel. The label of
this node may seem misleading since it allows pulling mattes from almost any channel. In some
respects, it is more accurate to call this node an all-purpose channel keyer, but its primary purpose is
for extracting alpha channels based on luminance

Inputs
The Luma Keyer node includes four inputs in the Node Editor.
– Input: The orange input accepts a 2D image that contains the luminance values you want to be
keyed for transparency.
– Garbage Matte: The gray garbage matte input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input causes
areas of the image that fall within the matte to be made transparent. The garbage matte is
applied directly to the alpha channel of the image.
– Solid Matte: The white solid matte input accepts a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input causes areas
of the image that fall within the matte to be fully opaque.
– Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input limits
the pixels where the luminance key occurs. An effects mask is applied to the tool after the
tool is processed.

Basic Node Setup

The Luma Keyer takes an input from an image with significant luminance difference to extract a key.
You can then use the output of the Luma Keyer into any mask input.

A Luma Keyer output connecting into an effect mask on a Merge node

Chapter 47 Matte Nodes 1125

Inspector

The Luma Keyer Controls tab

Controls Tab
The Controls tab in the Luma Keyer contains all the parameters for adjusting the quality of the matte.

Channel
This menu selects the color channel used for creating the matte. Select from the Red, Green, Blue,
Alpha, Hue, Luminance, Saturation, and Depth (Z-buffer) channels.

Threshold
This range slider sets the lower threshold using the handle on the left and sets the upper threshold
using the handle on the right. Adjusting them defines a range of luminance values to create a matte.
A value below the lower threshold setting becomes black or transparent in the matte.
Any value above the upper threshold setting becomes white or opaque in the matte.
The values within the range create a grayscale matte.

Filter
This control selects the filtering algorithm used when applying a blur to the matte.
– Box: This is the fastest method but at reduced quality. Box is best suited for minimal
amounts of blur.
– Bartlett: Otherwise known as a Pyramid filter, Bartlett makes a good compromise between speed
and quality.
– Multi-box: When selecting this filter, the Num Passes slider appears and lets you control the
quality. At 1 and 2 passes, results are identical to Box and Bartlett, respectively. At 4 passes and
above, results are usually as good as Gaussian, in less time and with no edge “ringing.”
– Gaussian: The Gaussian filter uses a true Gaussian approximation and gives excellent results, but
it is a little slower than the other filters. In some cases, it can produce an extremely slight edge
“ringing” on floating-point pixels.

Chapter 47 Matte Nodes 1126

Blur
This blurs the edge of the matte using the method selected in the Filter menu. A value of zero results
in a sharp, cutout-like hard edge. The higher the value, the more blur.

Clipping Mode
This option determines how edges are handled when performing domain of definition rendering. This
is profoundly important when blurring the matte, which may require samples from portions of the
image outside the current domain.
– Frame: The default option is Frame, which automatically sets the node’s domain of definition
to use the full frame of the image, effectively ignoring the current domain of definition. If the
upstream DoD is smaller than the frame, the remaining area in the frame is treated as black/
transparent.
– Domain: Setting this option to Domain respects the upstream domain of definition when applying
the node’s effect. This can have adverse clipping effects in situations where the node employs a
large filter.
– None: Setting this option to None does not perform any source image clipping at all. This means
that any data required to process the node’s effect that is usually outside the upstream DoD is
treated as black/transparent.

Contract/Expand
This slider shrinks or grows the semitransparent areas of the matte. Values above 0.0 expand the
matte, while values below 0.0 contract it.
This control is usually used in conjunction with the blur to take the hard edge of a matte and reduce
fringing. Since this control affects only semitransparent areas, it has no effect on a matte’s hard edge.

Gamma
Matte Gamma raises or lowers the values of the matte in the semitransparent areas. Higher values
cause the gray areas to be more opaque, and lower values cause the gray areas to be more
transparent. Wholly black or white regions of the matte remain unaffected.

Invert
Selecting this checkbox inverts the matte, causing all transparent areas to be opaque and all opaque
areas to be transparent.

Solid Matte
Solid mattes are mask nodes or images connected to the solid matte input on the node. The solid
matte is applied directly to the alpha channel of the image. Generally, solid mattes are used to hold out
keying in areas you want to remain opaque, such as someone with blue eyes against a blue screen.
Enabling Invert inverts the solid matte before it is combined with the source alpha.

Garbage Matte
Garbage mattes are mask nodes or images connected to the garbage matte input on the node. The
garbage matte is applied directly to the alpha channel of the image. Generally, garbage mattes are
used to remove unwanted elements that cannot be keyed, such as microphones and booms. They are
also used to fill in areas that contain the color being keyed but that you wish to maintain.
Garbage mattes of different modes cannot be mixed within a single tool. A Matte Control node is often
used after a Keyer node to add a garbage matte with the opposite effect of the matte applied to
the keyer.
Enabling Invert inverts the garbage matte before it is combined with the source alpha.

Chapter 47 Matte Nodes 1127

Post-Multiply Image
Select this option to cause the keyer to multiply the color channels of the image against the alpha
channel it creates for the image. This option is usually enabled and is on by default.
Deselect this checkbox and the image can no longer be considered premultiplied for purposes of
merging it with other images. Use the Subtractive option of the Merge node instead of the
Additive option.
For more information on these Merge node settings, see Chapter 35, “Composite Nodes” in the Fusion
Reference Manual or Chapter 96 in the DaVinci Resolve Reference Manual.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Matte nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Matte Control [MAT]

The Matte Control node

Matte Control Node Introduction

Keyer nodes are used to create an alpha channel on an image that does not have one. The Matte
Control node is used to combine and manipulate the alpha channels embedded in images as well as
masks created by masking tools.
Typically, you add this node to copy a color channel or alpha channel from the foreground input to the
background input, or to combine alpha channels from the two images.

Inputs
The Matte Control node includes four inputs in the Node Editor.
– Background: The orange background input accepts a 2D image that receives the foreground
image alpha channel (or some other channel you want to copy to the background).
– Foreground: The green foreground input accepts a 2D image that contains an alpha channel
(or some other channel) you want to be applied to the background image.
– Garbage Matte: The gray garbage matte input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input
causes areas of the foreground/background combination that fall within the matte to be made
transparent.
– Solid Matte: The white solid matte input accepts a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input causes areas
of the foreground/background combination that fall within the matte to be fully opaque.
– Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input limits
the pixels where the matte control occurs. An effects mask is applied to the tool after the
tool is processed.

Chapter 47 Matte Nodes 1128

Basic Node Setup
Below, the Matte Control node is set up to copy the foreground (green) input’s alpha channel into the
background (orange) input. The output of the Matte Control is then an image with an alpha channel
used as the foreground composite in the Merge node.

A Matte Control embedded an alpha from the foreground input to the background input

Inspector

The Matte Control Matte tab

Matte Tab
The Matte tab combines and modifies alpha or color channels from an image in the foreground input
with the background image.

Combine
Use this menu to select which operation is applied. The default is set to None for no operation.
– None: This causes the foreground image to be ignored.
– Combine Red: This combines the foreground red channel to the background alpha channel.
– Combine Green: This combines the foreground green channel to the background alpha channel.

Chapter 47 Matte Nodes 1129

 – Combine Blue: This combines the foreground blue channel with the background alpha channel.
– Combine Alpha: This combines the foreground alpha channel with the background alpha channel.
– Solid: This causes the background alpha channel to become completely opaque.
– Clear: This causes the background alpha channel to become completely transparent.

Combine Operation
Use this menu to select the method used to combine the foreground channel with the background.
– Copy: This copies the foreground source over the background alpha, overwriting any existing
alpha in the background.
– Add: This adds the foreground source to the background alpha.
– Subtract: This subtracts the foreground source from the background alpha.
– Inverse Subtract: This subtracts the background alpha from the foreground source.
– Maximum: This compares the foreground source and the background alpha and takes the value
from the pixel with the highest value.
– Minimum: This compares the foreground source and the background alpha and takes the value
from the pixel with the lowest value.
– And: This performs a logical AND on the two values.
– Or: This performs a logical OR on the values.
– Merge Over: This merges the foreground source channel over the background alpha channel.
– Merge Under: This merges the foreground source channel under the background alpha channel.

Filter
Selects the Filter that is used when blurring the matte.
– Box Blur: This option applies a Box Blur effect to the whole image. This method is faster than the
Gaussian blur but produces a lower-quality result.
– Bartlett: Bartlett applies a more subtle, anti-aliased blur filter.
– Multi-Box: Multi-Box uses a box filter layered in multiple passes to approximate a Gaussian shape.
With a moderate number of passes (e.g., four), a high-quality blur can be obtained, often faster
than the Gaussian filter and without any ringing.
– Gaussian: Gaussian applies a smooth, symmetrical blur filter, using a sophisticated constant-time
Gaussian approximation algorithm. In extreme cases, this algorithm may exhibit ringing; see below
for a discussion of this. This mode is the default filter method.

Blur
This blurs the edge of the matte using a standard, constant speed Gaussian blur. A value of zero
results in a sharp, cutout-like hard edge. The higher the value, the more blur is applied to the matte.

Clipping Mode
This option determines how edges are handled when performing domain of definition rendering. This
is profoundly important when blurring the matte, which may require samples from portions of the
image outside the current domain.
– Frame: The default option is Frame, which automatically sets the node’s domain of definition
to use the full frame of the image, effectively ignoring the current domain of definition. If the
upstream DoD is smaller than the frame, the remaining area in the frame is treated as black/
transparent.

Chapter 47 Matte Nodes 1130

 – Domain: Setting this option to Domain respects the upstream domain of definition when applying
the node’s effect. This can have adverse clipping effects in situations where the node employs a
large filter.
– None: Setting this option to None does not perform any source image clipping at all. This means
that any data required to process the node’s effect that is usually outside the upstream DoD is
treated as black/transparent.

Contract/Expand
This shrinks or grows the matte similar to an Erode Dilate node. Contracting the matte reveals more of
the foreground input, while expanding the matte reveals more of the background input. Values above
0.0 expand the matte, and values below 0.0 contract it.

Gamma
This raises or lowers the values of the matte in the semitransparent areas. Higher values cause the
gray areas to become more opaque, and lower values cause the gray areas to become more
transparent. Completely black or white regions of the matte remain unaffected.

Threshold
Any value below the lower threshold becomes black or transparent in the matte. Any value above the
upper threshold becomes white or opaque in the matte. All values within the range maintain their
relative transparency values.

Restore Fringe
This restores the edge of the matte around the keyed subject. Often when keying, the edge of the
subject where you have hair is clipped out. Restore Fringe brings back that edge while keeping the
matte solid.

Invert Matte
When this checkbox is selected, the alpha channel of the image is inverted, causing all transparent
areas to be opaque and all opaque areas to be transparent.

Solid Matte
Solid mattes are mask nodes or images connected to the solid matte input on the node. The solid
matte is applied directly to the alpha channel of the image. Generally, solid mattes are used to hold out
areas you want to remain opaque, such as someone with blue eyes against a blue screen.
Enabling Invert inverts the solid matte before it is combined with the source alpha.

Garbage Matte
Garbage mattes are mask nodes or images connected to the garbage matte input on the node. The
garbage matte is applied directly to the alpha channel of the image. Generally, garbage mattes are
used to remove unwanted elements that cannot be keyed, such as microphones and booms. They are
also used to fill in areas that contain the color being keyed but that you wish to maintain.
Garbage mattes of different modes cannot be mixed within a single tool. A Matte Control node is often
used after a Keyer node to add a garbage matte with the opposite effect of the matte applied to
the keyer.
Enabling Invert inverts the garbage matte before it is combined with the source alpha.

Post-Multiply Image
Selecting this option multiplies the color channels of the image against the alpha channel it creates for
the image. This option is usually enabled and is on by default.
Deselect this checkbox and the image can no longer be considered premultiplied for purposes of
merging it with other images. Use the Subtractive option of the Merge node instead of the
Additive option.

Chapter 47 Matte Nodes 1131

For more information on these Merge node settings, see Chapter 35, “Composite Nodes” in the Fusion
Reference Manual or Chapter 96 in the DaVinci Resolve Reference Manual.

The Matte Control Spill tab

Spill Tab
The Spill tab handles spill suppression in the Matte Control. Spill suppression is a form of color
correction that attempts to remove the screen color from the fringe of the matte.
Spill is the transmission of the screen color through the semitransparent areas of the alpha channel. In
the case of blue- or green-screen keying, this usually causes the color of the background to become
apparent in the edges of the foreground subject.

Spill Color
This menu selects the color used as the base for all spill suppression techniques.

Spill Suppression
When this slider is set to 0, no spill suppression is applied to the image. Increasing the slider increases
the strength of the spill method.

Spill Method
This selects the strength of the algorithm used to apply spill suppression to the image.
– None: None is selected when no spill suppression is required.
– Rare: This removes very little of the spill color and is the lightest of all methods.
– Medium: This works best for green screens.
– Well Done: This works best for blue screens.
– Burnt: This works best for blue screen. Use this mode only for very troublesome shots.

Fringe Gamma
This control can be used to adjust the brightness of the fringe or halo that surrounds the keyed image.

Fringe Size
This expands and contracts the size of the fringe or halo surrounding the keyed image.

Fringe Shape
Fringe Shape presses the fringe toward the external edge of the image or pulls it toward the inner
edge of the fringe. Its effect is most noticeable while the Fringe Size value is large.

Chapter 47 Matte Nodes 1132

Cyan/Red, Magenta/Green, and Yellow/Blue
Use these three controls to color correct the fringe of the image.
This is useful for correcting semitransparent pixels that still contain color from the original background
to match the new background.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Matte nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Primatte [Pri]

The Primatte node

NOTE: Primatte is only available in Fusion Studio.

Primatte Node Introduction

Primatte is an advanced keying tool for Fusion Studio. To use Primatte effectively, you must
understand how it works. Using a series of selection buttons, Primatte assigns RGB pixels into one of
the four specific zones.
– Zone 1: Complete background image.
– Zone 2: Foreground image with spill suppression and transparency.
– Zone 3: Foreground image with spill suppression only.
– Zone 4: Complete foreground image.

Depending on the type of blue- or green-screen content, you may find that the Delta Keyer or the
Primatte keyer handles the specific keying task better. There is no one-solution-fits-all when it comes
to keying, and in some cases, the combination of the two keyers may prove to be the best solution.

NOTE: Primatte is distributed and licensed by IMAGICA Corp. of America, Los Angeles, CA,
USA. Primatte was developed by and is a trademark of IMAGICA Corp., Tokyo, Japan.

Inputs
The Primatte node includes six inputs in the Node Editor. Unlike every other tool in Fusion, the primary
orange input is labeled as the Foreground input, since it accepts the green-screen or blue-screen
image. The background input on the Primatte node is the green input; this is an optional input that
allows Primatte to create the final merged composite.

Chapter 47 Matte Nodes 1133

 – Foreground Input: The orange input accepts a 2D image that contains blue or green screen.
– Background Input: The green (optional) input accepts a 2D image layered as the background
in the composite. If no image is connected, Primatte outputs the keyed foreground. Connecting
an image to the background input activates Primatte’s advanced edge blending options.
– Replacement Image: The magenta (optional) input accepts a 2D image used as a source of
Primatte’s spill suppression color correction.
– Garbage Matte: The gray garbage matte input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input causes
areas of the image that fall within the matte to be made transparent. The garbage matte is
applied directly to the alpha channel of the image.
– Solid Matte: The white solid matte input accepts a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input causes areas
of the image that fall within the matte to be fully opaque.
– Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input limits
the pixels where the keying occurs. An effects mask is applied to the tool after the
tool is processed.

NOTE: Connecting the background input without connecting the replacement image input
uses the background image as the replacement image surf spill suppression.

Basic Node Setup

A single Primatte keyer can rarely get perfect results because most green- or blue-screen shots have
problems the keyer is not made to handle. Keyers often need the help of garbage mattes or solid
mattes created with a Polygon or B-Spline node. Shots can also require more than just one keyer to
achieve perfect results. Below, the Primatte node has the blue-screen content connected to the
orange input. Unlike other Fusion nodes, the foreground gets connected to the orange input. The
result is an image with alpha that can then be connected into the foreground of a Merge node.

A Primatte node combined with a polygon matte as a garbage

matte and connected to the foreground of a Merge

Chapter 47 Matte Nodes 1134

Primatte Tab View Mode
At the top of the Inspector is the View Mode menu. The default selection shows the final Composite
result. You can change the view to see various intermediate stages of the keying process.

The Primatte tab View mode

– Black: Displays the foreground subject on a black or transparent background.

– Composite: The final keyed image with spill suppression, composited over the image connected
to the green Background Input on the node.
– Defocus Foreground: Displays the output of the Pre Matte key.
– Processed Foreground: Displays the alpha of the key before being combined with solid and
garbage masks. When displaying the matte, set the viewer to show the alpha channel.
– Hybrid Matte: Displays the matte generated when the Hybrid Rendering checkbox is enabled.
Best viewed when adjusting the Hybrid Blur and Hybrid Erode sliders.
– Lighting Foreground: Displays the foreground subject over the optimized artificial backing screen
that the Adjust Lighting mode creates.
– Lighting Background: Displays the optimized artificial backing screen that the Adjust Lighting
mode creates.

The Primatte tab Inspector

Chapter 47 Matte Nodes 1135

Primatte Tab
The core functionality for Primatte is found in the Primatte tab. The basic workflow is based on
selecting one of the operational mode buttons and then scrubbing over areas in the viewer.

Lock Color Picking

Activate this button once you finished adjusting your key to prevent making accidental changes in
the viewer.

Auto Compute
The Auto Compute button is likely the first button pressed when starting to key your footage. Primatte
automagically analyzes the original foreground image, determines the backing color, and sets it as the
central backing color. Then, using that information, another analysis determines the foreground areas.
A Clean FG Noise operation is performed using the newly determined foreground areas, and Primatte
renders the composite.

NOTE: The Auto Compute button may make the next three buttons—Select Background
Color, Clean Background Noise, and Clean Foreground Noise—unnecessary and make your
keying operation much more straightforward. Clicking Auto Compute automatically senses
the backing screen color, eliminates it, and even gets rid of some foreground and background
noise. If you get good results, then jump ahead to the Spill Removal tools. If you don’t get
satisfactory results, continue from this point using the three buttons described below.

Select Background Color

Clicking the Select Background Color button allows you to select the screen color by scrubbing in the
viewer. It uses the traditional Primatte method of taking the sampled backing screen color, projecting a
line in the opposite direction on the hue wheel, and generating artificial pixels that may represent the
FG object. Then, using the artificially generated foreground pixels, it internally does the Clean FG
Noise operation and creates the shape of the middle and outer polyhedrons. It then renders the
composite using these generated polyhedrons. This does not automatically use the Adjust Lighting
functionality, as it must be selected in a separate operation.

Clean Background Noise

Clicking this button helps to remove any white regions in the dark screen area (“noise”), or shades of
the screen color that did not get picked up on the first sample. Once you click the button, scrub the
mouse pointer over areas in the viewer to sample white-ish noise regions.

Clean Foreground Noise

If there are dark transparent regions in the middle of the mostly white opaque foreground object, click
the Clean Foreground Noise button and scrub over the dark pixels in the foreground area until that
area is as white as possible.

Spill Sponge
The Spill Sponge is the quickest method for removing color spill on your subject. Click the Spill
Sponge button and scrub the mouse pointer over a screen color pixel, and the screen color
disappears from the selected color region and is replaced by a complementary color, a selected color,
or a color from a replacement image. These options are set in the Replace tab. Additionally, use the
tools under the Fine Tuning tab or use the Spill(+) and Split(-) features to adjust the spill.

Chapter 47 Matte Nodes 1136

Matte Sponge
Sometimes in the Primatte operation, a 100% opaque, foreground area (all white) becomes slightly
transparent (gray). To clean those transparent areas, click the Matte Sponge button and scrub over the
transparent pixels. All the spill-suppression information remains intact.

Restore Detail
Clicking Restore Detail and scrubbing over background regions in the viewer turns completely
transparent areas translucent. This operation is useful for restoring lost hair details, thin wisps of
smoke, and the like.

Make Foreground Transparent

When this button is selected, the opaque foreground color region sampled in the viewer becomes
slightly translucent. This operation is useful for the subtle tuning of foreground subjects, which are
otherwise 100 percent covered with smoke or clouds. It can be used only one time on a particular
color. For a more flexible way to thin out a color region, and to be able to take multiple samples, use
the Matte(-) tool.

Spill(+)
Clicking the Spill(+) button returns the color spill to the sampled pixel color (and all colors like it) in the
amount of one Primatte increment. This tool can be used to move the sampled color more in the
direction of the color in the original foreground image. It can be used to nullify a Spill(-) step.

Spill(-)
Clicking the Spill(-) button removes from the sampled pixel color (and all colors like it) in the amount of
one Primatte increment. If spill color remains, another click using this operational mode tool removes
more of the color spill. Continue using this tool until all color spill has been removed from the sampled
color region.

Matte(+)
Clicking the Matte(+) button makes the matte more opaque for the sampled pixel color (and all colors
like it) in the amount of one Primatte increment. If the matte is still too translucent or thin, another click
using this operational mode tool makes the sampled color region even more opaque. This can be
used to thicken smoke or make a shadow darker to match shadows in the background imagery. It can
only make these adjustments to the density of the color region on the original foreground image. It can
be used to nullify a Matte(-) step.

Matte(-)
Clicking the Matte(+) button makes the matte more translucent for the sampled pixel color (and all
colors like it) in the amount of one Primatte increment. If the matte is still too opaque, another click
using this operational mode tool makes the sampled color region even more translucent. This can be
used to thin out smoke or make a shadow thinner to match shadows in the background imagery.

Detail(+)
When this button is selected, the foreground detail becomes less visible for the sampled pixel color
(and all colors like it) in the amount of one Primatte increment. If there is still too much detail, another
click using this operational mode tool makes more of it disappear. This can be used to remove smoke
or wisps of hair from the composite. Sample where detail is visible, and it disappears. This is for
moving color regions into the 100% background region. It can be used to nullify a Detail(-) step.

Chapter 47 Matte Nodes 1137

Detail(-)
When this button is selected, foreground detail becomes more visible for the sampled pixel color (and
all colors like it) in the amount of one Primatte increment. If detail is still missing, another click using this
operational mode tool makes detail more visible. This can be used to restore lost smoke or wisps of
hair. Sample where the smoke or hair just disappears and it returns to visibility. Use this for restoring
color regions that were moved into the 100% background region. It may start to bring in background
noise if shooting conditions were not ideal on the foreground image.

Algorithms
There are three keying algorithms available in the Primatte keyer:
– Primatte: The Primatte algorithm mode delivers the best results and supports both the Solid Color
and the Complement Color spill suppression methods. This algorithm uses three multifaceted
polyhedrons (as described later in this section) to separate the 3D RGB colorspace. It is also the
default algorithm mode and, because it is computationally intensive, it may take the longest to
render.
– Primatte RT: Primatte RT is the simplest algorithm and therefore the fastest. It uses only a single
planar surface to separate the 3D RGB colorspace (as described later in this section) and, as
a result, does not separate the foreground from the backing screen as carefully as the above
Primatte algorithm. Another disadvantage of the Primatte RT algorithm is that it does not work well
with less saturated backing screen colors, and it does not support the Complement Color spill
suppression method.
– Primatte RT+: Primatte RT+ is in between the above two options. It uses a six planar surface color
separation algorithm (as described later in this section) and delivers results in between the other
two options in both quality and performance. Another disadvantage of the Primatte RT+ algorithm
is that it does not work well with less saturated backing screen colors, and it does not support the
Complement Color spill suppression method.

Hybrid Rendering
After sampling the backing screen color and producing acceptable edges around the foreground
object, you sometimes find a transparent area within the foreground subject. This can occur when the
foreground subject contains a color that is close to the backing screen color. Removing this
transparency with the Clean FG Noise mode can cause the edge of the foreground subject to pick up
a fringe that is close to the backing screen color. Removing the fringe is very difficult without
sacrificing quality somewhere else on the image. The Hybrid Render mode internally creates two
keying operations: Body and Edge. The optimized Edge operation gets the best edge around the
foreground subject without any fringe effect. The Body operation deals with transparency within the
foreground subject. The resultant matte is created by combining these two mattes, and then blurring
and eroding the foreground subject in the Body matte and combining it with the edge matte.
To use Hybrid Rendering, start by keying the main foreground area using the Select Background Color
mode (or any of the other Primatte backing screen detection methods). Activate the Hybrid Rendering
checkbox. Lastly, select the Clean FG Noise button and scrub over the transparent area. The Hybrid
Render mode performs the “Body/Edge” operation. The result is a final composite with perfect edges
around the foreground subject with a solid foreground subject.

Hybrid Blur
Blurs the Body matte that has been automatically generated when Hybrid Rendering is activated.

Hybrid Erode
This slider dilates or erodes the Hybrid matte. You can view the results by selecting Hybrid matte in
the View Mode menu.

Chapter 47 Matte Nodes 1138

Adjust Lighting
Before applying the Adjust Lighting operation, it is necessary to determine the backing screen color
using Auto Compute or Select Background Color. After performing one of those operations, click on
the Adjust Lighting button. Primatte generates an artificial clean plate and uses it to generate an
evenly lit backing screen behind the foreground object. The default setting should detect all the areas
that contain foreground pixels and deliver a smooth backing screen for the keying.

Lighting Threshold
Should Adjust Lighting fail to produce a smoother backing screen, adjust the Lighting Threshold slider
while viewing the Lighting Background setting in the View Mode menu. This displays the optimized
artificial backing screen that the Adjust Lighting mode creates.

Crop
This button reveals the Crop sliders to create a rectangular garbage matte with the Primatte node. As
opposed to Fusion’s Crop tool, this does not change the actual image size.

Reset
Resets all the Primatte key control data back to a blue- or green-screen.

Soft Reset
Resets just the Primatte parameters used since the Select Background Color operation was last
completed.

The Primatte Fine Tuning tab

Fine Tuning Tab

The Fine Tuning tab can make refined adjustments to the spill suppression, density of the matte, and
semitransparent areas. These sliders provide a bit more granularity over the Spill(+)(-), Matte(+)(-) and
Detail(+)(-) buttons in the Primatte tab.

Selected Color
This shows the color selected (or registered) by the scrubbing in the viewer while the Fine Tuning tab
is selected.

Chapter 47 Matte Nodes 1139

Fine Tuning Sliders
The color of the scrubbed pixel is registered as a reference color for fine tuning. It is displayed in the
Color swatch. To perform the tuning operation, sample a color region on the image, and adjust one of
the Fine Tuning sliders to achieve the desired effect.

Spill
The Spill slider can be used to remove spill from the selected color region. The more to the right the
slider moves, the more spill is removed. The more to the left the slider moves, the closer the color
component of the selected region is to the color in the original foreground image. If moving the slider
to the right does not remove the spill, resample the color region and move the slider again.
These slider operations are additive. The result achieved by moving the slider to the right can also be
achieved by clicking on the color region using the Spill(-) operational mode.

Transparency
The Transparency slider makes the matte more translucent in the selected color region. Moving this
slider to the right makes the selected color region more transparent. Moving the slider to the left
makes the matte more opaque. If moving the slider to the right does not make the color region
translucent enough, resample the color region and again move the slider to the right. These slider
operations are additive. The result achieved by moving the slider to the right can also be achieved by
clicking on the color region using the Matte(-) operational mode.

Detail
The Detail slider can be used to restore lost detail. After selecting a color region, moving this slider to
the left makes the selected color region more visible. Moving the slider to the right makes the color
region less visible. If moving the slider to the left does not make the color region visible enough,
resample the color region and again move the slider to the left.
These slider operations are additive. This result achieved by moving the slider to the left can also be
achieved by clicking on the color region using the Detail(-) operational mode.

The Primatte Replace tab

Replace Tab
The Replace tab allows you to choose between the three methods of color spill replacement as
covered in detail in the Spill Sponge section above. There are three options for the replacement color
when removing the spill. These options are selected from the Replace mode menu.

Replace Mode
– Complement: Replaces the spill color with the complement of the screen color. This mode
maintains fine foreground detail and delivers the best-quality results. If foreground spill is not a
significant problem, this mode is the one that should be used. However, If the spill intensity on
the foreground image is rather significant, this mode may often introduce serious noise in the
resultant composite.
– Image: Replaces the spill color with colors from a defocused version of the background image or
the Replace image, if one is connected to the Replace input (magenta ) on the node. This mode
results in a good color tone on the foreground subject even with a high-contrast background.
On the negative side, the Image mode occasionally loses the fine edge detail of the foreground

Chapter 47 Matte Nodes 1140

 subjects. Another problem can occur if you later change the size of the foreground image against
the background. Since the background/foreground alignment would change, the applied color
tone from the defocused image might not match the new alignment.
– Color: Replaces the spill color with a solid color. When this option is selected, a color swatch
and R,G,B sliders are displayed for selecting the color. Changing the palette color for the solid
replacement, you can select a good spill replacement that matches the composite background. Its
strength is that it works fine with even severe spill conditions. On the negative side, when using
the Solid Color Replacement mode, fine detail on the foreground edge tends to be lost. The single
palette color sometimes cannot make a good color tone if the background image has some high-
contrast color areas.

The Primatte Degrain tab

Degrain Tab
The Degrain tab is used when a foreground image is highly compromised by film grain. As a result of
the grain, when backing screen noise is completely removed, the edges of the foreground object
often become harsh and jagged, leading to a poor key.

Grain Size
The Grain Size selector provides a range of grain removal from Small to Large. If the foreground image
has a large amount of film grain-induced pixel noise, you may lose a good edge to the foreground
object when trying to clean all the grain noise with the Clean Background Noise Operation Mode.
These tools clean up the grain noise without affecting the quality of the key.
– None: No degraining is performed.
– Small: The average color of a small region of the area around the sampled pixel. This should be
used when the grain is very dense.
– Medium: The average color of a medium-sized region of the area around the sampled pixel. This
should be used when the grain is less dense.
– Large: The average color of a larger region of the area around the sampled pixel. This should be
used when the grain is very loose.

Grain Tolerance
Adjusting this slider increases the effect of the Clean Background Noise tool without changing the
edge of the foreground object.

Chapter 47 Matte Nodes 1141

The Primatte Matte tab

Matte Tab
The Matte tab refines the alpha of the key, combined with any solid and garbage masks connected to
the node. When using the Matte tab, set the viewer to display the alpha channel of Primatte’s
final output.

Filter
This control selects the filtering algorithm used when applying blur to the matte.
– Box: This is the fastest method but at reduced quality. Box is best suited for minimal amounts of
blur.
– Bartlett: Otherwise known as a Pyramid filter, Bartlett makes a good compromise between speed
and quality.
– Multi-Box: When selecting this filter, the Num Passes slider appears and lets you control the
quality. At 1 and 2 passes, results are identical to Box and Bartlett, respectively. At 4 passes and
above, results are usually as good as Gaussian, in less time and with no edge “ringing.”
– Gaussian: The Gaussian filter uses a true Gaussian approximation and gives excellent results, but
it is a little slower than the other filters. In some cases, it can produce an extremely slight edge
“ringing” on floating-point pixels.

Blur
Matte Blur blurs the edge of the matte based on the Filter menu setting. A value of zero results in a
sharp, cutout-like hard edge. The higher the value, the more blur applied to the matte.

Blur Inward
Activating the Blur Inward checkbox generates the blur toward the center of the foreground subject.
Conventional blurring or defocus affects the matte edges in both directions (inward and outward) and
sometimes introduces a halo artifact around the edge in the composite view. Blur Inward functions
only in the inward direction of the foreground subject (toward the center of the white area). The final
result removes small and dark noise in the screen area without picking them up again in the Clean
Background Noise mode. It can sometimes result in softer, cleaner edges on the foreground objects.

Contract/Expand
This slider shrinks or grows the semitransparent areas of the matte. Values above 0.0 expand the
matte, while values below 0.0 contract it.

Chapter 47 Matte Nodes 1142

This control is usually used in conjunction with the Matte Blur to take the hard edge of a matte and
reduce fringing. Since this control affects only semitransparent areas, it will have no effect on a matte’s
hard edge.

Gamma
Matte Gamma raises or lowers the values of the matte in the semitransparent areas. Higher values
cause the gray areas to become more opaque, and lower values cause the gray areas to become
more transparent. Completely black or white regions of the matte remain unaffected.
Since this control affects only semitransparent areas, it will have no effect on a matte’s hard edge.

Threshold
This range slider sets the lower threshold using the handle on the left and sets the upper threshold
using the handle on the right.
Any value below the lower threshold setting becomes black or transparent in the matte.
Any value above the upper threshold setting becomes white or opaque in the matte. All values within
the range maintain their relative transparency values.
This control is often used to reject salt and pepper noise in the matte.

Restore Fringe
This restores the edge of the matte around the keyed subject. Often when keying, the edge of the
subject where you have hair is clipped out. Restore Fringe brings back that edge while keeping the
matte solid.

Invert Matte
When this checkbox is selected, the alpha channel created by the keyer is inverted, causing all
transparent areas to be opaque and all opaque areas to be transparent.

Solid Matte
Solid mattes are mask nodes or images connected to the solid matte input on the node. The solid
matte is applied directly to the alpha channel of the image. Generally, solid mattes are used to hold out
keying in areas you want to remain opaque, such as someone with blue eyes against a blue screen.
Enabling Invert inverts the solid matte before it is combined with the source alpha.

Garbage Matte
Garbage mattes are mask nodes or images connected to the garbage matte input on the node. The
garbage matte is applied directly to the alpha channel of the image. Generally, garbage mattes are
used to remove unwanted elements that cannot be keyed, such as microphones and booms. They are
also used to fill in areas that contain the color being keyed but that you wish to maintain.
Garbage mattes of different modes cannot be mixed within a single tool. A Matte Control node is often
used after a Keyer node to add a garbage matte with the opposite effect of the matte applied to
the keyer.
Enabling Invert inverts the garbage matte before it is combined with the source alpha.

Post-Multiply Image
Select this option to cause the keyer to multiply the color channels of the image against the alpha
channel it creates for the image. This option is usually enabled and is on by default.
Deselect this checkbox and the image can no longer be considered premultiplied for purposes of
merging it with other images. Use the Subtractive option of the Merge node instead of the
Additive option.
For more information on these Merge node settings, see Chapter 35, “Composite Nodes” in the Fusion
Reference Manual or Chapter 96 in the DaVinci Resolve Reference Manual.

Chapter 47 Matte Nodes 1143

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Matte nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

How to Key with Primatte

You begin keying with Primatte by connecting the blue- or green-screen shot to the orange
foreground input on the Primatte node and the background shot for the composite into the green
background input. Once the connections are made, there are four main steps to using the Primatte:
1 Select Background Color.
2 Clean the Background Noise.
3 Clean the Foreground Noise.
4 Remove Spill.

Selecting Background Color

– In the Inspector, click the Select Background Color button.
– Position the mouse pointer over the blue/green-screen area in the viewer, somewhere near the
foreground subject.
– Drag over the background color.
Primatte averages the pixels to get a single color. Sometimes Primatte works best when only a
single pixel is sampled instead of a range of pixels.
Should you have difficulties with your keying, try the Select Background Color operation again
with a single dark screen pixel or single light screen pixel.
Instead, If you want to make a rectangular selection, use the Box button in the top left-hand corner
of the viewer. The Median button is the same as Line selection, except that each point sampled is
the result of a 3 x 3 region based on where you click and then apply a median filter. This can
potentially reduce any noisy pixels.

Primatte viewer buttons

If the foreground image has a shadow in it that you want to keep in the composite, do not select
any of the dark screen pixels in the shadow. This keeps the shadow with the rest of the
foreground image.

Clean Background Noise

If there are any white or light gray regions in the dark screen area, this is referred to as “noise.”
Technically, it is varying shades of the screen color that did not get picked up on the first sample and
should be removed. You remove background noise using the Clean Background Noise button.
1 From the View Mode menu in the Inspector, select Black.
2 Above the viewer, click the Alpha Channel/RGB button.
The image displayed changes to a black and white “matte” view of the image.

Chapter 47 Matte Nodes 1144

3 Click the Clean Background Noise button.
4 Drag the mouse pointer through these white or light gray regions that should be pure black.
Primatte processes the selection and eliminates the noise.
5 Repeat this procedure as often as necessary to clear the noise from the background areas.
Selecting Gain/Gamma from the viewer’s Options menu to increase the brightness or gamma
allows you to see noise that would otherwise be invisible.

Primatte viewer Options menu

You do not need to remove every single white pixel to get good results. Most pixels displayed as a
dark color close to black in a key image are considered transparent and virtually allow the
background to be the final output in that area. Consequently, there is no need to eliminate all
noise in the screen portions of the image. In particular, if an attempt is made to remove noise
around the foreground subject meticulously, a smooth composite image is often difficult
to generate.

TIP: When clearing noise from around loose, flying hair or any background/foreground
transitional area, be careful not to select any of the areas near the edge of the hair. Leave
a little noise around the hair as this can be cleaned up later using the Fine Tuning tools.

Clean Foreground Noise

If there are dark regions in the middle of the mostly white foreground subject, the key is not 100% in
those areas. Using Clean Foreground Noise can make the matte as white as possible.
1 Keep the View Mode menu set to Black and the viewer set to the Alpha Channel.
2 Click the Clean Foreground Noise button.
3 Drag the mouse pointer through these dark pixels in the foreground that should be pure white.
Primatte processes the selection and eliminates the noise.
4 Repeat this procedure as often as necessary to clear the noise from the foreground areas.
5 If enabled, disable Gain/Gamma from the viewer’s Options menu to return to a regular viewer.

Removing Spill
The first three sections created a clean matte. At this point, the foreground can be composited onto
any background image. However, if there is color spill on the foreground subject, a final operation is
necessary to remove that screen spill for a more natural-looking composite.
1 From the View Mode menu, select Composite.
2 Above the viewer, click the Alpha/RGB toggle button to see the RGB image.

Chapter 47 Matte Nodes 1145

 There are two ways in Primatte to remove the spill color:
– Spill Sponge
The quickest method is to select the Spill Sponge button and then sample the spill areas away.
Additional spill removal can be done using the tools under the Fine Tuning tab or by using the
Spill(-) button.
– Fine Tuning Tab
To use the Fine Tuning tab for spill, first scrub over the spill color in the viewer. For most images,
adjusting the Spill slider is all that is required to remove any remaining spill.

NOTE: When using the slider in the Fine Tuning tab to remove spill, spill color
replacement is replaced based on the setting of the Spill Replacement options.

You can use the other two sliders in the same way for different key adjustments. The Detail slider
controls the matte softness for the color that is closest to the background color. For example, you
can recover lost rarefied smoke in the foreground by selecting the Fine Tuning mode, clicking on
the area of the image where the smoke starts to disappear and moving the Detail slider to the left.
The Transparency slider controls the matte softness for the color that is closest to the foreground
color. For example, if you have thick and opaque smoke in the foreground, you can make it
semitransparent by moving the Transparency slider to the right after selecting the pixels in the
Fine Tuning mode.

Ultra Keyer [UKY]

The Ultra Keyer node

Ultra Keyer Node Introduction

Like the newer Delta Keyer, the Ultra Keyer node has two keyers built in to it: a pre-matte keyer acts as
a garbage matte creator and the color difference keyer that extracts fine detail and transparency.
Generally, you start with the Delta Keyer as your first keyer of choice. If you do not get good results, try
Primatte if you are using Fusion Studio. A good third choice is to try the Ultra Keyer.

Inputs
The Ultra Keyer node includes four inputs in the Node Editor.
– Input: The orange input accepts a 2D image that contains the color you want to be keyed for
transparency.
– Garbage Matte: The gray garbage matte input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input causes
areas of the image that fall within the matte to be made transparent. The garbage matte is
applied directly to the alpha channel of the image.
– Solid Matte: The white solid matte input accepts a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input causes areas
of the image that fall within the matte to be fully opaque.

Chapter 47 Matte Nodes 1146

 – Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps masks. Connecting a mask to this input limits
the pixels where the keying occurs. An effects mask is applied to the tool after the
tool is processed.

Basic Node Setup

A single keyer can rarely get perfect results because most green- or blue-screen shots have problems
the keyer is not made to handle. Keyers often need the help of garbage mattes or solid mattes created
with a Polygon or B-Spline node. Shots can also require more than just one keyer to achieve perfect
results. Below, the Ultra Keyer node has the blue-screen content connected to the orange input. The
result is an image with alpha that can then be connected into the foreground of a Merge node.

An Ultra Keyer node combined with polygon matte as a

garbage matte and connected to the foreground of a Merge

Inspector

The Ultra Keyer Pre-Matte tab

Chapter 47 Matte Nodes 1147

Pre-Matte Tab
The Pre-Matte tab is where most keying begins. It is used to select the screen color and smooth out
the color of the screen.

Background Color
The Background Color is used to select the color of the blue or green screen of the images. It is good
practice to select the screen color close to the subject to be separated from the screen background.

Red Level, Green Level, Blue Level

These color sliders tune the level of the difference channels to help separate the color. When the
background color is green, Red and Blue level options are provided. When the background color is
blue, Red and Green level options are provided.

Background Correction
Depending on the background color selected above, the keyer iteratively merges the pre-keyed
image over either a blue or green background before processing it further.
In some instances, this leads to better, more subtle edges.

Matte Separation
Matte Separation performs a pre-process on the image to help separate the foreground from the
background before color selection. Generally, increase this control while viewing the alpha to eliminate
the bulk of the background, but stop just before it starts cutting holes in the subject or eroding fine
detail on the edges of the matte.

Pre-Matte Range
These R,G,B, and Luminance range controls update automatically to represent the current color
selection. Colors are selected by selecting the Ultra Keyer node’s tile in the node tree and dragging
the Eyedropper into the viewer to select the colors to be used to create the matte. These range
controls can be used to tweak the selection slightly, although selecting colors in the viewer is all that
is required.

Lock Color Picking

This checkbox prevents accidentally selecting more colors from the view. It is a good idea to activate
this checkbox once the color selection is made for the matte. All other controls in the node
remain editable.

Pre Matte Size

The Pre Matte Size control can be used to soften the general area around the keyed image. This is
used to close holes in the matte often caused by spill in semitransparent areas of the subject. This can
cause a small halo around the subject, which can be removed using the Matte Contract tools found
later in the tool.

Reset Pre Matte Ranges

This discards all color selection by resetting the ranges but maintains all other slider and
control values.

Chapter 47 Matte Nodes 1148

The Ultra Keyer Image tab

Image Tab
The Image tab handles the majority of spill suppression in the Ultra Keyer. Spill suppression is a form of
color correction that attempts to remove the screen color from the fringe of the matte.
Spill is the transmission of the screen color through the semitransparent areas of the alpha channel. In
the case of blue- or green-screen keying, this usually causes the color of the background to become
apparent in the edges of the foreground subject.

Spill Suppression
When this slider is set to 0, no spill suppression is applied to the image. Increasing the slider increases
the strength of the spill method.

Spill Method
This selects the strength of the algorithm used to apply spill suppression to the image.
– None: None is selected when no spill suppression is required.
– Rare: This removes very little of the spill color and is the lightest of all methods.
– Medium: This works best for green screens.
– Well Done: This works best for blue screens.
– Burnt: This works best for blue screens. Use this mode only for very troublesome shots.

Fringe Gamma
This control can be used to adjust the brightness of the fringe or halo that surrounds the keyed image.

Fringe Size
This expands and contracts the size of the fringe or halo surrounding the keyed image.

Fringe Shape
Fringe Shape presses the fringe toward the external edge of the image or pulls it toward the inner
edge of the fringe. Its effect is most noticeable while the Fringe Size value is large.

Cyan/Red, Magenta/Green, and Yellow/Blue

Use these three controls to color correct the fringe of the image.
This is useful for correcting semitransparent pixels that still contain color from the original background
to match the new background.

Chapter 47 Matte Nodes 1149

The Ultra Keyer Matte tab

Matte Tab
The Matte tab refines the alpha of the key, combined with any solid and garbage masks connected to
the node. When using the Matte tab, set the viewer to display the alpha channel of the Delta Keyer’s
final output.

Filter
This control selects the filtering algorithm used when applying blur to the matte.
– Box: This is the fastest method but at reduced quality. Box is best suited for minimal
amounts of blur.
– Bartlett: Otherwise known as a Pyramid filter, Bartlett makes a good compromise between speed
and quality.
– Multi-Box: When selecting this filter, the Num Passes slider appears and lets you control the
quality. At 1 and 2 passes, results are identical to Box and Bartlett, respectively. At 4 passes and
above, results are usually as good as Gaussian, in less time and with no edge “ringing.”
– Gaussian: The Gaussian filter uses a true Gaussian approximation and gives excellent results, but
it is a little slower than the other filters. In some cases, it can produce an extremely slight edge
“ringing” on floating-point pixels.

Blur
Matte Blur blurs the edge of the matte based on the Filter menu setting. A value of zero results in a
sharp, cutout-like hard edge. The higher the value, the more blur applied to the matte.

Clipping Mode
This option determines how edges are handled when performing domain of definition rendering. This
is profoundly important when blurring the matte, which may require samples from portions of the
image outside the current domain.
– Frame: The default option is Frame, which automatically sets the node’s domain of definition
to use the full frame of the image, effectively ignoring the current domain of definition. If
the upstream DoD is smaller than the frame, the remaining areas in the frame are treated
as black/transparent.

Chapter 47 Matte Nodes 1150

 – Domain: Setting this option to Domain respects the upstream domain of definition when applying
the node’s effect. This can have adverse clipping effects in situations where the node employs a
large filter.
– None: Setting this option to None does not perform any source image clipping at all. This means
that any data required to process the node’s effect that would normally be outside the upstream
DoD is treated as black/transparent.

Contract/Expand
This slider shrinks or grows the semitransparent areas of the matte. Values above 0.0 expand the
matte, while values below 0.0 contract it.
This control is usually used in conjunction with the Matte Blur to take the hard edge of a matte and
reduce fringing. Since this control affects only semitransparent areas, it has no effect on a matte’s
hard edge.

Gamma
Matte Gamma raises or lowers the values of the matte in the semitransparent areas. Higher values
cause the gray areas to become more opaque, and lower values cause the gray areas to become
more transparent. Completely black or white regions of the matte remain unaffected.
Since this control affects only semitransparent areas, it will have no effect on a matte’s hard edge.

Threshold
This range slider sets the lower threshold using the handle on the left and sets the upper threshold
using the handle on the right.
Any value below the lower threshold setting becomes black or transparent in the matte.
Any value above the upper threshold setting becomes white or opaque in the matte. All values within
the range maintain their relative transparency values.
This control is often used to reject salt and pepper noise in the matte.

Restore Fringe
This restores the edge of the matte around the keyed subject. Often when keying, the edge of the
subject where you have hair is clipped out. Restore Fringe brings back that edge while keeping the
matte solid.

Invert Matte
When this checkbox is selected, the alpha channel created by the keyer is inverted, causing all
transparent areas to be opaque and all opaque areas to be transparent.

Solid Matte
Solid mattes are mask nodes or images connected to the solid matte input on the node. The solid
matte is applied directly to the alpha channel of the image. Generally, solid mattes are used to hold out
keying in areas you want to remain opaque, such as someone with blue eyes against a blue screen.
Enabling Invert inverts the solid matte before it is combined with the source alpha.

Garbage Matte
Garbage mattes are mask nodes or images connected to the garbage matte input on the node. The
garbage matte is applied directly to the alpha channel of the image. Generally, garbage mattes are
used to remove unwanted elements that cannot be keyed, such as microphones and booms. They are
also used to fill in areas that contain the color being keyed but that you wish to maintain.
Garbage mattes of different modes cannot be mixed within a single tool. A Matte Control node is often
used after a Keyer node to add a garbage matte with the opposite effect of the matte applied to
the keyer.
Enabling Invert inverts the garbage matte before it is combined with the source alpha.

Chapter 47 Matte Nodes 1151

Post-Multiply Image
Select this option to cause the keyer to multiply the color channels of the image against the alpha
channel it creates for the image. This option is usually enabled and is on by default.
Deselect this checkbox and the image can no longer be considered premultiplied for purposes of
merging it with other images. Use the Subtractive option of the Merge node instead of the
Additive option.
For more information on these Merge node settings, see Chapter 35, “Composite Nodes” in the Fusion
Reference Manual or Chapter 96 in the DaVinci Resolve Reference Manual.

Subtract Background
This option color corrects the edges when the screen color is removed and anti-aliased to a black
background. By enabling this option, the edges potentially become darker. Disabling this option allows
you to pass on the color of the screen to use in other processes down the line.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other matte nodes. These common controls are
described in detail in the following “The Common Controls” section.

The Common Controls

Nodes that handle matte operations share a number of identical controls in the Inspector. This section
describes controls that are common among matte nodes.

Inspector

Common Matte settings Inspector

Chapter 47 Matte Nodes 1152

Settings Tab
The Settings tab in the Inspector can be found on every tool in the Matte category. The controls are
consistent and work the same way for each tool.

Blend
The Blend control is used to blend between the tool’s original image input and the tool’s final modified
output image. When the blend value is 0.0, the outgoing image is identical to the incoming image.
Usually, this causes the tool to skip processing entirely, copying the input straight to the output.

Process When Blend Is 0.0

The tool is processed even when the input value is zero. This can be useful if this node is scripted to
trigger another task, but the value is set to 0.0.

Red/Green/Blue/Alpha Channel Selector

These four buttons are used to limit the effect of the tool to specified color channels. This filter is often
applied after the tool has been processed.
For example, if the Red button on a Blur tool is deselected, the blur is first applied to the image, and
then the red channel from the original input is copied back over the red channel of the result.
There are some exceptions, such as tools where deselecting these channels causes the tool to skip
processing that channel entirely. In that case, there are a set of RGBA buttons on the Controls tab in
the tool. The buttons in the Settings and the Controls tabs are identical.

Apply Mask Inverted

Enabling the Apply Mask Inverted option inverts the complete mask channel for the tool. The mask
channel is the combined result of all masks connected to or generated in a node.

Multiply by Mask
Selecting this option causes the RGB values of the masked image to be multiplied by the mask
channel’s values. This causes all pixels of the image not included in the mask (i.e., set to 0) to become
black/transparent.

Use Object/Use Material (Checkboxes)

Some 3D software can render to file formats that support additional channels. Notably, the EXR file
format supports Object ID and Material ID channels, which can be used as a mask for the effect. These
checkboxes determine whether the channels are used, if present. The specific Material ID or Object ID
affected is chosen using the next set of controls.

Correct Edges
This checkbox appears only when the Use Object or Use Material checkboxes are selected. It toggles
the method used to deal with overlapping edges of objects in a multi-object image. When enabled,
the Coverage and Background Color channels are used to separate and improve the effect around the
edge of the object. If this option disabled (or no Coverage or Background Color channels are
available), aliasing may occur on the edge of the mask.
For more information on Coverage and Background Color channels, see Chapter 18, “Understanding
Image Channels” in the Fusion Reference Manual or Chapter 79 in the DaVinci Resolve
Reference Manual.

Object ID/Material ID (Sliders)

Use these sliders to select which ID is used to create a mask from the object or material channels of an
image. Use the Sample button in the same way as the Color Picker: to grab IDs from the image
displayed in the view. The image or sequence must have been rendered from a 3D software package
with those channels included.

Chapter 47 Matte Nodes 1153

Clipping Mode
This option determines how edges are handled when performing domain of definition rendering. This
is profoundly important when blurring the matte, which may require samples from portions of the
image outside the current domain.
– Frame: The default option is Frame, which automatically sets the node’s domain of definition
to use the full frame of the image, effectively ignoring the current domain of definition. If the
upstream DoD is smaller than the frame, the remaining area in the frame is treated as black/
transparent.
– Domain: Setting this option to Domain respects the upstream domain of definition when applying
the node’s effect. This can have adverse clipping effects in situations where the node employs a
large filter.
– None: Setting this option to None does not perform any source image clipping at all. This means
that any data required to process the node’s effect that would normally be outside the upstream
DoD is treated as black/transparent.

Use GPU
The Use GPU menu has three settings. Setting the menu to Disable turns off GPU hardware-
accelerated rendering. Enabled uses the GPU hardware for rendering the node. Auto uses a capable
GPU if one is available, and falls back to software rendering when a capable GPU is not available

Hide Incoming Connections

Enabling this checkbox can hide connection lines from incoming nodes, making a node tree appear
cleaner and easier to read. When enabled, empty fields for each input on a node are displayed in the
Inspector. Dragging a connected node from the node tree into the field hides that incoming
connection line as long as the node is not selected in the node tree. When the node is selected in the
node tree, the line reappears.

Comments
The Comments field is used to add notes to a tool. Click in the empty field and type the text. When a
note is added to a tool, a small red square appears in the lower-left corner of the node when the full
tile is displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the
note in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 47 Matte Nodes 1154

Chapter 48

Metadata Nodes
This chapter details the Metadata nodes available in Fusion.

The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Copy Metadata [META]  1156
Set Metadata [SMETA]  1157
Set Timecode [TCMETA]  1158
The Common Controls  1160

Chapter 48 Metadata Nodes 1155

Copy Metadata [META]

The Copy Metadata node

Copy Metadata Node Introduction

Copy Metadata combines, replaces, or clears the metadata in your image. Metadata can be viewed in
a subview of the viewer.

Inputs
The two inputs on the Copy Metadata node are used to connect two 2D images.
– Background Input: The orange background input is used for the primary 2D image that is
output from the node.
– Foreground Input: The green foreground input is used for the secondary 2D image that
contains metadata you want merge or overwrite onto the background image.

Basic Node Setup

The Copy Metadata node takes metadata from the foreground input (green) and copies it into the
background input (orange). The output is the background input with modified metadata.

A Copy Metadata node copies metadata from the foreground and embeds it into the background clip

Inspector

The Copy Metadata Controls tab

Controls Tab
The Controls tab configures how metadata coming from the foreground input image gets added to the
background input image.

Chapter 48 Metadata Nodes 1156

Operation
The Operation menu determines how the metadata of the foreground and background inputs
are treated.
– Merge (Replace Duplicates): All values are merged, but values with duplicate names are taken
from the foreground input.
– Merge (Preserve Duplicates): All values are merged, but values with duplicate names are taken
from the background input.
– Replace:  The metadata in the foreground replaces the entire metadata in the background.
– Clear: All metadata is discarded.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Metadata nodes. These common controls
are described in detail at the end of this chapter in “The Common Controls” section.

Set Metadata [SMETA]

The Set Metadata node

Set Metadata Node Introduction

Set Metadata allows you to create new Name = Value pairs in the metadata. Metadata can be viewed
in a subview of the viewer.

Inputs
The single input on the Set Metadata node is used to connect a 2D image that gets metadata added.
– Background Input: The orange background input is used for the primary 2D image that is
output from the node with the new metadata.

Basic Node Setup

The Set Metadata node embeds new metadata into the background input (orange). The output is the
background input with new metadata.

A Set Metadata node creates new metadata and embeds it into the background clip.

Chapter 48 Metadata Nodes 1157

Inspector

The Set Metadata Controls tab

Controls Tab
The Controls tab is where you set up the name of the metadata field and the value or information
regarding the metadata.

Field Name
The name of the metadata value. Do not use spaces.

Field Value
The value assigned to the name above.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Metadata nodes. These common controls
are described in detail at the end of this chapter in “The Common Controls” section.

Set Timecode [TCMETA]

The Set Timecode node

Set Timecode Node Introduction

Set Timecode inserts dynamic timecode values into the metadata table based on the FPS settings.

Inputs
The single input on the Set Timecode node is used to connect a 2D image that gets timecode added.
– Background Input: The orange background input is used for the primary 2D image that is
output from the node with the new timecode.

Basic Node Setup

The Set Timecode node embeds new timecode metadata into the background input (orange). The
output is the background input with updated timecode.

Chapter 48 Metadata Nodes 1158

A Set Timecode node inserts new timecode metadata into the background clip.

Inspector

The Set Timecode Controls tab

Controls Tab
The Controls tab sets the clip’s starting timecode metadata based on FPS, hours, minutes, seconds,
and frames.

FPS
You can choose from a variety of settings for frames per second.
Since this is a Fuse, you can easily adapt the settings to your needs by editing the appropriate piece of
code for the buttons:
MBTNC_StretchToFit = true,
{ MBTNC_AddButton = “24“ },
{ MBTNC_AddButton = “25“ },
{ MBTNC_AddButton = “30“ },
{ MBTNC_AddButton = “48“ },
{ MBTNC_AddButton = “50“ },
{ MBTNC_AddButton = “60“ },
})

as well as for the actual values:

local rates = { 24, 25, 30, 48, 50, 60 }

Hours/Minutes/Seconds/Frames Sliders
Define an offset from the starting frame of the current comp.

Chapter 48 Metadata Nodes 1159

Print to Console
Verbose output of the Timecode/Frame value in the Console.
The Timecode/Frames conversion is done according to the FPS settings. The result might look
like this:
TimeCode: 00:00:08:15
Frames: 207

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Metadata nodes. These common controls
are described in detail in the following “The Common Controls” section.

The Common Controls

Nodes that handle metadata operations share several identical controls in the Inspector. This section
describes controls that are common among Metadata nodes.

Inspector

The Common Metadata Settings tab

Settings Tab
The Settings tab in the Inspector can be found on every tool in the Metadata category. The controls
are consistent and work the same way for each tool.

Use Object/Use Material (Checkboxes)

Some 3D software can render to file formats that support additional channels. Notably, the EXR file
format supports Object ID and material ID channels, which can be used as a mask for the effect. These
checkboxes determine whether the channels are used, if present. The specific Material ID or Object ID
affected is chosen using the next set of controls.

Chapter 48 Metadata Nodes 1160

Correct Edges
This checkbox appears only when the Use Object or Use Material checkboxes are selected. It toggles
the method used to deal with overlapping edges of objects in a multi-object image. When enabled,
the Coverage and Background Color channels are used to separate and improve the effect around the
edge of the object. If this option disabled (or no Coverage or Background Color channels are
available), aliasing may occur on the edge of the mask.
For more information on the Coverage and Background Color channels, see Chapter 18,
“Understanding Image Channels” in the Fusion Reference Manual or Chapter 79 in the
DaVinci Resolve Reference Manual.

Object ID/Material ID (Sliders)

Use these sliders to select which ID is used to create a mask from the object or material channels of an
image. Use the Sample button in the same way as the Color Picker: to grab IDs from the image
displayed in the view. The image or sequence must have been rendered from a 3D software package
with those channels included.

Comments
The Comments field is used to add notes to a tool. Click in the empty field and type the text. When a
note is added to a tool, a small red square appears in the lower-left corner of the node when the full
tile is displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the
note in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 48 Metadata Nodes 1161

Chapter 49

Miscellaneous
Nodes
This chapter details miscellaneous nodes within Fusion.

The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Auto Domain [ADoD]  1163
Change Depth [CD]  1165
Custom Tool [CT]  1166
Fields [Flds]  1176
Frame Average [Avg]  1178
Keyframe Stretcher [KfS]  1180
Run Command [Run]  1182
Set Domain [DoD]  1185
Time Speed [TSpd]  1187
Time Stretcher [TSt]  1190
Wireless Link [Wire]  1193
The Common Controls  1194

Chapter 49 Miscellaneous Nodes 1162

Auto Domain [ADoD]

The Auto Domain node

Auto Domain Node Introduction

The Auto Domain node automatically sets the image’s domain of definition (DoD) based on bounds of
the input image’s background Canvas color. It does not change the image’s physical dimensions.
Some EXR images come with optimized DoDs already set, but other formats do not. For formats other
than EXR, this node can speed up compositions by optimizing the DoD based on the content rather
than the frame’s dimensions.
For example, a CG character rarely takes up the entire frame of an image. With this type of image, the
Auto Domain node sets the DoD to a rectangular region by comparing image pixels with the Canvas
color. The Canvas color indicates what color the pixels are outside the DoD. By default, unless a
Canvas color is set using the Set Canvas Color node, the color is set to black. This default works well
when an image has a premultiplied alpha channel. The result is a DoD that encompasses the portion
of the clip that contains only the character. The DoD is updated on each frame to accommodate
changes, such as a character walking closer to the camera. However, if a clip does not contain an
alpha channel, the Set Canvas Color node can be used to define the Canvas color as solid alpha with
a color that matches the solid background.
See Chapter 34, “Color Nodes” in the Fusion Reference Manual or Chapter 95 in the DaVinci Resolve
Reference Manual for more detail on the Set Canvas Color node.

NOTE: The Domain of Definition is a bounding box that encompasses pixels that
have a nonzero value. The DoD is used to limit image-processing calculations and
speeds up rendering.

Inputs
The single input on the Auto Domain node is used to connect a 2D image and an effect mask, which
can be used to limit the blurred area.
– Input: The orange input is used for the primary 2D image that is blurred.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the blur to only
those pixels within the mask.

Basic Node Setup

The example below assumes that the image does not contain an alpha channel but a CG character
rendered against a black background. This is common for different render passes like specular or
shadows, for example. The image is connected to a Set Canvas Color node, which sets the Canvas
color to black with a solid alpha. The Set Canvas Color then connects to the Auto Domain node, which
detects the pixels and sets the DoD. If the original image contained a premultiplied alpha channel, the
Set Canvas Color would not be needed, and the image could be connected directly into the Auto
Domain node.

Chapter 49 Miscellaneous Nodes 1163

An Auto Domain node automatically limits the area of image processing.

Inspector

The Auto Domain Controls tab

Controls Tab
In most cases, the Auto Domain node automatically calculates the DoD bounding box; however,
the rectangular shape can be modified using the Controls tab in the Inspector.

Left
Defines the left border of the search area of the ADoD. Higher values on this slider move the left
border toward the right, excluding more data from the left margin.
1 represents the right border of the image; 0 represents the left border. The slider defaults
to 0 (left border).

Bottom
Defines the bottom border of the search area of the ADoD. Higher values on this slider move the
bottom border toward the top, excluding more data from the bottom margin.
1 represents the top border of the image; 0 represents the bottom border. The slider defaults
to 0 (bottom border).

Right
Defines the right border of the search area of the ADoD. Higher values on this slider move the right
border toward the left, excluding more data from the right margin.
1 represents the right border of the image; 0 represents the left border. The slider defaults
to 1 (right border).

Top
Defines the top border of the search area of the ADoD. Higher values on this slider move the top
border toward the bottom, excluding more data from the top margin.
1 represents the top border of the image; 0 represents the bottom border. The slider defaults
to 1 (top border).

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other miscellaneous nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Chapter 49 Miscellaneous Nodes 1164

Change Depth [CD]

The Change Depth node

Change Depth Node Introduction

The Change Depth node has one simple use, and that is to change the bits per color channel used to
process a node. This node is often used after color correcting 32-bit floating-point image files,
converting them from float processing to 16-bit per channel to preserve memory and performance.
It can also be useful if, from a certain point in your node tree, you feel the need to process your images
in a higher bit depth than their original one or to reduce the bit depth to save memory.

Inputs
The single input on the Change Depth node is used to connect a 2D image and an effect mask, which
can be used to limit the blurred area.
– Input: The orange input is used for the primary 2D image to be converted.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes or bitmaps from other tools. Connecting a mask to this input limits the blur to only
those pixels within the mask.

Basic Node Setup

Below, a Loader node in Fusion Studio is color corrected and then down converted from a floating-
point image to a 16-bit image to save image-processing time and memory.

A Change Depth node placed after color correction is done on a floating-point image.

Inspector

The Change Depth Controls tab

Chapter 49 Miscellaneous Nodes 1165

Controls Tab
The two controls for this node are the Depth menu and the Dither menu. These two menus are used to
convert and adjust the color depth of the image.

Depth
The Keep setting doesn‘t do anything to the image but instead keeps the input depth. The other
options change the bit depth of the image to the respective value.

Dither
When down converting from a higher bit depth, it can be useful to add Error Diffusion or Additive
Noise to camouflage artifacts that result from problematic (high-contrast) areas.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other miscellaneous nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Custom Tool [CT]

The Custom Tool node

Custom Tool Node Introduction

The Custom Tool node is the most complex and the most powerful node in Fusion. It is used to create
custom expressions and filters to modify an image. In addition to providing three image inputs, the
Custom Tool node allows for the connection of up to eight numeric inputs and as many as four XY
position values from other controls and parameters in the node tree.
Per-pixel calculations can be performed on the Red, Green, Blue, Alpha, Z, Z-Coverage, UV texture
coords, XYZ Normals, RGBA background color, and XY motion vector channels of the images.
You should be moderately experienced with scripting, or C++ programming, to understand the
structure and terminology used by the Custom Tool node.

Inputs
The Custom Tool node has three image inputs, a matte input, and an effect mask input.
– Input: The orange, green, and magenta inputs combine 2D images to make your composite.
When entering them into the Custom Tool fields, they are referred to as c1, c2 and c3 (c
standard for all three R, G, B channels)
– Matte Input: The white input is for a matte created by polylines, basic primitive shapes, paint
strokes, or bitmaps from other tools. Connecting a matte to this input allows a matte to be
combined into any equation. When entering the matte into the Custom Tool fields, it is referred
to as m1.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the Custom
Tool effect to only those pixels within the mask.

Chapter 49 Miscellaneous Nodes 1166

Basic Node Setup
The Custom Tool below takes two image inputs and a matte input, and then combines them using
some calculation. The result can be output to a Merge or other image-processing nodes.

A Custom Tool is used to build your own effects using C++ and scripting.

Inspector

The Custom Tool Controls tab

Controls Tab
Point in 1-4, X and Y
These four controls are 2D X and Y center controls that are available to expressions entered in the
Setup, Intermediate, and Channels tabs as variables p1x, p1y, ...., p4x, p4y. They are normal positional
controls and can be animated or connected to modifiers as any other node might.

Number in 1-8
The values of these controls are available to expressions entered in the Setup, Intermediate, and
Channels tabs as variables n1, n2, n3, ..., n8. They are normal slider controls and can be animated or
connected to modifiers exactly as any other node might.

Chapter 49 Miscellaneous Nodes 1167

LUT in 1-4
The Custom Tool node provides 4 LUT splines. The values of these controls are available to
expressions entered in the Setup, Intermediate, and Channels tabs using the getlut# function. For
example, setting the R, G, B, and A expressions to getlut1(r1), getlut2(g1), getlut3(b1), and getlut4(a1),
respectively, would cause the Custom Tool node to mimic the Color Curves node.
These controls can be renamed using the options in the Config tab to make their meanings more
apparent, but expressions still see the values as n1, n2, ..., n8.

Custom Tool Setup Tab

The Custom Tool Setup tab

Setup 1-4
Up to four separate expressions can be calculated in the Setup tab of the Custom Tool node. The
Setup expressions are evaluated once per frame before any other calculations are performed. The
results are then made available to the other expressions in the Custom Tool node as variables s1, s2,
s3, and s4.

NOTE: Because these expressions are evaluated once per frame only and not for each pixel,
it makes no sense to use per-pixel variables like X and Y or channel variables like r1, g1, b1.
Allowable values include constants, variables such as n1..n8, time, W and H, and functions like
sin() or getr1d().

Chapter 49 Miscellaneous Nodes 1168

Custom Tool Intermediate Tab

The Custom Tool Intermediate tab

Intermediate 1-4
An additional four expressions can be calculated in the Inter tab. The Inter expressions are evaluated
once per pixel after the Setup expressions are evaluated but before the Channel expressions are
evaluated. Per-pixel channel variables like r1, g1, b1, and a1 are allowable. Results are available as
variables i1, i2, i3, and i4.

Custom Tool Config Tab

The Custom Tool Config tab

Chapter 49 Miscellaneous Nodes 1169

Random Seed
Use this to set the seed for the rand() and rands() functions. Click the Randomize button to set the
seed to a random value. This control may be needed if multiple Custom Tool nodes are required with
different random results for each.

Number Controls
There are eight sets of Number controls, corresponding to the eight Number In sliders in the Controls
tab. Uncheck the Show Number checkbox to hide the corresponding Number In slider, or edit the
Name for Number text field to change its name.

Point Controls
There are four sets of Point controls, corresponding to the four Point In controls in the Controls tab.
Uncheck the Show Point checkbox to hide the corresponding Point In control and its crosshair in the
viewer. Similarly, edit the Name for Point text field to change the control’s name.

Custom Tool Channels Tab

The Custom Tool Channel tab

RGBA, Z, UV Expressions, and XYZ Normal Expressions

The Channel tab is used to set up one expression per each available channel of the image. Each
expression is evaluated once per pixel. The result creates the value for that pixel in the output of
the image.
Color Channel expressions (RGBA) should generally return floating-point values between 0.0 and 1.0.
Values beyond this are clipped if the destination image is an integer. Other expression fields should
produce values appropriate to their channel (e.g., between -1.0 and 1.0 for Vector and Normal fields,
0.0 to 1.0 for Coverage, or any value for Depth). The Channel expressions may use the results from
both the Setup expressions (as variables s1–s4) and Inter expressions (as variables i1–i4).

Custom Tool Node Syntax

Value Variables
n1..n8 Numeric Inputs

p1x..p4x Position Values (X-axis)

p1y..p4y Position Values (Y-axis)

s1..s4 Setup Expression Results

Chapter 49 Miscellaneous Nodes 1170

Value Variables
i1..i4 Inter Expression Results

time Current Frame

x Horizontal co-ordinate of the current pixel, between 0.0 and 1.0

y Vertical co-ordinate of the current pixel, between 0.0 and 1.0

w (or w1..w3) Width of Image (for image1..image3)

h (or h1..h3) Height of Image (for image1..image3)

ax (or ax1..ax3) Image Aspect X (for image1..image3)

ay (or ay1..ay3) Image Aspect Y (for image1..image3)

NOTE: Use w and h and ax and ay without a following number to get the dimensions and
aspect of the primary image.

Channel (Pixel) Variables

c1..c3 Current Channel (for image1..image3)

r1..r3 Red (for image1..image3)

g1..g3 Green (for image1..image3)

b1..b3 Blue (for image1..image3)

a1..a3 Alpha (for image1..image3)

z1..z3 Z-Buffer (for image1..image3)

cv1..cv3 Z Coverage (for image1..image3)

u1..u3 U Coordinate (for image1..image3)

v1..v3 nx1..nx3 V Coordinate (for image1..image3) X Normal (for image1..image3)

ny1..ny3 Y Normal (for image1..image3)

nz1..nz3 Z Normal (for image1..image3)

bgr1..bgr3 Background Red (for image1..image3)

bgg1..bgg3 Background Green (for image1..image3)

bgb1..bgb3 Background Blue (for image1..image3)

bga1..bga3 Background Alpha (for image1..image3)

vx1..vx3 X Vector (for image1..image3)

vy1..vy3 Y Vector (for image1..image3)

nz1..nz3 Z Normal (for image1..image3)

NOTE: Use c1, c2, c3 to refer to the value of a pixel in the current channel. This makes
copying and pasting expressions easier. For example, if c1/2 is typed as the red expression,
the result would be half the value of the red pixel from image 1, but if the expression is copied
to the blue channel, now it would have the value of the pixel from the blue channel.

Chapter 49 Miscellaneous Nodes 1171

To refer to the red value of the current pixel in input 1, type r1. For the image in input 2, it would be r2.
– get[ch][#]b(x, y) Read pixel at x,y, or 0 if out of bounds—e.g., getr1b(0,0)
– get[ch][#]d(x, y) Read pixel at x,y or edge pixel if out of bounds—e.g., getr1d(0,0)
– get[ch][#]w(x, y) Read pixel at x,y or wrap if out of bounds—e.g., getr1w(0,0)

NOTE: There are a variety of methods used to refer to pixels from locations other than the
current one in an image.

In the above description, [ch] is a letter representing the channel to access. The [#] is a number
representing the input image. So to get the red component of the current pixel (equivalent to r), you
would use getr1b(x,y). To get the alpha component of the pixel at the center of image 2, you would use
geta2b(0.5, 0.5).
– getr1b(x,y) Output the red value of the pixel at position x, y, if there were a valid pixel present. It
would output 0.0 if the position were beyond the boundaries of the image (all channels).
– getr1d(x,y) Output the red value of the pixel at position x, y. If the position specified were outside
of the boundaries of the image, the result would be from the outer edge of the image (RGBA only).
– getr1w(x,y) Output the red value of the pixel at position x, y. If the position specified were outside
of the boundaries of the image, the x and y coordinates would wrap around to the other side of the
image and continue from there (RGBA only).

To access other channel values with these functions, substitute the r in the above examples with the
correct channel variable (r, g, b, a and, for the getr1b() functions only, z, and so on), as shown above.
Substitute the 1 with either 2 or 3 in the above examples to access the images from the other
image inputs.

Mathematical Expressions
pi The value of pi

e The value of e

log(x) The base-10 log of x

ln(x) The natural (base-e) log of x

sin(x) The sine of x (x is degrees)

cos(x) The cosine of x (x is degrees)

tan(x) The tangent of x (x is degrees)

asin(x) The arcsine of x, in degrees

acos(x) The arccosine of x, in degrees

atan(x) The arctangent of x, in degrees

atan2(x,y) The arctangent of x,y, in degrees

abs(x) The absolute (positive) value of x

int(x) The integer (whole) value of x

frac(x) The fractional value of x

sqrt(x) The Square Root of x

Chapter 49 Miscellaneous Nodes 1172

Mathematical Expressions
rand(x,y) A random value between x and y

rands(x,y,s) A random value between x and y, based on seed s

min(x,y) The minimum (lowest) of x and y

max(x,y) The maximum (highest) of x and y

dist(x1,y1,x2,y2) The distance between point x1,y2 and x2,y2

dist3d(x1,y1,z1,x2,y2,z2) The distance between 3D points x1,y2,z1 and x2,y2,z2

noise(x) A smoothly varying Perlin noise value based on x

noise2(x, y) A smoothly varying Perlin noise value based on x and y

noise3(x, y, z) A smoothly varying Perlin noise value based on x, y and z

if(c, x, y) returns x if c not 0, otherwise y

Mathematical Operators
!x 1.0 if x = 0, otherwise 0.0

-x (0.0 - x)

+x (0.0 + x) i.e. effectively does nothing

x^y x raised to the power of y

x*y x multiplied by y

x/y x divided by y

x%y x modulo y, i.e. remainder of (x divided by y)

x+y x plus y

x-y x minus y

x<y 1.0 if x is less than y, otherwise 0.0

x>y 1.0 if x is greater than y, otherwise 0.0

x <= y 1.0 if x is less than or equal to y, otherwise 0.0

x >= y 1.0 if x is greater than or equal to y, otherwise 0.0

x=y 1.0 if x is exactly equal to y, otherwise 0.0

x == y 1.0 if x is exactly equal to y, otherwise 0.0, identical to above

x <> y 1.0 if x is not equal to y, otherwise 0.0

x != y 1.0 if x is not equal to y, otherwise 0.0, i.e. identical to above

x&y 1.0 if both x and y are not 0.0, otherwise 0.0

x && y 1.0 if both x and y are not 0.0, otherwise 0.0, i.e. identical to above

x|y 1.0 if either x or y (or both) are not 0.0, otherwise 0.0

x||y 1.0 if either x or y (or both) are not 0.0, otherwise 0.0

Chapter 49 Miscellaneous Nodes 1173

Example
The following examples are intended to help you understand the various components of the
Custom Tool node.

Rotation
To rotate an image, we need the standard equations for 2D rotation:

x’ = x * cos(theta) - y * sin(theta) y’ = x * sin(theta) + y *

cos(theta)

Using the n1 slider for the angle theta, and a sample function, we get (for the red channel):

getr1b(x * cos(n1) - y * sin(n1), x * sin(n1) + y * cos(n1))

This calculates the current pixel’s (x,y) position rotated around the origin at (0,0) (the bottom-
left corner), and then fetches the red component from the source pixel at this rotated position.
For centered rotation, we need to subtract 0.5 from our x and y coordinates before we rotate
them, and add 0.5 back to them afterward:

getr1b((x-.5) * cos(n1) - (y-.5) * sin(n1) + .5, (x-.5) * sin(n1) + (y-.5)

* cos(n1) + .5)

Which brings us to the next lesson: Setup and Intermediate Expressions. These are useful for
speeding things up by minimizing the work that gets done in the Channel expressions. The
Setup expressions are executed only once, and their results don‘t change for any pixel, so you
can use these for s1 and s2, respectively.

cos(n1) sin(n1)

Intermediate expressions are executed once for each pixel, so you can use these for i1 and i2:

(x-.5) * s1 - (y-.5) * s2 + .5

(x-.5) * s2 + (y-.5) * s1 + .5

These are the x and y parameters for the getr1b() function from above, but with the Setup
results, s1 and s2, substituted so that the trig functions are executed only once per frame, not
every pixel. Now you can use these intermediate results in your Channel expressions:

getr1b(i1, i2)

getg1b(i1, i2)

getb1b(i1, i2)

geta1b(i1, i2)

With the Intermediate expressions substituted in, we only have to do all the additions,
subtractions, and multiplications once per pixel, instead of four times. As a rule of thumb, if it
doesn‘t change, do it only once.
This is a simple rotation that doesn’t take into account the image aspect at all. It is left as an
exercise for you to include this (sorry). Another improvement could be to allow rotation around
points other than the center.

Filtering
Our second example duplicates the functionality of a 3 x 3 Custom Filter node set to average
the current pixel together with the eight pixels surrounding it. To duplicate it with a Custom
Tool node, add a Custom Tool node to the node tree, and enter the following expressions into
the Setup tab.

Chapter 49 Miscellaneous Nodes 1174

(Leave the node disconnected to prevent it from updating until we are ready.)

S1

1.0/w1

S2

1.0/h1

These two expressions are evaluated at the beginning of each frame. S1 divides 1.0 by the
current width of the frame, and S2 divides 1.0 by the height. This provides a floating-point
value between 0.0 and 1.0 that represents the distance from the current pixel to the next pixel
along each axis.
Now enter the following expression into the first text control of the Channel tab (r).

(getr1w(x-s1, y-s2) + getr1w(x, y-s2) + getr1w(x+s1, y-s2) + getr1w(x+s1,

y) + getr1w(x-s1, y) + r1 +getr1w(x-s1, y+s2) + getr1w(x, y+s2) +
getr1w(x+s1, y+s2)) / 9

This expression adds together the nine pixels above the current pixel by calling the getr1w()
function nine times and providing it with values relative to the current position. Note that we
referred to the pixels by using x+s1, y+s2, rather than using x+1, y+1.
Fusion refers to pixels as floating-point values between 0.0 and 1.0, which is why we created
the expressions we used in the Setup tab. If we had used x+1, y+1 instead, the expression
would have sampled the same pixel over and over again. (The function we used wraps the
pixel position around the image if the offset values are out of range.)
That took care of the red channel; now use the following expressions for the green, blue, and
alpha channels.

(getg1w(x-s1, y-s2) + getg1w(x, y-s2) + getg1w(x+s1, y-s2) + getg1w(x+s1,

y) + getg1w(x-s1, y) + g1 +getg1w(x-s1, y+s2) + getg1w(x, y+s2) +
getg1w(x+s1, y+s2)) / 9

(getb1w(x-s1, y-s2) + getb1w(x, y-s2) + getb1w(x+s1, y-s2) + getb1w(x+s1,

y) + getb1w(x-s1, y) + b1 +getb1w(x-s1, y+s2) + getb1w(x, y+s2) +
getb1w(x+s1, y+s2)) / 9

(geta1w(x-s1, y-s2) + geta1w(x, y-s2) + geta1w(x+s1, y-s2) + geta1w(x+s1,

y) + geta1w(x-s1, y) + a1 + geta1w(x-s1, y+s2) + geta1w(x, y+s2) +
geta1w(x+s1, y+s2)) / 9

It’s time to view the results. Add a Background node set to a solid color and change the color
to a pure red. Add a hard-edged Rectangular effects mask and connect it to the expression
just created.
For comparison, add a Custom Filter node and duplicate the settings from the image above.
Connect a pipe to this node from the background to the node and view the results. Alternate
between viewing the Custom Tool node and the Custom Filter while zoomed in close to the
top corners of the effects mask.
Of course, the Custom Filter node renders a lot faster than the Custom Tool node we created,
but the flexibility of the Custom Tool node is its primary advantage. For example, you could
use an image connected to input 2 to control the median applied to input one by changing all
instances of getr1w, getg1w, and getb1w in the expression to getr2w, getg2w, and getb2w, but
leaving the r1, g1, and b1s as they are.
This is just one example; the possibilities of the Custom Tool node are limitless.

Chapter 49 Miscellaneous Nodes 1175

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other miscellaneous nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Fields [Flds]

The Fields node

Fields Node Introduction

The Fields node is a robust multipurpose utility offering several functions related to interlaced video
frames. It interpolates video fields into frames or video frames into fields. Although the interlace
preference and method type is defined in the MediaIn or Loader, and generators, this node can be
used to assist in the standards conversion of PAL to NTSC and provides the ability to process fields
and frames for specific portions of a node tree.
This node can also interlace two separate images together into a single interlace image.
The background input is the dominate field 1 and the foreground is field 2.

Inputs
The single input on the Fields node is used to connect a 2D image and an effect mask, which can be
used to limit the blurred area.
– Stream1 Input: The orange background input is used for the primary 2D image that is
interpolated or converted.
– Stream2 Input: The optional green foreground input is only used when merging two interlaced
images together.

Basic Node Setup

The Fields node below is used to convert the background image from a PAL interlaced format to
progressive frames.

The background image can be modified to various interlaced formats.

Chapter 49 Miscellaneous Nodes 1176

Inspector

The Fields Controls tab

Controls Tab
The Controls tab includes two menus. The Operation menu is used to select the type of field
conversion performed. The Process Mode menu is used to select the field’s format for the
output image.

Operatiion Mode

Operation Menu
– Do Nothing: This causes the images to be affected by the Process Mode selection exclusively.
– Strip Field 2: This removes field 2 from the input image stream, which shortens the image to half
of the original height.
– Strip Field 1: This removes field 1 from the input image stream, which shortens the image to half of
the original height.
– Strip Field 2 and Interpolate: This removes field 2 from the input image stream and inserts a
field interpolated from field 1 so that image height is maintained. Should be supplied with frames,
not fields.
– Strip Field 1 and Interpolate: This removes field 1 from the input image stream and inserts a
field interpolated from field 2 so that image height is maintained. Should be supplied with frames,
not fields.
– Interlace: This combines fields from the input image stream(s). If supplied with one image
stream, each pair of frames are combined to form half of the number of double-height frames.
If supplied with two image streams, single frames from each stream are combined to form
double-height images.
– De-Interlace: This separates fields from one input image stream. This will produce double the
amount of half-height frames.

Chapter 49 Miscellaneous Nodes 1177

Reverse Field Dominance
When selected, the Field Order or Dominance of the image will be swapped.

Process Mode

Process Mode Menu

– Full Frames: This forces Frame Processing. Useful for processing frames in a part of a node tree
that is otherwise field processing.
– NTSC Fields: This forces NTSC Field Processing. Useful for processing fields in a part of a node
tree that is otherwise frame processing.
– PAL Fields: This forces PAL Field Processing. Useful for processing fields in a part of a node tree
that is otherwise frame processing.
– PAL Fields (Reversed): This forces PAL-swapped Field Processing.
– NTSC Fields (Reversed): This forces NTSC-swapped Field Processing.
– Auto: This attempts to match the mode of its input images. Fields are used if the
input types are mixed.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other miscellaneous nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Frame Average [Avg]

The Frame Average node

Frame Average Node Introduction

The Frame Average node averages together a series of frames to simulate clips shot with long shutter
speeds. Aside from motion blur-style effects, it can be useful for time warps or noise removal.

Chapter 49 Miscellaneous Nodes 1178

Inputs
The single input on the Frame Average node is used to connect a 2D image that will have the
averaging applied.
– Input: The orange input is used for the primary 2D image that will be averaged.

Basic Node Setup

The image connected to the orange input is frame averaged based on the settings in the Inspector.

A Frame Average node blends the input image’s frames.

Inspector

The Frame Average Controls tab

Controls Tab
The Controls tab contains the parameters for setting the duration and guidance of the
averaged frames.

Sample Direction
The Sample Direction menu determines if the averaged frames are taken before the current frame,
after, or a mix of the two.
– Forward: Averages the number of frames set by the Frames slider after the current frame.
– Both: Averages the number of frames set by the Frames slider, taking frames before and after the
current frame.
– Backward: Averages the number of frames set by the Frames slider before the current frame.

Missing Frames
This control determines the behavior if a frame is missing from the clip.
– Duplicate Original: Uses the last original frame until a new frame is available.
– Blank Frame: Leaves missing frames blank.

Frames
This slider sets the number of frames that are averaged.

Chapter 49 Miscellaneous Nodes 1179

Keyframe Stretcher [KfS]

The Keyframe Stretcher node

Keyframe Stretcher Node Introduction

The Keyframe Stretcher node is inserted after animated nodes, so the keyframes stretch and the
comp’s duration is modified. It is used to scale the keyframes on the animation curve to the current
duration of the clip. This is particularly useful when creating title templates in Fusion for use in
DaVinci Resolve’s Edit or Cut page.

TIP: The Keyframe Stretcher can be used on a single parameter by applying the
Keystretcher modifier.

Inputs
The single input on the Keyframe Stretcher node is used to connect a 2D image that contains
keyframe animation.
– Input: The orange input is used for any node with keyframed animation. The input can be
a Merge node that is not animated but contains foreground and background nodes that are
animated.

Basic Node Setup

The Keyframe Stretcher is added just before the Media Out or Saver node. All nodes that include
animation before the Keyframe Stretcher are modified if the comp changes duration.

A Keyframe Stretcher changing the animation of two text nodes.

The diagram below shows the original 50-frame animation added to a parameter. The Keyframe
Stretcher Start and End would be set to 0 and 50. The second keyframe is set at frame 10, and the
third keyframe is set at frame 40. Setting the Stretch Start to frame 11 and the Stretch End to frame 39
will keep the existing keyframes at the same speed (number of frames.) The middle will be stretched.

Chapter 49 Miscellaneous Nodes 1180

Original 50-frame animation

In the below example, the duration of the clip is extended to 75 frames. The first 10 frames and the last
10 frames of the animation run at the same speed as the original animation, while any animation in the
middle is stretched to fill the difference.

NOTE: The actual Spline Editor will show only the original keyframe positions. The splines
are not changed by the Keyframe Stretcher; only the animation is changed.

Animation modified to 75 frames but stretching only the middle of the animation

Inspector

The Keyframe Stretcher Keyframes tab

Keyframes Tab
The Keyframes tab includes Source controls for setting the source duration and Stretch controls for
setting the area of the animation that gets modified.

Source Start/Source End

A source range is specified using the Source Start and Source End controls. These are typically set to
match the full range of the animation spline on the Keyframes control.

Chapter 49 Miscellaneous Nodes 1181

Stretch Start/Stretch End
The Stretch Start and Stretch End controls let you specify a middle zone where keyframes will be
stretched or squished. Handles outside the range will not get scaled. Any keyframes outside the
Stretch Start and End range always remain the same number of frames from the Start and End.
Any keyframe adjustments to the original control will be correspondingly scaled back to the source
curve and will match the original timing as expected.

Stretch Edges Instead

Enabling the Stretch Edges Instead checkbox overrides the Stretch Start and Stretch End controls and
stretches the edges of the animation.

Run Command [Run]

The Run Command node

Run Command Node Introduction

The Run Command node is used to execute an external command or batch file at specific points
during a render. You can choose to run a command at the start or the end of a render. Or you can have
the command execute once for each frame.
The Run Command can be used to net render other command line applications using the Fusion
Render Manager, as well as a host of other useful functions.

Inputs
The single input on the Run Command node is used to pass through a 2D image.
– Input: The optional orange image input is not required for this node to operate. However, if it is
connected to a node‘s output, the Run Command will only launch after the connected node has
finished rendering. This is often useful when connected to a Saver, to ensure that the output
frame has been fully saved to disk first. If the application launched returns a non-zero result, the
node will also fail.

Basic Node Setup

The Run Command node can be connected after a Saver and run once the final frame is completed.

Run Command placed after a Saver node.

Chapter 49 Miscellaneous Nodes 1182

Inspector

The Run Command Frame tab

Frame Tab
The Frame tab is where the command to execute is selected and modified.

Hide
Enable the Hide checkbox to prevent the application or script from displaying a window when it
is executed.

Wait
Enable this checkbox to cause the node to wait for a remote application or tool to exit before
continuing. If this checkbox is disabled, the Fusion continues rendering without waiting for the external
application.

Frame Command
This field is used to specify the path for the command to be run after each frame is rendered. The
Browse button can be used to identify the path.

Interactive
This checkbox determines whether the launched application should run interactively, allowing
user input.

Number A (%a) and Number B (%b)

Various wildcards can be used with the frame commands; these wildcards will be substituted at render
time with the correct values.
– %a: Outputs the number from the Number A thumbwheel control.
– %b: Outputs the number from the Number B thumbwheel control.
– %t: Outputs the current frame number (without zero padding).
– %s: Substitutes using the text from the large text entry field.
If you want to add zero paddings to the numbers generated by %t, refer to the wildcard with %0x,
where x is the number of characters with which to pad the value. This also works for %a and %b.

Chapter 49 Miscellaneous Nodes 1183

For example, test%04t.tga would return the following values at render time:
test0000.tga
test0001.tga
test0009.tga
test0010.tga

You may also pad a value with spaces by calling the wildcard as %x, where x is the number of spaces
with which you would like to pad the value.

The Run Command Start tab The Run Command End tab

Start and End Tabs

The Start and End tabs contain a file browser for a command to be run when the composition starts to
render and when the composition is done rendering.

Example
To copy the saved files from a render to another directory as each frame is rendered, save the
following text in a file called copyfile.bat to your C\ directory (the root folder).

@echo off
set parm=%1 %2
copy %1 %2 set parm=

Create or load any node tree that contains a Saver. The following example assumes a Saver is
set to output D\ test0000.exr, test0001.exr, etc. You may have to modify the example to match.
Add a Run Command node after the Saver to ensure the Saver has finished saving first. Now
enter the following text into the Run Command node’s Frame Command text box:

C\copytest.bat D\test%04f.exr C\

Select the Hide Frame command checkbox to prevent the Command Prompt window from
appearing briefly after every frame.
When this node tree is rendered, each file will be immediately copied to the C\ directory as it
is rendered.
The Run Command node could be used to transfer the files via FTP to a remote drive on the
network, to print out each frame as it is rendered, or to execute a custom image-
processing tool.
The Run Command node is not restricted to executing simple batch files. FusionScript,
VBScript, Jscript, CGI, and Perl files could also be used, as just a few examples.

Chapter 49 Miscellaneous Nodes 1184

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other miscellaneous nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Set Domain [DoD]

The Set Domain node

Set Domain Node Introduction

Set Domain is used to adjust or set the active area of an image or the area of the image considered to
have valid data.
It does not change the image‘s physical dimensions. Downstream nodes will not process anything
outside the Domain of Definition (DoD), thus speeding up rendering of computation-intensive nodes.
This node provides an absolute mode, for setting the domain of definition manually, and a relative
mode for adjusting the existing domain of definition.

Inputs
The two inputs on the Set Domain node are used to connect 2D images.
– Input: The orange background input must be connected. It accepts a 2D image with the DoD
you want to replace or adjust.
– Foreground: The green image input is optional but also accepts a 2D image as its input. When
the foreground input is connected, the Set Domain node will replace the Background input’s
domain of definition with the foreground’s DoD.

Basic Node Setup

The example below assumes an image is connected to a Set Domain node to manually
configure the DoD.

A Set Domain node manually sets the area to limit image processing.

Chapter 49 Miscellaneous Nodes 1185

Inspector

The Set Domain Controls tab in Set mode

Controls Tab
Mode
The Mode menu has two choices depending on whether you want to adjust or offset the existing
domain or set precise values for it.
The same operations can be performed in Set or in Adjust mode. In Adjust mode, the sliders default to
0, marking their respective full extent of the image. Positive values shrink the DoD while negative
values expand the DoD to include more data.
Set mode defaults to the full extent of the visible image. Sliders default to a scale of 0-1 from left to
right and bottom to top.

Left
Defines the left border of the DoD. Higher values on this slider move the left border toward the right,
excluding more data from the left margin.
1 represents the right border of the image; 0 represents the left border. The slider defaults to 0
(left border).

Bottom
Defines the bottom border of the DoD. Higher values on this slider move the bottom border toward the
top, excluding more data from the bottom margin.
1 represents the top border of the image; 0 represents the bottom border. The slider defaults to 0
(bottom border).

Right
Defines the right border of the DoD. Higher values on this slider move the right border toward the left,
excluding more data from the right margin.
1 represents the right border of the image; 0 represents the left border. In Set mode, the slider defaults
to 1 (right border).

Top
Defines the top border of the DoD. Higher values on this slider move the top border toward the
bottom, excluding more data from the top margin.
1 represents the top border of the image; 0 represents the bottom border. In Set mode, the slider
defaults to 1 (top border).

Chapter 49 Miscellaneous Nodes 1186

The Set Domain Controls tab in Adjust mode

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other miscellaneous nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Time Speed [TSpd]

The Time Speed node

Time Speed Node Introduction

The Time Speed node allows image sequences to be sped up, slowed down, reversed, or delayed.
Image Interpolation offers smooth, high-quality results. Time Speed should be used for static speed
changes or to introduce delays in the footage. To apply animated changes in time, such as
accelerating or decelerating time, use a Time Stretcher instead.
When operating in Flow mode, Optical Flow data is required. This node does not generate optical flow
directly. You have to create it upstream using an Optical Flow node or by loading the forward/reverse
vector channels from the image.
TimeSpeed does not interpolate the aux channels but instead destroys them. In particular, the Vector/
BackVector channels are consumed and destroyed after computation.
Add an Optical Flow after the Time Speed node if you want to generate flow vectors for the
retimed footage.

Inputs
The single input on the Time Speed node is used to connect a 2D image that will be retimed.
– Input: The orange input is used for the primary 2D image that will be retimed.

Chapter 49 Miscellaneous Nodes 1187

Basic Node Setup
The Time Speed node setup is as simple as connecting a 2D image into the orange background input
of the node.

A MediaIn node having its speed changed in the Time Speed node.

Inspector

The Time Speed Controls tab

Speed
This control is used to adjust the Speed, in percentage values, of the outgoing image sequence.
Negative values reverse the image sequence. 200% Speed is represented by a value of 2.0, 100% by
1.0, 50% by 0.5, and 10% by 0.1.
The Speed control cannot be animated.

Delay
Use this control to Delay the outgoing image sequence by the specified number of frames. Negative
numbers offset time back, and positive numbers advance time.

Interpolate Mode
This menu determines the how the time speed is processed in order to improve its visual playback
quality, especially in the case of clips that are slowed down. There are three choices in the menu.
– Nearest: The most processor efficient and least sophisticated method of processing; frames are
either dropped for fast motion or duplicated for slow motion.
– Blend: Also processor efficient, but can produce smoother results; adjacent duplicated frames are
dissolved together to smooth out slow or fast motion effects.
– Flow: The most processor intensive but highest quality method of speed effect processing.
Using vector channels pre-generated from an Optical Flow node, new frames are generated to
create slow or fast motion effects. The result can be exceptionally smooth when motion in a clip
is linear. However, two moving elements crossing in different directions or unpredictable camera
movement can cause unwanted artifacts.

Chapter 49 Miscellaneous Nodes 1188

Sample Spread
This slider is displayed only when Interpolation is set to Blend. The slider controls the strength of the
interpolated frames on the current frame. A value of 0.5 blends 50% of the frame before and 50% of
the frame ahead and 0% of the current frame.

Depth Ordering
This menu is displayed only when Interpolation is set to Flow. The Depth Ordering is used to
determine which parts of the image should be rendered on top. This is best explained by example.
In a locked-off camera shot where a car is moving through the frame, the background does not move,
so it produces small, or slow, vectors. The car produces larger, or faster, vectors.
The Depth Ordering, in this case, is Fastest on Top, since the car draws over the background.
In a shot where the camera pans to follow the car, the background has faster vectors, and the car has
slower vectors, so the Depth ordering method would be Slowest on Top.

Clamp Edges
This checkbox is displayed only when Interpolation is set to Flow. Under certain circumstances, this
option can remove the transparent gaps that may appear on the edges of interpolated frames. Clamp
Edges can cause a stretching artifact near the edges of the frame that is especially visible with objects
moving through it or when the camera is moving.
Because of these artifacts, it is a good idea to use clamp edges only to correct small gaps around the
edges of an interpolated frame.

Edge Softness
This slider is only displayed when Interpolation is set to Flow and Clamp Edges is enabled. It helps to
reduce the stretchy artifacts that might be introduced by Clamp Edges.
If you have more than one of the Source Frame and Warp Direction checkboxes turned on, this can
lead to doubling up of the stretching effect near the edges. In this case, you’ll want to keep the
softness rather small at around 0.01. If you have only one checkbox enabled, you can use a larger
softness at around 0.03.

Source Frame and Warp Direction

These checkboxes are displayed only when Interpolation is set to Flow. These controls determine
which frames and which vectors are used to create the in-between frames. Each method ticked on will
be blended into the result.
– Prev Forward: Takes the previous frame and uses the Forward vector to interpolate
the new frame.
– Next Forward: Takes the next frame in the sequence and uses the Forward vector to interpolate
the new frame.
– Prev Backward: Takes the previous frame and uses the Back Forward vector to interpolate the
new frame.
– Next Backward: Takes the next frame in the sequence and uses the Back vector to interpolate the
new frame.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other miscellaneous nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Chapter 49 Miscellaneous Nodes 1189

Time Stretcher [TSt]

The Time Stretcher node

Time Stretcher Node Introduction

The Time Stretcher node is similar to the Time Speed node but permits the speed of the clip to be
animated. Full spline control of the effect is provided, including smoothing. As a result, the Time
Stretcher can be used to animate a single clip to 200, back to normal speed, pause for a second, and
then play backward (like a VCR rewinding).
Image interpolation offers smooth, high-quality results, all using a spline curve to adjust time
nonlinearly. To apply constant time changes such as frame rate changes, use a Time Speed instead.
When operating in Flow mode, Optical Flow data is required. This node does not generate optical flow
directly; you must create it manually upstream using an Optical Flow node or by loading the forward/
reverse vector channels from disk.
Flow Stretcher does not interpolate the aux channels but instead destroys them. In particular, the
Vector/BackVector channels are consumed/destroyed. Add an Optical Flow after the Flow Stretcher if
you want to generate flow vectors for the retimed footage.

Inputs
The single input on the Time Stretcher node is used to connect a 2D image that will be time stretched.
– Input: The orange input is used for the primary 2D image that will be time stretched.

Basic Node Setup

Like the Time Speed node, the Time Stretcher setup is as simple as connecting a 2D image into the
orange background input of the node.

A MediaIn node having its time ramped to various speeds in the Time Stretcher node.

Inspector

The Time Stretcher controls

Chapter 49 Miscellaneous Nodes 1190

Source Time
This control designates from which frame in the original sequence to begin sampling.
When a Time Stretcher node is added to the node tree, the Source Time control already contains a
Bézier spline with a single keyframe set to 0.0. The keyframe position is determined by the current
time when the node is added to the node tree.

NOTE: The Source Time spline may not be immediately visible until Edit is selected from the
Source Time’s contextual menu, or Display all Splines is selected from the Spline Window’s
contextual menu.

Interpolate Mode
This menu determines the how the time speed is processed in order to improve its visual playback
quality, especially in the case of clips that are slowed down. There are three choices in the menu.
– Nearest: The most processor efficient and least sophisticated method of processing; frames are
either dropped for fast motion or duplicated for slow motion.
– Blend: Also processor efficient but can produce smoother results; adjacent duplicated frames are
dissolved together to smooth out slow or fast motion effects.
– Flow: The most processor intensive but highest quality method of speed effect processing.
Using vector channels pre-generated from an Optical Flow node, new frames are generated to
create slow or fast motion effects. The result can be exceptionally smooth when motion in a clip
is linear. However, two moving elements crossing in different directions or unpredictable camera
movement can cause unwanted artifacts.

Sample Spread
This slider is displayed only when Interpolation is set to Blend. The slider controls the strength of the
interpolated frames on the current frame. A value of 0.5 blends 50% of the frame before and 50% of
the frame ahead and 0% of the current frame.

Depth Ordering
This menu is displayed only when Interpolation is set to Flow. The Depth Ordering is used to
determine which parts of the image should be rendered on top. This is best explained by example.
In a locked-off camera shot where a car is moving through the frame, the background does not move,
so it produces small, or slow, vectors. The car produces larger, or faster, vectors.
The Depth Ordering in this case is Fastest on Top, since the car draws over the background.
In a shot where the camera pans to follow the car, the background has faster vectors, and the car has
slower vectors, so the Depth ordering method would be Slowest on Top.

Clamp Edges
This checkbox is displayed only when Interpolation is set to Flow. Under certain circumstances, this
option can remove the transparent gaps that may appear on the edges of interpolated frames. Clamp
Edges can cause a stretching artifact near the edges of the frame that is especially visible with objects
moving through it or when the camera is moving.
Because of these artifacts, it is a good idea to use clamp edges only to correct small gaps around the
edges of an interpolated frame.

Edge Softness
This slider is displayed only when Interpolation is set to Flow and Clamp Edges is enabled. It helps to
reduce the stretchy artifacts that might be introduced by Clamp Edges.

Chapter 49 Miscellaneous Nodes 1191

If you have more than one of the Source Frame and Warp Direction checkboxes turned on, this can
lead to doubling up of the stretching effect near the edges. In this case, you’ll want to keep the
softness rather small at around 0.01. If you have only one checkbox enabled, you can use a larger
softness at around 0.03.

Source Frame and Warp Direction

These checkboxes are displayed only when Interpolation is set to Flow. These controls determine
which frames and which vectors are used to create the in-between frames. Each method ticked on will
be blended into the result.
– Prev Forward: Takes the previous frame and uses the Forward vector to
interpolate the new frame.
– Next Forward: Takes the next frame in the sequence and uses the Forward vector to
interpolate the new frame.
– Prev Backward: Takes the previous frame and uses the Back Forward vector to
interpolate the new frame.
– Next Backward: Takes the next frame in the sequence and uses the Back vector to
interpolate the new frame.

Example
Make sure that the current time is either the first or last frame of the clip to be affected in the
project. Add the Time Stretcher node to the node tree. This will create a single point on the
Source Time spline at the current frame. The value of the Source Time will be set to zero for
the entire Global Range.
Set the value of the Source Time to the frame number to be displayed from the original source,
at the frame in time it will be displayed in during the project.
To shrink a 100-frame sequence to 25 frames, follow these steps:
1 Change the Current Time to frame 0.
2 Change the Source Time control to 0.0.
3 Advance to frame 24.
4 Change the Source Time to 99.
5 Check that the spline result is linear.
6 Fusion will render 25 frames by interpolating down the 100 frames to a length of 25.
7 Hold the last frame for 30 frames, and then play the clip backward at regular speed.
Continue the example from above and follow the steps below.
8 Advance to frame 129.
9 Right-click on the Source Time control and select Set Key from the menu.
10 Advance to frame 229 (129 + 100).
11 Set the Source time to 0.0.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other miscellaneous nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Chapter 49 Miscellaneous Nodes 1192

Wireless Link [Wire]

The Wireless Link node

Wireless Link Node Introduction

The Wireless Link node helps manage the tangle of connection lines in a node tree by wirelessly
connecting one 2D node to another 2D node.
Although Wireless Links can be helpful, try to keep as much of a node tree as visible as possible;
otherwise, you lose one of the main benefits of a node tree.

Inputs
There are no inputs on this node.

Basic Node Setup

There is no setup for this node. It is a free-standing node that connects “wirelessly” using the control
in the Inspector.

Inspector

The Wireless Link Controls tab

Controls Tab
The Controls tab in the Wireless Link node contains a single Input field for the linked node.

Input
To use the Wireless Link node, in the Node Editor, drag the 2D node into the Input field of the Wireless
Link node. Any change you make to the original node is wirelessly replicated in the Wireless Link
node. You can use the output from the Wireless Link node to connect to a nearby node.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other miscellaneous nodes. These common
controls are described in detail in the following “The Common Controls” section.

Chapter 49 Miscellaneous Nodes 1193

The Common Controls
Nodes that handle miscellaneous operations share a number of identical controls in the Inspector. This
section describes controls that are common among miscellaneous nodes.

Inspector

The Common Miscellaneous Settings tab

Settings Tab
The Settings tab in the Inspector can be found on every tool in the miscellaneous nodes. The controls
are consistent and work the same way for each tool.

Blend
The Blend control is used to blend between the tool’s original image input and the tool’s final modified
output image. When the blend value is 0.0, the outgoing image is identical to the incoming image.
Normally, this will cause the tool to skip processing entirely, copying the input straight to the output.

Process When Blend Is 0.0

The tool is processed even when the input value is zero. This can be useful if processing of this node
is scripted to trigger another task, but the value of the node is set to 0.0.

Chapter 49 Miscellaneous Nodes 1194

Red/Green/Blue/Alpha Channel Selector
These four buttons are used to limit the effect of the tool to specified color channels. This filter is often
applied after the tool has been processed.
For example, if the Red button on a Blur tool is deselected, the blur will first be applied to the image,
and then the red channel from the original input will be copied back over the red channel of the result.
There are some exceptions, such as tools for which deselecting these channels causes the tool to skip
processing that channel entirely. Tools that do this will generally possess a set of identical RGBA
buttons on the Controls tab in the tool. In this case, the buttons in the Settings and the Controls tabs
are identical.

Apply Mask Inverted

Enabling the Apply Mask Inverted option inverts the complete mask channel for the tool. The mask
channel is the combined result of all masks connected to or generated in a node.

Multiply by Mask
Selecting this option will cause the RGB values of the masked image to be multiplied by the mask
channel’s values. This will cause all pixels of the image not included in the mask (i.e., set to 0) to
become black/transparent.

Use Object/Use Material (Checkboxes)

Some 3D software can render to file formats that support additional channels. Notably, the EXR file
format supports Object ID and Material ID channels, which can be used as a mask for the effect. These
checkboxes determine whether the channels will be used, if present. The specific Material ID or
Object ID affected is chosen using the next set of controls.

Correct Edges
This checkbox appears only when the Use Object or Use Material checkboxes are selected. It toggles
the method used to deal with overlapping edges of objects in a multi-object image. When enabled,
the Coverage and Background Color channels are used to separate and improve the effect around the
edge of the object. If this option is disabled (or no Coverage or Background Color channels are
available), aliasing may occur on the edge of the mask.
For more information on Coverage and Background Color channels, see Chapter 18, “Understanding
Image Channels” in the Fusion Reference Manual or Chapter 79 in the DaVinci Resolve
Reference Manual.

Object ID/Material ID (Sliders)

Use these sliders to select which ID will be used to create a mask from the object or material channels
of an image. Use the Sample button in the same way as the Color Picker: to grab IDs from the image
displayed in the view. The image or sequence must have been rendered from a 3D software package
with those channels included.

Use GPU
The Use GPU menu has three settings. Setting the menu to Disable turns off GPU hardware-
accelerated rendering. Enabled uses the GPU hardware for rendering the node. Auto uses a capable
GPU if one is available and falls back to software rendering when a capable GPU is not available.

Chapter 49 Miscellaneous Nodes 1195

Motion Blur
– Motion Blur: This toggles the rendering of Motion Blur on the tool. When this control is toggled
on, the tool’s predicted motion is used to produce the motion blur caused by the virtual camera’s
shutter. When the control is toggled off, no motion blur is created.
– Quality: Quality determines the number of samples used to create the blur. A quality setting of 2
will cause Fusion to create two samples to either side of an object’s actual motion. Larger values
produce smoother results but increase the render time.
– Shutter Angle: Shutter Angle controls the angle of the virtual shutter used to produce the motion
blur effect. Larger angles create more blur but increase the render times. A value of 360 is the
equivalent of having the shutter open for one whole frame exposure. Higher values are possible
and can be used to create interesting effects.
– Center Bias: Center Bias modifies the position of the center of the motion blur. This allows for the
creation of motion trail effects.
– Sample Spread: Adjusting this control modifies the weighting given to each sample. This affects
the brightness of the samples.

Hide Incoming Connections

Enabling this checkbox can hide connection lines from incoming nodes, making a node tree appear
cleaner and easier to read. When enabled, empty fields for each input on a node will be displayed in
the Inspector. Dragging a connected node from the node tree into the field will hide that incoming
connection line as long as the node is not selected in the node tree. When the node is selected in the
node tree, the line will reappear.

Comments
The Comments field is used to add notes to a tool. Click in the empty field and type the text. When a
note is added to a tool, a small red square appears in the lower-left corner of the node when the full
tile is displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the
note in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 49 Miscellaneous Nodes 1196

Chapter 50

Optical Flow
This chapter details the Optical Flow nodes available in Fusion.

The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Optical Flow [OF]  1198
Repair Frame [Rep]  1201
Smooth Motion [SM]  1203
Tween [Tw]  1205
The Common Controls  1208

Chapter 50 Optical Flow 1197

Optical Flow [OF]

The OpticalFlow node

Optical Flow Node Introduction

This node analyzes a clip connected to its input using an Optical Flow algorithm. Think of optical flow
as a per-pixel motion vector that matches up features over several frames.
The computed optical flow is stored within the Vector and Back Vector aux channels of the output.
These channels can be used in other nodes like the Vector Motion Blur or Vector Distort. However,
Optical Flow must render twice when connecting it to a Time Stretcher or Time Speed node. These
nodes require the channels A. FwdVec and B. BackVec in that order, but Optical Flow generates A.
BackVec and A. FwdVec when it processes.
If you find that optical flow is too slow, consider rendering it out into OpenEXR files using a
Saver node.

TIP: If the footage input flickers on a frame-by-frame basis, it is a good idea to deflicker the
footage beforehand.

Inputs
The Optical Flow node includes a single orange image input.
– Input: The orange background input accepts a 2D image. This is the sequence of frames for
which you want to compute optical flow. The output of the Optical Flow node includes the
image and vector channels. The vector channels can be displayed by right-clicking in the
viewer and choosing Channel > Vectors and then Options > Normalize Color Range.

Basic Node Setup

The Optical Flow node analyzes the frames connected to the background input. Trimming a Loader or
MediaIn to only the range you need prevents analyzing unnecessary frames. The output of the node
can then be connected to any node that takes advantage of vector channels, such as a Time Stretcher.

TIP: When analyzing Optical Flow vectors, consider adding a Smooth Motion node afterward
with smoothing for forward/ backward vectors enabled.

An Optical Flow node generating motion vectors on-the-fly.

Chapter 50 Optical Flow 1198

Alternatively, if you find the Optical Flow node too slow to analyze the frames, consider rendering it
out to an OpenEXR format using a Saver node. Then import the rendered EXR file as your new image
with embedded vector channels.

An Optical Flow node rendered out through a Saver node.

Inspector

Optical Flow advanced controls

Controls Tab (Advanced)

When you add an Optical Flow, Repair Frame, or Tween node to a Comp, a Method drop-down menu
in the Inspector allows you to choose between an Advanced GPU-based algorithm or a Classic
CPU-based algorithm. This Advanced method is the same Optical Flow algorithm used in other
DaVinci Resolve pages.

Warp Count
Decreasing this slider makes the optical flow computations faster. To understand what this option
does, you must understand that the optical flow algorithm progressively warps one image until it
matches with the other image. After some point, convergence is reached, and additional warps
become a waste of computational time. You can tweak this value to speed up the computations, but it
is good to watch what the optical flow is doing at the same time.

Iteration Count
Decreasing this slider makes the computations faster. In particular, just like adjusting the Warp Count,
adjusting this option higher will eventually yield diminishing returns and not produce significantly
better results. By default, this value is set to something that should converge for all possible shots and
can be tweaked lower fairly often without reducing the disparity’s quality.

Smoothness
This controls the smoothness of the optical flow. Higher smoothness helps deal with noise, while lower
smoothness brings out more detail.

Half Resolution
The Half Resolution checkbox is used purely to speed up the calculation of the optical flow. The input
images are resized down and tracked to produce the optical flow.

Chapter 50 Optical Flow 1199

Controls Tab (Classic)
By choosing Classic from the Method drop-down menu in the Inspector, you can use the older
CPU-based algorithm to maintain compatibility with Comps created in previous versions. This method
may also be better suited for some Stereo3D processing.
When using the Classic method, a single slider at the top of the Inspector improves performance by
generating proxies. The remaining Advanced section parameters tune the Optical Flow vector
calculations. The default settings serve as a good standard. In most cases, tweaking of the advanced
settings is not needed. Many deliver small or diminishing returns. However, depending on the settings,
rendering time can easily vary by 10x. If you’re interested in reducing process time, it is best to start by
experimenting with the Proxy, Number of Iterations, and Number of Warps sliders and changing the
filtering to Bilinear.

Proxy (for Tracking)

The Proxy slider is used purely to speed up the calculation of the optical flow. The input images are
resized down by the proxy scale and tracked to produce the optical flow. The computational time is
roughly proportional to the number of pixels in the image. This means a proxy scale of 2 will give a
4x speedup, and a proxy scale of 3 will give a 9x speedup.

Smoothness
This controls the smoothness of the optical flow. Higher smoothness helps deal with noise, while lower
smoothness brings out more detail.

Edges
This slider is another control for smoothness but applies it based on the color channel. It tends to have
the effect of determining how edges in the flow follow edges in the color images. When it is set to a
low value, the optical flow becomes smoother and tends to overshoot edges. When it is set to a high
value, details from the color images start to slip into the optical flow, which is not desirable. Edges in
the flow end up more tightly aligning with the edges in the color images. This can result in streaked-
out edges when the optical flow is used for interpolation. As a rough guideline, if you are using the
disparity to produce a Z-channel for post effects like Depth of Field, then set it lower in value. If you are
using the disparity to perform interpolation, you might want it to be higher in value.

Match Weight
This control sets a threshold for how neighboring groups of foreground/background pixels are
matched over several frames. When set to a low value, large structural color features are matched.
When set to higher values, small sharp variations in the color are matched. Typically, a good value for
this slider is in the [0.7, 0.9] range. When dealing with stereo 3D, setting this option higher tends to
improve the matching results in the presence of differences due to smoothly varying shadows or local
lighting variations between the left and right images. The user should still perform a color match or
deflickering on the initial images, if necessary, so they are as similar as possible. This option also helps
with local variations like lighting differences due to light passing through a mirror rig.

Mismatch Penalty
This option controls how the penalty for mismatched regions grows as they become more dissimilar.
The slider provides a choice between a balance of Quadratic and Linear penalties. Quadratic strongly
penalizes large dissimilarities, while Linear is more robust to dissimilar matches. Moving this slider
toward Quadratic tends to give a disparity with more small random variations in it, while Linear
produces smoother, more visually pleasing results.

Warp Count
Decreasing this slider makes the optical flow computations faster. In particular, the computational time
depends linearly upon this option. To understand what this option does, you must understand that the
optical flow algorithm progressively warps one image until it matches with the other image. After some
point, convergence is reached, and additional warps become a waste of computational time. The
default value in Fusion is set high enough that convergence should always be reached. You can tweak
this value to speed up the computations, but it is good to watch what the optical flow is doing at the
same time.
Chapter 50 Optical Flow 1200
Iteration Count
Decreasing this slider makes the computations faster. In particular, the computational time depends
linearly upon this option. Just like adjusting the Warp Count, adjusting this option higher will eventually
yield diminishing returns and not produce significantly better results. By default, this value is set to
something that should converge for all possible shots and can be tweaked lower fairly often without
reducing the disparity’s quality.

Filtering
This option controls filtering operations used during flow generation. Catmull-Rom filtering will
produce better results, but at the same time, turning on Catmull-Rom will increase the computation
time steeply.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Optical Flow nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Repair Frame [Rep]

The Repair Frame node

Repair Frame Node Introduction

Repair Frame replaces damaged or missing frames or portions of frames with scratches or other
temporally transient artifacts. It requires three frames: the repair frame and two neighboring frames. An
Optical Flow node is not required for generating motion vectors since the Repair Frame node
computes the optical flow. However, this can make it slow to process.
Repair Frame will not pass through, but rather destroys, any aux channels after the
computation is done.
See the Optical Flow node for controls and settings information.

TIP: If your footage varies in color from frame to frame, sometimes the repair can be
noticeable because, to fill in the hole, Repair Frame must pull color values from adjacent
frames. Consider using deflickering, color correction, or using a soft-edged mask to help
reduce these kinds of artifacts.

Inputs
There are two inputs on the Repair Frame node. One is used to connect a 2D image that will be
repaired and the other is for an effect mask.
– Input: The orange input is used for the primary 2D image that will be repaired.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the repairs to
certain areas.

Chapter 50 Optical Flow 1201

Basic Node Setup
The Repair Frame node analyzes the incoming MediaIn node and repairs single frame issues like dust
or scratches.

A Repair Frame node set up to analyze a MediaIn node using internal optical flow analysis.

Inspector

The Repair Frame Controls tab

Controls Tab
The Controls tab includes options for how to repair the frames. It also includes controls for adjusting
the optical flow analysis, identical to those controls in the Optical Flow node.

Depth Ordering
The Depth Ordering determines which parts of the image should be rendered on top by selecting
either Fastest On Top or Slowest On Top. The examples below best explain these options.
In a locked-off camera shot where a car is moving through the frame, the background does not move,
so it produces small, or slow, vectors, while the car produces larger, or faster, vectors.
The depth ordering in this case is Fastest On Top since the car draws over the background.
In a shot where the camera pans to follow the car, the background has faster vectors, and the car has
slower vectors, so the Depth Ordering method is Slowest On Top.

Clamp Edges
Under certain circumstances, this option can remove the transparent gaps that may appear on the
edges of interpolated frames. Clamp Edges causes a stretching artifact near the edges of the frame
that is especially visible with objects moving through it or when the camera is moving.
Because of these artifacts, it is a good idea to use clamp edges only to correct small gaps around the
edges of an interpolated frame.

Chapter 50 Optical Flow 1202

Edge Softness
This slider is displayed only when Clamp Edges is enabled. The slider helps to reduce the stretchy
artifacts that might be introduced by Clamp Edges.
If you have more than one of the Source Frame and Warp Direction checkboxes turned on, this can
lead to doubling up of the stretching effect near the edges. In this case, you’ll want to keep the
softness rather small at around 0.01. If you have only one checkbox enabled, you can use a larger
softness at around 0.03.

Source Frame and Warp Direction

These checkboxes allow you to choose which frames and vectors create the in-between frames.
Each method ticked on will be blended into the result.
– Prev Forward: Takes the previous frame and uses the Forward vector to interpolate
the new frame.
– Next Forward: Takes the next frame in the sequence and uses the Forward vector to interpolate
the new frame.
– Prev Backward: Takes the previous frame and uses the Back Forward vector to interpolate
the new frame.
– Next Backward: Takes the next frame in the sequence and uses the Back vector to interpolate
the new frame.

Optical Flow Options

These settings tweak the optical flow analysis. See the Classic and Advanced Controls section for the
Optical Flow node earlier in this chapter.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Optical Flow nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Smooth Motion [SM]

The Smooth Motion node

Smooth Motion Node Introduction

The Smooth Motion node smooths various AOV (Arbitrary Output Variables) channels in a clip using
optical flow to look at neighboring frames. It can be used for smoothing the Disparity channel in a
stereo 3D clip, where it helps reduce temporal edge/fringing artifacts, but it can also smooth a wide
range of channels like vectors, normals, and Z.
It is required that the image connected to the input on the node have precomputed Vector and Back
Vector channels; otherwise, this tool prints error messages in the Console window.
Check on the channels you want to temporally smooth. Be aware that if a channel selected for
smoothing is not present, Smooth Motion will not fail, nor will it print any error messages.

Chapter 50 Optical Flow 1203

It can also be used to smooth the Vector and Back Vector channels; however, sometimes, this can
make the interpolated results worse if there are conflicting motions or objects in the shot that move
around erratically, jitter, or bounce rapidly.

TIP: You can use two or more Smooth Motion nodes in sequence to get additional smoothing.
With one Smooth Motion node, the previous, current, and next frames are examined for a
total of 3; with two Smooth Motion nodes, 5 frames are examined; and with three Smooth
Motion nodes, 7 frames are examined.

Another technique using two Smooth Motion nodes is to use the first Smooth Motion node to smooth
the Vector and Back Vector channels. Use the second Smooth Motion to smooth the channels you
want to smooth (e.g., Disparity). This way, you use the smoothed vector channels to smooth Disparity.
You can also try using the smoothed motion channels to smooth the motion channels.

Inputs
The Smooth Motion node includes a single orange image input.
– Input: The orange image input accepts a 2D image. This is the sequence of images for which
you want to compute smooth motion. This image must have precomputed Vector and Back
Vector channels either generated from an Optical Flow node or saved in EXR format with
vector channels.

Basic Node Setup

The Smooth Motion node takes the output of the Optical Flow node for the required Vector and Back
Vector channels. The Smooth Motion node can then be used to smooth those channels or
AO channels.

A Smooth Motion node using Vector and Back Vector channels from the Optical Flow node.

Inspector

The Smooth Motion Controls tab

Chapter 50 Optical Flow 1204

Controls Tab
The Controls tab includes checkboxes for the channels you want to smooth. If a channel selected for
smoothing is not available in the input image, Smooth Motion will not fail, nor will it print any error
messages to the Console.

Channel
Smooth Motion can be applied to more than just the RGBA channels. It can also be applied to the
other AOV channels.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Optical Flow nodes. These common
controls are described in detail at the end of this chapter in “The Common Controls” section.

Tween [Tw]

The Tween node

Tween Node Introduction

Tween reconstructs a missing frame by interpolating between two neighboring frames using the
optical flow. Tween is nearly identical in functionality to Time Speed and Time Stretcher. The major
difference is that it works on two images that are not serial members of a sequence. As a
consequence, it cannot use the Vector or Back Vector aux channels stored in the images. The Tween
node manually generates the optical flow, so there is no need to add an Optical Flow node before the
Tween node. The generated optical flow is thrown away and is not stored back into the output frames.
Since optical flow is based on color matching, it is a good idea to color correct your images to match
ahead of time. Also, if you are having trouble with noisy images, it may also help to remove some of
the noise ahead of time.
Tween destroys any input aux channels. See the Optical Flow node for controls and settings
information.

Inputs
There are two image inputs on the Tween node and an effects mask input.
– Input 0: The orange input, labeled input 0, is the previous frame to the one you are generating.
– Input 1: The green input, labeled input 1, is the next frame after the one you are generating.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the Tween to
certain areas.

Chapter 50 Optical Flow 1205

Basic Node Setup
The Tween node receives two inputs for the two neighboring frames to the one you are generating.
Below, the previous frame, frame 01, is connected to the orange input 0. The next frame, frame 03, is
connected to the green input 1. The Tween node will generate frame 02 and output the sequence.

The Tween node receives two neighboring frames and generates the middle one.

Inspector

The Tween Controls tab

Controls Tab
The Controls tab includes options for how to tween frames. It also includes controls for adjusting the
optical flow analysis, identical to those controls in the Optical Flow node.

Interpolation Parameter
This option determines where the frame you are interpolating is, relative to the two source frames A
and B. An Interpolation Parameter of 0.0 will result in frame A, a parameter of 1.0 will result in frame B,
and a parameter of 0.5 will yield a result halfway between A and B.

Depth Ordering
The Depth Ordering determines which parts of the image should be rendered on top by selecting
either Fastest On Top or Slowest On Top. The examples below best explain these options.
In a locked-off camera shot where a car is moving through the frame, the background does not move,
so it produces small, or slow, vectors, while the car produces larger, or faster, vectors.

Chapter 50 Optical Flow 1206

The Depth Ordering in this case is Fastest On Top since the car draws over the background.
In a shot where the camera pans to follow the car, the background has faster vectors, and the car has
slower vectors, so the Depth Ordering method is Slowest On Top.

Clamp Edges
Under certain circumstances, this option can remove the transparent gaps that may appear on the
edges of interpolated frames. Clamp Edges causes a stretching artifact near the edges of the frame
that is especially visible with objects moving through it or when the camera is moving.
Because of these artifacts, it is a good idea to use Clamp Edges only to correct small gaps around the
edges of an interpolated frame.

Edge Softness
This slider is displayed only when Clamp Edges is enabled. The slider helps to reduce the stretchy
artifacts that might be introduced by Clamp Edges.
If you have more than one of the Source Frame and Warp Direction checkboxes turned on, this can
lead to doubling up of the stretching effect near the edges. In this case, you‘ll want to keep the
softness rather small at around 0.01. If you have only one checkbox enabled, you can use a larger
softness at around 0.03.

Source Frame and Warp Direction

These checkboxes allow you to choose which frames and vectors create the in-between frames.
Each method ticked on will be blended into the result.
– Prev Forward: Takes the previous frame and uses the Forward vector to interpolate
the new frame.
– Next Forward: Takes the next frame in the sequence and uses the Forward vector to interpolate
the new frame.
– Prev Backward: Takes the previous frame and uses the Back Forward vector to interpolate
the new frame.
– Next Backward: Takes the next frame in the sequence and uses the Back vector to interpolate
the new frame.

Optical Flow Options

These settings tweak the optical flow analysis. See the Class and Advanced Controls section for the
Optical Flow node earlier in this chapter.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Optical Flow nodes. These common
controls are described in detail in the following “The Common Controls” section.

Chapter 50 Optical Flow 1207

The Common Controls
Nodes that handle optical flow operations share a number of identical controls in the Inspector. This
section describes controls that are common among Optical Flow nodes.

Inspector

r
The Common Optical Flow Settings tab

Settings Tab
The Settings tab in the Inspector can be found on every tool in the Optical Flow category. The controls
are consistent and work the same way for each tool.

Blend
The Blend control is used to blend between the tool’s original image input and the tool’s final modified
output image. When the blend value is 0.0, the outgoing image is identical to the incoming image.
Normally, this causes the tool to skip processing entirely, copying the input straight to the output.

Process When Blend Is 0.0

The tool is processed even when the input value is zero. This can be useful when this node process is
scripted to trigger another task, but the value of the node is set to 0.0.

Red/Green/Blue/Alpha Channel Selector

These four buttons are used to limit the effect of the tool to specified color channels. This filter is often
applied after the tool has been processed.
For example, if the Red button on a Blur tool is deselected, the blur is first applied to the image, and
then the red channel from the original input will be copied back over the red channel of the result.

Chapter 50 Optical Flow 1208

There are some exceptions, such as tools for which deselecting these channels causes the tool to skip
processing that channel entirely. Tools that do this generally possess a set of identical RGBA buttons
on the Controls tab in the tool. In this case, the buttons in the Settings and the Controls tabs are
identical.

Apply Mask Inverted

Enabling the Apply Mask Inverted option inverts the complete mask channel for the tool. The mask
channel is the combined result of all masks connected to or generated in a node.

Multiply by Mask
Selecting this option causes the RGB values of the masked image to be multiplied by the mask
channel’s values. This causes all pixels of the image not included in the mask (i.e., set to 0) to become
black/transparent.

Use Object/Use Material (Checkboxes)

Some 3D software can render to file formats that support additional channels. Notably, the EXR file
format supports Object ID and Material ID channels, which can be used as a mask for the effect. These
checkboxes determine whether the channels are used, if present. The specific Material ID or Object ID
affected is chosen using the next set of controls.

Correct Edges
This checkbox appears only when the Use Object or Use Material checkboxes are selected. It toggles
the method used to deal with overlapping edges of objects in a multi-object image. When enabled,
the Coverage and Background Color channels are used to separate and improve the effect around the
edge of the object. If this option disabled (or no Coverage or Background Color channels are
available), aliasing may occur on the edge of the mask.
For more information on the Coverage and Background Color channels, see Chapter 18,
“Understanding Image Channels” in the Fusion Studio Reference Manual or Chapter 79 in the
DaVinci Resolve Reference Manual.

Object ID/Material ID (Sliders)

Use these sliders to select which ID is used to create a mask from the object or material channels of an
image. Use the Sample button in the same way as the Color Picker: to grab IDs from the image
displayed in the view. The image or sequence must have been rendered from a 3D software package
with those channels included.

Hide Incoming Connections

Enabling this checkbox can hide connection lines from incoming nodes, making a node tree appear
cleaner and easier to read. When enabled, empty fields for each input on a node are displayed in the
Inspector. Dragging a connected node from the node tree into the empty field hides that incoming
connection line as long as the node is not selected in the node tree. When the node is selected in the
node tree, the line reappears.

Comments
The Comments field is used to add notes to a tool. Click in the empty field and type the text. When a
note is added to a tool, a small red square appears in the lower-left corner of the node when the full
tile is displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the
note in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, see the Fusion scripting documentation.

Chapter 50 Optical Flow 1209

Chapter 51

Paint Node
This chapter details the Paint node available in Fusion.

The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Paint  1211
Paint Node Modifiers  1217
Keyboard Shortcuts  1218

Chapter 51 Paint Node 1210

Paint

The Paint node

Paint Node Introduction

Paint is an extremely flexible, stroke-based tool for wire and rig removal, image cloning, or to create
custom masks and mattes rapidly. Fusion’s paint can even be used to create new images and artistic
elements from scratch.
Each Paint node is made up of a series of brush strokes. These strokes are vector shapes with
editable brush, size, and effect. A wide range of apply modes and brush types are available.
Most Brushstrokes styles are editable polylines for fine control. They can be animated to change
shape, length, and size over time. The opacity and size of a stroke can be affected by velocity and
pressure when used with a supported tablet.
Unlimited undo and redo of paint provides the ability to experiment before committing changes to an
image sequence. Paint strokes can be reordered, deleted, and modified with virtually infinite flexibility.

Inputs
The two inputs on the Paint node are used to connect a 2D image and an effect mask which can be
used to limit the painted area.
– Input: It is required to connect the orange input with a 2D image that creates the size
of the “canvas” on which you paint.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the Paint to
only those pixels within the mask.

Basic Node Setup

The Paint node always needs an input connection. The simplest setup is to paint directly on the
incoming MediaIn node.

A MediaIn node connected directly to the Paint input

A more flexible setup is to use a Background node to set the size that matches the image you are
painting on. In the Inspector, the background would be set to be fully transparent. Then, the Paint tool
can be merged as the foreground over the actual image you want to paint on.

Chapter 51 Paint Node 1211

A Paint node merged over the top of a MediaIn for more flexibility

Types of Paint Strokes

Paint stroke types located in the viewer toolbar.

To begin working with the Paint tool, first select the paint stroke type from the Paint toolbar above the
viewer. There are ten stroke types to choose from as well as two additional tools for selecting and
grouping paint strokes. The stroke types and tools are described below in the order they appear in
the toolbar.
– Multistroke: Although this is the default selection and the first actual brush type in the toolbar,
Multistroke is not typically the stroke type most often used. However, it’s perfect for those
100-strokes-per-frame retouching paint jobs like removing tracking markers. Multistroke is much
faster than the Stroke type but is not editable after it is created. By default, Multistroke lasts for
one frame and cannot be modified after it has been painted. Use the Duration setting in the Stroke
controls to set the number of frames before painting. A shaded area of the Multistroke duration
is visible but not editable in the Keyframes Editor. While Multistrokes aren’t directly editable, they
can be grouped with the PaintGroup modifier, then tracked, moved, and rotated by animating the
PaintGroup instead.
– Clone Multistroke: Similar to Multistroke but specifically meant to clone elements from one
area or image to the other. Perfect for those 100-strokes-per-frame retouching paint jobs like
removing tracking markers. Clone Multistroke is faster than the Stroke type but is not editable
after it is created. By default, Clone Multistroke lasts for one frame and cannot be modified after
it has been painted. Use the Duration setting in the Stroke controls to set the number of frames
before painting. A shaded area of the Clone Multistroke duration is visible but not editable in the
Keyframes Editor.
– Stroke: In most cases, the Stroke tool is what people think of when they think of paint and is the
tool of choice for most operations. It is a fully animatable and editable vector-based paint stroke. It
can become slow if hundreds of strokes are used in an image; when creating a lot of paint strokes,
it is better to use Multistroke. The Stroke type has a duration of the entire global range. However,
you can edit its duration at any time in the Keyframes Editor. When the painting is complete,
choose the Select button in the Paint toolbar to avoid accidentally adding new strokes.
– Polyline Stroke: This provides the ability to create and manipulate a stroke in the same way that
a Bézier path or polygon mask might be created. To add a Polyline Stroke, select the Polyline
button and click in the viewer to add the first point. Continue clicking to add additional points to
the polyline. This click append style is the default, but polyline strokes can also be created in draw
append mode. Polylines can be tracked or connected to existing polylines like masks or animation
paths. The Polyline Stroke has a default duration of the entire global range. However, you can edit
its duration at any time in the Keyframes Editor.

Chapter 51 Paint Node 1212

 – Circle: Creates a circular shape with animatable control over radius and center. The Circle type
has a duration of the entire global range. However, you can edit its duration at any time in the
Keyframes Editor.
– Rectangle: Creates a rectangular area. The Rectangle type has a duration of the entire global
range. However, you can edit its duration at anytime in the Keyframes Editor.
– Copy Polyline: Allows you to create a closed Polyline area with animatable offset to clone
elements from one area or image to the other. The Copy Polyline type has a duration of the entire
global range. However, you can edit its duration at any time in the Keyframes Editor.
– Copy Circle/Rectangle: A circular or rectangular shape area with animatable offset to clone
elements from one image to the other. The Copy Circle/Rectangle type has a duration of the entire
global range. However, you can edit its duration at any time in the Keyframes Editor.
– Fill: Similar to the Wand mask tool. This tool fills similarly colored adjacent pixels with a fill color
based on the selected color channel. The tool has a duration of the entire global range. However,
you can edit its duration at any time in the Keyframes Editor
– Paint Group: Allows you to group multiple paint strokes with full control over center and size. Since
Multistroke and Clone Multistroke styles cannot be tracked, grouping these paint stroke types and
tracking the paint group is one common use of groups.

Editing Options Toolbar

Paint edit options are displayed in the viewer after a Polyline stroke is created.

Polyline-based paint strokes include a second toolbar in the viewer to select different editing options.
The paint strokes that include this second toolbar are Polyline Stroke and Copy Polyline. The Stroke
style also displays this toolbar after the stroke is selected and the Make Editable button is clicked in
the Inspector.
– Click Append: This is the default option when creating a polyline stroke. It works more like a
Bézier pen drawing tool than a paintbrush tool. Clicking sets a control point and appends the next
control point when you click again in a different location.
– Draw Append: This is a freehand drawing tool. It paints a stroke similar to drawing with a pencil on
paper. You can create a new Polyline Stroke or Copy Polyline Stroke using the Draw tool, or you
can extend a Stroke style after clicking the Make Editable button in the Inspector.
– Insert: Insert adds a new control point along the paint stroke spline.
– Modify: Modify allows you to safely move or smooth any exiting point along a spline without
worrying about adding a new point accidentally.
– Done: Prevents any point along the spline from being moved or modified. Also, new points cannot
be added. You can, however, move and rotate the entire spline.
– Closed: Closes an open polyline.
– Smooth: Changes the selected stroke or control point from a linear to a smooth curve.
– Linear: Changes the selected stroke or control point from a smooth curve to linear.
– Select All: Selects all the control points on the polyline.

Chapter 51 Paint Node 1213

 – Keys: Shows or hides the control points along the polyline.
– Handles: Shows or hides the Bézier handles along the polyline.
– Shape: Places a reshape rectangle around the selected polyline control points. Using the reshape
rectangle, you can deform groups of polyline control points or entire shapes much easier than
modifying each point.
– Delete: Deletes the selected control point(s).
– Reduce: Opens a Freehand precision window that can be used to reduce the number of control
points on a polyline. This can make the paint stroke easier to modify, especially if it has been
created using the Draw tool.
– Publish: You can use the Publish menu to either publish control points or the path. Publishing is
a form of parameter linking. It makes the selected item available for use by other controls, or to
attach a control point to a tracker.
– Follow Points: Allows a selected point to follow the path of a published point. The point follows
the published point using an offset position.
– Roto Assist: Enable the Roto Assist button when you begin painting with the Polyline Stroke tool.
The polyline points snap to the closest edge as you click to add points to the shape.  A cyan
outline indicates the points that have snapped to an edge. There are three main Roto Assist
options selectable through the drop-down menu:
– Multiple Points: When enabled, a single click on a high-contrast edge will add multiple points
to define the entire edge, instead of having to add each point individually. This is a one time
only click. The second click reverts to single point edge detection.
– Distance 8: Opens a dialog where you can set the pixel range within which searching for an
edge will take place.
– Reset: Used for resetting the snap attribute of all snapped points. After resetting, the points will
become unavailable for tracking.

Inspector

The Paint Controls tab

Chapter 51 Paint Node 1214

Controls Tab
Not all of the controls described here appear in all modes. Some controls are useful only in a specific
Paint mode and do not appear when they are not applicable. The Controls tab is used to configure
your paint settings before painting. Once a paint stroke is created, except for the Multistroke and
Clone Multistroke, you can select the stroke in the viewer and update the controls.

Brush Controls
Brush Shape
The brush shape buttons select the brush tip shape. Except for the single pixel shape, you can modify
the size of the brush shape in the viewer by holding down the Command or Ctrl key while dragging
the mouse.
– Soft Brush: The Soft Brush type is a circular brush tip with soft edges.
– Circular Brush: A Circular Brush is a brush tip shape with hard edges.
– Image Brush: The Image Brush allows images from any node in the node tree,
or from a file system, to be used as a brush tip.
– Single Pixel Brush: The Single Pixel Brush is perfect for fine detail work, creating a brush tip
precisely one pixel in size. No anti-aliasing is applied to the single pixel brush.
– Square Brush: A Square Brush is a brush tip with hard edges.

Vary Size
Vary size settings change the stroke size based on speed or a pressure-sensitive pen and tablet.
– Constant: The brush tip remains a constant size over the stroke.
– With Pressure: The stroke size varies with the actual applied pressure.
– With Velocity: The stroke size varies with the speed of painting. The faster the stroke,
the thinner it is.

Vary Opacity
Vary opacity settings change the stroke opacity based on speed or a pressure-sensitive pen
and tablet.
– Constant: The brush tip remains at a constant transparency setting over the entire stroke.
– With Pressure: The stroke transparency varies with the applied pressure.
– With Velocity: The stroke transparency varies with the speed of painting. The faster the stroke,
the more transparent it is.

Softness
Use this control to increase or decrease the Softness of a soft brush.

Image Source
When using the Image Source brush type, select between three possible source brush images.
– Node: The image source is derived from the output of a node in the node tree. Drag the node into
the Inspector’s Source node input field to set the source.
– Clip: The image source is derived from an image or sequence on disk. Any file supported by
Fusion’s Loader or MediaIn node can be used.
– Brush: Select an image to use as a brush from the menu. Images located in the Fusion > Brushes
directory are used to populate the menu.

Chapter 51 Paint Node 1215

Color Space
When the Fill tool is selected, a Color Space menu selects the color space when sampling colors
around the Fill tool center for inclusion in the fill range.

Channel
When the Fill tool is selected, a Channel menu selects which color channel is used in the fill paint. For
example, with alpha selected, the fill occurs on contiguous pixels of the alpha channel.

Apply Controls
Apply Mode
The Apply Modes are buttons that change a brush’s painting functionality.
– Color: The Color Apply Mode paints simple colored strokes. When used in conjunction with an
image brush, it can also be used to tint the image.
– Clone: The Clone Apply Mode copies an area from the same image using adjustable positions and
time offsets. This mode can also copy portions of one image into another image. Any image from
the node tree can be used as the source image.
– Emboss: The Emboss Apply Mode embosses the portions of the image covered by the
brush stroke.
– Erase: Erase reveals the underlying image through all other strokes, effectively erasing portions of
the strokes beneath it without actually destroying the strokes.
– Merge: This Apply Mode effectively merges the brush onto the image. This mode behaves in
much the same way as the Color Apply Mode but has no color controls. It is best suited for use
with the image brush type.
– Smear: Smear the image using the direction and strength of the brushstroke as a guide.
– Stamp: Stamps the brush onto the image, completely ignoring any alpha channel or transparency
information. This mode is best suited for applying decals to the target image.
– Wire: This Wire Removal Mode is used to remove wires, rigging, and other small elements in the
frame by sampling adjacent pixels and draw them in toward the stroke.

Stroke Controls
The stroke controls contain parameters that adjust the entire stroke of paint as well as control it
over time.
– Size: This control adjusts the size of the brush when the brush type is set to either Soft Brush or
Circle. The diameter of the brush is drawn in the viewer as a small circle surrounding the mouse
pointer. The size can also be adjusted interactively in the viewer by holding the Command or Ctrl
key while dragging the mouse pointer.
– Spacing: The Spacing slider determines the distance between dabs (samples used to draw a
continuous stroke along the underlying vector shape). Increasing this value increases the density
of the stroke, whereas decreasing this value causes the stroke to assume the appearance of a
dotted line.
– Stroke Animation: The Stroke Animation menu provides several pre-built animation effects
that can be applied to a paint stroke. This menu appears only for vector strokes like Stroke and
Polyline Stroke.
– All Frames: This default displays the stroke for all frames of the image connected to the orange
background input of the Paint node.
– Limited Duration: This exists on the number of frames specified by the Duration slider.

Chapter 51 Paint Node 1216

 – Write On: When Write On is selected, an animation spline is added to the paint stroke that
precisely duplicates the timing of the paint stroke’s creation. The stroke is written on the image
exactly as it was drawn. To adjust the timing of the Write On effect, switch to the Spline Editor
and use the Time Stretcher node to adjust the overall length of the animation spline. To smooth
or manually adjust the motion, try reducing the points in the animation spline.
– Write Off: Write Off performs the reverse of Write On, drawing the stroke starting from the end
and working backward to the start of the stroke.
– Write On Then Off: This menu option applies a Write On and then a Write Off animation mode
to the stroke.
– Trail: Selecting the Trail option from the menu causes both the start and end points of the
stroke to be animated simultaneously, offset from each other by the amount specified in the
Duration control. This has the effect of creating a painted segment that follows the stroke path.
As with the Write On and Write Off effects, this starts on the current frame when the animation
mode is selected. The timing of the animation can be adjusted manually using the Spline or
Keyframes Editors.

– Duration: Duration sets the duration of each stroke in frames. This control is present only for
Multistroke and Clone Multistroke, or when the stroke animation mode is set to Limited Duration. It
is most commonly employed for frame-by-frame rotoscoping through a scene.
Each Vector stroke applied to a scene has a duration in the Keyframes Editor that can be trimmed
independently from one stroke to the next. The duration can be set to 0.5, which allows each
stroke to last for a single field only when the node tree is processing in Fields mode.
– Write On and Write Off: This range slider appears when the Stroke Animation is set to one of
the Write On and Write Off methods. The range represents the beginning and end points of the
stroke. Increase the Start value from 0.0 to 1.0 to erase the stroke, or increase the End value from
0.0 to 1.0 to draw the stroke on the screen. This control can be animated to good effect. It works
most effectively when automatically animated through the use of the Write On and Write Off
modes of the Stroke Animation menu.
– Make Editable: This button appears only for Vector strokes. Clicking on Make Editable turns the
current stroke into a polyline spline so that the shape can be adjusted or animated.

Paint Node Modifiers

Every paint stroke created in the viewer creates an associated modifier stroke. These modifier strokes
are represented as a list of paint stroke operations in the Modifiers tab of the Inspector. Each stroke
you create can be modified or deleted, or applied in a different order using the modifier stack.

NOTE: The MultiStroke tools are built for speed and can contain many strokes internally
without creating a huge list stack in the modifiers

Each Paint modifier stroke contains Brush controls, Apply controls, and Stroke controls identical to
those found in the main Controls tab of the Inspector.

Chapter 51 Paint Node 1217

Keyboard Shortcuts
Keyboard shortcuts allow you to adjust painting styles and color without having to navigate menus.

While painting:
Hold Command or Ctrl while left-dragging to change brush size.
Hold Option or Alt while clicking to pick a color in the viewer.

While cloning:
Option-click or Alt-click to set the clone source position. Strokes start cloning
from the selected location.
Hold O to temporarily enable a 50% transparent overlay of the clone source
(% can be changed with pref Tweaks.CloneOverlayBlend).
Press P to toggle an opaque overlay of the clone source.

While overlay is showing:

Arrow keys change the clone source position; you can also drag crosshairs and adjust angle
control or size sliders.
Option + Left/Right or Alt + Left/Right Arrow keys change the clone source angle.
Option + Up/Down or Alt + Up/Down Arrow keys change the clone source size.
Shift + Command or Shift + Ctrl can be used with the above for greater or lesser adjustments.
Left and right square brackets, [ and ], change the clone source Time Offset (this requires a
specific Clone Source node to be set in the Source Node field).

Copy Rect/Ellipse:
Shift + drag out the source to constrain the shape.

With a single stroke selected (not available on multi or polyline strokes):

Press X or Y to flip the stroke.

Paint Groups:
Command + drag or Ctrl + drag to change the position of a group’s crosshair, without
changing the position of the group.

Chapter 51 Paint Node 1218

Chapter 52

Particle Nodes
This chapter details the Particle nodes available in Fusion.

The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents

Particle Nodes  1220 pGradientForce [pGF]  1243

pAvoid [pAv]  1220 pImage Emitter [pLE]  1245

pBounce [pBn]  1222 pKill [pKI]  1248

pChangeStyle [pCS]  1224 pMerge [pMg]  1249

pCustom [pCu]  1226 pPoint Force [pPF]  1250

pCustomForce [pCF]  1230 pRender [pRn]  1251

pDirectionalForce [pDF]  1231 pSpawn [pSp]  1258

pEmitter [pEm]  1233 pTangent Force [pTF]  1260

pFlock [pFl]  1238 pTurbulence [pTr]  1261

pFollow [pFo]  1240 pVortex [pVt]  1263

pFriction [pFr]  1242 The Common Controls  1264

Chapter 52 Particle Nodes 1219

Particle Nodes
The Particle nodes are used to generate a large number of duplicated objects that automatically
animate. They are used to create elements like falling rain, fireworks, smoke, pixie dust, and much
more. There are endless possibilities. Particles in Fusion consist of a set of nodes that are strung
together in a chain for generating, modifying, and rendering particles in a 2D or 3D scene.
To begin, every particle system you create must contain two fundamental nodes:
– pEmitter: Used to generate the particles and control their basic look, motion. and behavior.
– pRender: Used to render the output of the pEmitter into a 2D or 3D scene. When creating
particles, you only ever view the pRender node.

Particle Nodes

The remaining particle nodes modify the pEmitter results to simulate natural phenomena like gravity,
flocking, and bounce. The names of particle nodes all begin with a lowercase p to differentiate them
from non-particle nodes. They can be found in the particles category in the Effects Library.

pAvoid [pAv]

The pAvoid node

pAvoid Node Introduction

The pAvoid node is used to create a region or area within the image that affected particles attempt to
avoid entering and crossing.
It has two primary controls. The first determines the distance from the region a particle should be
before it begins to move away from the region. The second determines how strongly the particle
moves away from the region.
A pAvoid node creates a “desire” in a particle to move away from a specific region. If the velocity of
the particle is stronger than the combined distance and strength of the pAvoid region, the particle’s
desire to avoid the region does not overcome its momentum and the particle crosses that
region regardless.

Chapter 52 Particle Nodes 1220

Inputs
The pAvoid node has a single orange input by default. Like most particle nodes, this orange input
accepts only other particle nodes. A green bitmap or mesh input appears on the node when you set
the Region menu in the Region tab to either Bitmap or Mesh.
– Input: The orange input takes the output of other particle nodes.
– Region: The green or magenta region input takes a 2D image or a 3D mesh depending on
whether you set the Region menu to Bitmap or Mesh. The color of the input is determined by
whichever is selected first in the menu. The 3D mesh or a selectable channel from the bitmap
defines the area particles avoid.

Basic Node Setup

The pAvoid node is placed in between the pEmitter and pRender. A Shape 3D node is used to create
the region the particles will avoid.

A pAvoid node using a Shape 3D node as the region to avoid

Inspector

The pAvoid controls

Randomize
The Random Seed slider and Randomize button are presented whenever a Fusion node relies on a
random result. Two nodes with the same seed values will produce the same random results. Click the
Randomize button to randomly select a new seed value, or adjust the slider to manually select a new
seed value.

Distance
Determines the distance from the region a particle should be before it begins to move away from
the region.

Strength
Determines how strongly the particle moves away from the region. Negative values make the particles
move toward the region instead.

Chapter 52 Particle Nodes 1221

Common Controls
Conditions, Style, Region, and Settings Tabs
The Conditions, Style, Region, and Settings tabs are common to all Particle nodes, so their
descriptions can be found in “The Common Controls” section at the end of this chapter.

pBounce [pBn]

The pBounce node

pBounce Node Introduction

The pBounce tool is used to create a region from which affected particles will bounce off when they
come into contact.

Inputs
The pBounce node has a single orange input by default. Like most particle nodes, this orange input
accepts only other particle nodes. A green or magenta bitmap or mesh input appears on the node
when you set the Region menu in the Region tab to either Bitmap or Mesh.
– Input: The orange input takes the output of other particle nodes.
– Region: The green or magenta region input takes a 2D image or a 3D mesh depending on
whether you set the Region menu to Bitmap or Mesh. The color of the input is determined by
whichever is selected first in the menu. The 3D mesh or a selectable channel from the bitmap
defines the area particles bounce off.

Basic Node Setup

The pBounce node is placed in between the pEmitter and pRender. A Shape 3D node is used to
create the region the particles bounce off.

A pBounce node using a Shape 3D node as the region on which particles bounce off

Chapter 52 Particle Nodes 1222

Inspector

The pBounce controls

Randomize
The Random Seed slider and Randomize button are presented whenever a Fusion node relies on a
random result.
Two nodes with the same seed values will produce the same random results. Click the Randomize
button to randomly select a new seed value, or adjust the slider to manually select a new seed value.

Elasticity
Elasticity affects the strength of a bounce, or how much velocity the particle will have remaining after
impacting upon the Bounce region. A value of 1.0 will cause the particle to possess the same velocity
after the bounce as it had entering the bounce. A value of 0.1 will cause the particle to lose 90% of its
velocity upon bouncing off of the region.
The range of this control is 0.0 to 1.0 by default, but greater values can be entered manually. This will
cause the particles to gain momentum after an impact, rather than lose it. Negative values will be
accepted but do not produce a useful result.

Variance
By default, particles that strike the Bounce region will reflect evenly off the edge of the Bounce region,
according to the vector or angle of the region. Increasing the Variance above 0.0 will introduce a
degree of variation to that angle of reflection. This can be used to simulate the effect of a
rougher surface.

Spin
By default, particles that strike the region will not have their angle or orientation affected in any way.
Increasing or decreasing the Spin value will cause the Bounce region to impart a spin to the particle
based on the angle of collision, or to modify any existing spin on the particle. Positive values will
impart a forward spin, and negative values impart a backward spin. The larger the value, the faster the
spin applied to the particle will be.

Roughness
This slider varies the bounce off the surface to slightly randomize particle direction.

Surface Motion
This slider makes the bounce surface behave as if it had motion, thus affecting the particles.

Surface Motion Direction

This thumbwheel control sets the angle relative to the bounce surface.

Chapter 52 Particle Nodes 1223

Common Controls
Conditions, Style, Region, and Settings Tabs
The Conditions, Style, Region, and Settings tabs are common to all Particle nodes, so their
descriptions can be found in “The Common Controls” section at the end of this chapter.

pChangeStyle [pCS]

The pChange Style node

pChange Style Node Introduction

The pChange Style node provides a mechanism for changing the appearance or style of particles that
interact with a defined region. The primary controls mirror those found in the Style tab of the pEmitter
node. Particles that intersect or enter the defined region change based on the parameters of
this node.
Except for the pCustom node, this is the only node that modifies the particles’ appearance rather than
its motion. It is often used to trigger a change in the appearance in response to some event, such as
striking a barrier.

Inputs
The pChange Style node has a single orange input by default. Like most particle nodes, this orange
input accepts only other particle nodes. A green or magenta bitmap or mesh input appears on the
node when you set the Region menu in the Region tab to either Bitmap or Mesh.
– Input: The orange input takes the output of other particle nodes.
– Region: The green or magenta region input takes a 2D image or a 3D mesh depending on
whether you set the Region menu to Bitmap or Mesh. The color of the input is determined by
whichever is selected first in the menu. The 3D mesh or a selectable channel from the bitmap
defines the area where the custom particle node takes effect.

Basic Node Setup

Opposite of what you may think, to create a change in style that appears to be caused by some
physical event, the pChange Style node should be placed before the node that creates the event. For
example, below, the particles generated by the Emitter node change style after bouncing off a
pBounce. Both the pChange Style and pBounce use the same Shape 3D node as the region. The
pChange Style must be placed before the pBounce. If the pChange Style node is placed after the
pBounce, the particles bounce off the region before the pChange Style calculates its effect. The
particle will never get to intersect with the pChange Style node’s region, and so the style
never changes.

Chapter 52 Particle Nodes 1224

A pChange Style node placed before the pBounce node

Inspector

The pChange Style controls

Randomize
The Random Seed slider and Randomize button are presented whenever a Fusion node relies on a
random result. Two nodes with the same seed values will produce the same random results. Click the
Randomize button to randomly select a new seed value, or adjust the slider to manually select a new
seed value.

Change Sets
This option allows the user to change the particle’s Set to become influenced by forces other than the
original particle. See “The Common Controls” in this chapter to learn more about Sets.

Chapter 52 Particle Nodes 1225

Style
This option allows the user to change the particle’s Style and thus the look. See
“The Common Controls” in this chapter to learn more about Styles.

Common Controls
Conditions, Style, Region, and Settings Tabs
The Conditions, Style, Region, and Settings tabs are common to all Particle nodes, so their
descriptions can be found in “The Common Controls” section at the end of this chapter.

pCustom [pCu]

The pCustom node

pCustom Node Introduction

The pCustom node is used to create custom expressions that affect the properties of particles. This
node is similar to the Custom node, except the calculations affect particles rather than pixels.

Inputs
The pCustom node has three inputs. Like most particle nodes, this orange input accepts only other
particle nodes. The green and magenta inputs are 2D image inputs for custom image calculations.
Optionally, there are teal or white bitmap or mesh inputs, which appear on the node when you set the
Region menu in the Region tab to either Bitmap or Mesh.
– Input: The orange input takes the output of other particle nodes.
– Image 1 and 2: The green and magenta image inputs accept 2D images that are used for per
pixel calculations and compositing functions.
– Region: The teal or white region input takes a 2D image or a 3D mesh depending on whether
you set the Region menu to Bitmap or Mesh. The color of the input is determined by whichever
is selected first in the menu. The 3D mesh or a selectable channel from the bitmap defines the
area where the custom particle node takes effect.

Basic Node Setup

The pCustom node is placed in between the pEmitter and pRender. A Shape 3D node is used to
create the region where the Custom particle event occurs.

A pCustom node using a Shape 3D node as the region where the custom event occurs

Chapter 52 Particle Nodes 1226

Inspector
All the same operators, functions, and conditional statements described for the Custom node apply to
the pCustom node as well, including Pixel-read functions for the two image inputs (e.g., getr1w(x,y),
getz2b(x,y), and so on).

The pCustom controls

Number 1-8
Numbers are variables with a dial control that can be animated or connected to modifiers exactly as
any other control might. The numbers can be used in equations on particles at current time: n1, n2, n3,
n4, … or at any time: n1_at(float t), n2_at(float t), n3_at(float t), n4_at(float t), where t is the time you
want. The values of these controls are available to expressions in the Setup and Intermediate tabs.

pCustom Position tab

Position 1-8
These eight point controls include 3D X,Y,Z position controls. They are normal positional controls and
can be animated or connected to modifiers as any other node might. They are available to expressions
entered in the Setup, Intermediate, and Channels tabs.

Chapter 52 Particle Nodes 1227

pCustom Setup tab

Setup 1-8
Up to eight separate expressions can be calculated in the Setup tab of the pCustom node. The Setup
expressions are evaluated once per frame, before any other calculations are performed. The results
are then made available to the other expressions in the node as variables s1, s2, s3, and s4.
Think of them as global setup scripts that can be referenced by the intermediate and channel scripts.

pCustom Intermediate tab

Chapter 52 Particle Nodes 1228

Inter 1-8
An additional eight expressions can be calculated in the Intermediate tab. The Intermediate
expressions are evaluated once per frame, after the Setup expressions are evaluated. Results are
available as variables i1, i2, i3, i4, i5, i6, i7, i8, which can be referenced by channel scripts.

Particle
Particle position, velocity, rotation, and other controls are available in the Particle tab.
The following particle properties are exposed to the pCustom control:

The following particle properties are exposed to the pCustom control:

px, py, pz particle position on the x, y, and z axis

vx, vy, vz particle velocity on the x, y and z axis

rx, ry, rz particle rotation on the x, y, and z axis

sx, sy, sz particle spin on the x, y, and z axis

pxi1, pyi1 the 2d position of a particle, corrected for image 1’s aspect

pxi2, pyi2 the 2d position of a particle, corrected for image 2’s aspect

mass not currently used by anything

size the current size of a particle

id the particle’s identifier

r, g, b, a the particles red, green, blue and alpha color values

rgnhit this value is 1 if the particle hit the pCustom node’s defined region

rgndist this variable contains the particles distance from the region

condscale the strength of the region at the particle’s position

rgnix, rgniy, rgniz values representing where on the region the particle hit

rgnnx, rgnny, rgnnz region surface normal of the particle when it hit the region

w1, h1 image 1 width and height

w2 h2 image 2 width and height

i1, i2, i3, i4 the result of the intermediate calculations 1 through 4

s1, s2, s3, s4 the result of the setup calculations 1 through 4

n1..n8 the values of numeric inputs 1 through 8

p1x, p1y, p1z .. p4x, p4y,

the values of position inputs 1 through 4
p4z

time the current time or frame of the compositions

age the current age of the particle

lifespan the lifespan of the current particle

Common Controls
Conditions, Style, Region, and Settings Tabs
The Conditions, Style, Region, and Settings tabs are common to all Particle nodes, so their
descriptions can be found in “The Common Controls” section at the end of this chapter.

Chapter 52 Particle Nodes 1229

pCustomForce [pCF]

The pCustom Force node

pCustom Force Node Introduction

The pCustom Force node allows you to change the forces applied to a particle system or subset. This
node is one of the most complex and the most powerful node in Fusion. If you are experienced with
scripting or C++ programming, you should find the structure and terminology used by the Custom
Force node to be familiar.
The forces on a particle within a system can have their positions and rotations affected by forces. The
position in XYZ and the Torque, which is the spin of the particle, are controlled by independent custom
equations. The Custom Force node is used to create custom expressions and filters to modify the
behavior. In addition to providing three image inputs, this node will allow for the connection of up to
eight numeric inputs and as many as four XY position values from other controls and parameters in the
node tree.

Inputs
The pCustom Force node has three inputs. Like most particle nodes, this orange input accepts only
other particle nodes. A green and magenta are 2D image inputs for custom image calculations.
Optionally there are teal or white bitmap or mesh inputs, which appear on the node when you set the
Region menu in the Region tab to either Bitmap or Mesh.
– Input: The orange input takes the output of other particle nodes.
– Image 1 and 2: The green and magenta image inputs accept 2D images that are used for per-
pixel calculations and compositing functions.
– Region: The teal or white region input takes a 2D image or a 3D mesh depending on whether
you set the Region menu to Bitmap or Mesh. The color of the input is determined by whichever
is selected first in the menu. The 3D mesh or a selectable channel from the bitmap defines the
area where the pCustom Force takes effect.

Basic Node Setup

The pCustom Force node is inserted between a pEmitter and pRender node to serve as a catalyst for
particles using advanced C++ and scripting.

A pCustom Force is applied to the particles generated by the pEmitter.

Chapter 52 Particle Nodes 1230

Inspector

The pCustom Force controls

The tabs and controls located in the Inspector are similar to the controls found in the pCustom node.
Refer to the pCustom node in this chapter for more information.

Common Controls
Conditions, Style, Region, and Settings Tabs
The Conditions, Style, Region, and Settings tabs are common to all Particle nodes, so their
descriptions can be found in “The Common Controls” section at the end of this chapter.

pDirectionalForce [pDF]

The pDirectional Force node

pDirectional Force Node Introduction

This node applies a unidirectional force that pulls the affected particles in a specified direction.
Its primary controls affect the strength of the force, and the angle of the force’s pull along the
X, Y, and Z axis.
Since the most common use of this node is to simulate gravity, the default direction of the pull is down
along the Y axis (-90 degrees), and the default behavior is to ignore regions and affect all particles.

Inputs
The pDirectional Force node has a single orange input by default. Like most particle nodes, this
orange input accepts only other particle nodes. A green or magenta bitmap or mesh input appears on
the node when you set the Region menu in the Region tab to either Bitmap or Mesh.
– Input: The orange input takes the output of other particle nodes.
– Region: The green or magenta region input takes a 2D image or a 3D mesh depending on
whether you set the Region menu to Bitmap or Mesh. The color of the input is determined by
whichever is selected first in the menu. The 3D mesh or a selectable channel from the bitmap
defines the area where the directional force takes effect.

Chapter 52 Particle Nodes 1231

Basic Node Setup
The pDirectional Force node is placed in between the pEmitter and pRender and is often used to
create gravity.

A pDirectional Force node placed between the pEmitter and pRender nodes

Inspector

The pDirectional Force controls

Randomize
The Random Seed slider and Randomize button are presented whenever a Fusion node relies on a
random result. Two nodes with the same seed values will produce the same random results. Click the
Randomize button to select a new seed value randomly, or adjust the slider to select a new seed
value manually.

Strength
Determines the power of the force. Positive values will move the particles in the direction set by the
controls; negative values will move the particles in the opposite direction.

Direction
Determines the direction in X/Y space.

Direction Z
Determines the direction in Z space.

Chapter 52 Particle Nodes 1232

Common Controls
Conditions, Style, Region, and Settings Tabs
The Conditions, Style, Region, and Settings tabs are common to all Particle nodes, so their
descriptions can be found in “The Common Controls” section at the end of this chapter.

pEmitter [pEm]

The pEmitter node

pEmitter Node Introduction

The pEmitter node is the main source of particles (pImage Emitter is another) and will usually be the
first node used in any new particle system. This node contains controls for setting the initial position,
orientation, and motion of the particles, as well as controls for the visual style of each particle.
Like all other Particle nodes (with the exception of the pRender node), the pEmitter produces a particle
set, not a visible image, and therefore cannot be displayed directly in a viewer. To view the output of a
particle system, add a pRender node after the pEmitter.

Inputs
By default, the pEmitter node has no inputs at all. You can enable an image input by selecting Bitmap
from the Style menu in the Style tab. Also, two region inputs, one for bitmap and one for mesh, appear
on the node when you set the Region menu in the Region tab to either Bitmap or Mesh. The colors of
these inputs change depending on the order in which they are enabled.
– Style Bitmap Input: This image input accepts a 2D image to use as the particles’ image. Since
this image duplicates into potentially thousands of particles, it is best to keep these images
small and square—for instance, 256 x 256 pixels.
– Region: The region inputs takes a 2D image or a 3D mesh depending on whether you set the
Region menu to Bitmap or Mesh. The color of the input is determined by whichever is selected
first in the menu. The 3D mesh or a selectable channel from the bitmap defines the area where
the particles are emitted.

Basic Node Setup

The pEmitter node starts the branch of a particle system that always ends with a pRender node. The
pEmitter can feed directly into a pRender node to feed other particle nodes.

A pEmitter node connected to a pRender node is a typical setup for more particle systems.

Chapter 52 Particle Nodes 1233

Inspector
The pEmitter inspector is divided into four main tabs and a common settings tab. The controls tab is
the first tab displayed and it contains settings that affect the general setup of the particle cells emitted
by the node. These settings do not directly affect the appearance of the cells or shape of the emitter
region. They modify fundamental behaviors like quantity, duration, speed, and rotation of the
particle cells.

The pEmitter controls

Randomize and Random Seed

The Random Seed slider is used to seed all the variance and random number generators used by the
node when creating the particle system. Two pEmitter nodes with exactly the same settings for all
controls and the same random seed will generate exactly the same particle system. Changing the
random seed will cause variation between the nodes. Click the Randomize button to automatically set
a randomly chosen value for the Random Seed.

Chapter 52 Particle Nodes 1234

Number
This control is used to set the amount of new particles generated on each frame. A value of 1 would
cause one new particle to be generated each frame. By frame 10, there would be a total of 10 particles
in existence (unless Particle Lifespan was set to fewer than 10 frames).
Animate this parameter to specify the number of particles generated in total. For example, if only 25
particles in total are desired, animate the control to produce five particles on frame 0–4, then set a key
on frame five to generate zero particles for the remainder of the project.

Number Variance
This modifies the amount of particles generated for each frame, as specified by the Number control.
For example, if Number is set to 10.0 and Number Variance is set to 2.0, the emitter will produce
anywhere from 9-11 particles per frame. If the value of Number Variance is more than twice as large as
the value of Number, it is possible that no particles will be generated for a given frame.

Lifespan
This control determines how long a particle will exist before it disappears or ‘dies.’ The default value of
this control is 100 frames, although this can be set to any value. The timing of many other particle
controls is relative to the Lifespan of the particle. For example, the size of a particle can be set to
increase over the last 80% of its life, using the Size Over Life graph in the Style tab of the pEmitter.

Lifespan Variance
Like Number Variance, the Lifespan Variance control allows the Lifespan of particles produced to be
modified. If Particle Lifespan was set to 100 frames and the Lifespan Variance to 20 frames, particles
generated by the emitter would have a lifespan of 90–110 frames.

Color
This provides the ability to specify from where the color of each particle is derived. The default setting
is Use Style Color, which will provide the color from each particle according to the settings in the Style
tab of the pEmitter node.
The alternate setting is Use Color From Region, which overrides the color settings from the Style tab
and uses the color of the underlying bitmap region.
The Use Color From Region option only makes sense when the pEmitter region is set to use a bitmap
produced by another node in the composition. Particles generated in a region other than a bitmap
region will be rendered as white when the Use Color From Region option is selected.

Position Variance
This control determines whether or not particles can be ‘born’ outside the boundaries of the pEmitter
region. By default, the value is set to zero, which will restrict the creation area for new particles to the
exact boundaries of the defined region. Increasing this control’s value above 0.0 will allow the particle
to be born slightly outside the boundaries of that region. The higher the value, the ‘softer’ the region’s
edge will become.

Temporal Distribution
In general, an effect is processed per frame, based on the comp frame rate. However, processing
some particles only at the exact frame boundaries can cause pulsing. To make the behavior subtly
more realistic, the particles can be birthed in subframe increments.
The default, At The Same Time setting renders on frame boundaries, where as the other two settings
take advantage of sub frame rendering. Randomly Distributed randomizes birth times +/- around the
frame number, eg birth 10 particles at random sub times 24.1 24.85, 24.21, 24.37 etc. one particle at a
time. Evenly Distributed births particles at regular sub times, eg 10 particles, birth 1 at at time at 24.0,
24.1, 24.2, 24.3, 24.4, 24.5 ... 24.8, 24.9.

Chapter 52 Particle Nodes 1235

These settings are influenced by the Sub Frame Accuracy setting in the pRender node. The Sub
Frame Accuracy slider controls how many in-between frames are calculated between each frame. The
higher the value the more accurate the particle calculation but the longer the render times.

Velocity
The controls in the Velocity section determine the speed and direction of the particle cells as the are
generated from the emitter region.

Velocity and Velocity Variance

These determine the initial speed or velocity of new particles. By default, the particle has no velocity
and will not move from its point of origin unless acted upon by outside forces. A velocity setting of 10.0
would cause the particle to cross the entire width of the image in one step so a velocity of 1.0 would
cause the particle to cross the width of the image over 10 frames.
Velocity Variance modifies the velocity of each particle at birth, in the same manner described in
Lifespan Variance and Number Variance above.

Inherit
Inherit Velocity passes the emitter region’s velocity on to the particles. This slider has a wide range
that includes negative and positive values. A negative value causes the particles to move in the
opposite direction, a value of 1 will cause the particles to move with a velocity that matches the emitter
region’s velocity, and a value of 2 causes the particles to move ahead of the emitter region.

Angle and Angle Variance

This determines the angle at which particles with velocity applied will be heading at their birth.

Angle Z and Angle Z Variance

This is as above, except this control determines the angle of the particles along the Z space axis
(toward or away from the camera).

Rotation
Rotation controls are used to set the orientation of particle cells and animating that orientation
over time .

Rotation Mode
This menu control provides two options to help determine the orientation of the particles emitted.
When the particles are spherical, the effect of this control will be unnoticeable.
– Absolute Rotation: The particles will be oriented as specified by the Rotation controls, regardless
of velocity and heading.
– Rotation Relative To Motion: The particles will be oriented in the same direction as the
particle is moving. The Rotation controls can now be used to rotate the particle‘s orientation away
from its heading.

Rotation XYZ and Rotation XYZ Variance

These controls allow for Rotation of the individual particles. This can be particularly useful when
dealing with a bitmap particle type, as the incoming bitmap may not be oriented in the desired
direction.
Rotation XYZ Variance can be used to randomly vary the rotation by a specified amount around the
center of the Rotation XYZ value to avoid having every particle oriented in the exact same direction.

Spin
Spin controls are auto animated controls that change the orientation of particle cells over time.

Chapter 52 Particle Nodes 1236

Spin XYZ and Spin Variance
These provide a spin to be applied to each particle at birth. The particles will rotate ‚x‘ degrees each
frame, as determined by the value of Spin XYZ.
The Spin XYZ variances will vary the amount of rotation applied to each frame in the manner
described by Number Variance and Lifespan Variance documented above.

Sets Tab
This tab contains settings that affect the physics of the particles emitted by the node. These settings
do not directly affect the appearance of the particles. Instead, they modify behavior like velocity, spin,
quantity, and lifespan.

The pEmitter Sets tab

Set 1-32
To assign the particles created by a pEmitter to a given set, simply select the checkbox of the set
number you want to assign. A single pEmitter node can be assigned to one or multiple sets. Once they
are assigned in the pEmitter, you can enable sets in other particle nodes so they only affect particles
from specific pEmitters.

Style Tab
The Style tab provides controls that affect the appearance of the particles. For detailed information
about the style Tab, see the “The Common Controls” section at the end of this chapter.

The pEmitter Style tab

Chapter 52 Particle Nodes 1237

Region Tab
The Region tab controls the shape, size, and location of the the area that emits the particle cells. This
is often called the Emitter. Only one emitter region can be set for a single pEmitter node. If the
pRender is set to 2D, then the emitter region will produce particles along a flat plane in Z Space. 3D
emitter regions possess depth and can produce particles inside a user-defined, three-dimensional
region. For more detail on the Region tab, see “The Common Controls” section at the end of
this chapter.

Common Controls
Conditions, Style, Region, and Settings Tabs
The Conditions, Style, Region, and Settings tabs are common to all Particle nodes, so their
descriptions can be found in “The Common Controls” section at the end of this chapter.

pFlock [pFl]

The pFlock node

pFlock Node Introduction

The pFlock node can be used to simulate the behavior of organic systems, such as a flock of birds or a
colony of ants. Its use can make an otherwise mindless particle system appear to be motivated, or
acting under the direction of intelligence.
The pFlock node works through two basic principles. Each particle attempts to stay close to other
particles and each particle attempts to maintain a minimum distance from other particles.
The strength of these “desires” produces the seemingly motivated behavior perceived by the viewer.

Inputs
The pFlock node has a single orange input by default. Like most particle nodes, this orange input
accepts only other particle nodes. A green or magenta bitmap or mesh input appears on the node
when you set the Region menu in the Region tab to either Bitmap or Mesh.
– Input: The orange background input takes the output of other particle nodes.
– Region: The green or magenta region input takes a 2D image or a 3D mesh depending on
whether you set the Region menu to Bitmap or Mesh. The color of the input is determined by
whichever is selected first in the menu. The 3D mesh or a selectable channel from the bitmap
defines the area where the flocking takes effect.

Basic Node Setup

When combined with pFollow, the pFlock node can produce natural swarming behaviors that change
direction.

Chapter 52 Particle Nodes 1238

A pFlock node applying more herd-type mentality to particles

Inspector

The pFlock controls

Randomize
The Random Seed slider and Randomize button are presented whenever a Fusion node relies on a
random result. Two nodes with the same seed values will produce the same random results. Click the
Randomize button to randomly select a new seed value, or adjust the slider to manually select a new
seed value.

Flock Number
The value of this control represents the number of other particles that the affected particle will attempt
to follow. The higher the value, the more visible “clumping” will appear in the particle system and the
larger the groups of particles will appear.

Follow Strength
This value represents the strength of each particle’s desire to follow other particles. Higher values will
cause the particle to appear to expend more energy and effort to follow other particles. Lower values
increase the likelihood that a given particle will break away from the pack.

Chapter 52 Particle Nodes 1239

Attract Strength
This value represents the strength of attraction between particles. When a particle moves farther from
other particles than the Maximum Space defined in the pFlock node, it will attempt to move closer to
other particles. Higher values cause the particle to maintain its spacing energetically, resolving
conflicts in spacing more rapidly.

Repel Strength
This value represents the force applied to particles that get closer together than the distance defined
by the Minimum Space control of the pFlock node. Higher values will cause particles to move away
from neighboring particles more rapidly, shooting away from the pack.

Minimum/Maximum Space
This range control represents the distance each particle attempts to maintain between it and other
particles. Particles will attempt to get no closer or farther than the space defined by the Minimum/
Maximum values of this range control. Smaller ranges will give the appearance of more organized
motion. Larger ranges will be perceived as disorganized and chaotic.

Common Controls
Conditions, Style, Region, and Settings Tabs
The Conditions, Style, Region, and Settings tabs are common to all Particle nodes, so their
descriptions can be found in “The Common Controls” section at the end of this chapter.

pFollow [pFo]

The pFollow node

pFollow Node Introduction

Inserting the pFollow node into a particle branch causes the particles to spring back and forth toward
a follow object. The follow object can be positioned in 3D or animated to create a new motion path for
the particles.

Inputs
The pFollow node has a single orange input by default. Like most particle nodes, this orange
background input accepts only other particle nodes. A green bitmap or mesh input appears on the
node when you set the Region menu in the Region tab to either Bitmap or Mesh.
– Input: The orange input takes the output of other particle nodes.
– Region: The green or magenta region input takes a 2D image or a 3D mesh depending on
whether you set the Region menu to Bitmap or Mesh. The color of the input is determined by
whichever is selected first in the menu. The 3D mesh or a selectable channel from the bitmap
defines the area where particles will follow the position point.

Chapter 52 Particle Nodes 1240

Basic Node Setup
When combined with pFlock, the pFollow node can produce natural swarming behaviors that change
direction.

A pFollow node introduces a follow object that influences the particles’ motion.

Inspector

The pFollow Controls tab

Random Seed
The Random Seed slider and Randomize button are presented whenever a Fusion node relies on a
random result. Two nodes with the same seed values will produce the same random results. Click the
Randomize button to randomly select a new seed value, or adjust the slider to manually select a new
seed value.

Position XYZ
The position controls are used to create the new path by positioning the follow object. Moving the
XYZ parameters displays the onscreen position of the follow object. Animating these parameters
creates the new path the particles will be influenced by.

Spring
The Spring setting causes the particles to move back and forth along the path. The spread of the
spring motion increases over the life of the particles depending on the distance between the
particles and the follow object. Higher spring settings increase the elasticity, while lower settings
decrease elasticity.

Dampen
This value attenuates the spring action. A lower setting offers less resistance to the back and forth
spring action. A higher setting applies more resistance.

Common Controls
Conditions, Style, Region, and Settings Tabs
The Conditions, Style, Region, and Settings tabs are common to all Particle nodes, so their
descriptions can be found in “The Common Controls” section at the end of this chapter.

Chapter 52 Particle Nodes 1241

pFriction [pFr]

The pFriction node

pFriction Node Introduction

The pFriction node applies resistance to the motion of a particle, slowing the particle’s motion through
a defined region. This node produces two types of Friction. One type reduces the velocity of any
particle intersecting/crossing the defined region, and one reduces or eliminates spin and rotation.

Inputs
The pFriction node has a single orange input by default. Like most particle nodes, this orange input
accepts only other particle nodes. A green or magenta bitmap or mesh input appears on the node
when you set the Region menu in the Region tab to either Bitmap or Mesh.
– Input: The orange input takes the output of other particle nodes.
– Region: The green or magenta region input takes a 2D image or a 3D mesh depending on
whether you set the Region menu to Bitmap or Mesh. The color of the input is determined by
whichever is selected first in the menu. The 3D mesh or a selectable channel from the bitmap
defines the area where the friction occurs.

Basic Node Setup

The pFriction node is placed in between the pEmitter and pRender. A Shape 3D node is used to
create the region the particles bounce off.

A pFriction node using a Shape 3D node as the region where friction is introduced to the particles.

Inspector

The pFriction controls

Chapter 52 Particle Nodes 1242

Random Seed
The Random Seed slider and Randomize button are presented whenever a Fusion node relies on a
random result. Two nodes with the same seed values will produce the same random results. Click the
Randomize button to randomly select a new seed value, or adjust the slider to manually select a new
seed value.

Velocity Friction
This value represents the Friction force applied to the particle’s Velocity. The larger the value, the
greater the friction, thus slowing down the particle.

Spin Friction
This value represents the Friction force applied to the particle’s Rotation or Spin. The larger the value,
the greater the friction, thus slowing down the rotation of the particle.

Common Controls
Conditions, Style, Region, and Settings Tabs
The Conditions, Style, Region, and Settings tabs are common to all Particle nodes, so their
descriptions can be found in “The Common Controls” section at the end of this chapter.

pGradientForce [pGF]

The pGradient Force node

pGradient Force Node Introduction

The particles are affected by a force generated by the gradients in the alpha values of the input image.
Particles will accelerate along the gradient, moving from white to black (high values to low values).
This node can be used to give particles the appearance of moving downhill or following the contour of
a provided shape.

Inputs
The pGradient Force node accepts two inputs: the default orange input from a particle node and one
from a bitmap image with an alpha channel gradient. A magenta or teal bitmap or mesh input appears
on the node when you set the Region menu in the Region tab to either Bitmap or Mesh.
– Input: The orange input takes the output of other particle nodes.
– Input: The green input takes the 2D image that contains the alpha channel gradient.
– Region: The magenta or teal region input takes a 2D image or a 3D mesh depending on
whether you set the Region menu to Bitmap or Mesh. The color of the input is determined by
whichever is selected first in the menu. The 3D mesh or a selectable channel from the bitmap
defines the area where the gradient force occurs.

Chapter 52 Particle Nodes 1243

Basic Node Setup
The pGradient Force node is placed in between the pEmitter and pRender nodes. A Fast Noise node
is used to create the alpha gradient used to modify the velocity of the particles.

A pGradient Force node using a Fast Noise node as the gradient to modify the particles’ motion

Inspector

The pGradient Force controls

Randomize
The Random Seed slider and Randomize button are presented whenever a Fusion node relies on a
random result.
Two nodes with the same seed values will produce the same random results. Click the Randomize
button to randomly select a new seed value, or adjust the slider to manually select a new seed value.

Strength
Gradient Force has only one specific control, which affects the strength of the force and acceleration
applied to the particles. Negative values on this control will cause the Gradient Force to be applied
from black to white (low values to high values).

Common Controls
Conditions, Style, Region, and Settings Tabs
The Conditions, Style, Region, and Settings tabs are common to all Particle nodes, so their
descriptions can be found in “The Common Controls” section at the end of this chapter.

Chapter 52 Particle Nodes 1244

pImage Emitter [pLE]

The pImage Emitter node

pImage Emitter Node Introduction

The pImage Emitter node takes an input image and treats each pixel of the image as if it were a
particle. The main differences between the pImage Emitter and the normal pEmitter is that instead of
emitting particles randomly within a given region, this node emits pixels in a regular 2D grid with
colors based on the input image.

Inputs
The pImage Emitter node has three inputs. Like most particle nodes, the orange input accepts only
other particle nodes. Green and magenta inputs are 2D image inputs for custom image calculations.
Optionally, there are teal or white bitmap or mesh inputs, which appear on the node when you set the
Region menu in the Region tab to either Bitmap or Mesh.
– Input: Unlike most other particle nodes, the orange input on the pImage Emitter accepts a 2D
image used as the emitter of the particles. If a region is defined for the emitter, this input is used
to define the color of the particles.
– Style Bitmap Input: This image input accepts a 2D image to use as the particles’ image. Since
this image duplicates into potentially thousands of particles, it is best to keep these images
small and square—for instance, 256 x 256 pixels.
– Region: The teal or white region input takes a 2D image or a 3D mesh depending on whether
you set the Region menu to Bitmap or Mesh. The color of the input is determined by whichever
is selected first in the menu. The 3D mesh or a selectable channel from the bitmap defines the
area where the particles are emitted.

Basic Node Setup

The pImage Emitter node is placed at the start of a particle branch, replacing the location of a pEmitter
node. Below, a MediaIn node is used to emit particles using the colors from the clip.

A pImage Emitter node emits particles based on an image connected to the orange input.

Chapter 52 Particle Nodes 1245

Inspector

The pImage Emitter controls

The great majority of controls in this node are identical to those found in the pEmitter, and those
controls are documented in that previous section. Below are the descriptions of the controls unique to
the pImage Emitter node.

X and Y Density
The X and Y Density sliders are used to set the mapping of particles to pixels for each axis. They
control the density of the sampling grid. A value of 1.0 for either slider indicates 1 sample per pixel.
Smaller values will produce a looser, more pointillistic distribution of particles, while values
above 1.0 will create multiple particles per pixel in the image.

Alpha Threshold
The Alpha Threshold is used for limiting particle generation so that pixels with semitransparent alpha
values will not produce particles. This can be used to harden the edges of an otherwise soft alpha
channel. The higher the threshold value, the more opaque a pixel must be before it will generate a
particle. Note that the default threshold of 0.0 will create particles for every pixel, regardless of alpha,
although many may be transparent and invisible.

Lock Particle Color to Initial Frame

Select this checkbox to force the particles to keep the color with which they were born throughout the
life of the particle. If this is off, and the input image changes on successive frames, the particles will
also change color to match the image. This allows video playback on a grid of particles.

Create Particles Every Frame

Enabling this creates a whole new set of particles every frame, instead of just one set on the frame.
This can lead to very large particle systems but allows some interesting effects—for example, if the
particles are given some initial velocity or if emitting from an animated source. Try a small velocity, an
Angle Z of -90, and a seething Fast Noise as a source to get smoothly varying clouds of particles that
you could fly through. Note that if this checkbox is left off, only one set of particles is ever created, and
thus animating any of the emitter’s other controls will have no effect.

Chapter 52 Particle Nodes 1246

X/Y/Z Pivot
These controls allow you to position the grid of emitted particles.

Use Z Channel for Particle Z

If the input image used to generate the particles has a Z depth channel, that channel can be used to
determine the initial position of the particle in Z space. This can have an interesting hollow shell effect
when used in conjunction with camera rotation in the pRender node.

Common Controls
Conditions, Style, Region, and Settings Tabs
The Conditions, Style, Region, and Settings tabs are common to all Particle nodes, so their
descriptions can be found in “The Common Controls” section at the end of this chapter.

Sets Tab

The pImage Emitter Sets tab

NOTE: Pixels with a black (transparent) alpha channel will still generate invisible particles,
unless you raise the Alpha Threshold above 0.0. This can slow down rendering significantly.
An Alpha Threshold value of 1/255 = 0.004 is good for eliminating all fully transparent pixels.
The pixels are emitted in a fixed-size 2D grid on the XY plane, centered on the Pivot position.
Changing the Region from the default of All allows you to restrict particle creation to more
limited areas. If you need to change the size of this grid, use a Transform 3D node after
the pRender.
Remember that the various emitter controls apply only to particles when they are emitted.
That is, they set the initial state of the particle and do not affect it for the rest of its lifespan.
Since pImageEmitter (by default) emits particles only on the first frame, animating these
controls will have no effect. However, if the Create Particles Every Frame checkbox is turned
on, new particles will be emitted each frame and will use the specified initial settings for
that frame.

Chapter 52 Particle Nodes 1247

pKill [pKI]

The pKill node

pKill Node Introduction

The Kill node is used to destroy (kill) any particle that crosses or intersects its region. It has no specific
controls, as it has only one possible affect on a particle. The controls found in the Region tab are
normally used to limit this node by restricting the effect to particles that fall within a certain region, age,
set, or by reducing the probability of the node applying to a given particle.

Inputs
The pKill node has a single orange input by default. Like most particle nodes, this orange input
accepts only other particle nodes. A green bitmap or mesh input appears on the node when you set
the Region menu in the Region tab to either Bitmap or Mesh.
– Input: The orange input takes the output of other particle nodes.
– Region: The green or magenta region input takes a 2D image or a 3D mesh depending on
whether you set the Region menu to Bitmap or Mesh. The color of the input is determined by
whichever is selected first in the menu. The 3D mesh or a selectable channel from the bitmap
defines the area particles are killed.

Basic Node Setup

The pKill node is placed in between the pEmitter and pRender nodes. A Shape 3D node is used to
create the region where the particles die.

A pKill node using a Shape 3D node as the region where particles die

Inspector
This node only contains common controls in the Conditions and Regions tabs. The Conditions and
Regions controls are used to define the location, age, and set of particles that are killed.

Chapter 52 Particle Nodes 1248

Common Controls
Conditions, Style, Region, and Settings Tabs
The Conditions, Style, Region, and Settings tabs are common to all Particle nodes, so their
descriptions can be found in “The Common Controls” section at the end of this chapter.

pMerge [pMg]

The pMerge node

pMerge Node Introduction

This node has no controls whatsoever. It serves to combine particles from two streams. Any nodes
downstream of the pMerge node will treat the two streams as one.
The combined particles will preserve any sets assigned to them when they were created, making it
possible for nodes downstream of the pMerge to isolate specific particles when necessary.

Inputs
The pMerge node has two identical inputs, one orange and one green. These two inputs accept only
other particle nodes.
– Particle 1 and 2 Input: The two inputs accept two streams of particles and merge them.

Basic Node Setup

The pMerge node connects two pEmitter nodes. The output of the pMerge can go on to feed other
particle nodes or to a pRender.

A pMerge node combining two pEmitter nodes.

Chapter 52 Particle Nodes 1249

pPoint Force [pPF]

The pPoint Force node

pPoint Force Node Introduction

This node applies a force to the particles that emanate from a single point in 3D space. The pPoint
Force can either attract or repel particles within its sphere of influence. There are four controls specific
to the pPoint Force node.

Inputs
The pPoint Force node has a single orange input by default. Like most particle nodes, this orange
input accepts only other particle nodes. A green bitmap or mesh input appears on the node when you
set the Region menu in the Region tab to either Bitmap or Mesh.
– Input: The orange input takes the output of other particle nodes.
– Region: The green or magenta region input takes a 2D image or a 3D mesh depending on
whether you set the Region menu to Bitmap or Mesh. The color of the input is determined by
whichever is selected first in the menu. The 3D mesh or a selectable channel from the bitmap
defines the area where the point force affects the particles.

Basic Node Setup

The pPoint Force node is inserted between a pEmitter and a pRender node.

The pPoint Force node positions a tangent force that particles are attracted to or repelled from.

Inspector

The pPoint Force controls

Chapter 52 Particle Nodes 1250

Randomize
The Random Seed slider and Randomize button are presented whenever a Fusion node relies on a
random result. Two nodes with the same seed values will produce the same random results. Click the
Randomize button to randomly select a new seed value, or adjust the slider to manually select a new
seed value.

Strength
This parameter sets the Strength of the force emitted by the node. Positive values represent attractive
forces; negative values represent repellent forces.

Power
This determines the degree to which the Strength of the force falls off over distance. A value of zero
causes no falloff of strength. Higher values will impose an ever-sharper falloff in strength of the force
with distance.

Limit Force
The Limit Force control is used to counterbalance potential problems with temporal sub-sampling.
Because the position of a particle is sampled only once a frame (unless sub-sampling is increased in
the pRender node), it is possible that a particle can overshoot the Point Force’s position and end up
getting thrown off in the opposite direction. Increasing the value of this control reduces the likelihood
that this will happen.

X, Y, Z Center Position
These controls are used to represent the X, Y, and Z coordinates of the point force in 3D space.

Common Controls
Conditions, Style, Region, and Settings Tabs
The Conditions, Style, Region, and Settings tabs are common to all Particle nodes, so their
descriptions can be found in “The Common Controls” section at the end of this chapter.

pRender [pRn]

The pRender node

pRender Node Introduction

The pRender node converts the particle system to either an image or geometry. The default is a 3D
particle system, which must be connected to a Renderer 3D to produce an image. This allows the
particles to be integrated with other elements in a 3D scene before they are rendered.

Inputs
The pRender node has one orange input, a green camera input, and a blue effects mask input. Like
most particle nodes, this orange input accepts only other particle nodes. A green bitmap or mesh
input appears on the node when you set the Region menu in the Region tab to either Bitmap or Mesh.

Chapter 52 Particle Nodes 1251

 – Input: The orange input takes the output of other particle nodes.
– Camera Input: The optional green camera input accepts a camera node directly or a 3D scene
with a camera connected that is used to frame the particles during rendering.
– Effect Mask: The optional blue input expects a mask shape created by polylines, basic
primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this input for
2D particles crops the output of the particles so they are seen only within the mask.

Basic Node Setup

The pRender node is always placed at the end of a particle branch. If the pRender is set to 2D, then
the output connects to other 2D nodes like a Merge node. If the pRender is set to 3D, the output
connects to a 3D node like a Merge 3D.

All particle branches end with a pRender node.

Inspector

The pRender controls

Output Mode (2D/3D)

While the pRender defaults to 3D output, it can be made to render a 2D image instead. This is done
with the 3D and 2D buttons on the Output Mode control. If the pRender is not connected to a 3D-only
or 2D-only node, you can also switch it by selecting View > 2D Viewer from the viewer’s pop-up menu.

Chapter 52 Particle Nodes 1252

In 3D mode, the only controls in the pRender node that have any effect at all are Restart, Pre-Roll and
Automatic Pre-Roll, Sub-Frame Calculation Accuracy, and Pre-Generate frames. The remaining
controls affect 2D particle renders only. The pRender node also has a Camera input on the node tree
that allows the connection of a Camera 3D node. This can be used in both 2D and 3D modes to allow
control of the viewpoint used to render an output image.

Render and the Viewers

When the pRender node is selected in a node tree, all the onscreen controls from Particle nodes
connected to it are presented in the viewers. This provides a fast, easy-to-modify overview of the
forces applied to the particle system as a whole.

Pre-Roll Options
Particle nodes generally need to know the position of each particle on the last frame before they can
calculate the effect of the forces applied to them on the current frame. This makes changing current
time manually by anything but single frame intervals likely to produce an inaccurate image.
The controls here are used to help accommodate this by providing methods of calculating the
intervening frames.

Restart
This control also works in 3D. Clicking on the Restart button will restart the particle system at the
current frame, removing any particles created up to that point and starting the particle system from
scratch at the current frame.

Pre-Roll
This control also works in 3D. Clicking on this button causes the particle system to recalculate, starting
from the beginning of the render range up to the current frame. It does not render the image
produced. It only calculates the position of each particle. This provides a relatively quick mechanism to
ensure that the particles displayed in the views are correctly positioned.
If the pRender node is displayed when the Pre-Roll button is selected, the progress of the pre-roll is
shown in the viewer, with each particle shown as point style only.

Automatic Pre-Roll
Selecting the Automatic Pre-Roll checkbox causes the particle system to automatically pre-roll the
particles to the current frame whenever the current frame changes. This prevents the need to manually
select the Pre-Roll button whenever advancing through time in jumps larger than a single frame. The
progress of the particle system during an Automatic Pre-Roll is not displayed to the viewers to prevent
distracting visual disruptions.

About Pre-Roll
Pre-Roll is necessary because the state of a particle system is entirely dependent on the last known
position of the particles. If the current time were changed to a frame where the last frame particle state
is unknown, the display of the particle is calculated on the last known position, producing
inaccurate results.
To demonstrate:
1 Add a pEmitter and a pRender node to the composition.
2 View the pRender in one of the viewers.
3 Set the Velocity of the particles to 0.1.
4 Place the pEmitter on the left edge of the screen.
5 Set the Current Frame to 0.
6 Set a Render Range from 0–100 and press the Play button.

Chapter 52 Particle Nodes 1253

7 Observe how the particle system behaves.
8 Stop the playback and return the current time to frame 0.
9 In the pRender node, disable the Automatic Pre-Roll option.
10 Use the current time number field to jump to frame 10, and then to frames 60 and 90.

Notice how the particle system only adds to the particles it has already created and does not try to
create the particles that would have been emitted in the intervening frames. Try selecting the Pre-Roll
button in the pRender node. Now the particle system state is represented correctly.
For simple, fast-rendering particle systems, it is recommended to leave the Automatic Pre-Roll option
on. For slower particle systems with long time ranges, it may be desirable to only Pre-Roll manually,
as required.
– Only Render in Hi-Q
Selecting this checkbox causes the style of the particles to be overridden when the Hi-Q
checkbox is deselected, producing only fast rendering Point-style particles. This is useful when
working with a large quantity of slow Image-based or Blob-style particles. To see the particles as
they would appear in a final render, simply enable the Hi-Q checkbox.
– View
This drop-down list provides options to determine the position of the camera view in a 3D particle
system. The default option of Scene (Perspective) will render the particle system from the
perspective of a virtual camera, the position of which can be modified using the controls in the
Scene tab. The other options provide orthographic views of the front, top, and side of the
particle system.
It is important to realize that the position of the onscreen controls for Particle nodes is unaffected
by this control. In 2D mode, the onscreen controls are always drawn as if the viewer were showing
the front orthographic view. (3D mode gets the position of controls right at all times.)
The View setting is ignored if a Camera 3D node is connected to the pRender node’s Camera
input on the node tree, or if the pRender is in 3D mode.

Conditions
Blur, Glow, and Blur Blend
When generating 2D particles, these sliders apply a Gaussian blur, glows, and blur blending to the
image as it is rendered, which can be used to soften the particles and blend them. The result is no
different than adding a Blur after the pRender node in the node tree.

Sub-Frame Calculation Accuracy

This determines the number of sub-samples taken between frames when calculating the particle
system. Higher values increase the accuracy of the calculation but also increase the amount of time to
render the particle system.

Pre-Generate Frames
This control is used to cause the particle system to pre-generate a set number of frames before its first
valid frame. This is used to give a particle system an initial state from which to start.
A good example of when this might be useful is in a shot where particles are used to create the smoke
rising from a chimney. Set Pre-Generate Frames to a number high enough to ensure that the smoke is
already present in the scene before the render begins, rather than having it just starting to emerge
from the emitter for the first few frames.

Chapter 52 Particle Nodes 1254

Kill Particles That Leave the View
Selecting this checkbox control automatically destroys any particles that leave the visible boundaries
of the image. This can help to speed render times. Particles destroyed in this fashion never return,
regardless of any external forces acting upon them.

Generate Z Buffer
Selecting this checkbox causes the pRender node to produce a Z Buffer channel in the image. The
depth of each particle is represented in the Z Buffer. This channel can then be used for additional
depth operations like Depth Blur, Depth Fog, and Downstream Z Merging.
Enabling this option is likely to increase the render times for the particle system dramatically.

Depth Merge Particles

Enabling this option causes the particles to be merged using Depth Merge techniques, rather than
layer-based techniques.

Scene Tab

The pRender Scene tab

Z Clip
The Z Clip control is used to set a clipping plane in front of the camera. Particles that cross this plane
are clipped, preventing them from impacting on the virtual lens of the camera and dominating
the scene.

Grid Tab
These controls do not apply to 3D particles.

Chapter 52 Particle Nodes 1255

The grid is a helpful, non-rendering display guide used to orient the 2D particles in 3D space. The grid
is never seen in renders, just like a center crosshair is never seen in a render. The width, depth,
number of lines, and grid color can be set using the controls found in this tab.
These controls cannot be animated.

The pRender Grid tab

Image Tab
The controls in this tab are used to set the resolution, color depth, and pixel aspect of the rendered
image produced by the node.

The pRender Image tab

Chapter 52 Particle Nodes 1256

Process Mode
Use this menu control to select the Fields Processing mode used by Fusion to render changes to the
image. The default option is determined by the Has Fields checkbox control in the Frame Format
preferences.

Use Frame Format Settings

When this checkbox is selected, the width, height, and pixel aspect of the rendered images by the
node will be locked to values defined in the composition’s Frame Format preferences. If the Frame
Format preferences change, the resolution of the image produced by the node will change to match.
Disabling this option can be useful to build a composition at a different resolution than the eventual
target resolution for the final render.

Width/Height
This pair of controls is used to set the Width and Height dimensions of the image to be rendered
by the node.

Pixel Aspect
This control is used to specify the Pixel Aspect ratio of the rendered particles. An aspect ratio of 1:1
would generate a square pixel with the same dimensions on either side (like a computer display
monitor), and an aspect of 0.9:1 would create a slightly rectangular pixel (like an NTSC monitor).

NOTE: Right-click on the Width, Height, or Pixel Aspect controls to display a menu listing the
file formats defined in the preferences Frame Format tab. Selecting any of the listed options
will set the width, height, and pixel aspect to the values for that format, accordingly.

Depth
The Depth menu is used to set the pixel color depth of the particles. 32-bit pixels require 4X the
memory of 8-bit pixels but have far greater color accuracy. Float pixels allow high dynamic range
values outside the normal 0…1 range, for representing colors that are brighter than white or darker
than black.

Source Color Space

You can use the Source Color Space menu to set the Color Space of the footage to help achieve a
linear workflow. Unlike the Gamut tool, this doesn‘t perform any actual color space conversion, but
rather adds the source space data into the metadata, if that metadata doesn‘t exist. The metadata can
then be used downstream by a Gamut tool with the From Image option, or in a Saver, if explicit output
spaces are defined there. There are two options to choose from:
– Auto: Automatically reads and passes on the metadata that may be in the image.
– Space: Displays a Color Space Type menu where you can choose the correct color
space of the image.

Source Gamma Space

Using the Curve type menu, you can set the Gamma Space of the footage and choose to remove it by
way of the Remove Curve checkbox when working in a linear workflow. There are three choices in the
Curve type menu:
– Auto: Automatically reads and passes on the metadata that may be in the image.
– Space: Displays a Gamma Space Type menu where you can choose the correct
gamma curve of the image.
– Log: Brings up the Log/Lin settings, similar to the Cineon tool. For more information, see
Chapter 38, “Film Nodes” in the Fusion Studio Reference Manual or Chapter 99 in the
DaVinci Resolve Reference Manual.

Chapter 52 Particle Nodes 1257

Remove Curve
Depending on the selected Gamma Space or on the Gamma Space found in Auto mode, the Gamma
Curve is removed from, or a log-lin conversion is performed on, the material, effectively converting it
to a linear output space.

Motion Blur
As with other 2D nodes in Fusion, Motion Blur is enabled from within the Settings tab. You may set
Quality, Shutter Angle, Sample Center, and Bias, and Blur will be applied to all moving particles.

NOTE: Motion Blur on 3D mode particles (rendered with a Renderer 3D) also requires that
identical motion blur settings are applied to the Renderer 3D node.

pSpawn [pSp]

The pSpawn node

pSpawn Node Introduction

The pSpawn node makes each affected particle act as an emitter that can produce one or more
particles of its own. The original particle continues until the end of its lifespan, and each of the
particles it emits becomes wholly independent with a lifespan and properties of its own.
As long as a particle falls under the effect of the pSpawn node, it will continue to generate particles. It
is important to restrict the effect of the node with limiters like Start and End Age, Probability, Sets and
Regions, and by animating the parameters of the emitter so that the node is operative only
when required.

Inputs
By default, the pSpawn node has a single orange input. Like most particle nodes, this orange input
accepts only other particle nodes. You can enable an image input by selecting Bitmap from the Style
menu in the Style tab. Also, two region inputs, one for bitmap and one for mesh, appear on the node
when you set the Region menu in the Region tab to either Bitmap or Mesh. The colors of these inputs
change depending on the order they are enabled.
– Input: The orange input accepts the output of other particle nodes.
– Style Bitmap Input: This image input accepts a 2D image to use as the particles’ image. Since
this image duplicates into potentially thousands of particles, it is best to keep these images
small and square—for instance, 256 x 256 pixels.
– Region: The region inputs take a 2D image or a 3D mesh depending on whether you set the
Region menu to Bitmap or Mesh. The color of the input is determined by whichever is selected
first in the menu. The 3D mesh or a selectable channel from the bitmap defines the area where
the particles are emitted.

Chapter 52 Particle Nodes 1258

Basic Node Setup
The pSpawn node is placed between the pEmitter and pRender nodes. Using the Age parameter in
the pSpawn’s Conditions tab, you can spawn new particles as the old ones die off. This is one way to
have a trail of particles shoot up in the air like a rocket and burst into sparkling fireworks.

A pSpawn node used to generate new particles at specific points in the old particles’ life

Inspector
The pSpawn node has a large number of controls, most of which exactly duplicate those found within
the pEmitter node. There are a few controls that are unique to the pSpawn node, and their effects are
described below.

The pSpawn controls

Affect Spawned Particles

Selecting this checkbox causes particles created by spawning to also become affected by the pSpawn
node on subsequent frames. This can exponentially increase the number of particles in the system,
driving render times up to an unreasonable degree. Use this checkbox cautiously.

Velocity Transfer
This control determines how much velocity of the source particle is transferred to the particles it
spawns. The default value of 1.0 causes each new particle to adopt 100 percent of the velocity and
direction from its source particle. Lower values will transfer less of the original motion to the
new particle.

Common Controls
Conditions, Style, Region, and Settings Tabs
The Conditions, Style, Region, and Settings tabs are common to all Particle nodes, so their
descriptions can be found in “The Common Controls” section at the end of this chapter.

Chapter 52 Particle Nodes 1259

pTangent Force [pTF]

The pTangent Force node

pTangent Force Node Introduction

This node is used to apply a tangential force to the particles—a force that is applied perpendicularly to
the vector between the pTangent Force’s region and the particle it is affecting.

Inputs
The pTangent Force node has a single orange input by default. Like most particle nodes, this orange
input accepts only other particle nodes. A green bitmap or mesh input appears on the node when you
set the Region menu in the Region tab to either Bitmap or Mesh.
– Input: The orange input takes the output of other particle nodes.
– Region: The green or magenta region input takes a 2D image or a 3D mesh depending on
whether you set the Region menu to Bitmap or Mesh. The color of the input is determined by
whichever is selected first in the menu. The 3D mesh or a selectable channel from the bitmap
defines the area where the tangent force effects the particles.

Basic Node Setup

The pTangent Force node is inserted between a pEmitter and a pRender node.

The pTangent Force node positions a tangent force that particles maneuver around.

Inspector

The pTangent Force controls

Chapter 52 Particle Nodes 1260

The controls for this node are used to position the offset in 3D space and to determine the strength of
the tangential force along each axis independently.

Randomize
The Random Seed slider and Randomize button are presented whenever a Fusion node relies on a
random result.
Two nodes with the same seed values will produce the same random results. Click the Randomize
button to randomly select a new seed value, or adjust the slider to manually select a new seed value.

X, Y, Z Center Position
These controls are used to represent the X, Y, and Z coordinates of the Tangent force in 3D space.

X, Y, Z Center Strength
These controls are used to determine the Strength of the Tangent force in 3D space.

Common Controls
Conditions, Style, Region, and Settings Tabs
The Conditions, Style, Region, and Settings tabs are common to all Particle nodes, so their
descriptions can be found in “The Common Controls” section at the end of this chapter.

pTurbulence [pTr]

The pTurbulence node

pTurbulence Node Introduction

The pTurbulence node imposes a frequency-based chaos on the position of each particle, causing the
motion to become unpredictable and uneven. The controls for this node affect the strength and
density of the Turbulence along each axis.

Inputs
The pTurbulence node has a single orange input by default. Like most particle nodes, this orange input
accepts only other particle nodes. A green bitmap or mesh input appears on the node when you set
the Region menu in the Region tab to either Bitmap or Mesh.
– Input: The orange input takes the output of other particle nodes.
– Region: The green or magenta region input takes a 2D image or a 3D mesh depending on
whether you set the Region menu to Bitmap or Mesh. The color of the input is determined by
whichever is selected first in the menu. The 3D mesh or a selectable channel from the bitmap
defines the area of turbulence.

Chapter 52 Particle Nodes 1261

Basic Node Setup
The pTurbulence node is inserted between a pEmitter and a pRender node.

The pTurbulence node disturbs the rigid flow of particles for a more natural motion.

Inspector

The pTurbulence controls

Randomize
The Random Seed slider and Randomize button are presented whenever a Fusion node relies on a
random result. Two nodes with the same seed values will produce the same random results. Click the
Randomize button to randomly select a new seed value, or adjust the slider to manually select a new
seed value.

X, Y, and Z Strength
The Strength control affects the amount of chaotic motion imparted to particles.

Strength Over Life

This mini Spline Editor control can be used to control the amount of turbulence applied to a particle
according to its age. For example, a fire particle may originally have very little turbulence applied at
the start of its life, and as it ages, the turbulence increases.

Density
Use this control to adjust the density in the turbulence field. Lower values causes more particle cells to
be affected similarly, almost as if “waves” of the turbulence field run through the particles, affecting
groups of cells at the same time. Higher values add finer variations to more individual particle cells
causing more of a spread in the turbulence field.

Common Controls
Conditions, Style, Region, and Settings Tabs
The Conditions, Style, Region, and Settings tabs are common to all Particle nodes, so their
descriptions can be found in “The Common Controls” section at the end of this chapter.

Chapter 52 Particle Nodes 1262

pVortex [pVt]

The pVortex node

pVortex Node Introduction

The pVortex node applies a rotational force to each particle, causing them to be drawn toward the
source of the Vortex. In addition to the Common Particle Controls, the pVortex node also has the
following controls.

Inputs
The pVortex node has a single orange input by default. Like most particle nodes, this orange input
accepts only other particle nodes. A green bitmap or mesh input appears on the node when you set
the Region menu in the Region tab to either Bitmap or Mesh.
– Input: The orange input takes the output of other particle nodes.
– Region: The green or magenta region input takes a 2D image or a 3D mesh depending on
whether you set the Region menu to Bitmap or Mesh. The color of the input is determined by
whichever is selected first in the menu. The 3D mesh or a selectable channel from the bitmap
defines the area of the vortex.

Basic Node Setup

The pVortex node is placed in between the pEmitter and pRender nodes.

A pVortex node creates a spiraling motion for particles that fall within its pull.

Inspector

The pVortex controls

Chapter 52 Particle Nodes 1263

Randomize
The Random Seed slider and Randomize button are presented whenever a Fusion node relies on a
random result. Two nodes with the same seed values will produce the same random results. Click the
Randomize button to randomly select a new seed value, or adjust the slider to manually select a new
seed value.

Strength
This control determines the Strength of the Vortex Force applied to each particle.

Power
This control determines the degree to which the Strength of the Vortex Force falls off with distance.

X, Y, and Z Offset
Use these sliders to set the amount by which the vortex Offsets the affected particles.

Size
This is used to set the Size of the Vortex Force.

Angle X and Y
These sliders control the amount of rotational force applied by the Vortex along the X and Y axes.

Common Controls
Conditions, Style, Region, and Settings Tabs
The Conditions, Region, and Settings tabs are common to all Particle nodes, so their descriptions can
be found in the following “The Common Controls” section.

The Common Controls

Particle nodes share a number of identical controls in the inspector. This section describes the Style,
Conditions, Region, and Settings tabs that are common among particle nodes.

Inspector

The pEmitter Style tab

Style Tab
The Style Tab is common to the pEmitter, pSpawn, pChangeStyle, and pImage Emitter. It controls the
appearance of the particles using general controls like type, size, and color.

Chapter 52 Particle Nodes 1264

Style
The Style menu provides access to the various types of particles supported by the Particle Suite.
Each style has its specific controls, as well as controls it will share with other styles.
– Point: This option produces particles precisely one pixel in size. Controls that are specific to Point
Style are Apply Mode and Sub Pixel Rendered.
– Apply Mode: This menu applies only to 2D particles. 3D particle systems are not affected.
It includes Apply modes for Add and Merge. Add combines overlapping particles by adding
together the color values of each particle. Merge uses a simple over operation to combine
overlapping particles.
– Sub Pixel Rendered: This checkbox determines whether the point particles are rendered with
Sub Pixel precision, which provides smoother-looking motion but blurrier particles that take
slightly longer to render.

– Bitmap: This style produces particle cells based on an image file or another node in the Node
editor. When this option is selected an orange image input appears on the node in the node
editor. There are several controls for affecting the appearance and animation. In addition to the
controls in the Style section, a Merge section is displayed at the bottom of the inspector when
Bitmap is selected as the Style. The Merge section includes controls for additive or subtractive
merges when the particle cells overlap.
– Animate Over Time: This menu includes three options for determining how movie files play
when they are used as particle cell bitmaps. The Over Time setting plays the movie file
sequentially. For instance, when the comps is on frame 2, frame 2 of the movie file is displayed,
when the comp is on frame 3, frame 3 of the movie files is displayed and so on. If a particle
cell is not generated until frame 50, it begins with frame 50 of the movie file. This causes all
particle cells to use the same image on any give frame of the comp. The Particle Age setting
causes each particle cell to begin with the first frame of the movie file, regardless of when the
particle cell is generated. The Particle Birth Time setting causes each particle to begin with the
frame that coincides with the frame of the particle cell birth time. For instance, if the particle
is generated on frame 25, then it uses frame 25 of the movie file for the entire comp. Unlike
the other two options, the Particle Birth Time setting holds the same frame for the duration
of the comp
– Time Offset: This dial is used to slip or offset the starting frame used from the movie file.
For instance, setting it to 10 will cause the movie file to start at frame 10 instead of frame 1.
– Time Scale: This slider is a multiplier on the frame. Instead of using an offset, it changes the
starting frame by multiplying the frame by the value selected with the slider. For instance, if a
value of 2 is selected then when the playhead reaches frame 2, the movie files displays frame 4
(2x2=4) and when the playhead reaches frame 8, the movie file displays frame 16 (8x2=16).
– Gain: The gain slider is a multiplier of the pixel value. It is used to apply a correction to the
overall Gain of the Bitmap. Let’s say you have a bitmap particle cell that contains a pixel value
of R0.5 G0.5 B0.4 and you add a Gain of 1.2, you end up with a pixel value of R0.6 G0.6, B0.48
(i.e., 0.4 * 1.2 = 0.48) while leaving black pixels unaffected. Higher values produce a brighter
image, whereas lower values reduce both the brightness and the transparency of the image.
– Style Bitmap: This control appears when the Bitmap style is selected, along with an orange
Style Bitmap input on the node’s icon in the Node view. Connect a 2D node to this input to
provide images to be used for the particles. You can do this on the Node view, or you may
drag and drop the image source node onto the Style Bitmap control from the Node Editor or
Timeline, or right-click on the control and select the desired source from the Connect To menu.

Chapter 52 Particle Nodes 1265

 – Blob: This option produces large, soft spherical particles, with controls for Color, Size, Fade timing,
Merge method, and Noise.
– Noise: This slider only applies to 2D Blob particles. The noise slider is used to introduce a
computer generated Perlin noise pattern into the blob particles in order to give the blobs more
texture. A setting of 0 introduces no noise to the Blob particles and a setting of 1 introduces the
maximum amount of noise.

– Brush: This styles produces particle cells based on any image file located in the brushes directory.
There are numerous controls for affecting the appearance and animation.
– Gain: The gain slider is a multiplier of the pixel value. It is used to apply a correction to the
overall Gain of the image that is used as the Brush. Let’s say you have a brush particle cell that
contains a pixel value of R0.5 G0.5 B0.4 and you add a Gain of 1.2, you end up with a pixel
value of R0.6 G0.6, B0.48 (i.e., 0.4 * 1.2 = 0.48) while leaving black pixels unaffected. Higher
values produce a brighter image, whereas lower values reduce both the brightness and the
transparency of the image.
– Brush: This menu shows the names of any image files stored in the Brushes directory. The
location of the Brushes directory is defined in the Preferences dialog, under Path Maps. The
default is the Brushes subdirectory within Fusion’s install folder.
– Use Aspect From: The Use Aspect From menu includes three settings for the aspect ratio of
the brush image. You can choose image format to use the brush image’s native aspect ration.
Choose Frame Format to use the aspect ratio set in the Frame Format Setting in the Fusion
Preferences, or choose Custom to enter your own Pixel X and Y dimensions.

– Line: This style produces straight line-type particles with optional “falloff.” The Size to Velocity
control described below (under Size Controls) is often useful with this Line type. The Fade control
adjusts the amount of falloff over the length of the line.
– Point Cluster: This style produces small clusters of single-pixel particles. Point Clusters are similar
to the Point style; however, they are more efficient when a large quantity of particles is required.
This style shares parameters with the Point style. Additional controls specific to Point Cluster style
are Number of Points and Number Variance.
– Sub Pixel Rendered: This checkbox determines whether the point particles are rendered with
Sub Pixel precision, which provides smoother-looking motion but blurrier particles that take
slightly longer to render.
– Number of Points and Variance: The value of this control determines how many points are in
each Point Cluster.

Color Controls
The Color Controls select the color and Alpha values of the particles generated by the emitter.

Color Variance
These range controls provide a means of expanding the colors produced by the pEmitter. Setting the
Red variance range at -0.2 to +0.2 will produce colors that vary 20% on either side of the red channel,
for a total variance of 40%. If the pEmitter is set to produce R0.5, G0.5, B0.5 (pure gray), the variance
shown above will produce points with a color range between R0.3, G0.5, B0.5, and R0.7, G0.5, B0.5.
To visualize color space as values between 0-256 or as 0-65535, change the values used by Fusion
using the Show Color As option provided in the General tab within the Preferences dialog.

Chapter 52 Particle Nodes 1266

Lock Color Variance
This checkbox locks the color variance of the particles. Unlocking this allows the color variance to be
applied differently to each color channel, giving rise to a broader range of colors.

Particles Color Over Life controls

Color Over Life

This standard gradient control allows for the selection of a range of color values to which the particle
will adhere over its lifetime.
The left point of the gradient represents the particle color at birth. The right point shows the color of
the particle at the end of its lifespan.
Additional points can be added to the gradient control to cause the particle color to shift
throughout its life.
This type of control can be useful for fire-type effects (for example, the flame may start blue, turn
orange, and end a darker red). The gradient itself can be animated over time by right-clicking on the
control and selecting Animate from the contextual menu. All points on the gradient will be controlled
by a single Color Over Life spline, which controls the speed at which the gradient itself changes. You
may also use the From Image modifier, which produces a gradient from the range of colors in an image
along a line between two points.

Particles Size and Fade controls

Chapter 52 Particle Nodes 1267

Size Controls
The majority of the Size Controls are self-explanatory. The Size and Size Variance controls are used to
determine the size and degree of size variation for each particle. It is worth noting that the Point style
does not have size controls (each point is a single pixel in size, and there is no additional control).
When a Bitmap Particle style is used, a value of 1.0 indicates that each particle should be the same
size as the input bitmap. A value of 2.0 will scale the particle up in size by 200%. For the best quality
particles, always try to make the input bitmap as big, or bigger, than the largest particle produced by
the system.
For the Point Cluster style, the size control adjusts the density of the cluster, or how close together
each particle will get.
There are additional size controls that can be used to adjust further the size of particles based on
velocity and depth.

Size to Velocity
This increases the size of each particle relative to the velocity or speed of the particle. The velocity of
the particle is added to the size, scaled by the value of this control.
1.0 on this control, such as for a particle traveling at 0.1, will add another 0.1 to the size (velocity * size to
velocity + size = new size). This is most useful for Line styles, but the control can be used to adjust the
size of any style.

Size Z Scale
This control measures the degree to which the size of each particle changes according to its Z
position. The effect is to exaggerate or reduce the impact of perspective. The default value is 1.0,
which provides a relatively realistic perspective effect.
Objects on the focal plane (Z = 0.0) will be actual-sized. Objects farther along Z will become smaller.
Objects closer along Z will get larger.
A value of 2.0 will exaggerate the effect dramatically, whereas a value of 0.0 will cancel the effects of
perspective entirely.

Size Over Life

This spline control determines the size of a particle throughout its lifespan. The vertical scale
represents a percentage of the value defined by the Size control, from 0 to 200%. The horizontal scale
represents a percentage of the particle’s lifespan (0 to 100%).
This graph supports all the features available to a standard spline editor. These features can be
accessed by right-clicking on the graph. It is also possible to view and edit the graph spline in the
larger Spline Editor.

Fade Controls
This simple range slider provides a mechanism for fading a particle at the start and end of its lifetime.
Increasing the Fade In value will cause the particle to fade in at the start of its life. Decreasing the Fade
Out value will cause the particle to fade out at the end of its life.
This control’s values represent a percentage of the particle’s overall life, therefore, setting the Fade In
to 0.1 would cause the particle to fade in over the first 10% of its total lifespan. For example, a particle
with a life of 100 frames would fade in from frame 0…10.

Chapter 52 Particle Nodes 1268

Merge Controls
This set of particle controls affects the way individual particles are merged together. The Subtractive/
Additive slider works as documented in the standard Merge node. The Burn-In control will cause the
particles to overexpose, or “blow out,” when they are combined.
None of the Merge controls will have any effect on a 3D particle system.

Particles Blur controls

Blur Controls
This set of particle controls can be used to apply a Blur to the individual particles. Blurring can be
applied globally, by age, or by Z depth position.
None of the Blur controls will have any effect on a 3D particle system.

Blur (2D) and Blur Variance (2D)

These controls apply blur to each particle. Unlike the Blur in the pRender node, this is applied to each
particle independently before the particles are merged together. The Blur Variance slider modifies the
amount of blur applied to each particle.

Blur Over Life

This spline graph controls the amount of blur that is applied to the particle over its life. The vertical
scale represents a percentage of the value defined by the Blur control. The horizontal scale represents
a percentage of the particle’s lifespan.
This graph supports all of the features available to a standard Spline Editor. These features can be
accessed by right-clicking on the graph. It is also possible to view and edit the spline in the larger
Spline editor.

Z Blur (DoF) (2D) and DoF Focus

This slider control applies blur to each particle based on its position along the Z axis.
The DoF Focus range control is used to determine what area of the image remains in focus. Lower
values along Z are closer to the camera. Higher values are farther away. Particles within the range will
remain in focus. Particles outside that range will have the blur defined by the Z Blur control
applied to them.

Chapter 52 Particle Nodes 1269

Conditions

pVortex Conditions tab

Conditions Tab
The Conditions tab limits the particles that are affected by the node’s behavior. You can limit the
particle using probability or more specifically using sets.

Probability
The Probability slider determines the percentage of chance that the node affects any given particle.
The default value of 1.0 affects all particles. A setting of 0.6 would mean that each particle has a 60
percent chance of being affected by the control.
Probability is calculated for each particle on each frame. For example, a particle that is not affected by
a force on one frame has the same chance of being affected on the next frame.

Start/End Age
This range control can be used to restrict the effect of the node to a specified percentage of the
particle lifespan.
For example, to restrict the effect of a node to the last 20 percent of a particle’s life, set the Start value
to 0.8, and the End value remains at 1.0. The node on frames 80 through 100 only affects a particle
with a lifespan of 100 frames.

Set Mode Menu

The Set Mode menu drives how the particle node influences the active particle sets. There are three
options from this menu:
– Ignore Sets: The particle node disregards the state of the Set checkboxes and applies to all nodes.
– Affect Specified Sets: The particle node applies its behavior to the active Set checkboxes only.
– Ignore Specified Sets: The particle node applies its behavior to the inactive Set checkboxes only.

Chapter 52 Particle Nodes 1270

Set #
The state of a Set # checkbox determines if the Particle node’s effect will be applied to the particles in
the set. It allows you to limit the effects of some nodes to a subset of particles.
Sets are assigned by the nodes that create particles. These include the pEmitter, pImage Emitter,
pChangeStyle, and the pSpawn nodes.

Region Tab
The Region tab is used to restrict the node’s effect to a geometric region or plane, and to determine
the area where particles are created if it’s a pEmitter node or where the behavior of a node has
influence.

pEmitter Regions tab set to mesh

The Region tab is common to almost all particle nodes. In the pEmitter node Emitter Regions are used
to determine the area where particles are created. In most other tools it is used to restrict the tool’s
effect to a geometric region or plane. There are seven types of regions, each with its own controls.
Only one emitter region can be set for a single pEmitter node. If the pRender is set to 2D, then the
emitter region will produce particles along a flat plane in Z Space. 3D emitter regions possess depth
and can produce particles inside a user-defined, three-dimensional region.

Region Mode Menu

The Region Mode menu includes seven types of regions to define the area, each with its controls.
– All: In 2D, the particles will be created anywhere within the boundaries of the image. In 3D, this
region describes a cube 1.0 x 1.0 x 1.0 units in size.
– Bézier: Bézier mode uses a user-created polyline to determine the region where particles are
created. The Bézier mode works in both 2D and 3D modes; however, the Bézier polyline region
can only be created in 2D.
To animate the shape of the polyline over time or to connect it to another polyline, right-click the
Shape animation label at the bottom of the inspector and select the appropriate option from the
drop-down menu.
– Bitmap: A Bitmap source from one of the other nodes in the composition will be used as the
region where particles are born.
– Cube: A full 3D Cube is used to determine the region within which particles are created.
The height, width, depth, and XYZ positions can all be determined by the user and be animated
over time.
– Line: A simple line control determines where particles are created. The line is composed of two
end-points, which can be connected to Paths or Trackers, as necessary. This type of emitter
region includes X, Y, and Z position controls for the start and end of the line
– Mesh: Any 3D Mesh can be used as a region. In Mesh mode, the region can also be restricted by
the Object ID using the ObjectID slider. See below for a more in-depth explanation of how mesh
regions work.

Chapter 52 Particle Nodes 1271

 – Rectangle: The Rectangle region type is like the Cube type, except that this region has no depth
in Z space. Unlike other 2D emitter regions, this region can be positioned and rotated in Z space.
– Sphere: This is a spherical 3D emitter region with Size and Center Z controls. Sphere (3D) is the
default region type for a new pEmitter node.

Mesh Regions
Region Type
The Region Type drop-down menu allows you to choose whether the region will include the inner
volume or just the surface. For example, with a pEmitter mesh region, this determines if the particles
emit from the surface or the full volume.

Winding Rule and Winding Ray Direction

The Winding Rule and Winding Ray Direction parameters determine how the mesh region handles
particle creation with meshes that are not closed, as is common in many meshes imported from
external applications. This scenario is common with imported mesh geometry, and even geometry that
appears closed will frequently appear to “leak” thanks to improperly welded vertices.
To determine if a particle is in the interior of an object, a ray is cast from infinity through that particle
and then out to -infinity. The Winding Ray Direction determines which direction this ray is cast in. Each
time a surface is pierced by the ray, it is recorded and added onto a total to generate a winding
number. Going against a surfaces normal counts as +1, and going with the normal counts as -1.
The Winding Rule is then used to determine what is inside/outside. For example, setting the Winding
Rule to Odd means that only particles with odd values for the winding number are kept when creating
the particles. The exact same approach is used to ensure that polylines that intersect themselves are
closed properly.
For example, the following node tree and image show two image planes being used as a mesh region
for particle creation.

Chapter 52 Particle Nodes 1272

By setting the region’s Winding Ray Direction to the Z (blue) axis, this mesh can then be treated as a
closed volume for purposes of particle creation, as pictured below.

Limit By ObjectID
Selecting this checkbox allows the Object ID slider to select the ObjectID used as part of the region.

Style Tab
The Style tab exists in the pEmitter, pSpawn, pChangeStyle, and pImage Emitter. It controls the
appearance of the particles, allowing the look of the particles to be designed and animated over time.

Style
The Style menu provides access to the various types of particles supported by the Particle Suite. Each
style has its specific controls, as well as controls it will share with other styles.
– Point Style: This option produces particles precisely one pixel in size. Controls that are specific to
Point style are Apply Mode and Sub Pixel Rendered.
– Bitmap Style and Brush Style: Both the Bitmap and Brush styles produce particles based on an
image file. The Bitmap style relies on the image from another node in the node tree, and the Brush
style uses image files in the Brushes directory. They both have numerous controls for affecting
their appearance and animation, described below.
– Blob Style: This option produces large, soft spherical particles, with controls for Color, Size, Fade
timing, Merge method, and Noise.
– Line Style: This style produces straight line-type particles with optional “falloff.” The Size to
Velocity control described below (under Size Controls) is often useful with this Line type. The Fade
control adjusts the amount of falloff over the length of the line.
– Point Cluster Style: This style produces small clusters of single-pixel particles. Point Clusters are
similar to the Point style; however, they are more efficient when a large quantity of particles is
required. This style shares parameters with the Point style. Additional controls specific to Point
Cluster style are Number of Points and Number Variance.

Style Options
The following options appear only on some of the styles, as indicated below.
Apply Mode (Point and Point Cluster)
This control applies only to 2D particles; 3D particle systems are not affected.
– Add: Overlapping particles are combined by adding together the color values of each particle.
– Merge: Overlapping particles are merged.

Chapter 52 Particle Nodes 1273

Sub Pixel Rendered (Point and Point Cluster)
This checkbox determines whether the point particles are rendered with Sub Pixel precision, which
provides smoother-looking motion but blurrier particles that take slightly longer to render.

Number of Points and Variance (Point Cluster)

The value of this control determines how many points are in each Point Cluster.

Animate (Bitmap Style)

If the Bitmap source is a movie file or image sequence, this menu determines which frame is grabbed
from the source and applied to newly-created particles.
– Over Time: All particles use the image produced by the Style Bitmap node at the current time, and
change to each successive image together in step, as time increases. A particle created at frame
1 will contain the image at frame 1 of the Style Bitmap. At frame 2, the original particle will use the
image from frame 2, and so will any new particles. All created particles will share the same bitmap
image from their source at all times.
– Particle Age: Each particle animates through the sequence of images provided by the Style
Bitmap node, independently of other particles. In other words, an individual particle’s appearance
is taken from the Style Bitmap node at successive times, indexed by its age.
– Particle Birth Time: New particles take the image from the Style Bitmap node at the current time
and keep it unchanged until the end of the particle’s lifespan. Thus, particles generated on a given
frame will all have the same appearance and will stay that way.

Time Offset (Bitmap Style)

This control allows the Bitmap source frame to be offset in time from the current frame.

Time Scale (Bitmap Style)

This control scales the time range of the source bitmap images by a specified amount. For example, a
scale of 2 will cause the particle created at frame 1 to be read from the bitmap source at frame 2.

Gain (Bitmap and Brush Style)

This control applies a gain correction to the image used as the bitmap. Higher values produce a
brighter image, whereas lower values reduce both the brightness and the transparency of the image.

Style Bitmap (Bitmap Style)

This control appears when the Bitmap style is selected, along with an orange Style Bitmap input on
the node’s icon in the Node view. Connect a 2D node to this input to provide images to be used for
the particles. You can do this on the Node view, or you may drag and drop the image source node
onto the Style Bitmap control from the Node Editor or Timeline, or right-click on the control and select
the desired source from the Connect To menu.

Brush (Brush Style)

This menu shows the names of any image files stored in the Brushes directory. The location of the
Brushes directory is defined in the Preferences dialog, under Path Maps. The default is the Brushes
subdirectory within Fusion’s install folder. If no images are in this directory, the only option in the menu
will be None, and no particles will be rendered.

Noise (Blob Style)

Increasing this control’s value will introduce grain-type noise to the blobby particle.

Fade (Line Style)

The Fade control adjusts the falloff over the line particle’s length.
The default value of 1.0 causes the line to fade out completely by the end of the length.

Chapter 52 Particle Nodes 1274

Color Controls
The Color Controls select the color and Alpha values of the particles generated by the emitter.

Color Variance
These range controls provide a means of expanding the colors produced by the pEmitter. Setting the
Red variance range at -0.2 to +0.2 will produce colors that vary 20% on either side of the red channel,
for a total variance of 40%. If the pEmitter is set to produce R0.5, G0.5, B0.5 (pure gray), the variance
shown above will produce points with a color range between R0.3, G0.5, B0.5, and R0.7, G0.5, B0.5.
To visualize color space as values between 0-256 or as 0-65535, change the values used by Fusion
using the Show Color As option provided in the General tab within the Preferences dialog.

Lock Color Variance

This checkbox locks the color variance of the particles. Unlocking this allows the color variance to be
applied differently to each color channel, giving rise to a broader range of colors.

Particles Color Over Life controls

Color Over Life

This standard gradient control allows for the selection of a range of color values to which the particle
will adhere over its lifetime.
The left point of the gradient represents the particle color at birth. The right point shows the color of
the particle at the end of its lifespan.
Additional points can be added to the gradient control to cause the particle color to shift
throughout its life.
This type of control can be useful for fire-type effects (for example, the flame may start blue, turn
orange and end a darker red). The gradient itself can be animated over time by right-clicking on the
control and selecting Animate from the contextual menu. All points on the gradient will be controlled
by a single Color Over Life spline, which controls the speed at which the gradient itself changes. You
may also use the From Image modifier, which produces a gradient from the range of colors in an image
along a line between two points.

Chapter 52 Particle Nodes 1275

Chapter 53

Position Nodes
This chapter details the Position nodes available in Fusion.

The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Volume Fog [VLF]  1277
Volume Mask [VLM]  1284
Z to World Pos [Z2W]  1288
WPP Concept  1289
The Common Controls  1291

Chapter 53 Position Nodes 1276

Volume Fog [VLF]

The Volume Fog node

Volume Fog Node Introduction

The Volume Fog node is used to create sophisticated volumetric fog on images containing XYZ
Position channels.
As opposed to 3D-rendered volumetric fog, it works on 2D images and delivers much faster results
and interactive feedback when setting up the fog. See the “WPP Concept” section at the end of this
chapter for further explanation of how this technology works and to learn about the required imagery.

Basic Node Setup

The Volume Fog node takes an image input; in the example below, it is a Renderer 3D with World
Position enabled in the output channels. Another input is the 3D scene, which contains the camera.
A Fast Noise node generates the fog texture.

Volume Fog Node Structure

A 3D scene and a rendered scene connected to Volume Fog

Inputs
The following inputs appear on the Volume Fog node in the Node Editor.
– Image: The orange input accepts the primary image where the fog will be applied. This image
contains a World Position Pass in the XYZ Position channels.
– Fog Image: The green Fog image input is for creating volumetric fog with varying depth and
extent; a 2D image can be connected here. A good starting point is to use a Fast Noise at a
small resolution of 256 x 256 pixels.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the fog to
certain areas.
– Scene Input: The magenta scene input accepts a 3D scene containing a 3D Camera.

Chapter 53 Position Nodes 1277

Inspector

The Volume Fog Shape tab

Shape Tab
The Shape tab defines the size and location of the fog volume. You can either use the Pick buttons to
select the location and orientation in the viewer or use the Translation, Rotation, and Scale controls.

Shape
This menu switches between a basic spherical or rectangular volume to be placed in your image.
These volumes can then be further refined using the Fog image and effect mask.

Pick
Drag the Pick button into the viewer to select the XYZ coordinates from any 3D scene or 2D image
containing XYZ values, such as a rendered World Pass, to position the center of the Volume object.
When picking from a 2D image, make sure it’s rendered in 32-bit float to get full precision.

X, Y, Z Offset
These controls can be used to position the center of the fog volume manually or can be animated or
connected to other controls in Fusion.

Rotation Pick
Drag the Pick button into the viewer to select the rotational values from any 3D Scene or 2D image
containing those values, like an XYZ-Normal-Pass, to reorient the fog volume.
When picking from a 2D image, like an XYZ Normal pass, make sure it’s rendered in 32-bit float to get
full precision and accurate rotational values.

X, Y, Z Rotation
Use these controls to rotate the fog volume around its center.

X, Y, Z Scale
Scale the fog volume in any direction from its center to refine further the overall Size value
specified below.

Chapter 53 Position Nodes 1278

Size
The overall size of the fog volume created.

Soft Edge
Controls how much the fog volume is faded toward the center from its perimeter to achieve a
softer look.

The Volume Fog Color tab

Color Tab
The Color tab controls the detail and color of the fog.

Adaptive Samples
Volumes images consist of multiple layers, so there may be 64 layers in a volume. This checkbox
adjusts the rendering algorithm for how to best blend those layers.
Dither: Applies a form of noise to improve the blending and hide visible layer differences.

Samples
Determines how many times a “ray” shot into the volume will be evaluated before the final image is
created. Not unlike raytracing, higher values lead to more detail inside the volume but also increase
render times.

Z Slices
The higher the Z Slices value, the more images from the connected Fog image sequence will be used
to form the depth of the volume.
You can, for example, use a Fast Noise with a high Seethe Rate to create such a sequence of images.
Be careful with the resolution of the images. Higher resolutions can require a large amount of memory.
As a rule of thumb, a resolution of 256 x 256 pixels with 256 Z Slices (i.e., forming a 256 x 256 x 256
cubic volume, which will use up to 256 MB for full color 32-bit float data) should give you a good
starting point.

Chapter 53 Position Nodes 1279

First Slice Time
Determines which frame of the Global Range is used to deliver the first slice from the connected fog
image sequence.
Make sure that both Global In and Global Out, as well as the valid range of your source node, fall
within the range of First Slice Time + Z Slices.

Color
Allows you to modify the color of the fog generated. This will multiply over any color provided by the
connected Fog image.

Gain
Increases or decreases the intensity of the fog. More Gain will lead to a stronger glow and less
transparency in the fog. Lower values let the fog appear less dense.

Subtractive/Additive Slider
Similar to the Merge node, this value controls whether the fog is composed onto the image in Additive
or Subtractive mode, leading to a brighter or dimmer appearance of the fog.

Fog Only
This option outputs the generated fog on a black background, which then can be composited
manually or used as a mask on a Color Corrector for further refinement.

The Volume Fog Noise tab

Noise Tab
The Noise tab controls the shape and pattern of the noise added to the fog.

Detail
Increase the value of this slider to produce a greater level of detail in the noise result. Larger values
add more layers of increasingly detailed noise without affecting the overall pattern. High values take
longer to render but can produce a more natural result.

Gain
This control increases or decreases the brightest parts of the noise map.

Brightness
This control adjusts the overall brightness of the noise map, before any gradient color mapping is
applied. In Gradient mode, this produces a similar effect to the Offset control.

Chapter 53 Position Nodes 1280

Translation
Use the Translation coordinate control to pan and move the noise pattern.

Noise Rotation
Use the Rotation controls to orient the noise pattern in 3D.

Seethe
Adjust this thumbwheel control to interpolate the noise map against a different noise map. This will
cause a crawling shift in the noise, like it was drifting or flowing. This control must be animated to
affect the noise over time.

Discontinuous
Normally, the Noise function interpolates between values to create a smooth, continuous gradient of
results. Enable this checkbox to create hard discontinuity lines along some of the noise contours. The
result will be a dramatically different effect.

Inverted
Select this checkbox to invert the noise, creating a negative image of the original pattern. This is most
effective when Discontinuous is also enabled.

The Volume Fog Camera tab

Camera Tab
For a perfect evaluation of a fog volume, a camera or 3D scene can be connected to the Scene input
of the node.

Camera
If multiple cameras are available in the connected Scene input, this menu allows the selection of the
correct camera needed to evaluate the fog volume. Instead of connecting a camera, position values
can be provided manually or by connecting the XYZ values to other controls.

Translation Pick
Drag the Pick button into the viewer to select XYZ coordinates from any 3D scene or 2D image
containing XYZ values, like a rendered World Pass, to define the center of the camera. When picking
from a 2D image, make sure it’s rendered in 32-bit float to get full precision.

X, Y, Z Offset
These controls can be used to define the center of the camera manually or can be animated or
connected to other controls in Fusion.

Chapter 53 Position Nodes 1281

The Volume Fog Light tab

Light Tab
To utilize the controls in the Light tab, you must have actual lights in your 3D scene. Connect that
scene, including Camera and Lights, to the 3D input of the node.

Do Lighting
Enables or disables lighting calculations. Keep in mind that when not using OpenCL (i.e., rendering on
the CPU), these calculations may become a bit slow.

Do In-Scattering
Enables or disables light-scattering calculations. The volume will still be lit according to the state of the
Do Lighting checkbox, but scattering will not be performed.

Light Samples
Determines how accurate the lighting is calculated. Higher values mean more accurate calculation at
the expense of longer render times.

Density
This is similar to scattering in that it makes the fog appear thicker. With a high amount of scattering,
though, the light will be scattered out of the volume before it has had much chance to travel through
the fog, meaning it won’t pick up a lot of the transmission color. With a high density instead, the fog still
appears thicker, but the light gets a chance to be transmitted, thus picking up the transmission color
before it gets scattered out. Scattering is affected by the light direction when Asymmetry is not 0.0.
Density is not affected by light direction at all.

Chapter 53 Position Nodes 1282

Scattering
Determines how much of the light bouncing around in the volume ends up scattering the light out of
the fog. If the light scatters more, or more accurately, then there’s a higher probability of the light being
scattered out of the volume, hence less light is left to continue through the fog. This option can make
the fog seem denser.

Asymmetry
Determines in what direction the light is scattered. A value of 0 produces uniform, or isotropic,
scattering, meaning all directions have equal probability. A value greater than 0 causes “forward
scattering,” meaning the light is scattered more into the direction of the light rays. This is similar to
what happens with water droplets in clouds. A value smaller than 0 produces “back scattering,” where
the light is more scattered back toward the original light source.

Transmission
Defines the color that is transmitted through the fog. The light that doesn’t get scattered out will tend
toward this color. It is a multiplier, though, so if you have a red light, but blue transmission, you won’t
see any blue.

Reflection
Changes the intensity of the light that is scattered out. Reflection can be used to modify the overall
color before Emission is added. This will be combined with the color channels of the volume texture
and then used to scale the values. The color options and the color channels of the volume texture are
multiplied together, so if the volume texture were red, setting the Reflection color options to blue
would not make the result blue. In such a case, they will multiply together to produce black.

Emission
This adds a bit of “glowing” to the fog, adding energy/light back into the calculation. If there are no
lights in the scene, and the fog emission is set to be 1.0, the results are similar to no lighting, like
turning off the Do Lighting option. Glowing can also be done while producing a different kind of look,
by having a Transmission greater than 1. This, however, would never happen in the real world.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Position nodes. These common controls
are described in detail at the end of this chapter in “The Common Controls” section.

Examples
In these examples, we are looking at a volume from the outside. On the left, you see how the
Volume Fog looks with straight accumulation. That means the Do Lighting option is turned off.

On the right, you see the same volume with lighting/scattering turned on, and a single
point light.

Chapter 53 Position Nodes 1283

 Here, we have a slightly more complex Volume.

On the left with straight accumulation; in the middle with lighting, scattering, and a single point
light; and on the right, the light in the scene has been moved, which also influences the look of
the volume.

Volume Mask [VLM]

The Volume Mask node

Volume Mask Node Introduction

The Volume Mask node is used to create volumetric masks from images containing XYZ
Position channels.
This can, for example, be used to isolate objects for color correction without the need to track or
rotoscope the scene. See the “WPP Concept” section later in this chapter for further explanation on
how this technology works and to learn about the required imagery.

Inputs
The following three inputs appear on the Volume Mask node in the Node Editor:
– Image: The orange image input accepts a 2D image containing a World Position Pass in the
XYZ Position channels.
– Mask Image: An image can be connected to the green mask image input for refining the mask.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the volume
mask to certain areas.

Basic Node Setup

Below, a MediaIn labeled RGBA contains the main rendered image from a 3D scene. A World Position
Pass from a 3D-rendered scene is labeled WPP_PASS. The Channel Booleans node is used to map
the X position pass to the red channel, Y to the green channel, and Z to the blue channel. The Volume
Mask tool extracts the 3D position information from the World Position Pass to place a mask in 3D

Chapter 53 Position Nodes 1284

space. The mask is then used as an effect mask on a color corrector to selectively color correct
objects in a 3D scene.

A Volume Mask tool takes advantage of World Position Pass for color correction in a 3D scene

Inspector

The Volume Mask Shape tab

Shape Tab
The Shape tab defines the size and location of the Volume Mask. You can either use the Pick buttons
to select the location and orientation in the viewer or use the Translation, Rotation, and Scale controls.

Shape
This menu switches between a spherical or rectangular mask to be placed in your image. The mask
can be further refined using the mask image input.

Translation Pick
Drag the Pick button into the viewer to select XYZ coordinates from any 3D scene or 2D image
containing XYZ values, like a rendered World Pass, to position the center of the Volume Mask. When
picking from a 2D image, make sure it’s rendered in 32-bit float to get full precision.

Chapter 53 Position Nodes 1285

X, Y, Z Offset
These controls can be used to position the center of the mask manually or can be animated or
connected to other controls in Fusion.

Rotation Pick
Drag the Pick button into the viewer to select rotational values from any 3D scene or 2D image
containing those values, like an XYZ Normal pass, to reorient the mask.
When picking from a 2D image, like an XYZ Normal pass, make sure it’s rendered in 32-bit float, and
use World Space coordinates to get full precision and the correct rotational values.

X, Y, Z Rotation
Use these controls to rotate the mask around its center.

X, Y, Z Scale
Scale the mask in any direction from its center to further refine the overall Size value specified below.

Size
The overall size, in X, Y, and Z, of the mask created.

Soft Edge
Controls how much the Volume is faded toward the center from its perimeter to achieve a softer look.

r
The Volume Mask Color tab

Color Tab
The Color tab controls the color and blending of the mask image.

Color
Allows you to modify the color of the generated Volume Mask. This will add to any color provided by
the connected mask image.

Subtractive/Additive Slider
Similar to the Merge node, this value controls whether the mask is composed onto the image in
Additive or Subtractive mode, leading to a brighter or dimmer appearance of the mask.

Chapter 53 Position Nodes 1286

Mask Only
Outputs the generated mask on a black background, which then can be used as a mask on a Color
Corrector for further refinement.

The Volume Mask Camera tab

Camera Tab
For a perfect evaluation of a Volume, a camera or 3D scene can be connected to the Scene input
of the node.

Camera
If multiple cameras are available in the connected Scene input, this drop-down menu allows you to
choose the correct camera needed to evaluate the Volume.
Instead of connecting a camera, position values can also be provided manually or by connecting the
XYZ values to other controls.

Translation Pick
Drag the Pick button into the viewer to select XYZ coordinates from any 3D scene or 2D image
containing XYZ values, like a rendered World Pass, to define the center of the camera.
When picking from a 2D image, make sure it’s rendered in 32-bit float to get full precision.

X, Y, Z Offset
These controls can be used to define the center of the camera manually or can be animated or
connected to other controls in Fusion.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Position nodes. These common controls
are described in detail at the end of this chapter in “The Common Controls” section.

Chapter 53 Position Nodes 1287

Z to World Pos [Z2W]

The Z to World Position node

Z to World Pos Node Introduction

The Z to World Pos node is used to either generate a World Position Pass from a Z channel and a 3D
Camera or a Z channel from a World Position Pass and a 3D Camera.
Creating a World Position Pass from Z-depth can be useful when your 3D application is not capable of
creating a WPP.
It can also be used when a 3D-tracking software outputs a per-pixel Z-depth together with the 3D
Camera. Thus, the Volume Mask and Volume Fog could be applied to real-world scenes. The quality of
the resulting WPP depends mainly on the quality of the incoming Z channel.
See the “WPP Concept” section for further explanation on how this technology works and to learn
about the required imagery.

Inputs
The following inputs appear on the node tile in the Node Editor:
– Image: The orange image input accepts an image containing a World Position Pass or a
Z-depth pass, depending on the desired operation.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the World
Position Pass to certain areas.
– Scene Input: The magenta scene input accepts a 3D scene input containing a 3D Camera.

Basic Node Setup

Below, a MediaIn labeled RGBA contains the main rendered image from a 3D scene. A Z-depth pass
from a 3D-rendered scene is labeled Z_PASS. The Channel Booleans node is used to map the Aux Z
channel into either the red, green, or blue channel. The Z to World Position node is placed after the
Channel Booleans node, and an imported 3D camera that matches the RGBA image is connected to
the 3D camera input on the Z to World Position node. A Channel Booleans node is placed after the Z
to World Position node, which can remap the X, Y, and Z positions for use in other nodes.

A Z to World Position node creates a World Position Pass from a Z-depth pass

Chapter 53 Position Nodes 1288

Inspector

The Z to World Position Controls tab

Controls Tab
The Controls tab determines whether you are creating a World Position Pass or a Z channel. If there is
more than one camera in the connected scene, this tab also selects the camera to use for the
calculation.

Mode
This menu switches between creating a Z channel from a World Position Pass or vice versa.

Camera
If multiple cameras are available in the connected Scene input, this drop-down menu allows you to
choose the correct camera needed to evaluate the image.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Position nodes. These common controls
are described in detail at the end of this chapter in “The Common Controls” section.

WPP Concept
The Position nodes in Fusion offer an entirely new way of working with masks and Volumetrics for
footage containing XYZ Position channels. Z to World offers the option to create those channels out of
a Z channel and 3D Camera information. For this overview, we refer to the World Position Pass as WPP.

What Is a WPP?
The WPP interprets each pixel’s XYZ position in 3D space as an RGB color value.
For instance, if a pixel sits at 0/0/0, the resulting pixel has an RGB value of 0/0/0 and thus will be black.
If the pixel sits at 1/0/0 in the 3D scene, the resulting pixel is entirely red. Of course, if the coordinates
of the pixel are something like -60/75/123, WPP interprets those values as RGB color values as well.
Due to the potentially enormous size of a 3D scene, the WPP channel should always be rendered in
32-bit floating point to provide the accuracy needed. The image below shows a 3D rendering of a
scene with its center sitting at 0/0/0 in 3D Space and the related WPP. For better visibility, the WPP is
normalized in this example.

Chapter 53 Position Nodes 1289

Different Coordinate Spaces
Rendering WPPs can occur in different Coordinate Spaces. These include World Space, Eye Space,
and Object Space. The image below depicts how those different spaces look, although the nodes in
Fusion require the WPP rendering to occur in World Space.

The Scene Input

The nodes offer a Scene input, which can either be a 3D camera or a 3D scene containing a camera.
While the camera is vital for the Z to World node, Volume Mask and Volume Fog can generate their
output without any camera attached or with the camera position set to 0/0/0.
However, connecting a camera that lines up with the original camera the WPP has been rendered
from, or setting the camera’s position manually, dramatically improves the accuracy and look of the
resulting fog or mask.

The “Invisible Sphere”

The example scene shown so far has an empty background, meaning there is nothing in the scene
apart from the ground plane and the cubes.
If applying fog to a scene like that, which is larger than the ground plane, the result will look similar to
the “w/o Sphere” example shown below because, with no WPP information outside the ground plane,
the resulting value is 0/0/0, and the fog fills that area as well.
To get around that, you can add an invisible bounding sphere to your scene to create “dummy” WPP
values to help the Fog node to create the correct volume as shown in the “with Sphere”
example below.

Chapter 53 Position Nodes 1290

The Common Controls
Nodes that handle Position operations share several identical controls in the Inspector. This section
describes controls that are common among Position nodes.

Inspector

Position nodes Common Settings tab

Settings Tab
The Settings tab in the Inspector can be found on every tool in the Position category. The controls are
consistent and work the same way for each tool.

Blend
The Blend control is used to blend between the tool’s original image input and the tool’s final modified
output image. When the blend value is 0.0, the outgoing image is identical to the incoming image.
Normally, this will cause the tool to skip processing entirely, copying the input straight to the output.

Process When Blend Is 0.0

The tool is processed even when the input value is zero. This can be useful if the node is scripted to
trigger a task, but the node’s value is set to 0.0.

Red/Green/Blue/Alpha Channel Selector

These four buttons are used to limit the effect of the tool to specified color channels. This filter is often
applied after the tool has been processed.
For example, if the Red button on a Blur tool is deselected, the blur will first be applied to the image,
and then the red channel from the original input will be copied back over the red channel of the result.

Chapter 53 Position Nodes 1291

There are some exceptions, such as tools for which deselecting these channels causes the tool to skip
processing that channel entirely. Tools that do this will generally possess a set of identical RGBA
buttons on the Controls tab in the tool. In this case, the buttons in the Settings and the Controls tabs
are identical.

Apply Mask Inverted

Enabling the Apply Mask Inverted option inverts the complete mask channel for the tool. The mask
channel is the combined result of all masks connected to or generated in a node.

Multiply by Mask
Selecting this option will cause the RGB values of the masked image to be multiplied by the mask
channel’s values. This will cause all pixels of the image not included in the mask (i.e., set to 0) to
become black/transparent.

Use Object/Use Material (Checkboxes)

Some 3D software can render to file formats that support additional channels. Notably, the EXR file
format supports Object and Material ID channels, which can be used as a mask for the effect. These
checkboxes determine whether the channels will be used if present. The specific Material ID or Object
ID affected is chosen using the next set of controls.

Correct Edges
This checkbox appears only when the Use Object or Use Material checkboxes are selected. It toggles
the method used to deal with overlapping edges of objects in a multi-object image. When enabled,
the Coverage and Background Color channels are used to separate and improve the effect around the
edge of the object. If this option is disabled (or no Coverage or Background Color channels are
available), aliasing may occur on the edge of the mask.
For more information on the Coverage and Background Color channels, see Chapter 18,
“Understanding Image Channels and Node Processing” in the Fusion Reference Manual or Chapter 79
in the DaVinci Resolve Reference Manual.

Object ID/Material ID (Sliders)

Use these sliders to select which ID will be used to create a mask from the object or material channels
of an image. Use the Sample button in the same way as the Color Picker: to grab IDs from the image
displayed in the viewer. The image or sequence must have been rendered from a 3D software
package with those channels included.

Motion Blur
– Motion Blur: This toggles the rendering of Motion Blur on the tool. When this control is toggled
on, the tool’s predicted motion is used to produce the motion blur caused by the virtual camera’s
shutter. When the control is toggled off, no motion blur is created.
– Quality: Quality determines the number of samples used to create the blur. A quality setting of 2
will cause Fusion to create two samples to either side of an object’s actual motion. Larger values
produce smoother results but increase the render time.
– Shutter Angle: Shutter Angle controls the angle of the virtual shutter used to produce the motion
blur effect. Larger angles create more blur but increase the render times. A value of 360 is the
equivalent of having the shutter open for one full frame exposure. Higher values are possible and
can be used to create interesting effects.
– Center Bias: Center Bias modifies the position of the center of the motion blur. This allows for the
creation of motion trail effects.
– Sample Spread: Adjusting this control modifies the weighting given to each sample. This affects
the brightness of the samples.

Chapter 53 Position Nodes 1292

Use GPU
The Use GPU menu has three settings. Setting the menu to Disable turns off GPU hardware-
accelerated rendering. Enabled uses the GPU hardware for rendering the node. Auto uses a capable
GPU if one is available and falls back to software rendering when a capable GPU is not available.

Hide Incoming Connections

Enabling this checkbox can hide connection lines from incoming nodes, making a node tree appear
cleaner and easier to read. When enabled, empty fields for each input on a node will be displayed in
the Inspector. Dragging a connected node from the node tree into the field will hide that incoming
connection line as long as the node is not selected in the node tree. When the node is selected in the
node tree, the line will reappear.

Comments
The Comments field is used to add notes to a tool. Click in the empty field and type the text. When a
note is added to a tool, a small red square appears in the lower-left corner of the node when the full
tile is displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the
note in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 53 Position Nodes 1293

Chapter 54

Resolve Connect
This chapter details the single node found in the Resolve Connect category, available
only in standalone Fusion Studio.

The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
External Matte Saver [EMS]  1295

Chapter 54 Resolve Connect 1294

External Matte Saver [EMS]

The External Matte Saver node

NOTE: The Resolve Connect category and External Matte Saver node are available only in
Fusion Studio.

External Matte Saver Node Introduction

The External Matte Saver node renders multiple mattes into multiple channels of an EXR file. This file
is intended to import into DaVinci Resolve’s Color page as an efficient way to deliver multiple mattes
for color grading. To perform the same operation with a traditional Saver node, a Channel Boolean is
needed to place each matte into a channel, and then name that channel in the Saver. It requires a bit
more of a set up. This node streamlines the process by providing multiple inputs and naming the
channel based on the node.

Inputs
By default, the node provides a single input for a 2D image you want to save as a matte.
– Input: Although initially there is only a single orange input for a matte to connect, the Inspector
provides an Add button for adding additional inputs. Each input uses a new color, but all accept
2D RGBA images.

Basic Node Setup

The External Matte Saver node can be placed at the end of any node tree branch for saving mattes.
Below, it is connected to the Delta Keyer as well as three other mattes. Each matte beyond the initial
one is connected by clicking the Add button in the Inspector first, and then a new input is provided
on the node.

An External Matte Saver node added as a separate branch in a node tree to render the mattes

Chapter 54 Resolve Connect 1295

Inspector

The External Matte Saver Controls tab

Controls Tab
The Controls tab is used to name the saved file and determine where on your hard drive the file
is stored.

Filename
Enter the name you want to use for the EXR file in the Filename field. At the end of the name, append
the .exr extension to ensure that the file is saved as an EXR file.

Browse
Clicking the Browse button opens a standard file browser window where you can select the location
to save the file.

The External Matte Saver Mattes tab

Mattes Tab
The Mattes tab is where you set up the number of mattes saved in the file, the name for each channel,
and the RGBA channels saved from each input.

Channels menu
The Channels menu allows you to select which channels are saved in the matte. You can choose the
alpha channel, the RGB channels, or the RGBA channels.

Channels Name
The Channels Name field allows you to customize the name of the matte channel you are saving. This
name is displayed in DaVinci Resolve’s Color page.

Node Name
The Node Name field displays the source of the matte. This is automatically populated when you
connect a node to the input.

Add
Clicking the Add button adds an input on the node and another set of fields for you to configure and
name the new matte channel.

Chapter 54 Resolve Connect 1296

The External Matte Saver Settings tab

Settings Tab
The Settings Tab in the Inspector is similar to settings found in the Saver tool. The controls are
consistent and work the same way as the Settings in other tools.

Blend
The Blend control is used to blend between the tool’s original image input and the tool’s final modified
output image. When the blend value is 0.0, the outgoing image is identical to the incoming image.
Normally, this causes the tool to skip processing entirely, copying the input straight to the output.

Process When Blend Is 0.0

The tool is processed even when the input value is zero. This can be useful if processing of this node
is scripted to trigger another task, but the value of the node is set to 0.0.

Red/Green/Blue/Alpha Channel Selector

These four buttons are used to limit the effect of the tool to specified color channels. This filter is often
applied after the tool has been processed.
For example, if the Red button on a Blur tool is deselected, the blur is first applied to the image, and
then the red channel from the original input is copied back over the red channel of the result.
There are some exceptions, such as tools for which deselecting these channels causes the tool to skip
processing that channel entirely. Tools that do this generally possess a set of identical RGBA buttons
on the Controls tab in the tool. In this case, the buttons in the Settings and the Controls tabs are
identical.

Chapter 54 Resolve Connect 1297

Apply Mask Inverted
Enabling the Apply Mask Inverted option inverts the complete mask channel for the tool. The mask
channel is the combined result of all masks connected to or generated in a node.

Multiply by Mask
Selecting this option causes the RGB values of the masked image to be multiplied by the mask
channel’s values. This causes all pixels of the image not included in the mask (i.e., set to 0) to become
black/transparent.

Use Object/Use Material (Checkboxes)

Some 3D software can render to file formats that support additional channels. Notably, the EXR file
format supports Object ID and Material ID channels, which can be used as a mask for the effect. These
checkboxes determine whether the channels are used, if present. The specific Material ID or Object ID
affected is chosen using the next set of controls.

Correct Edges
This checkbox appears only when the Use Object or Use Material checkboxes are selected. It toggles
the method used to deal with overlapping edges of objects in a multi-object image. When enabled,
the Coverage and Background Color channels are used to separate and improve the effect around the
edge of the object. If this option is disabled (or no Coverage or Background Color channels are
available), aliasing may occur on the edge of the mask.
For more information on the Coverage and Background Color channels, see Chapter 18,
“Understanding Image Channels and Node Processing” in the Fusion Reference Manual or Chapter 79
in the DaVinci Resolve Reference Manual.

Object ID/Material ID (Sliders)

Use these sliders to select which ID is used to create a mask from the object or material channels of an
image. Use the Sample button in the same way as the Color Picker: to grab IDs from the image
displayed in the viewer. The image or sequence must have been rendered from a 3D software
package with those channels included.

Motion Blur
– Motion Blur: This toggles the rendering of Motion Blur on the tool. When this control is toggled
on, the tool’s predicted motion is used to produce the motion blur caused by the virtual camera’s
shutter. When the control is toggled off, no motion blur is created.
– Quality: Quality determines the number of samples used to create the blur. A quality setting of
2 causes Fusion to create two samples to either side of an object’s actual motion. Larger values
produce smoother results but increase the render time.
– Shutter Angle: Shutter Angle controls the angle of the virtual shutter used to produce the motion
blur effect. Larger angles create more blur but increase the render times. A value of 360 is the
equivalent of having the shutter open for one full frame exposure. Higher values are possible and
can be used to create interesting effects.
– Center Bias: Center Bias modifies the position of the center of the motion blur. This allows for the
creation of motion trail effects.
– Sample Spread: Adjusting this control modifies the weighting given to each sample. This affects
the brightness of the samples.

Chapter 54 Resolve Connect 1298

Hide Incoming Connections
Enabling this checkbox can hide connection lines from incoming nodes, making a node tree appear
cleaner and easier to read. When enabled, empty fields for each input on a node will be displayed in
the Inspector. Dragging a connected node from the node tree into the field will hide that incoming
connection line as long as the node is not selected in the node tree. When the node is selected in the
node tree, the line will reappear.

Comments
The Comments field is used to add notes to a tool. Click in the empty field and type the text. When a
note is added to a tool, a small red square appears in the lower-left corner of the node when the full
tile is displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the
note in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 54 Resolve Connect 1299

Chapter 55

Shape Nodes
This chapter details the Shape nodes available in Fusion.

Contents
sBoolean  1301
sDuplicate  1304
sEllipse  1306
sExpand  1309
sGrid  1310
sJitter  1311
sMerge  1313
sNGon  1314
sOutline  1317
sRectangle  1319
sRender  1322
sStar  1324
sTransform  1327
Common Controls  1329

Chapter 55 Shape Nodes 1300

sBoolean

The sBoolean node

The sBoolean node combines or excludes overlapping areas of two shapes based on a menu of
boolean operations.
Like almost all shape nodes, you can only view the sBoolean node’s results through a sRender node.

External Inputs
The following inputs appear on the node’s tile in the Node Editor. Except when using the subtract
boolean operation, which shape you connect into which input does not matter.
– Input1: [orange, required] This input accepts the output of another shape node. This input is
used as the base shape when the subtract boolean operation is chosen.
– Input2: [green, optional] This input accepts the output of another shape node. This input is
used to cut the base shape hole when the subtract boolean operation is chosen.

Basic Node Setup

The sBoolean node is used to combine two shape nodes. The output of the sBoolean can then be
output to another shape node or a sRender node for viewing.

Star and ellipse shapes combined in an sBoolean node, then

output to an sRender for viewing and further processing

Inspector

The sBoolean Controls tab

Chapter 55 Shape Nodes 1301

Controls
The Controls tab is primarily used to select the boolean operation that determines how the two shapes
are combined.

Operation
The operation menu includes four boolean operations:
– Intersection: Sometimes called an AND operation, this setting will only show areas where the two
shapes overlap. The result is only where input 1 AND input 2 overlap.

Star and ellipse shapes with an sBoolean node set to intersection

– Union: Sometimes called an OR operation, this setting will only show areas where either of the
two shapes exists. The result is where either input 1 OR input 2 exists. The Union setting is similar
to the result of the sMerge node.

Star and ellipse shapes with an sBoolean node set to union

– Subtract: Sometimes called a NOT operation, this setting outputs the shape of input 1 but
eliminates the areas where input 2 overlaps. The result is input 1 minus input 2.

Star and ellipse shapes with an sBoolean node set to subtract

– Xor: Sometimes called an AND NOT operation, this setting outputs the shape of input 1 or input 2
but eliminates the areas where they overlap. The result is (input 1 minus input 2) + (input 2 minus
input 1).

Chapter 55 Shape Nodes 1302

 Star and ellipse shapes with an sBoolean node set to xor

Style Mode
The Style mode menu only includes one option. The Replace setting replaces the color and alpha
level of the incoming shapes with the color set in the Style tab.

Style Tab

The sBoolean Style tab

Style
Any color assigned to the individual shape nodes is replaced by the color set using the Style
tab controls.

Color
The color controls determine the color of the output shape from the sBoolean node. To choose a
shape color, you can click the color disclosure arrow, use the color swatch, or drag the eyedropper
into the viewer to select a color from an image. The RGBA sliders or number fields can be used to
enter each color channel’s value or the strength of the alpha channel.

Allow Combining
When this checkbox is enabled, the alpha channel value is maintained even when passing through
other nodes downstream that may cause the shape to overlap with copies of itself. When disabled, the
alpha channel value may increase when the shape overlaps itself.
For instance, if an ellipse’s alpha channel is set to .5, enabling the Allow Combining checkbox
maintains that value even if the shape passes through a duplicate or grid node that causes the shape
to overlap. Disabling the checkbox causes the alpha channel values to be compounded at each

Chapter 55 Shape Nodes 1303

overlapping area. When using the sBoolean node, the individual shape node checkboxes are ignored,
and the sBoolean node’s checkbox determines the alpha channel’s behavior.

Allow Combining Enabled (left), Allow Combining Disabled (right)

Common Controls
Settings tab
The Settings tab in the Inspector is common to all Shape nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

sDuplicate

The sDuplicate node

The sDuplicate node creates copies of the input shape, offsetting each copy’s position, size, and
rotation. Like almost all shape nodes, you can only view the sDuplicate node’s results through a
sRender node.

External Inputs
The following input appears on the node’s tile in the Node Editor:
– Input1: [orange, required] This input accepts the output of another Shape node. The shape
connected to this input is copied and offset based on the controls in the Inspector.

Basic Node Setup

The sDuplicate node takes a single input that is most often from a single or compound shape from a
sMerge or sBoolean node. The sDuplicate node creates copies of the incoming shape and offsets
them to create a pattern. The output of the sDuplicate can then be output to another shape node or to
a sRender node for viewing or compositing into the greater node tree.

An sEllipse shape connected to an sDuplicate and then output to

an sRender for viewing and combining with other elements

Chapter 55 Shape Nodes 1304

Inspector

The sDuplicate Controls tab

Controls
The Controls tab is used to determine the number of copies and set their position, size, and
rotation offset.

Copies
This slider determines the number of copies created by the node. The number does not include the
original shape, so entering a value of five will produce five copies plus the original.

X and Y Offset
These sliders set the X and Y distance between each of the copies. Each copy is offset from the
previous copy by the value entered in the X and Y number fields. The copies all start at 0, the center of
the original shape, and are offset from there. Using Fusion’s normalized coordinate system, entering X
Offset at 0.5 would move each copy half the frame’s width to the right. Entering -1.0 would move each
copy to the left by the width of the frame.

X and Y Size
Sets the X and Y size offset based on the previous shape size. For instance, an X and Y value of 1.0
creates copies identical in size to the original but. Entering a value of X and Y of 0.5 will cause each
copy to be half the size of the copy before it.

Axis Mode
The Axis mode menu provides four options for determining how each copy determines its rotational
pivot point.
– Absolute: Allows you to set an X and Y position for the axis of rotation based on the original
shape’s location. The axis of rotation is then copied and offset with each duplicated shape.
– Origin Relative: Each copy uses its center point as its axis of rotation.
– Origin Absolute: Each copy uses the center of the original shape as its axis of rotation.
– Progressive: Compounds each shape copy by progressively transforming each copy based on the
previous shape’s position, rotation, and scale.

X and Y Pivot
The X and Y pivot controls are displayed when the Axis mode is set to Absolute. You can use these
position controls to place the axis of rotation.

Rotation
Determines an offset rotation applied to each copy. The rotation is calculated from the offset rotation
of the previous copy. To rotate all copies identically, use the Angle parameter on the original shape or
use a sTransform node.

Chapter 55 Shape Nodes 1305

Common Controls
Settings tab
The Settings tab in the Inspector is common to all Shape nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

sEllipse

The sEllipse node

The sEllipse node is used to create circular shapes. Like almost all shape nodes, you can only view the
sEllipse node’s results through a sRender node.

External Inputs
This node generates shapes and does not have any inputs.

Basic Node Setup

The sEllipse node is a shape generator, meaning it generates a shape and therefore has no input. The
output of the sEllipse can go into a sRender for viewing and further compositing or, more likely,
connect to another shape node like sGrid or sDuplicate.

An sEllipse node connecting to an sGrid node, and then viewed using an sRender node

Inspector

The sEllipse Controls tab

Chapter 55 Shape Nodes 1306

Controls
The Controls tab is used to define the elliptical shape characteristics, including fill, border, size,
and position.

Solid
When enabled, the Solid checkbox fills the elliptical shape with the color defined in the Style tab.
When disabled, an outline created by the Border Width control is displayed, and the center is made
transparent.

Border Width
This parameter expands or contracts the border around the shape. Although it can be used when the
Solid checkbox is enabled, it is primarily used to determine the outline thickness when the checkbox
is disabled.

Cap style
When the Solid checkbox is disabled, three cap style options are displayed. The cap styles can create
lines with flat, rounded, or squared ends. Flat caps have flat, squared ends, while rounded caps have
semi-circular ends. Squared caps have projecting ends that extend half the line width beyond the end
of the line.
The caps are not visible unless the length is below 1.0.

Position
The position parameter is only displayed when the Solid checkbox is disabled. It allows you to position
the starting point of the shape. When used in conjunction with the length parameter, it positions the
gap in the ellipse outline.

Length
The length parameter is only displayed when the Solid checkbox is disabled. A length of 1.0 is a
closed shape. Setting the length below 1.0 creates an opening or gap in the outline. Keyframing the
length parameters allows you to create write-on style animations.

X and Y Offset
These parameters are used to position the shape left, right, up, and down in the frame. The shape
starts in the center of the frame, and the parameters are used to offset the position. The offset
coordinates are normalized based on the width of the frame. An X offset of 0.0 is centered, and a
value of 0.5 places the center of the shape directly on the right edge of the frame.

Width/Height
The width and height determine the vertical and horizontal size of the ellipse. If the values are
identical, then you have a perfect circle.

Angle
The angle rotates the shape, which on a perfect circle doesn’t change the image all that much, but if
you create an oval or an outline with a short length, you can rotate the shape based on the center axis.

Chapter 55 Shape Nodes 1307

Style Tab

The sEllipse Style tab

Style
The Style tab is used to assign a color to the shape and control its transparency.

Color
The color controls determine the color of the fill and border. To choose a shape color, you can click the
color disclosure arrow, use the color swatch, or drag the eyedropper into the viewer to select a color
from an image. The RGBA sliders or number fields can be used to enter each color channel’s value or
the strength of the alpha channel.

Allow Combining
When this checkbox is enabled, the alpha channel value is maintained even when passing through
other nodes downstream that may cause the shape to overlap with copies of itself. When disabled, the
alpha channel value may increase when the shape overlaps itself.
For instance, if an ellipse’s alpha channel is set to .5, enabling the Allow Combining checkbox
maintains that value even if the shape passes through a Duplicate or Grid node that causes the shape
to overlap. Disabling the checkbox causes the alpha channel values to be compounded at each
overlapping area.

Allow Combining Enabled (left), Allow Combining Disabled (right)

Common Controls
Settings tab
The Settings tab in the Inspector is common to all Shape nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 55 Shape Nodes 1308

sExpand

The sExpand node

The sExpand node is used to dilate or erode shapes. Like almost all Shape nodes, you can only view
the sExpand node’s results through a sRender node.

External Inputs
The following input appears on the node’s tile in the Node Editor.
– Input1: [orange, required] This input accepts the output of another shape node. This shape or
compound shape connected to this input is either eroded or dilated.

Basic Node Setup

The sExpand node takes a single input that is most often from a compound shape. However, it can be
used on single shapes like sStars and sNgons. The output of the sExpand can then be output to
another shape node or to a sRender node for viewing or compositing into the greater node tree.

Star and ellipse shapes combined in an sBoolean node, then output to an sExpand for dilating or eroding

Inspector

The sExpand Controls tab

Controls
The Controls tab includes all of the parameters for the sExpand node.

Amount
A positive value dilates the shape while a negative value erodes it.

Chapter 55 Shape Nodes 1309

Border Style
The border style controls how the expanded or contracted shapes join at the corners. There are four
styles provided as options. Bevel squares off the corners. Round creates rounded corners. Miter and
Miter Clip maintain pointed edges, until a certain threshold. The Threshold is set by the Miter
limit slider.

Miter Limit
The Miter parameter is only displayed when the Miter or Miter Clip border style is selected. The miter
limit determines when the pointed edges become beveled based on the shape’s thickness.

Common Controls
Settings tab
The Settings tab in the Inspector is common to all Shape nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

sGrid

The sGrid node

The sGrid node replicates the shape on an X and Y grid and adds the ability to offset the rows and
columns. Like almost all Shape nodes, you can only view the sGrid node’s results through a
sRender node.

External Inputs
The following input appears on the node’s tile in the Node Editor.
– Input1: [orange, required] This input accepts the output of another Shape node. The shape
connected to this input is replicated on a custom grid.

Basic Node Setup

The sGrid node takes a single input from a single or compound shape from am sMerge or sBoolean
node. The sGrid node places the incoming shape on a grid of rows and columns. The output of the
sGrid can then be output to another Shape node or a sRender node for viewing or compositing into
the greater node tree.

An sEllipse shapes connected to an sGrid and then output to an sRender for viewing and combining with other elements

Chapter 55 Shape Nodes 1310

Inspector

The sGrid Controls tab

Controls
The Controls tab is used to determine the number of grid cells and their offset position.

Grid Cells X and Y

These parameters set the number of cells on the grid, both horizontally and vertically. For instance,
entering 5 in the X and Y number field creates five rows of the shape and five columns.

X and Y Offset
Sets the X and Y distance between the rows and columns. An offset value of 0.0 will have all the rows
and columns on top of each other. Entering X Offset at 1.0 would spread the columns the width to
the frame.

Common Controls
Settings tab
The Settings tab in the Inspector is common to all Shape nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

sJitter

The sJitter node

The sJitter node is most often used to randomly position an array of shapes generated from a sGrid or
sDuplicate node. However, it includes an auto-animating random mode that can be used to distort and
randomly jitter single shapes.
Like almost all Shape nodes, you can only view the sJitter node’s results through a sRender node.

External Inputs
The following input appears on the node’s tile in the Node Editor.
– Input1: [orange, required] This input accepts the output of another Shape node. The shape
connected to this input is offset, distorted, and animated based on the sJitter node settings.

Chapter 55 Shape Nodes 1311

Basic Node Setup
The sJitter node takes an array of shapes from a sGrid or sDuplicate node and randomly changes their
position, size, and rotation. The output of the sExpand can then be output to another Shape node or to
a sRender node for viewing or compositing into the greater node tree.

An array of shapes created by the sGrid node input into an sJitter node to randomly offset or scale the shapes

Inspector

The sJitter Controls tab

Controls
The Controls tab offers range sliders that determine the variation amount for offset, size, and rotation.
The Point Jitter parameters are used to offset the invisible points that create the vector shapes.

Jitter Mode
The Jitter Mode menu allows you to choose between static position and size offsets or enabling an
auto-animation mode. Leaving the default Fixed selection allows you to offset a grid of shapes,
animating with keyframes or modifiers if needed. The Random menu selection auto-animates the
parameters based on the range you define using the range sliders. If all the range sliders are left in the
default position, no random animation is created. Increasing the range on any given parameter will
randomly animate that parameter between the range slider values.

Shape X and Y Offset

These parameters set the horizontal and vertical offset from the shape array’s original position. This is
done randomly, so not all shapes in the array will offset by the same amount.

Chapter 55 Shape Nodes 1312

Shape X and Y Size
These parameters set the horizontal and vertical scaling for each shape in an array. The left range
value decreases the scale, and the right range value increases the scale. This is done randomly, so not
all shapes in the array will scale by the same amount.

Shape Rotate
This parameter rotates each shape in an array.

Point Jitter
The X and Y Point Jitter parameters use the vector control points to distort the shape. This can be
used to give a distressed appearance to ellipses or wobbly animation to other shapes.

Common Controls
Settings tab
The Settings tab in the Inspector is common to all Shape nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

sMerge

The sMerge node

The sMerge node combines shapes similar to a standard Merge node, except the sMerge node can
accept more than two shape inputs.
Like almost all Shape nodes, you can only view the sMerge node’s results through a sRender node.

External Inputs
The node displays only two inputs first, but as each shape node is connected, a new input appears on
the node, assuring there is always one free to add a new shape into the composite.
– Input[#]: These multi-colored inputs are used to connect multiple Shape node. There is no
limit to the number of inputs this node can accept. The node dynamically adds more inputs as
needed, ensuring that there is always at least one input available.

Basic Node Setup

The sMerge node is used to combine two Shape nodes. In terms of layering, each subsequent input is
placed over the previous input. For instance, the first shape connected to the orange input is the
bottom-most shape, the green input is layered over it, and if a third shape is connected to the pink
input, that is the topmost layer.

Chapter 55 Shape Nodes 1313

Three shapes are combined in an sMerge node, then output to an sRender for viewing and further processing.

Inspector

The sMerge Controls tab

Controls
The only control for the sMerge node is the Override Axis checkbox, which overrides the shape’s axis.

Common Controls
Settings tab
The Settings tab in the Inspector is common to all Shape nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

sNGon

The sNGon node

The sNGon node is used to create multi-sided shapes like triangles, pentagons, and octagons. Like
almost all Shape nodes, you can only view the sNGon node’s results through a sRender node.

External Inputs
This node generates shapes and does not have any inputs.

Basic Node Setup

The sNGon node is a shape generator, meaning it generates a shape and therefore has no inputs. The
output of the sNGon can go into a sRender for viewing and further compositing or, more likely, connect
to another Shape node like sGrid or sDuplicate.

Chapter 55 Shape Nodes 1314

An sNGon node connecting to an sDuplicate node, and then viewed using an sRender node.

Inspector

The sNGon Controls tab

Controls
The Controls tab is used to define multi-sided shape characteristics, including fill, border, size,
and position.

Solid
When enabled, the Solid checkbox fills the NGon shape with the color defined in the Style tab. When
disabled, an outline created by the Border Width control is displayed, and the center is made
transparent.

Border Width
This parameter expands or contracts the border around the shape. Although it can be used when the
Solid checkbox is enabled, it is primarily used to determine the outline thickness when the checkbox
is disabled.

Border Style
The Border Style parameter controls how the sides of the NGon join at the corners. There are three
styles provided as options. Bevel squares off the corners. Round creates rounded corners. Miter
maintains pointed corners.

Cap style
When the Solid checkbox is disabled, three cap style options are displayed. The cap styles can create
lines with flat, rounded, or squared ends. Flat caps have flat, squared ends, while rounded caps have
semi-circular ends. Squared caps have projecting ends that extend half the line width beyond the end
of the line.
The caps are not visible unless the length is below 1.0.

Chapter 55 Shape Nodes 1315

Position
The Position parameter is only displayed when the Solid checkbox is disabled. It allows you to position
the starting point of the shape. When used in conjunction with the Length parameter, it positions the
gap in the outline.

Length
The Length parameter is only displayed when the Solid checkbox is disabled. A length of 1.0 is a
closed shape. Setting the length below 1.0 creates an opening or gap in the outline. Keyframing the
Length parameters allows you to create write-on style animations.

X and Y Offset
These parameters are used to position the shape left, right, up, and down in the frame. The shape
starts in the center of the frame, and the parameters are used to offset the position. The offset
coordinates are normalized based on the width of the frame. So, an X offset of 0.0 is centered and a
value of 0.5 places the center of the shape directly on the right edge of the frame.

Width/Height
The Width and Height parameters determine the vertical and horizontal size of the ellipse. If the values
are identical, then all sides are of equal length.

Angle
The Angle parameter rotates the shape based on the center axis.

Style Tab

The sNGon Style tab

Style
The Style tab is used to assign a color to the shape and control its transparency.

Color
The Color controls determine the color of the fill and border. To choose a shape color, you can click
the color disclosure arrow, use the color swatch, or drag the eyedropper into the viewer to select a
color from an image. The RGBA sliders or number fields can be used to enter each color channel’s
value or the strength of the alpha channel.

Chapter 55 Shape Nodes 1316

Allow Combining
When this checkbox is enabled, the alpha channel value is maintained even when passing through
other nodes downstream that may cause the shape to overlap with copies of itself. When disabled, the
alpha channel value may increase when the shape overlaps itself.
For instance, if a NGon alpha channel is set to .5, enabling the Allow Combining checkbox maintains
that value even if the shape passes through a duplicate or grid node that causes the shape to overlap.
Disabling the checkbox causes the alpha channel values to be compounded at each overlapping area.

Allow Combining Enabled (left), Allow Combining Disabled (right)

Settings tab
The Settings tab in the Inspector is common to all Shape nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

sOutline

The sOutline node

The sOutline node is used to create outlines from merged or boolean compound shapes. The
individual shapes retain their own style, color, size, position, and other characteristics. The only
difference is the border thickness, border style, position, and length are applied to all incoming shapes
uniformly in the sOutline node.
Like almost all shape nodes, you can only view the sOutline node’s results through a sRender node.

External Inputs
The following input appears on the node’s tile in the Node Editor:
– Input1: [orange, required] This input accepts the another shape node’s output, but more likely a
compound shape from a sMerge or sBoolean. An outline is created from the compound shape
connected to this input.

Basic Node Setup

The sOutline node takes a single input that is most often from a compound shape, however, it can
sometimes be useful on single shapes to create double outlines. The output of the sOutline can then
be output to another Shape node or to a sRender node for viewing or compositing into the greater
node tree.

Chapter 55 Shape Nodes 1317

A compound shape from an sBoolean node is connected to an sOutline for creating a complex outlined shape

Inspector

The sOutline Controls tab

Controls
The Controls tab is used to define the outline thickness, border and cap style, position, and length that
is applied to the compound shape connected to the input.

Thickness
This parameter controls the width of the outline.

Border Style
The Border Style parameter controls how the outline joins at the corners. There are three styles
provided as options. Bevel squares off the corners. Round creates rounded corners. Miter maintains
pointed corners.

Cap style
Three Cap Style options are used to create lines with flat, rounded, or squared ends. Flat caps have
flat, squared ends, while rounded caps have semi-circular ends. Squared caps have projecting ends
that extend half the line width beyond the end of the line.
The caps are not visible unless the length is below 1.0.

Position
The Position parameter allows you to position the starting point of the shape. When used in
conjunction with the Length parameter, it positions the gap in the outline.

Length
The Length parameter controls the end position of the outline. A length of 1.0 is a closed shape.
Setting the length below 1.0 creates an opening or gap in the outline. Keyframing the Length
parameters allows you to create write-on style animations.

Chapter 55 Shape Nodes 1318

Common Controls
Settings tab
The Settings tab in the Inspector is common to all Shape nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

sRectangle

The sRectangle node

The sRectangle node is used to create rectangular shapes. Like almost all shape nodes, you can only
view the sRectangle node’s results through a sRender node.

External Inputs
This node generates shapes and does not have any inputs.

Basic Node Setup

The sRectangle node is a shape generator, meaning it generates a shape and therefore has no inputs.
The output of the sRectangle can go into a sRender for viewing and further compositing or, more likely,
connect to another Shape node like sGrid or sDuplicate.

An sRectangle node connecting to an sDuplicate node, and then viewed using an sRender node

Inspector

The sRectangle Controls tab

Chapter 55 Shape Nodes 1319

Controls
The Controls tab is used to define the rectangle characteristics, including fill, border, size, and position.

Solid
When enabled, the Solid checkbox fills the rectangle shape with the color defined in the Style tab.
When disabled, an outline created by the Border Width control is displayed, and the center is made
transparent.

Border Width
This parameter expands or contracts the border around the shape. Although it can be used when the
Solid checkbox is enabled, it is primarily used to determine the outline thickness when the checkbox
is disabled.

Border Style
The Border Style parameter controls how the sides of the rectangle join at the corners. There are
three styles provided as options. Bevel squares off the corners. Round creates rounded corners. Miter
maintains pointed corners.

Cap style
When the Solid checkbox is disabled, three Cap Style options are displayed. The cap styles can create
lines with flat, rounded or squared ends. Flat caps have flat, squared ends, while rounded caps have
semi-circular ends. Squared caps have projecting ends that extend half the line width beyond the end
of the line.
The caps are not visible unless the length is below 1.0.

Position
The Position parameter is only displayed when the Solid checkbox is disabled. It allows you to position
the starting point of the shape. When used in conjunction with the Length parameter, it positions the
gap in the outline.

Length
The Length parameter is only displayed when the Solid checkbox is disabled. A length of 1.0 is a
closed shape. Setting the length below 1.0 creates an opening or gap in the outline. Keyframing the
Length parameters allows you to create write-on style animations.

X and Y Offset
These parameters are used to position the shape left, right, up, and down in the frame. The shape
starts in the center of the frame, and the parameters are used to offset the position. The offset
coordinates are normalized based on the width of the frame. So an X offset of 0.0 is centered and a
value of 0.5 places the center of the shape directly on the right edge of the frame.

Width/Height
The Width and Height parameters determine the vertical and horizontal size of the rectangle. If the
values are identical, then you have a square.

Corner Radius
This parameter determines if the corners of the rectangle are sharp or rounded. A value of 0.0
produces sharp corners, while a value of 1.0 will create a circle from a staring square shape or a pill
shape from a rectangle.

Angle
The Angle parameter rotates the shape based on the center axis.

Chapter 55 Shape Nodes 1320

Style Tab

The sRectangle Style tab

Style
The Style tab is used to assign color to the shape and control its transparency.

Color
The Color parameter controls determine the color of the fill and border from the sRectangle node.
To choose a shape color, you can click the color disclosure arrow and use the color swatch, or drag
the eye dropper into the viewer to select a color from an image. The RGBA sliders or number fields
can be used to enter the value of each color channel or the strength of the alpha channel.

Allow Combining
When this checkbox is enabled, the alpha channel value is maintained even when passing through
other nodes downstream that may cause the shape to overlap with copies of itself. When disabled, the
alpha channel value may increase when the shape overlaps itself. For instance, if a rectangle alpha
channel is set to .5, enabling the Allow Combining checkbox maintains that value even if the shape
passes through a duplicate or grid node that causes the shape and alpha channel to overlap. Disabling
the checkbox causes the alpha channel values to be compounded at each overlapping area.

Allow Combining Enabled (left), Allow Combining Disabled (right)

Common Controls
Settings tab
The Settings tab in the Inspector is common to all Shape nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 55 Shape Nodes 1321

sRender

The sRender node

The sRender node converts the vector shapes to an image. The output of the sRender allows the
vector shapes to be integrated with other elements in a composite.

Inputs
There is one input on the Background node for an Effect Mask input.
– Input1: [orange, required] This input accepts the output of your final shape node. A rendered
bitmap image is created from the sRender node for composting into the rest of your comp.
– Effect Mask: The optional blue effect mask input accepts a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the displayed area to only those pixels within the mask.

Basic Node Setup

The sRender node is always placed at the end of a string of Shape nodes. The output connects to
other 2D nodes like a Soft Glow node.

Multiple Shape nodes connected to the sRender node and then processed and composited with a title

Chapter 55 Shape Nodes 1322

Inspector

The sRender Image tab

Image Tab
The controls in this tab are used to set the resolution, color depth, and pixel aspect of the image
produced by the sRender node.

Process Mode
Use this menu control to select the Fields Processing mode used by Fusion to render the resulting
image. The default Full Frames option is appropriate for progressive formats.

Width/Height
This pair of controls are used to set the Width and Height dimensions of the image to be created by
the sRender node.

Pixel Aspect
This control is used to specify the Pixel Aspect ratio of the created images. An aspect ratio of 1:1 would
generate a square pixel with the same dimensions on either side (like a computer display monitor), and
an aspect of 0.9:1 would create a slightly rectangular pixel (like an NTSC monitor).

NOTE: Right-click on the Width, Height, or Pixel Aspect controls to display a menu listing the
file formats defined in the preferences Frame Format tab. Selecting any of the listed options
will set the width, height, and pixel aspect to the values for that format, accordingly.

Auto Resolution
When this checkbox is selected, the width, height, and pixel aspect of the image created by the node
will be locked to values defined in the composition’s Frame Format preferences. If the Frame Format
preferences change, the resolution of the image produced by the node will change to match. Disabling
this option can be useful to build a composition at a different resolution than the eventual target
resolution for the final render.

Depth
The Depth button array is used to set the pixel color depth of the image created by the Creator node.
32-bit pixels require 4X the memory of 8-bit pixels but have far greater color accuracy. Float pixels
allow high dynamic range values outside the normal 0..1 range, for representing colors that are brighter
than white or darker than black.

Chapter 55 Shape Nodes 1323

Source Color Space
You can use the Source Color Space menu to set the Color Space of the footage to help achieve a
linear workflow. Unlike the Gamut tool, this doesn‘t perform any actual color space conversion, but
rather adds the source space data into the metadata, if that metadata doesn‘t exist. The metadata can
then be used downstream by a Gamut tool with the From Image option, or in a Saver, if explicit output
spaces are defined there. There are two options to choose from:
– Auto: Automatically reads and passes on the metadata that may be in the image.
– Space: Displays a Color Space Type menu where you can choose the correct color
space of the image.

Source Gamma Space

Using the Curve Type menu, you can set the Gamma Space of the footage and choose to remove it by
way of the Remove Curve check box when working in a linear workflow. There are three choices in the
Curve Type menu
– Auto: Automatically reads and passes on the metadata that may be in the image.
– Space: Displays a Gamma Space Type menu where you can choose the correct gamma curve of
the image.
– Log: Brings up the Log/Lin settings, similar to the Cineon tool. For more information, see Chapter
38, “Film Nodes” in the Fusion Reference Manual or Chapter 99 in the the DaVinci Resolve
Reference Manual.

Remove Curve
Depending on the selected Gamma Space or on the Gamma Space found in Auto mode, the Gamma
Curve is removed from, or a log-lin conversion is performed on, the material, effectively converting it
to a linear output space.

Common Controls
Settings tab
The Settings tab in the Inspector is common to all Shape nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

sStar

The sStar node

The sStar node is used to create multi-point star shapes. Like almost all Shape nodes, you can only
view the sStar node’s results through a sRender node.

External Inputs
This node generates shapes and does not have any inputs.

Chapter 55 Shape Nodes 1324

Basic Node Setup
The sStar node is a shape generator, meaning it generates a shape and therefore has no inputs. The
output of the sStar can go into a sRender for viewing and further compositing or, more likely, connect
to another shape node like sGrid or sDuplicate.

An sStar node connecting to an sDuplicate node, and then viewed using an sRender node

Inspector

The sStar Controls tab

Controls
The Controls tab is used to define the star shape’s characteristics, including number of points, depth,
fill, border, size, and position.

Points
This slider determines the number of points or arms on the star.

Depth
The depth slider controls the inner radius or width of the arms. A depth of 0.001 makes hair-thin arms,
while a depth of 1.0 makes a faceted circle.

Solid
When enabled, the Solid checkbox fills the star shape with the color defined in the Style tab. When
disabled, an outline created by the Border Width control is displayed, and the center is made
transparent.

Border Width
This parameter expands or contracts the border around the shape. Although it can be used when the
Solid checkbox is enabled, it is primarily used to determine the outline thickness when the checkbox
is disabled.

Chapter 55 Shape Nodes 1325

Border Style
The Border Style parameter controls how the sides of the star join at the corners. There are three
styles provided as options. Bevel squares off the corners. Round creates rounded corners. Miter
maintains pointed corners.

Cap style
When the Solid checkbox is disabled, three cap style options are displayed. The cap styles can create
lines with flat, rounded or squared ends. Flat caps have flat, squared ends while rounded caps have
semi-circular ends. Squared caps have projecting ends that extend half the line width beyond the end
of the line.
The caps are not visible unless the length is below 1.0.

Position
The Position parameter is only displayed when the Solid checkbox is disabled. It allows you to position
the starting point of the shape. When used in conjunction with the length parameter, it positions the
gap in the outline.

Length
The Length parameter is only displayed when the Solid checkbox is disabled. A length of 1.0 is a
closed shape. Setting the length below 1.0 creates an opening or gap in the outline. Keyframing the
Length parameters allows you to create write-on style animations.

X and Y Offset
These parameters are used to position the shape left, right, up, and down in the frame. The shape
starts in the center of the frame, and the parameters are used to offset the position. The offset
coordinates are normalized based on the width of the frame. So an X offset of 0.0 is centered and a
value of 0.5 places the center of the shape directly on the right edge of the frame.

Width/Height
The Width and Height parameters determine the vertical and horizontal size of the star. If the values
are identical, then all arms of the star are of equal length.

Angle
The Angle parameter rotates the shape based on the center axis.

Style Tab

The sStar Style tab

Chapter 55 Shape Nodes 1326

Style
The Style tab is used to assign color to the shape and control its transparency.

Color
The Color controls determine the color of the fill and border from the sStar node. To choose a shape
color, you can click the color disclosure arrow and use the color swatch, or drag the eye dropper into
the viewer to select a color from an image. The RGBA sliders or number fields can be used to enter
the value of each color channel or the strength of the alpha channel.

Allow Combining
When this checkbox is enabled, the alpha channel value is maintained even when passing through
other nodes downstream that may cause the shape to overlap with copies of itself. When disabled, the
alpha channel value may increase when the shape overlaps itself. For instance, if a star alpha channel
is set to .5, enabling the Allow Combining checkbox maintains that value even if the shape passes
through a duplicate or grid node that causes the shape and alpha channel to overlap. Disabling the
checkbox causes the alpha channel values to be compounded at each overlapping area.

Allow Combining Enabled (left), Allow Combining Disabled (right)

Common Controls
Settings tab
The Settings tab in the Inspector is common to all Shape nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

sTransform

The sTransform node

The sTransform node is used to add an additional set of transform controls to the existing controls that
are contained in Shape nodes. These additional transforms can be used to create hierarchical
animations. For instance, you can use a sStar’s built-in Angle control to spin the star around. The star
can then be output to an sTransform node. The rotation control in the sTransform can be used to orbit
the star around the frame.
Like almost all Shape nodes, you can only view the sStar node’s results through a sRender node.

Chapter 55 Shape Nodes 1327

External Inputs
The following input appears on the node’s tile in the Node Editor:
– Input1: [orange, required] This input accepts the output of another Shape node. The shape
connected to this input is moved, scaled, and rotated based on the sTransform settings.

Basic Node Setup

The sTransform node takes the input from another Shape node and adds another set of transforms or
hierarchical animation. The output of the sTransform can go into a sRender for viewing and further
compositing.

An sStar node connecting to an sTransform node, and then viewed using an sRender node

Inspector

The sTransform Controls tab

Controls
The Controls tab is used to define the add a set of transform controls to the incoming shape.

X and Y Offset
These parameters are used to position the shape left, right, up, and down in the frame. The shape
starts in the center of the frame, and the parameters are used to offset the position. The offset
coordinates are normalized based on the width of the frame. So an X offset of 0.0 is centered and a
value of 0.5 places the center of the shape directly on the right edge of the frame.

X and Y Size
The X and Y Size determine the vertical and horizontal scaling of the incoming shape. If the values are
different then the shape will be skewed from its original design.

Rotation
The dial rotates the shape based on the pivot controls.

Chapter 55 Shape Nodes 1328

X and Y Pivot
These parameters position the axis of rotation for the incoming shape. The pivot point is visible in the
viewer as a red X. The X can be dragged in the viewer for positioning.

Transform Axis
Check this box to apply the transform to the shape’s axis.

Common Controls
Settings tab
The Settings tab in the Inspector is common to all Shape nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Common Controls
Nodes that handle Shape operations share a number of identical controls in the Inspector. This section
describes controls that are common amongst Shape nodes.

Settings Tab
The Settings tab in the Inspector can be found on every Shape node. Most of the controls listed here
are only found in the sRender node but a few are common to all Shape nodes.

Blend (sRender only)

The Blend control is used to blend between the tool’s original image input and the tool’s final modified
output image. When the blend value is 0.0, the outgoing image is identical to the incoming. Normally,
this causes the tool to skip processing entirely, copying the input straight to the output.

Process When Blend Is 0.0 (sRender only)

The tool is processed even when the input value is zero. This can be useful if processing of this node
is scripted to trigger another task but the value of the node is set to 0.0.

Red/Green/Blue/Alpha Channel Selector (sRender only)

These four buttons are used to limit the effect of the tool to specified color channels. This filter is often
applied after the tool has been processed.
For example, if the red button on a Blur tool is deselected, the blur is first applied to the image, then
the red channel from the original input is copied back over the red channel of the result.
There are some exceptions, such as tools where deselecting these channels causes the tool to skip
processing that channel entirely. Tools that do this generally possess a set of identical RGBA buttons
on the Controls tab in the tool. In this case, the buttons in the Settings and the Control tabs are
identical.

Apply Mask Inverted (sRender only)

Enabling the Apply Mask Inverted option inverts the complete mask channel for the tool. The mask
channel is the combined result of all masks connected to or generated in a node.

Multiply By Mask (sRender only)

Selecting this option causes the RGB values of the masked image to be multiplied by the mask
channel’s values. This causes all pixels of the image not included in the mask (i.e., set to 0) to become
black/transparent.

Chapter 55 Shape Nodes 1329

Motion Blur (sRender only)
– Motion Blur: This toggles the rendering of motion blur on the tool. When this control is toggled
on, the tool’s predicted motion is used to produce the motion blur caused by the virtual camera’s
shutter. When the control is toggled off, no motion blur is created.
– Quality: Quality determines the number of samples used to create the blur. A quality setting of
2 causes Fusion to create two samples to either side of an object’s actual motion. Larger values
produce smoother results but increase the render time.
– Shutter Angle: Shutter Angle controls the angle of the virtual shutter used to produce the motion
blur effect. Larger angles create more blur but increase the render times. A value of 360 is the
equivalent of having the shutter open for one whole frame exposure. Higher values are possible
and can be used to create interesting effects.
– Center Bias: Center Bias modifies the position of the center of the motion blur. This allows for the
creation of motion trail effects.
– Sample Spread: Adjusting this control modifies the weighting given to each sample. This affects
the brightness of the samples.

Use GPU (sRender only)

The user GPU menu has three settings. Setting the menu to Disable turns off hard accelerated
rendering using the graphics card in your computer. Auto uses a capable GPU if one is available and
falls back to software rendering when a capable GPU is not available.

Hide Incoming Connections

Enabling this checkbox can hide connection lines from incoming nodes, making a node tree appear
cleaner and easier to read. When enabled, empty fields for each input on a node will be displayed in
the Inspector. Dragging a connected node from the node tree into the empty field will hide that
incoming connection line as long as the node is not selected in the node tree.. When the node is
selected in the node tree the line will reappear.

Comments
The Comments field is used to add notes to a tool. Click in the empty field and type the text. When a
note is added to a tool, a small red square appears in the lower left corner of the node when the full
tile is displayed or a small text bubble icon appears on the right when nodes are collapsed. To see the
note in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more information on
scripting nodes, see the Fusion scripting documentation.

Chapter 55 Shape Nodes 1330

Chapter 56

Stereo Nodes
This chapter details the Stereo nodes available in Fusion. Stereoscopic nodes are
available only in Fusion Studio and DaVinci Resolve Studio.

The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Anaglyph [Ana]  1332
Combiner [Com]  1335
Disparity [Dis]  1337
Disparity To Z [D2Z]  1340
Global Align [GA]  1344
New Eye [NE]  1346
Splitter [Spl]  1349
Stereo Align [SA]  1350
Z To Disparity [Z2D]  1356
The Common Controls  1359

Chapter 56 Stereo Nodes 1331

Anaglyph [Ana]

The Anaglyph node

NOTE: The Anaglyph node is available only in Fusion Studio and DaVinci Resolve Studio.

Anaglyph Node Introduction

The Anaglyph node is used to create stereoscopic images by combining separate left eye and right
eye images. It is most commonly used at the end of a stereoscopic workflow to display or deliver the
final result.

Inputs
The three inputs on the Anaglyph node are the left eye input, right eye input, and effect mask.
– Left Eye Input: The orange input is used to connect the 2D image representing the left eye in
the stereo comp.
– Right Eye Input: The green input is used to connect the 2D image representing the right eye in
the stereo comp.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive
shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the
stereoscopic creation to only those pixels within the mask.

Basic Node Setup

The Anaglyph node is usually placed at the end of a stereoscopic node tree to display the final result.
When using separate images for the left and right eye, the left eye image is connected to the orange
input, and the right eye image is connected to the green input of the node. When using either
horizontally or vertically stacked images containing both left eye and right eye information, these only
connect to the orange input.

An Anaglyph node using separate left and right eye inputs

Chapter 56 Stereo Nodes 1332

Inspector

The Anaglyph Controls tab

Controls Tab
Using the parameters in the Controls tab, the separate images are combined to create a
stereoscopic output.

Color Type Menu

The Color Type menu allows you to choose between different color encodings to fit your preferred
display device. To match your stereo glasses, you can choose between Red/Cyan, Red/Green, Red/
Blue, Amber/Blue, and Green/Magenta encoding; Red/Cyan is the most commonly used.

The Anaglyph Color Type menu

Method
In addition to the color used for encoding the image, you can also choose five different methods from
the Method menu: Monochrome, Half-color, Color, Optimized, and Dubois. These methods are
described below.

The Anaglyph Method menu

– Monochrome: Assuming you are using a Red/Cyan Color Type, the left eye contains the
luminance of the left image and is placed in the output of the red channel. The right eye contains
the luminance of the right image and is placed in the output green and blue channels.

Chapter 56 Stereo Nodes 1333

– Half-Color: Assuming you are using a Red/Cyan Color Type, the left eye contains the luminance
of the left image and is placed in the output of the red channel. The right eye contains the color
channels from the right image that match the glasses’ color for that eye.

Monochrome Half-Color

– Color: The left eye contains the color channels from the left image that match the glasses’ color
for that eye. The right eye contains the color channels from the right image that match the glasses’
color for that eye.
– Optimized: Used with red/cyan glasses, for example, the resulting brightness of what shows
through the left eye is substantially less than the brightness of the right eye. Using typical ITU-R
601 ratios for luminance as a guide, the red eye would give 0.299 brightness, while the cyan
eye would give 0.587+0.114=0.701 brightness—over twice as bright. The difference in brightness
between the eyes can produce what are referred to as retinal rivalry or binocular rivalry, which
can destroy the stereo effect. The Optimized method generates the right eye in the same fashion
as the Color method. The left eye also uses the green and blue channels but in combination with
increased brightness that reduces retinal rivalry. Since it uses the same two channels from each
of the source images, it doesn’t reproduce the remaining one. For example, 1.05× the green and
0.45× the blue channels of the left image is placed in the red output channel, and the green and
blue channels of the right image are placed in the output green and blue channels. Red from both
the left and right images is not used.

Color Optimized

– Dubois: Images with fairly saturated colors can produce retinal rivalry with the Half-color, Color,
and Optimized methods because the color is visible in only one eye. For example, with red/
cyan glasses, a saturated green object looks black in the red eye, and green in the cyan eye.
The Dubois method uses the spectral characteristics of (specifically) red/cyan glasses and CRT
(Trinitron) phosphors to produce a better anaglyph and in the end, tends to reduce retinal rivalry
caused by such color differences in each eye. It also tends to reduce ghosting produced when
one eye ‘leaks’ into the other eye. The particular calculated matrix in Fusion is designed for red/

Chapter 56 Stereo Nodes 1334

 cyan glasses and isn’t available for other glasses types. Since it is also derived from CRT color
primaries, it may not give the best results with a common LCD (though it’ll still likely produce less
retinal rivalry and ghosting than the other methods).

Dubois

Swap Eyes
Allows you to swap the left and right eye inputs easily.

Horiz Stack
Takes an image that contains both left and right eye information stacked horizontally. These images
are often referred to as “crosseyed” or “straight stereo” images. You only need to connect that one
image to the orange input of the node. It then creates an image half the width of the original input,
using the left half of the original image for the left eye and the right half of the original image for the
right eye. Color encoding takes place using the specified color type and method.

Vert Stack
Takes an image that contains both left and right eye information stacked vertically. You only need to
connect that one image to the orange input of the node. It then creates an image half the height of the
original input, using the bottom half of the original image for the left eye and the top half of the original
image for the right eye. Color encoding takes place using the specified color type and method.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Stereo nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Combiner [Com]

The Combiner node

NOTE: The Combiner node is available only in Fusion Studio and DaVinci Resolve Studio.

Chapter 56 Stereo Nodes 1335

Combiner Node Introduction
The Combiner node takes two stereoscopic inputs and creates so-called “stacked images” with the
left and right eye, either side by side or on top of each other. Stereoscopic nodes are available only in
Fusion Studio and DaVinci Resolve Studio.

Inputs
The two inputs on the Combiner node are used to connect the two images that get combined in a
stacked stereo image.
– Image 1 Input: The orange input is used to connect the 2D image representing the left eye in
the stereo comp.
– Image 2 Input: The green input is used to connect the 2D image representing the right eye in
the stereo comp.

Basic Node Setup

Below, a left eye image and right eye image are connected to the Combiner node to create a single
stacked stereo image. It can be more efficient to render out the stacked stereo images as EXR files
than to generate the disparity on-the-fly.

Left and right eye images are connected into a Combiner node to generate a stacked stereo image.

Inspector

The Combiner Controls tab

Controls Tab
To stack the images, the left eye image is connected to the orange input, and the right eye image is
connected to the green input of the node.

Combine
The Combine menu provides three options for how the two images are made into a stacked
stereo image.
– None: No operation will take place. The output image is identical to the left eye input.
– Horiz: Both images will be stacked horizontally, or side-by-side, with the image connected to the
left eye input on the left. This will result in an output image double the width of the input image.
– Vert: Both images will be stacked vertically, or on top of each other, with the image connected
to the left eye input on the bottom. This will result in an output image double the height of the
input image.

Chapter 56 Stereo Nodes 1336

Swap Eyes
Allows you to easily swap the left and right eye input.

Add Metadata
Metadata is carried along with the images and can be added to the existing metadata using this
checkbox. To view Metadata, use the viewer’s SubView menu set to Metadata.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Stereo nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Disparity [Dis]

The Disparity node

NOTE: The Disparity node is available only in Fusion Studio and DaVinci Resolve Studio.

Disparity Node Introduction

Disparity generates the left/right shift between the frames in a stereo pair. It also generates the
vertical disparity between the left/right images, which is usually a lot smaller than the horizontal
disparity and ideally should be 0 to minimize viewing discomfort. When viewing the output of the
Disparity node in the views, the human eye can distinguish quality/detail in the Disparity map better by
looking at either the grayscale X disparity or Y disparity rather than looking at the combined XY
disparity as a Red/Green color image.
The generated disparity is stored in the output image’s Disparity aux channel, where the left image
contains the left > right disparity, and the right image contains the right > left disparity. Because
disparity works based on matching regions in the left eye to regions in the right eye by comparing
colors and gradients of colors, colors in the two eyes must be as similar as possible. Thus, it is a good
idea to color correct ahead of time. It is also a good idea to crop away any black borders around the
frames, as this confuses the disparity tracking (and also causes problems if you are using the Color
Corrector’s histogram match ability to do the color matching).
In Stack mode, left and right outputs deliver the same image. If the left and right images have a global
vertical offset larger than a few pixels, it can help the disparity tracking algorithm if you vertically align
features in the left/right eyes ahead of time using a Transform node. Small details tend to get lost in the
tracking process when you have a large vertical offset between left/right eyes.
Consider using a SmoothMotion node to smooth your disparity channel. This can help reduce time-
dependent flickering when warping an eye. Also, think about whether you want to remove lens
distortion before computing disparity. If you do not, your Disparity map becomes a combined Disparity
and Lens Distortion map. This can have advantages and disadvantages.
One disadvantage is that if you then do a vertical alignment, you are also removing lens distortion
effects. When trying to reduce the computation time, start first with adjusting the Proxy and Number of
Iterations sliders.

Chapter 56 Stereo Nodes 1337

The Disparity node does not support RoI or DoD.

Inputs
The two inputs on the Disparity node are used to connect the left and right images.
– Left Input: The orange input is used to connect either the left eye image or the stacked image.
– Right Input: The green input is used to connect the right eye image. This input is available only
when the Stack Mode menu is set to Separate.

Outputs
Unlike most nodes in Fusion, Disparity has two outputs for the left and right eye.
Left Output: This holds the left eye image with a new disparity channel, or a Stacked Mode image with
a new disparity channel.
Right Output: This holds the right eye image with a new disparity channel. This output is visible only if
Stack Mode is set to Separate.

Basic Node Setup

Below, a left eye image and right eye image are connected to the Disparity node. The Disparity node
then outputs each eye to Saver nodes. It can be more efficient to render out the stereo images as EXR
files than to generate the disparity on-the-fly.

Left and right eye images are connected into a Disparity node to generate and render out a stereo image

Inspector

The Disparity Controls tab

Chapter 56 Stereo Nodes 1338

Proxy (for Tracking)
The input images are resized down by the proxy scale, tracked to produce the disparity, and then the
resulting disparities are scaled back up. This option is purely to speed up the calculation of the
disparity, which can be slow. The computational time is roughly proportional to the number of pixels in
the image. This means a proxy scale of 2 gives a 4x speedup, and a proxy scale of 3 gives a 9x
speedup. In general, 1:1 proxy will give the most detailed flow, but keep in mind that this is highly
dependent on the amount of noise and film grain. If noise is present in large quantities, it can
completely obliterate any gains moving from 2:1 to 1:1 proxy. In some situations, it can even make things
worse. You can think of the Proxy setting as acting as a simplistic low-pass filter for removing
noise/grain.

Advanced
The Advanced settings section has parameter controls to tune the Disparity map calculations. The
default settings have been chosen to be the best default values from experimentation with many
different shots and should serve as a good standard. In most cases, tweaking of the Advanced
settings is not needed.

Smoothness
This controls the smoothness of the disparity. Higher values help deal with noise, while lower values
bring out more detail.

Edges
This slider is another control for smoothness but applies it based on the color channel. It tends to have
the effect of determining how edges in the disparity follow edges in the color images. When it is set to
a lower value, the disparity becomes smoother and tends to overshoot edges. When it is set to a
higher value, edges in the disparity align more tightly with the edges in the color images, and details
from the color channels start to slip into the disparity, which is not usually desirable.
As a rough guideline, if you are using the disparity to produce a Z channel for post effects like depth of
field, experiment with higher values, but if you are using the disparity to do interpolation, you might
want to keep the values lower.
In general, if the Edges slider is set is too high, there can be problems with streaked out edges when
the disparity is used for interpolation.

Match Weight
This controls how matching is done between neighboring pixels in the left image and neighboring
pixels in the right image. When a lower value is used, large structural color features are matched.
When higher values are used, small sharp variations in the color are matched. Typically, a good value
for this slider is in the [0.7, 0.9] range. Setting this option higher tends to improve the matching results
in the presence of differences due to smoothly varying shadows or local lighting variations between
the left and right images. You should still color match the initial images so they are as similar as
possible; this option tends to help with local variations (e.g., lighting differences due to light passing
through a mirror rig).

Mismatch Penalty
This controls how the penalty for mismatched regions grows as they become more dissimilar. The
slider provides a choice between a balance of Quadratic (lower values) and Linear (higher values)
penalties. Lower value Quadratic settings strongly penalize large dissimilarities, while higher value
Linear settings are more robust to dissimilar matches. Moving this slider toward lower tends to give a
disparity with more small random variations in it, while higher values produce smoother, more visually
pleasing results.

Warp Count
Turning down the Warp Count makes the disparity computations faster. In particular, the computational
time depends linearly upon this option. To understand what this option does, you need to understand
that the Disparity algorithm progressively warps the left image until it matches with the right image.

Chapter 56 Stereo Nodes 1339

After some point, convergence is reached, and additional warps are just a waste of computational
time. The default value in Fusion is set high enough that convergence should always be reached. You
can tweak this value to speed up the computations, but it is good to watch how the disparity is
degrading in quality at the same time.

Iteration Count
Turning down the Iteration Count makes the disparity computations faster. In particular, the
computational time depends linearly upon this option. Just like adjusting Warp Count, at some point
adjusting this option higher will yield diminishing returns and will not produce significantly better
results. By default, this value is set to something that should converge for all possible shots and can be
tweaked lower fairly often without reducing the disparity’s quality.

Filtering
This menu determines the filtering operations used during flow generation. Catmull-Rom filtering will
produce better results, but at the same time, it increases the computation time steeply.

Stack Mode
This menu determines how the input images are stacked.
When set to Separate, the Right Input and Output will appear, and separate left and right images must
be connected.

Swap Eyes
Enabling this checkbox causes the left and right images to swap.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Stereo nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Disparity To Z [D2Z]

The DisparityToZ node

NOTE: The Disparity to Z node is available only in Fusion Studio and DaVinci Resolve Studio.

Disparity to Z Node Introduction

Disparity To Z takes a 3D camera and an image containing a disparity channel as inputs, and outputs
the same image but with a newly computed Z channel.
Optionally, this node can output Z into the RGB channels. Ideally, either a stereo Camera 3D or a
tracked stereo camera is connected into Disparity To Z. However, if no camera is connected, the node
provides artistic controls for determining a Z channel. The depth created by this node can be used for
post effects like fogging or depth of field (DoF).

Chapter 56 Stereo Nodes 1340

The Z values produced become more incorrect the larger (negative) they get. The reason is that
disparity approaches a constant value as Z approaches -infinity. So Z = -1000 and Z = -10000 and Z =
-100000 may map to D=142. 4563 and D=142. 4712 and D=142. 4713. As you can see, there is only
0.0001 in D to distinguish between 10,000 and 100,000 in Z. The maps produced by disparity are not
accurate enough to make distinctions like this.

Inputs
The three inputs on the Disparity To Z node are used to connect the left and right images and a
camera node.
– Left Input: The orange input is used to connect either the left eye image or the stack image.
– Right Input: The green input is used to connect the right eye image. This input is available only
when the Stack Mode menu is set to Separate.
– Stereo Camera: The magenta input is used to connect a stereo camera node.

Outputs
Unlike most nodes in Fusion, Disparity To Z has two outputs for the left and right eye.
– Left Output: This holds the left eye image with a new Z channel, or a Stacked Mode image with
a new disparity channel.
– Right Output: This holds the right eye image with a new Z channel. This output is visible only if
Stack Mode is set to Separate.

Basic Node Setup

Disparity To Z takes a 3D camera and stereo images containing a disparity channel as inputs. The
output is an image with a newly computed Z channel.

A Disparity to Z node creates an image with a Z channel.

Inspector

The Disparity To Z Controls tab

Chapter 56 Stereo Nodes 1341

Controls Tab
In addition to outputting Z values in the Z channel, this tab promotes the color channels to float32 and
outputs the Z values into the color channels as {z, z, z, 1}. This option is useful to get a quick look at the
Z channel.

NOTE: Z values are negative, becoming more negative the further you are from the camera.
The viewers only show 0.0 to 1.0 color, so to visualize other data it has to be converted via a
normalization method to fit in a display 0-1 range. To do this, right-click in the viewer and
choose Options > Show Full Color Range.

Output Z to RGB
Rather than keeping the Z values within the associated aux channel only, they will be copied into the
RGB channels for further modification with any of Fusion’s nodes.

Refine Z
The Enable checkbox refines the depth map based upon the RGB channels. The refinement causes
edges in the flow to align more closely with edges in the color channels. The downside is that
unwanted details in the color channels start to show up in the flow. You may want to experiment with
using this option to soften out harsh edges for Z-channel post effects like depth of field or fogging.

HiQ Only
Activating this checkbox causes the Refine Z option to process only when rendering is set to High
Quality. You can ensure High Quality is enabled by right-clicking to the left or right of the transport
controls in the main toolbar.

Strength
Increasing this slider does two things. It smooths out the depth in constant color regions and moves
edges in the Z channel to correlate with edges in the RGB channels.
Increasing the refinement has the undesirable effect of causing texture in the color channel to show up
in the Z channel. You will want to find a balance between the two.

Radius
This is the radius of the smoothing algorithm.

Stack Mode
This menu determines how the input images are stacked.
When set to Separate, the Right Input and Output will appear, and separate left and right images must
be connected.

Swap Eyes
Enabling this checkbox causes left and right images to be swapped.

The Disparity To Z Camera tab

Chapter 56 Stereo Nodes 1342

Camera Tab
If you need correct real-world Z values because you are trying to match some effect to an existing
scene, you should use the External Camera options to get precise Z values back. If any Z-buffer will
suffice and you are not that particular about the exact details of how it is offset and scaled, or if there is
no camera available, the Artistic option might be helpful.
– External Mode: An input is available on the node to connect an existing stereo Camera 3D. This
can either be a single stereo Camera 3D (i.e., its eye separation is set to non-zero), or a pair of
(tracked) Camera 3Ds connected via the Camera 3D > Stereo > Right Camera input.
– Artistic Mode: If you do not have a camera, you can adjust these controls to produce an “artistic”
Z channel whose values will not be physically correct but will still be useful. To reconstruct the
Disparity > Z Curve, pick (D, Z) values for a point in the foreground and a point in the background.

TIP: If artistic mode is a little too “artistic” for you and you want more physically-based
parameters to adjust (e.g., convergence and eye separation), create a dummy Camera 3D,
connect it into the Disparity To Z > Camera input, and then fiddle with the Camera 3D’s controls.

Foreground Disparity (Pick from Left Eye)

When the camera Mode is set to Artistic, a Foreground Disparity slider is available. This is the disparity
for the closest foreground object. It will get mapped to the depth value specified by the Foreground
Depth control. Any objects with disparity outside of the range [ForegroundDisparity,
BackgroundDisparity] will have their disparity values clipped to this range leading to flat areas in the Z
channel, so make sure that you pick values that enclose the actual disparity range.

Background Disparity (Pick from Left Eye)

When the camera Mode is set to Artistic, a Background Disparity is available. This is the disparity for
the furthest background object. It will get mapped to the depth value specified by the Background
Depth control. One way to think of this input is as the upper limit to disparity values for objects at
-infinity. This value should be for the left eye. The corresponding value in the right eye will be the same
in magnitude but negative.

Foreground Depth
This is the depth to which Foreground Disparity will be mapped. Think of this as the depth of the
nearest object. Note that values here are positive depth.

Background Depth
This is the depth to which Background Disparity will be mapped. Think of this as the depth of the most
distant object.

Falloff
Falloff controls the shape of the depth curve between the requested foreground and background
depths. When set to Hyperbolic, the disparity-depth curve behaves roughly like depth = constant/
disparity. When set to Linear, the curve behaves like depth = constant * disparity. Hyperbolic tends to
emphasize Z features in the foreground, while linear gives foreground/background features in the Z
channel equal weighting.
Unless there’s a specific reason, choose Hyperbolic, as it is more physically accurate, while Linear
does not correspond to nature and is purely for artistic effect.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Stereo nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 56 Stereo Nodes 1343

Global Align [GA]

The GlobalAlign node

NOTE: The Global Align node is available only in Fusion Studio and DaVinci Resolve Studio.

Global Align Node Introduction

As opposed to Stereo Align, this node does not utilize optical flow at all. It’s meant as a fast and
convenient way to do simple stereo alignment for both X and Y as well as rotation.
Global Align comes in handy at the beginning of the node chain to visually correct major differences
between the left and right eye before calculating Disparity.
Manual correction of large discrepancies between left and right, as well as applying an initial color
matching, helps Disparity generate more accurate results.

Inputs
The two inputs on the Global Align node are used to connect the left and right images.
– Left Input: The orange input is used to connect either the left eye image or the stack image.
– Right Input: The green input is used to connect the right eye image. This input is available only
when the Stack Mode menu is set to Separate.

Outputs
Unlike most nodes in Fusion, Global Align has two outputs for the left and right eye.
– Left Output: This outputs the newly aligned left eye image.
– Right Output: This outputs the newly aligned right eye image.

Basic Node Setup

Global Align is typically placed at the beginning of the node tree. Below it is inserted between the left
and right eye images and the Disparity node to visually correct major differences.

A Global Align node used to manually correct left and right eye discrepancies

Chapter 56 Stereo Nodes 1344

Inspector

The Global Align Controls tab

Controls Tab
The Controls tab includes translation and rotation controls to align the stereo images manually.

Translation X and Y
– Balance: Determines how the global offset is applied to the stereo footage.
– None: No translation is applied.
– Left Only: The left eye is shifted, while the right eye remains unaltered.
– Right Only: The right eye is shifted, while the left eye remains unaltered.
– Split Both: Left and right eyes are shifted in opposite directions.

Snap to Nearest Pixel

While adjusting the X or Y shift dial, this option ensures that the image is shifted in full pixel amounts
only to maintain optimum quality. This avoids sub-pixel rendering of the image, which could result in
subtle blurring.

Rotation
– Balance: Determines how the global rotation is applied to the stereo footage.
– None: No rotation is applied.
– Left Only: The left eye is rotated, while the right eye remains unaltered.
– Right Only: The right eye is rotated, while the left eye remains unaltered.
– Split Both: Left and right eyes are rotated in opposite directions.

Angle
This dial adjusts the angle of the rotation. Keep in mind that the result depends on the Balance
settings. If only rotating one eye by, for example, 10 degrees, a full 10-degree rotation will be applied
to that eye.
When applying rotation in Split mode, one eye will receive a -5 degree and the other eye a
+5 degree rotation.

Chapter 56 Stereo Nodes 1345

Translation Filter Method
This menu chooses the filter method that delivers the best results depending on the content of
your footage.

Visualization
This control allows for different color encodings of the left and right eye to conveniently examine the
results of the above controls without needing to add an extra Anaglyph or Combiner node.
Set this to None for final output.

Stack Mode
Determines how the input images are stacked.
When set to Separate, the right input and output will appear, and separate left and right images must
be connected.

Swap Eyes
With Stacked Mode, image stereo pairs’ left and right images can be swapped.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Stereo nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

New Eye [NE]

The New Eye node

NOTE: The New Eye node is available only in Fusion Studio and DaVinci Resolve Studio.

New Eye Node Introduction

The New Eye node constructs a new image by interpolating between two existing stereo images
using the embedded disparity channels. This node can also be used to replace one view with a
warped version of the other. In Stack Mode, L and R outputs will output the same image.
You can map the left eye onto the right eye and replace it. This can be helpful when removing errors
from certain areas of the frame.
New Eye does not interpolate the aux channels but instead destroys them. In particular, the disparity
channels are consumed/destroyed. Add another Disparity node after the New Eye if you want to
generate Disparity for the realigned footage.

Chapter 56 Stereo Nodes 1346

Inputs
The two inputs on the New Eye node are used to connect the left and right images.
– Left Input: The orange input is used to connect either the left eye image or the stack image.
– Right Input: The green input is used to connect the right eye image. This input is available only
when the Stack Mode menu is set to Separate.

Outputs
Unlike most nodes in Fusion, New Eye has two outputs for the left and right eye.
– Left Output: This outputs the left eye image with a new disparity channel, or a Stacked Mode
image with a new disparity channel.
– Right Output: This outputs the right eye image with a new disparity channel. This output is
visible only if Stack Mode is set to Separate.

Basic Node Setup

The New Eye node in the example below constructs a new image by interpolating between the two
stereo images using the embedded disparity channels.

A New Eye node creates a new stereo image using embedded disparity

Inspector

The New Eye Controls tab

Chapter 56 Stereo Nodes 1347

Controls Tab
The Controls tab is divided into identical parameters for the left eye and right eye. The parameters are
used to select which eye to recreate and the methods used for the interpolation.

Enable
The Enable checkbox allows you to activate the left or right eye independently. The New Eye will
replace enabled eye with an interpolated eye. For example, if the left eye is your “master” eye and you
are recreating the right eye, you would disable the left eye and enable the right eye.

Lock XY
Locks the X and Y interpolation parameters. When they are unlocked, you can provide separate
interpolation factors for using the X and Y disparity. For example, if you are working with the right eye
and you have the X Interpolation slider set to 1.0 and the Y Interpolation slider set to -1.0, you will be
effectively interpolating the left eye onto the right eye but vertically aligned to the left eye.

XY Interpolation Factor
Interpolation determines where the interpolated frame is positioned, relative to the two source frames:
A slider position of -1.0 outputs the frame Left and a slider position of 1.0 outputs the frame Right. A
slider position of 0.0 outputs a result that is halfway between Left and Right.

Depth Ordering
The Depth Ordering is used to determine which parts of the image should be rendered on top. When
warping images, there is often overlap. When the image overlaps itself, there are two options for which
values should be drawn on top.
– Largest Disparity On Top: The larger disparity values will be drawn on top in the overlapping
image sections.
– Smallest Disparity On Top: The smaller disparity values will be drawn on top in the overlapping
image sections.

Clamp Edges
Under certain circumstances, this option can remove the transparent gaps that may appear on the
edges of interpolated frames. Clamp Edges will cause a stretching artifact near the edges of the frame
that is especially visible with objects moving through it or when the camera is moving.
Because of these artifacts, it is a good idea to use Clamp Edges only to correct small gaps around the
edges of an interpolated frame.

Softness
Helps to reduce the stretchy artifacts that might be introduced by Clamp Edges.
If you have more than one of the Source Frame and Warp Direction checkboxes turned on, this can
lead to doubling up of the stretching effect near the edges. In this case, you’ll want to keep the
softness rather small at around 0.01. If you have only one checkbox enabled, you can use a larger
softness at around 0.03.

Source Frame and Warp Direction

The output of this node is generated by combining up to four different warps. You can choose to use
either the color values from the left or right frame in combination with the Forward (left > right)
Disparity or the Backward (right > left) Disparity. Sometimes you will want to replace an existing eye.
For example, if you want to regenerate the right eye, you would only use left eye warps.
It’s good to experiment with various options to see which gives the best effect. Using both the left and
right eyes can help fill in gaps on the left/right side of images. Using both the Forward/Backward
Disparity can give a doubling-up effect in places where the disparities disagree with each other.

Chapter 56 Stereo Nodes 1348

 – Left Forward: Takes the Left frame and uses the Forward Disparity to interpolate the new frame.
– Right Forward: Takes the Right frame and uses the Forward Disparity to interpolate the
new frame.
– Left Backward: Takes the Left frame and uses the Back Disparity to interpolate the new frame.
– Right Backward: Takes the Right frame and uses the Back Disparity to interpolate the new frame.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Stereo nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Splitter [Spl]

The Splitter node

NOTE: The Splitter node is available only in Fusion Studio and DaVinci Resolve Studio.

Splitter Node Introduction

The Splitter takes a stacked input image—for example, created with the Combiner—and provides two
output images: a left eye and a right eye.

Inputs
The two inputs on the Splitter node are used to connect the left and right images.
– Left Input: The orange input is used to connect a stacked stereo image.

Outputs
Unlike most nodes in Fusion, the Splitter node has two outputs for the left and right eye.
– Left Output: This outputs the left eye image.
– Right Output: This outputs the right eye image.

Basic Node Setup

Below, a stacked stereo image is connected to the input on a Splitter where a left and right eye
is output.

Chapter 56 Stereo Nodes 1349

A Splitter node creates a left and right image from a stacked stereo image

Inspector

l
The Splitter Controls tab

Controls Tab
The Controls tab is used to define the type of stacked image connected to the node’s input.

Split
The Split menu contains three options for determining the orientation of the stacked input image.
– None: No operation takes place. The output image on both outputs is identical to the input image.
– Horiz: The node expects a horizontally stacked image. This will result in two output images, each
being half the width of the input image.
– Vert: The node expects a vertically stacked image. This will result in two output images, each
being half the height of the input image.

Swap Eyes
Allows you to easily swap the left and right eye outputs.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Stereo nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Stereo Align [SA]

The StereoAlign node

NOTE: The Stereo Align node is available only in Fusion Studio and DaVinci Resolve Studio.

Chapter 56 Stereo Nodes 1350

Stereo Align Node Introduction
This extremely versatile node for fixing Stereo issues can be used for performing any of the following
actions or combinations thereof:
– Vertical alignment of one eye to the other
– Changing the convergence
– Changing the eye separation

By combining these operations in one node, you can execute them using only a single image
resampling. In essence, this node can be thought of as applying scales and translation to the
disparities and then using the modified disparities to interpolate between the views.

NOTE: Changing the eye separation can cause holes to appear, and it may not be possible to
fill them since the information needed may not be in either image. Even if the information is
there, the disparity may have mismatched the holes. You may need to fill the holes manually.
This node modifies only the RGBA channels.

TIP: Stereo Align does not interpolate the aux channels but instead destroys them. In
particular, the disparity channels are consumed/destroyed. Add another Disparity node after
the StereoAlign if you want to generate Disparity for the realigned footage.

Inputs
The two inputs on the Stereo Align node are used to connect the left and right images.
– Left Input: The orange input is used to connect either the left eye image or the stack image.
– Right Input: The green input is used to connect the right eye image. This input is available only
when the Stack Mode menu is set to Separate.

Outputs
Unlike most nodes in Fusion, Stereo Align has two outputs for the left and right eye.
– Left Output: This outputs the left eye image with a new disparity channel, or a Stacked Mode
image with a new disparity channel.
– Right Output: This outputs the right eye image with a new disparity channel. This output is
visible only if Stack Mode is set to Separate.

Basic Node Setup

Below, the Stereo Align receives the left and right eye images with disparity. Once the adjustments are
made, another Disparity node is added after it to generate disparity for the realigned footage.

Chapter 56 Stereo Nodes 1351

A Stereo Align node destroys the disparity channel, so another Disparity node is placed after it

Inspector

The Stereo Align Controls tab

Controls Tab
Vertical Alignment
This option determines how the vertical alignment is split between two eyes. Usually, the left eye is
declared inviolate, and the right eye is aligned to it to avoid resampling artifacts.

Chapter 56 Stereo Nodes 1352

When doing per pixel vertical alignment, it may be helpful to roughly pre-align the images by a global
Y-shift before disparity computation because the disparity generation algorithm can have problems
resolving small objects that move large distances.
Also, be aware that you must be careful about lens distortion because even if two cameras are
perfectly vertically aligned, they will still have vertical disparities due to lens distortion. As a best
practice, remove the lens distortion before computing the disparity. When a vertical alignment of the
right eye is done, you have essentially removed the Y-component of the lens distortion in the right
eye, and it will look wrong later when you try to distort it again.

Apply to
– Right: Only the right eye is adjusted.
– Left: Only the left eye is adjusted.
– Both: The vertical alignment is split evenly between the left and right eyes.

Mode
– Global: The eyes are simply translated up or down by the Y-shift to match up.
– Per Pixel: The eyes are warped pixel-by-pixel using the disparity to vertically align.

Keep in mind that this can introduce sampling artifacts and edge artifacts.

Y-shift
Y-shift is available only when the Mode menu is set to Global. You can either adjust the Y-shift
manually to get a match or drag the Sample button into the viewer, which picks from the disparity
channel of the left eye. Also remember, if you use this node to modify disparity, you can’t use the
Sample button while viewing the node’s output.

Snap to Whole Pixels

You can snap the global shift to whole pixels by enabling this option. When enabled, there is no
resampling of the image, but rather a simple shift is done so there will be no softening or image
degradation.

Convergence Point
The Convergence Point section is used as a global X-translation of L/R images.

Apply to
This menu determines which eyes are affected by convergence. You can choose to apply the
convergence to the left eye, right eye, or split between the two. In most cases, this will be set to Split. If
you set the eyes to Split, then the convergence will be shared 50-50 between both eyes. Sharing the
convergence between both eyes means you get half the shift in each eye, which in turn means smaller
holes and artifacts that need to be fixed later. The tradeoff is that you’ve resampled both eyes rather
than keeping one eye as a pure reference master.

X-shift
You can either use the slider to adjust the X-shift manually to get a match or use the Sample button to
pick from the disparity channels for easy point-to-feature alignment.

Snap
You can snap the global shift to whole pixels using this option. In this mode, there is no resampling of
the image, but rather a simple shift is done so there will be no softening or image degradation.

Chapter 56 Stereo Nodes 1353

Eye Separation
Eye Separation changes the distance between the left/right eyes, causing objects in the left/right eyes
to converge/diverge further depending on their distance from the camera.
This has the same effect as the Eye Separation option in the Camera 3D node.

Separation
This is a scale factor for eye separation.
– When set to 0.0, this leaves the eyes unchanged.
– Setting it to 0.1 increases the shifts of all objects in the scene by a factor of 10% in each eye.
– Setting it to 0.1 will scale the shifts of all objects 10% smaller.

Unlike the Split option for vertical alignment, which splits the alignment effect 50-50 between both
eyes, the Both option will apply 100-100 eye separation to both eyes. If you are changing eye
separation, it can be a good idea to enable per-pixel vertical alignment, or the results of interpolating
from both frames can double up.

Left/Right Eye Options

The left and right eye options contain depth ordering and warp direction controls independently for
the left and right eye.

Depth Ordering
The Depth Ordering is used to determine which parts of the image should be rendered on top. When
warping images, there is often overlap. When the image overlaps itself, there are two options for which
values should be drawn on top.
– Largest Disparity On Top: The larger disparity values will be drawn on top in the overlapping
image sections.
– Smallest Disparity On Top: The smaller disparity values will be drawn on top in the overlapping
image sections.

Clamp Edges
Under certain circumstances, this option can remove the transparent gaps that may appear on the
edges of interpolated frames. Clamp Edges will cause a stretching artifact near the edges of the frame
that is especially visible with objects moving through it or when the camera is moving.
Because of these artifacts, it is a good idea to use Clamp Edges only to correct small gaps around the
edges of an interpolated frame.

Edge Softness
Helps to reduce the stretchy artifacts that might be introduced by Clamp Edges.
If you have more than one of the Source Frame and Warp Direction checkboxes turned on, this can
lead to doubling up of the stretching effect near the edges. In this case, you’ll want to keep the
softness rather small at around 0.01. If you have only one checkbox enabled, you can use a larger
softness at around 0.03.

Source Frame and Warp Direction

The output of this node is generated by combining up to four different warps. You can choose to use
either the color values from the left or right frame in combination with the Forward (left > right)
Disparity or the Backward (right > left) Disparity. Sometimes you will want to replace an existing eye.
For example, if you want to regenerate the right eye, you would use only left eye warps.

Chapter 56 Stereo Nodes 1354

It’s good to experiment with various options to see which gives the best effect. Using both the left and
right eyes can help fill in gaps on the left/right side of images. Using both the Forward/Backward
Disparity can give a doubling-up effect in places where the disparities disagree with each other.
– Left Forward: Takes the Left frame and uses the Forward Disparity to interpolate the new frame.
– Right Forward: Takes the Right frame and uses the Forward Disparity to interpolate the new frame.
– Left Backward: Takes the Left frame and uses the Back Disparity to interpolate the new frame.
– Right Backward: Takes the Right frame and uses the Back Disparity to interpolate the new frame.

Stack Mode
In Stack Mode, L and R outputs will output the same image.
If High Quality is off, the interpolations are done using nearest-neighbor sampling, leading to a more
“noisy” result. To ensure High Quality is enabled, right-click under the viewers, near the transport
controls, and choose High Quality from the pop-up menu.

Swap Eyes
Allows you to easily swap the left and right eye outputs.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Stereo nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Example

Different settings for Eye Separation…

…and example settings for Convergence

Chapter 56 Stereo Nodes 1355

Z To Disparity [Z2D]

The Z To Disparity node

NOTE: The Z To Disparity node is available only in Fusion Studio and DaVinci Resolve Studio.

Z To Disparity Node Introduction

Z To Disparity takes a stereo camera and an image containing a Z channel and outputs the same
image but with disparity channels in it. This is useful for constructing a Disparity map from CG renders,
which will be more accurate than the Disparity map created from the Disparity node.

Inputs
The three inputs on the Z To Disparity node are used to connect the left and right images and a
camera node.
– Left Input: The orange input is used to connect either the left eye image or the stack image.
– Right Input: The green input is used to connect the right eye image. This input is available only
when the Stack Mode menu is set to Separate.
– Stereo Camera: The magenta input is used to connect a stereo perspective camera, which may
be either a Camera 3D with eye separation, or a tracked L/R Camera 3D.

Outputs
Unlike most nodes in Fusion, Z To Disparity has two outputs for the left and right eye.
– Left Output: This outputs the left eye image containing a new disparity channel, or a Stacked
Mode image with a new disparity channel.
– Right Output: This outputs the right eye image with a new disparity channel. This output is
visible only if Stack Mode is set to Separate.

Basic Node Setup

Below, a stereo camera and an image containing a Z channel is connected to a Z To Disparity node.
The same image is output with disparity channels.

A Z To Disparity node takes an image with a Z channel and creates a disparity channel

Chapter 56 Stereo Nodes 1356

Inspector

The Z To Disparity Controls tab

Controls Tab
The Controls tab includes settings that refine the conversion algorithm.

Output Disparity To RGB

In addition to outputting disparity values into the disparity channel, activating this checkbox causes Z
To Disparity to also output the disparity values into the color channels as {x, y, 0, 1}.
When activated, this option will automatically promote the RGBA color channels to float32. This option
is useful for a quick look to see what the disparity channel looks like.

Refine Disparity
This refines the Disparity map based on the RGB channels.

Strength
Increasing this slider does two things. It smooths out the depth in constant color regions and moves
edges in the Z channel to correlate with edges in the RGB channels. Increasing the refinement has the
undesirable effect of causing texture in the color channel to show up in the Z channel. You will want to
find a balance between the two.

Radius
This is the pixel-radius of the smoothing algorithm.

Stack Mode
In Stack Mode, L and R outputs will output the same image.
If HiQ is off, the interpolations are done using nearest-neighbor sampling, leading to a more
“noisy” result.

Swap Eyes
This allows you to easily swap the left and right eye outputs.

The Z To Disparity Camera tab

Chapter 56 Stereo Nodes 1357

Camera Tab
The Camera tab includes settings for selecting a camera and setting its conversion point if necessary.

Camera Mode
If you need correct real-world disparity values because you are trying to match some effect to an
existing scene, you should use the External setting to get precise disparity values back. When External
is selected, a magenta camera input is available on the node to connect an existing stereo Camera 3D
node, and use the Camera settings to determine the Disparity settings.
If you just want any disparity and do not particularly care about the exact details of how it is offset and
scaled, or if there is no camera available, then the Artistic setting might be helpful.

Camera
If you connect a Merge 3D node that contains multiple cameras to the camera input, the Camera menu
allows you to select the camera to use.

Artistic Camera mode

If you do not have a camera, you can adjust the artistic controls to produce a custom disparity channel
whose values will not be physically correct but will be good enough for compositing hacks. There are
two controls to adjust:

Convergence Point
This is the Z value of the convergence plane. This corresponds to the negative of the Convergence
Distance control that appears in Camera 3D. At this distance, objects in the left and right eyes are at
exactly the same position (i.e., have zero disparity).
Objects that are closer appear to pop out of the screen, and objects that are further appear behind
the screen.

Background Disparity (Sample from Left Eye)

This is the disparity of objects in the distant background. You can think of this as the upper limit to
disparity values for objects at infinity. This value should be for the left eye. The corresponding value in
the right eye will be the same in magnitude but negative.

Settings Tab
The Settings tab in the Inspector is also duplicated in other Stereo nodes. These common controls are
described in detail in the following “The Common Controls” section.

Chapter 56 Stereo Nodes 1358

The Common Controls
The Common Settings tab can be found on virtually every tool found in Fusion. The following controls
are specific settings for Stereo nodes.

Settings Tab

The Common Settings Stereo 3D Settings tab

Blend
The Blend control is used to blend between the tool’s original image input and the tool’s final modified
output image. When the blend value is 0.0, the outgoing image is identical to the incoming image.
Normally, this will cause the tool to skip processing entirely, copying the input straight to the output.

Process When Blend Is 0.0

The tool is processed even when the input value is zero. This can be useful if the node is scripted to
trigger a task, but the node’s value is set to 0.0.

Red/Green/Blue/Alpha Channel Selector

These four buttons are used to limit the effect of the tool to specified color channels. This filter is often
applied after the tool has been processed.
For example, if the Red button on a Blur tool is deselected, the blur will first be applied to the image,
and then the red channel from the original input will be copied back over the red channel of the result.
There are some exceptions, such as tools for which deselecting these channels causes the tool to skip
processing that channel entirely. Tools that do this will generally possess a set of identical RGBA
buttons on the Controls tab in the tool. In this case, the buttons in the Settings and the Controls tabs
are identical.

Chapter 56 Stereo Nodes 1359

Apply Mask Inverted
Enabling the Apply Mask Inverted option inverts the complete mask channel for the tool. The mask
channel is the combined result of all masks connected to or generated in a node.

Multiply by Mask
Selecting this option will cause the RGB values of the masked image to be multiplied by the mask
channel’s values. This will cause all pixels of the image not included in the mask (i.e., set to 0) to
become black/transparent.

Use Object/Use Material (Checkboxes)

Some 3D software can render to file formats that support additional channels. Notably, the EXR file
format supports Object ID and Material ID channels, which can be used as a mask for the effect. These
checkboxes determine whether the channels will be used, if present. The specific Material ID or
Object ID affected is chosen using the next set of controls.

Correct Edges
This checkbox appears only when the Use Object or Use Material checkboxes are selected. It toggles
the method used to deal with overlapping edges of objects in a multi-object image. When enabled,
the Coverage and Background Color channels are used to separate and improve the effect around the
edge of the object. If this option is disabled (or no Coverage or Background Color channels are
available), aliasing may occur on the edge of the mask.
For more information on Coverage and Background Color channels, see Chapter 18, “Understanding
Image Channels” in the Fusion Reference Manual or Chapter 79 in the DaVinci Resolve
Reference Manual.

Object ID/Material ID (Sliders)

Use these sliders to select which ID will be used to create a mask from the object or material channels
of an image. Use the Sample button in the same way as the Color Picker: to grab IDs from the image
displayed in the viewer. The image or sequence must have been rendered from a 3D software
package with those channels included.

Hide Incoming Connections

Enabling this checkbox can hide connection lines from incoming nodes, making a node tree appear
cleaner and easier to read. When enabled, empty fields for each input on a node will be displayed in
the Inspector. Dragging a connected node from the node tree into the field will hide that incoming
connection line as long as the node is not selected in the node tree. When the node is selected in the
node tree, the line will reappear.

Comments
The Comments field is used to add notes to a tool. Click in the empty field and type the text. When a
note is added to a tool, a small red square appears in the lower-left corner of the node when the full
tile is displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the
note in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 56 Stereo Nodes 1360

Chapter 57

Tracker Nodes
This chapter details the Tracker nodes available in Fusion.

The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Tracker [Tra]  1362
Planar Tracker Node [PTra]  1376
Planar Transform Node [PXF]  1386
Camera Tracker [CTra]  1388
The Common Controls  1403

Chapter 57 Tracker Nodes 1361

Tracker [Tra]

The Tracker node

Tracker Node Introduction

The Tracker is used to detect and follow one or more pixel patterns across frames in moving video.
The tracking data can then be used to control the position or values of other nodes in the composition
(for example, the center of a Light Rays node). Additionally, trackers can be used to stabilize an image
or to apply destabilization to one image based on the motion of another.
For more information, refer to Chapter 22, “Using the Tracker Node” in the Fusion Reference Manual or
Chapter 83 in the DaVinci Resolve Reference Manual.

Inputs
The Tracker has three inputs:
– Background: The orange image input accepts the main 2D image to be tracked.
– Foreground: The optional green foreground accepts a 2D image to be merged on top of the
background as a corner pin or match move.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the tracking to
certain areas.

Basic Node Setup

Tracker nodes can be applied inline with other nodes or as a branch from the clip you want to track.
When used inline, an image can be stabilized by connecting it to the orange background input. After
the image is tracked, setting the Tracker’s Operation menu to Match Move will apply a stabilization to
the connected image.

A Tracker node connected inline to an image for stabilization

When used as an offshoot from the node tree, an image can be tracked and then that tracking data is
published for use on another node somewhere else in the node tree. The output of the Tracker does
not need to connect to another node. The tracking data is published and can be used via the Connect
To contextual menu.

Chapter 57 Tracker Nodes 1362

A Tracker node branched from the node tree

The Tracker can also work as a replacement for a Merge tool in match-moving setups. Below, the
Tracker tracks the image connected to the orange background input and applies the tracking data to
the image connected to the foreground input. The same foreground-over-background merge
capabilities are available in the Tracker node.

A Tracker node set up to apply a match move to the foreground input

Tracker Onscreen Controls

Each pattern in the Tracker has its own set of onscreen controls used to select the pixels in the image
to be tracked. These controls are visible in the viewers whenever you select a tracker in the node tree.

Tracker onscreen controls

Pattern Rectangle
In the viewer, the tracker displays a solid-line red rectangle called the pattern rectangle. Every pixel
within the rectangle makes up the pattern used for tracking. You can resize the pattern if necessary by
dragging on the rectangle’s borders.

A pattern rectangle identifies the area to track

Chapter 57 Tracker Nodes 1363

Search Rectangle
Whenever the mouse moves over the pattern rectangle, a second rectangle with a dashed outline
appears. The dashed outline represents the search area, which determines how far away from the
current pattern the Tracker looks in the next frame. The search area should always be larger than the
pattern, and it should be large enough to encompass the largest frame-to-frame movement in the
scene. Faster moving objects require larger search areas, and slower moving objects can get away
with smaller search areas. The larger the search area, the longer it takes to track, so try not to make
the search area larger than necessary. If the selected Tracker has a custom name, the name of that
Tracker is displayed as a label at the bottom right of the search area rectangle.

The search rectangle is the area searched

from frame to frame to locate the pattern.

Repositioning the Tracker

The pattern rectangle has a small handle in the upper-left corner. Dragging on the handle repositions
the pattern. An enlarged view of the pattern is displayed under your mouse pointer to assist with the
precise positioning of the pattern. This thumbnail disappears when the mouse button is released. You
can adjust the magnification ratio in the Inspector’s Options tab.

Dragging the handle magnifies the pattern rectangle for precise placement.

TIP: There is no limit to the number of trackers that can be used in one composition, or the
number of objects that use the tracking data. There is also no limit to the number of patterns
that can be tracked by a single Tracker node. This chapter serves as a reference for the
various controls in the Tracker, but we strongly suggest you read the more general
information in Chapter 22, “Using the Tracking Node” in the Fusion Reference Manual or
Chapter 83 in the DaVinci Resolve Reference Manual.
The Tracker can be employed in two forms: as a node in the Node Editor or as a modifier
attached to a parameter. When used as a node in the Node Editor, the image tracked comes
from the input to the Tracker node. When used as a modifier, controls appear in the Modifiers
tab for the node with the connected control. Tracker Modifiers can track only one pattern, but
the image source can come from anywhere in the composition. Use this technique when
tracking a quick position for an element.

Chapter 57 Tracker Nodes 1364

Inspector

The Tracker node Trackers tab

Trackers Tab
The Trackers tab contains controls for creating, positioning, and initiating tracking operations. After
tracking, offset controls are used to improve the alignment of the image following the track.

Track Buttons
There are four buttons to initiate tracking, and one button in the middle used to stop a track in
progress. These buttons can track the current pattern forward or backward in time. Holding the pointer
over each button displays a tooltip with the name of each button.
The buttons operate as follows:
– Track Reverse: Clicking this button causes all active trackers to begin tracking, starting at the end
of the render range and moving backward through time until the beginning of the render range.
– Track Reverse From Current Time: Clicking this button causes all active trackers to begin
tracking, starting at the current frame and moving backward through time until the beginning of
the render range.
– Stop Tracking: Clicking this button or pressing ESC stops the tracking process immediately. This
button is active only when tracking is in process.
– Track Forward From Current Time: Clicking this button causes all active trackers to begin
tracking, starting at the current frame and moving forward through time until the end of the render
range.
– Track Forward: Clicking this button causes all active trackers to begin tracking, starting at the first
frame in the render range and moving forward through time until the end of the render range.

Chapter 57 Tracker Nodes 1365

Tracking Behavior Controls
The following controls all affect how trackers adapt to changing patterns, how the resulting track path
is defined, and how many keyframes are generated.

Frames Per Path Point

This slider determines how often the Tracker sets a keyframe on the path. The default is 1, which sets a
keyframe on the tracked path at every frame.
Increasing the value causes the tracked path to be less accurate. This may be desirable if the track is
returning fluctuating results, but under normal circumstances, leave this control at its default value.

TIP: If the project is field rendered, a value of 1 sets a keyframe on every field. Since the
Tracker is extremely accurate, this will result in a slight up-and-down jittering due to the
position of the fields. For better results when tracking interlaced footage in Field mode, set
the Frames Per Path Point slider to a value of 2, which results in one keyframe per frame of
your footage.

Adaptive Mode
Fusion is capable of reacquiring the tracked pattern, as needed, to help with complex tracks. This
menu determines the Adaptive tracking method.
– None: When set to None, the tracker searches for the original pattern in each frame.
– Every Frame: When set to Every Frame, the tracker reacquires the pattern every frame. This helps
the Tracker compensate for gradual changes in profile and lighting over time.
– Best Match: When set to Best Match, the tracker compares the original selected pattern to the
pattern acquired at each frame. If the variation between the two patterns exceeds the threshold
amount defined by the Match Tolerance control, the tracker does not reacquire the pattern on that
frame. This helps to avoid Tracker drift caused by transient artifacts that cross the pattern’s path
(such as a shadow).

Path Center
This menu determines how the Tracker behaves when repositioning a pattern. This menu is particularly
useful when a pattern leaves the frame or changes so significantly that it can no longer be tracked.
– Pattern Center: When Pattern Center is selected in the menu, the tracked path continues from the
center of the new path. This is appropriate when replacing an existing path entirely.
– Track Center (append): When Track Center (append) is selected in the menu, the path tracked by
a new pattern will be appended to the existing path. The path created is automatically offset by
the required amount. This setting is used to set a new tracking pattern when the original pattern
moves out of the frame or gets obscured by other objects. This technique work bests if the new
pattern is located close to the position of the original pattern to avoid any problems with parallax
or lens distortion.

Tracker List
A Tracker node can track multiple patterns. Each tracker pattern created in the current Tracker node is
managed in the Tracker List.

Tracker List
The Tracker List shows the names of all trackers created.
– Each tracker pattern appears in the list by name, next to a small checkbox. Clicking the name of
the tracker pattern will select that tracker pattern.

Chapter 57 Tracker Nodes 1366

 – The controls below the list will change to affect that tracker pattern only. Click a selected tracker
pattern once to rename the tracker pattern to something more descriptive.
– Clicking the checkbox changes the state of the tracker.

Tracker States
– Enabled (black checkbox): An enabled pattern will re-track each time the track is initiated. Its
path data is available for use by other nodes, and the data is available for Stabilization and Corner
Positioning.
– Suspended (white circle): A Suspended pattern does not re-track when the track is initiated.
The data is locked to prevent additional changes. The data from the path is still available for
other nodes, and the data is available for advanced Tracking modes like Stabilization and Corner
Positioning.
– Disabled (clear): A Disabled pattern does not create a path when tracking is initialized, and its
data is not available to other nodes or for advanced Tracking operations like Stabilization and
Corner Positioning.

Three tracking patterns from top to bottom:

enabled, suspended, and disabled

Add/Delete Tracker
Use these buttons to add or delete trackers from your Tracker List.

Show
This menu selects what controls are displayed in the Tracker node controls. They do not affect the
operation of the tracker; they only affect the lower half of the Inspector interface.
– Selected Tracker Details: When Selected Tracker Details is chosen, the controls displayed pertain
only to the currently selected tracker. You will have access to the Pattern window and the Offset
sliders.
– All Trackers: When All Trackers is selected, the pattern window for each of the added tracking
patterns is displayed simultaneously below the Tracker List.

Left Pattern Display

The pattern display has two side-by-side image windows and a series of status bars. The window on
the left shows the pattern initially selected, while the one on the right shows a real-time display of the
current pattern as tracking progresses.
As the onscreen controls move while tracking, the display in the leftmost window updates to show the
pattern. As the pattern moves, the vertical bars immediately to the right of the image indicate the
clarity and contrast of the image channels.
The best channel or channels get selected for tracking based on clarity. These channels have a gray
background in the vertical bar representing that channel. You can use the automatic tracking or
override the selection and choose the channel by selecting the button beneath the channel to track.

Chapter 57 Tracker Nodes 1367

Tracker pattern display

Under normal circumstances, the channel selected shows in the pattern display. If the selected
channel is blue, then a grayscale representation of the blue channel for the pattern appears. The
image is represented in color only when you activate the Full Color button.
Override this behavior by selecting the Show Full Color button beneath the pattern display instead of
the Show Selected Channel button.

TIP: Because Fusion looks for the channel with the highest contrast automatically, you might
end up tracking a noisy but high-contrast channel. Before tracking, it’s always a good idea to
zoom in to your footage and check the RGB channels individually.

Right Pattern Display

The pattern display on the right indicates the actual pattern acquired for tracking. This display is black
until tracking the selected pattern for the first time. The pattern display becomes active during
tracking, displaying the pattern that Fusion acquires from frame to frame.
As the tracking occurs, the pattern from each frame accumulates into a Flipbook, which can be played
back in the pattern window after tracking by using the transport controls at the bottom of the window.
While the track is progressing, the vertical bar immediately to the right of the pattern shows how
confident Fusion is that the current pattern matches the initially selected pattern. A green bar indicates
a high degree of confidence that the current pattern matches the original, a yellow bar indicates less
certainty, and a red bar indicates extreme uncertainty.
After tracking, the pattern display shows a small Flipbook of the track for that pattern to help identify
problem frames for the track.

Tracker Sizes
In addition to onscreen controls, each tracker has a set of sizing parameters that let you adjust the
pattern and search box.
– Pattern Width and Height: Use these controls to adjust the width and height of the selected
tracker pattern manually. The size of the tracker pattern can also be adjusted in the viewer, which
is the normal method, but small adjustments are often easier to accomplish with the precision of
manual controls.
– Search Width and Height: The search area defines how far Fusion will look in the image from
frame to frame to reacquire the pattern during tracking. As with the Pattern Width and Height, the
search area can be adjusted in the viewer, but you may want to make small adjustments manually
using these controls.

Tracked Center
This positional control indicates the position of the tracker’s center. To remove a previously tracked
path from a tracker pattern, right-click this parameter and select Remove Path from the
contextual menu.

Chapter 57 Tracker Nodes 1368

X and Y Offset
The Offset controls help to create a track for objects that may not provide very well defined or reliable
patterns. The Offset controls permit the tracking of something close to the intended object instead.
Use these Offsets to adjust the desired position of the path, while the tracker pattern rectangle is
positioned over the actual tracking location.
The Offset can also be adjusted directly in the viewer by activating the Offsets button in the
viewer toolbar.

The tracker offset icon in the

upper left of the viewer is used
to offset the tracking pattern
from the intended object.

The Tracker Operation tab

Operation Tab
While the Trackers tab controls let you customize how the Tracker node analyzes motion to create
motion paths, the Operation tab puts the analyzed motion data to use, performing image transforms of
various kinds.
The Tracker node is capable of performing a wide variety of functions, from match moving an object
into a moving scene, smoothing out a shaky camera movement, or replacing the content of a sign. Use
the options and buttons in the Operation tab to select the function performed by the Tracker node.

Operation Menu
The Operation menu contains four functions performed by the Tracker. The remaining controls in this
tab fine-tune the result of this selection.
– None: The Tracker performs no additional operation on the image beyond merely locating and
tracking the chosen pattern. This is the default mode, used to create a path that will then drive
another parameter on another node.

Chapter 57 Tracker Nodes 1369

 – Match Move: When only the orange background input is connected, this mode stabilizes the
image. When a foreground image is connected to the green foreground input, the foreground
image matches the position, rotation, and scaling based on the tracking patterns. Stabilizing and
match move require a minimum of one tracking pattern to determine position, and two or more to
determine scaling and rotation.
– Corner Positioning: The Corner Positioner mode tracks the four corners of a rectangular object
and replaces the contents with a new image. This function requires a minimum of four tracking
patterns. If there are not enough tracking patterns, new tracking patterns are added until the total
equals four.
– Perspective Positioning: This mode is the inverse of the Corner Positioning mode. Rather than
replacing the contents of the rectangle, the four trackers are mapped to the four corners of the
image. This is generally used to remove perspective from an image. Like the Corner Positioning
mode, this mode requires four tracking patterns, which automatically get added if there are
fewer patterns.

Additional Layering Controls

When you choose any operation other than None, a series of additional controls appear.

Merge
The Merge control determines what is done (if anything) with the image provided to the green
Foreground input of the Tracker. This menu appears when the operation is set to anything other
than None.
– BG Only: The foreground input is ignored; only the background is affected. This is used primarily
when stabilizing the background image.
– FG Only: The foreground input is transformed to match the movement in the background, and this
transformed image is passed through the Tracker’s output. This Merge technique is used when
match moving one layer’s motion to another layer’s motion.
– FG Over BG: The foreground image is merged over the background image, using the Merge
method described by the Apply Mode control that appears.
– BG Over FG: The background is merged over the foreground. This technique is often used when
tracking a layer with an Alpha channel so that a more static background can be applied behind it.

Apply Mode and Operator Menus

This menu provides a variety of options that determine how the two layers should be combined. The
options in this menu are identical to those found in the Merge node.
– Apply Modes: The Apply Mode setting determines the math used when blending or combining
the foreground and background pixels.
– Normal: The Default merge mode uses the foreground’s alpha channel as a mask to determine
which pixels are transparent and which are not. When this is active, another menu shows
possible operations, including Over, In, Held Out, Atop, and XOr.
– Screen: Screen merges the images based on a multiplication of their color values. The alpha
channel is ignored, and layer order becomes irrelevant. The resulting color is always lighter.
Screening with black leaves the color unchanged, whereas screening with white will always
produce white. This effect creates a similar look to projecting several film frames onto the same
surface. When this is active, another menu shows possible operations, including Over, In, Held
Out, Atop, and XOr.
– Dissolve: Dissolve mixes two image sequences together. It uses a calculated average of the
two images to perform the mixture.

Chapter 57 Tracker Nodes 1370

 – Multiply: Multiplies the values of a color channel. This will give the appearance of darkening
the image as the values are scaled from 0 to 1. White has a value of 1, so the result would be
the same. Gray has a value of 0.5, so the result would be a darker image or, in other words, an
image half as bright.
– Overlay: Overlay multiplies or screens the color values of the foreground image, depending
on the color values of the background image. Patterns or colors overlay the existing pixels
while preserving the highlights and shadows of the color values of the background image. The
background image is not replaced but is mixed with the foreground image to reflect the original
lightness or darkness of the background image.
– Soft Light: Soft Light darkens or lightens the foreground image, depending on the color values
of the background image. The effect is similar to shining a diffused spotlight on the image.
– Hard Light: Hard Light multiplies or screens the color values of the foreground image,
depending on the color values of the background image. The effect is similar to shining a harsh
spotlight on the image.
– Color Dodge: Color Dodge uses the foreground’s color values to brighten the background
image. This is similar to the photographic practice of dodging by reducing the exposure of an
area of a print.
– Color Burn: Color Burn uses the foreground’s color values to darken the background image.
This is similar to the photographic practice of burning by increasing the exposure of an area of
a print.
– Darken: Darken looks at the color information in each channel and selects the background or
foreground image’s color value, whichever is darker, as the result color. Pixels lighter than the
merged colors are replaced, and pixels darker than the merged color do not change.
– Lighten: Lighten looks at the color information in each channel and selects the background or
foreground image’s color values, whichever is lighter, as the result color value. Pixels darker
than the merged color are replaced, and pixels lighter than the merged color do not change.
– Difference: Difference looks at the color information in each channel and subtracts the
foreground color values from the background color values or the background from the
foreground, depending on which has the greater brightness value. Merging with white inverts
the color. Merging with black produces no change.
– Exclusion: Exclusion creates an effect similar to but lower in contrast than the Difference mode.
Merging with white inverts the base color values. Merging with black produces no change.
– Hue: Hue creates a result color with the luminance and saturation of the background color
values and the hue of the foreground color values.
– Saturation: Saturation creates a result color with the luminance and hue of the base color and
the saturation of the blend color.
– Color: Color creates a result color with the luminance of the background color value and the
hue and saturation of the foreground. This preserves the gray levels in the image and is useful
for coloring monochrome images.
– Luminosity: Luminosity creates a color with the hue and saturation of the background color
and the luminance of the foreground color. This mode creates an inverse effect from that of the
Color mode.

– Operator Modes: This menu is used to select the Operation Mode of the merge. It determines
how the foreground and background are combined to produce a result. This drop-down menu is
visible only when the Merge node’s Apply Mode is set to either Normal or Screen.

NOTE: For an excellent description of the math underlying the Operation modes, read
“Compositing Digital Images,” Porter, T., and T. Duff, SIGGRAPH 84 proceedings, pages
253-259. Essentially, the math is as described below.

Chapter 57 Tracker Nodes 1371

 TIP: Some modes not listed in the Operator drop-down menu (Under, In, Held In, Below)
are easily obtained by swapping the foreground and background inputs and choosing a
corresponding mode.

The formula used to combine pixels in the merge is always fg * x + bg * y. The different operations
determine exactly what x and y are, as shown in the description for each mode.
The Operator modes are as follows:
– Over: The Over mode adds the foreground layer to the background layer by replacing the
pixels in the background with the pixels from the Z wherever the foreground’s alpha channel is
greater than 1.

x = 1, y = 1-[foreground Alpha]

– In: The In mode multiplies the alpha channel of the background input against the pixels in
the foreground. The color channels of the foreground input are ignored. Only pixels from the
foreground are seen in the final output. This essentially clips the foreground using the mask
from the background.

x = [background Alpha], y = 0

– Held Out: Held Out is essentially the opposite of the In operation. The pixels in the foreground
image are multiplied against the inverted alpha channel of the background image. You can
accomplish exactly the same result using the In operation and a Matte Control node to invert
the matte channel of the background image.

x = 1-[background Alpha], y = 0

– ATop: ATop places the foreground over the background only where the background has
a matte.

x = [background Alpha], y = 1-[foreground Alpha]

– XOr: XOr combines the foreground with the background wherever either the foreground or the
background have a matte, but never where both have a matte.

x = 1-[background Alpha], y = 1-[foreground Alpha]

– Subtractive/Additive: This slider controls whether Fusion performs an Additive merge, a

Subtractive merge, or a blend of both. This slider defaults to Additive merging for most operations,
assuming the input images are premultiplied (which is usually the case). If you don’t understand
the difference between Additive and Subtractive merging, here’s a quick explanation:
– An Additive merge is necessary when the foreground image is premultiplied, meaning that the
pixels in the color channels have been multiplied by the pixels in the alpha channel. The result
is that transparent pixels are always black since any number multiplied by 0 always equals 0.
This obscures the background (by multiplying with the inverse of the foreground alpha), and
then simply adds the pixels from the foreground.
– A Subtractive merge is necessary if the foreground image is not premultiplied. The compositing
method is similar to an Additive merge, but the foreground image is first multiplied by its alpha
to eliminate any background pixels outside the alpha area.

In most software applications, you will find the Additive/Subtractive option displayed as a simple
checkbox. Fusion lets you blend between the Additive and Subtractive versions of the merge
operation, which is occasionally useful for dealing with problem composites with edges that are
calling attention to themselves as too bright or too dark.
For example, using a Subtractive setting on a premultiplied image may result in darker edges.
Using an Additive setting with a non-premultiplied image may result in lightening the edges.

Chapter 57 Tracker Nodes 1372

 By blending between Additive and Subtractive, you can tweak the edge brightness to be just right
for your situation.

Filter Method (Match Move)

Determines which filter to use to handle image transforms made using the Tracker node. This menu
appears only when the Operation Mode is set to Match Move.
– Box: This is a simple interpolation resize of the image.
– Linear: This uses a simplistic filter, which produces relatively clean and fast results.
– Quadratic: This filter produces a nominal result. It offers a good compromise between
speed and quality.
– Cubic: This produces better results with continuous-tone images but is slower than Bi‑Cubic.
If the images have fine detail in them, the results may be blurrier than desired.
– Catmull-Rom: This produces good results with continuous-tone images that are resized down. It
produces sharp results with finely detailed images.
– Gaussian: This is very similar in speed and quality to Bi-Cubic.
– Mitchell: This is similar to Catmull-Rom but produces better results with finely detailed images.
It is slower than Catmull-Rom.
– Lanczos: This is very similar to Mitchell and Catmull-Rom but is a little cleaner and also slower.
– Sinc: This is an advanced filter that produces very sharp, detailed results; however, it may produce
visible “ringing” in some situations.
– Bessel: This is similar to the Sinc filter but may be slightly faster.

Edges
This menu selects how the revealed edges are handled when the image is moved to match position
and scaling.
– Black Edges: Out-of-frame edges revealed by Stabilization are left black.
– Wrap: Portions of the image moved off frame to one side are used to fill edges that are revealed
on the opposite side.
– Duplicate: The last valid pixel on an edge is repeated to the edge of the frame.
– Mirror: Image pixels are mirrored to fill to the edge of the frame.

Position, Rotation, and Scaling Checkboxes (Match Move)

The Position, Rotation, and Scaling checkboxes appear only when the mode is set to Match Move.
They determine what components of motion that Stabilization will attempt to correct in the image. For
example, if only the Position checkbox is selected, no attempt will be made to correct for Rotation and
Scaling in the image.

Flatten Transformation (Match Move)

This checkbox appears only when the mode is set to Match Move. Like most transformations in Fusion,
Stabilization is concatenated with other sequential transformations by default. Selecting this checkbox
will flatten the transform, breaking any concatenation taking place and applying the transform
immediately.

Mapping Type
The Mapping Type control appears only in the Corner Positioning mode. There are two options
in the menu:
– Bi_Linear: The first method is Bi-Linear, where the foreground image is mapped into the
background without any attempt to correct for perspective distortion. This is identical to how
previous versions of Fusion operated.

Chapter 57 Tracker Nodes 1373

 – Perspective: The foreground image is mapped into the background taking perspective distortion
into account. This is the preferred setting since it maps better to the real world than the older Bi-
Linear setting.

Corner Selector (Corner or Perspective Positioning)

When the operation of the Tracker is set to either Corner or Perspective Positioning modes, four
drop-down menus appear. These options choose which trackers map to each of the four corners of
the rectangle. This is useful when a Tracker has more than four patterns selected, and you must
choose which patterns the positioners use.

Rotate Clockwise and Counter-Clockwise Buttons (Corner or Perspective Positioning)

These controls appear only when the operation of the Tracker is set to either Corner or Perspective
Positioning modes. They are used to rotate the foreground image by 90 degrees before it is applied to
the background.

Stabilize Settings
The Tracker node automatically outputs several steady and unsteady position outputs to which other
controls in the Node Editor can be connected. The Stable Position output provides X and Y
coordinates to match or reverse motion in a sequence. These controls are available even when the
operation is not set to Match Move, since the Stable Position output is always available for connection
to other nodes.

Match Move Settings

These settings determine how tracking data is correlated with the reference pattern for making
transforms.

Pivot Type
The Pivot type menu determines how the anchor point for rotation is selected.
– Tracker Average: Averages the location based on the tracking points.
– Selected Tracker: Provides a menu where one of the current trackers can be selected
as the pivot point.
– Manual: Displays X and Y position number fields where you can manually position the pivot points.

Reference
The Reference mode determines the “snapshot frame” based on the frame where the pattern is first
selected. All Stabilization is intended to return the image back to that reference.
– Select Time: Lets you select the current frame.
– Start: The Snapshot Frame is determined to be the first frame in the tracked path. All Stabilization
is intended to return the image back to that reference.
– Start and End: The Start and End Reference mode is somewhat different from all other Reference
modes. Where the others are intended to take a snapshot frame to which all stabilization returns,
immobilizing the image, the Start and End mode is intended to smooth existing motion, without
removing it. This mode averages the motion between the Start and End of the path, drawing a
straight line between those points.
When this mode is active, it reveals the Reference Intermediate Points control. Increasing the value
of this control increases the number of points in the path used by the Reference, smoothing the
motion from a straight line between Start and End without making it wholly linear.
– End: The Snapshot Frame is determined to be the last frame in the tracked path. All Stabilization is
intended to return the image back to that reference.

Chapter 57 Tracker Nodes 1374

 TIP: By default, the Tracker displays a single displacement path of the tracked data in the
Spline Editor. To view X and Y paths of the tracked points in the Spline Editor, go to
Preferences > Globals > Splines.

Display Options Tab

The Display Options tab lets you customize the look of onscreen controls in the viewer.

The Tracker Display Options tab

Show Pattern Names

This option defines whether the Tracker’s pattern names will be displayed in the viewer. Switch it off to
see the pattern rectangle instead.

Enlarged Pattern on Dragging

This option defines whether there is a magnified thumbnail view when positioning the pattern
rectangle.

Enlargement Scale
The zoom factor that is used when positioning the pattern rectangle when the above option is
activated.

TIP: The outputs of a tracker (seen in the Connect to… menu) can also be used by
scripts. They are:
– SteadyPosition: Steady Position
– UnsteadyPosition: Unsteady Position
– SteadyAxis: Steady Axis
– SteadySize: Steady Size
– UnsteadySize: Unsteady Size
– SteadyAngle: Steady Angle
– UnsteadyAngle: Unsteady Angle
– Position1: Tracker 1 Offset position
– PerspectivePosition1: Tracker 1 Perspective Offset position
– PositionX1: Tracker 1 Offset X position (3D Space)
– PositionY1: Tracker 1 Offset Y position (3D Space)
– PerspectivePositionX1: Tracker 1 Perspective Offset X position (3D Space)
– PerspectivePositionY1: Tracker 1 Perspective Offset Y position (3D Space)
– SteadyPosition1: Tracker 1 Steady Position
– UnsteadyPosition1: Tracker 1 Unsteady Position (likewise for the 2nd, 3rd, and so on)

Chapter 57 Tracker Nodes 1375

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Tracking nodes. These common controls
are described in detail at the end of this chapter in “The Common Controls” section.

Planar Tracker Node [PTra]

The Planar Tracker node

Planar Tracker Node Introduction

The Planar Tracker node is designed to solve a match-moving problem that commonly comes up
during post-production. As an example, live-action footage can often contain a planar surface such as
a license plate or a road sign that needs new numbers in the license plate or a new city’s name on the
road sign. Often, the problem is that the camera is moving in the shot, so the license plate or road sign
is continuously changing perspective. You cannot just merge a new license plate over the existing one
without accounting for the perspective distortions. A time-intensive way to solve this problem would
be to use a Corner Pin node and manually keyframe the four corners. The Planar Tracker automates
this keyframing process and tracks the perspective distortions of a planar surface over time. That
tracking data is then applied with those same perspective distortions to a different foreground.
For more information on using the Planar Tracker, see Chapter 23, “Planar Tracking” in the Fusion
Reference Manual or Chapter 84 in the DaVinci Resolve Reference Manual.

TIP: Part of using a Planar Tracker is also knowing when to give up and fall back to using
Fusion’s Tracker node or to manual keyframing. Some shots are simply not trackable, or the
resulting track suffers from too much jitter or drift. The Planar Tracker is a useful time-saving
node in the artist’s toolbox, but while it may track most shots, it is not a 100% solution.

What the Planar Tracker Saves

While the Planar Tracker does save the resulting final track in the composition on disk, it does not save
temporary tracking information such as the individual point trackers (compared with the Camera
Tracker, which does save the individual point trackers). Some consequences of this include:
– The point trackers no longer appear in the viewer when a comp containing a Planar Tracker
node is saved and reloaded.
– Tracking may not be resumed after a comp containing a Planar Tracker node has been
saved and reloaded. In particular, this also applies to auto saves. For this reason, it is good to
complete all planar tracking within one session.
– The size of composition files is kept reasonable (in some situations, a Planar Tracker can
produce hundreds of megabytes of temporary tracking data).
– Saving and loading of compositions is faster and more interactive.

Chapter 57 Tracker Nodes 1376

Inputs
The Planar Tracker has four inputs:
– Background: The orange background image input accepts a 2D image with the planar surface
to be tracked.
– Corner Pin 1: The green corner pin 1 input accepts a 2D image to be pinned on top of the
background. There may be multiple corner pin inputs, named Corner Pin 1, Corner Pin 2,…etc.
– Occlusion Mask: The white occlusion mask input is used to mask out regions that do not
need to be tracked. Regions where this mask is white will not be tracked. For example, a
person moving in front of and occluding bits of the pattern may be confusing the tracker, and a
quickly-created rough rotomask around the person can be used to tell the tracker to ignore the
masked-out bits.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the output of
the Planar Tracker to certain areas.

Basic Node Setup

A basic Planar Tracker setup consists of just two nodes: a MediaIn connects to the background input
and the Planar Tracker can be used as a separate branch from the rest of the node tree. Once the
tracking is completed, a Planar Transform node should be generated to use the planar tracking data.

A Planar Tracker can be isolated on its own branch of a node tree.

A Typical Planar Tracker Workflow

The following steps outline the workflow with the Planar Tracker:
1 Remove lens distortion: The more lens distortion in the footage, the more the resulting track will
slide and wobble.
2 Connect footage: Connect a Loader or MediaIn node that contains a planar surface to the orange
background input and view the Planar Tracker node in a viewer.
3 Select a reference frame: Move to a frame where the planar surface to be tracked is not occluded
and click the Set button to set this as a reference frame.
4 Choose the pattern: In the viewer, make sure the onscreen controls are visible, and draw a
polygon around the planar surface you want to track. This is called the “pattern.” In most cases,
this will probably be a rectangle, but an arbitrary closed polygon can be used. The pixels enclosed
by this region will serve as the pattern that will be searched for on other frames. Note that it is
important that the pattern is drawn on the reference frame. Do not confuse the pattern with the
region to corner pin (which always has four corners and is separately specified in Corner Pin mode).
5 Adjust render range: In the Keyframes Editor, adjust the render range to match the range of
frames where the planar surface is visible.
6 Adjust track options: Frequently changed options include Tracker, Motion Type, and
Track Channel.

Chapter 57 Tracker Nodes 1377

7 Mask out occluders: If moving objects partially cover up the planar surface, you may wish to
connect an occlusion mask to the Planar Tracker. When using the Hybrid tracker, providing a
mask to deal with occluding objects is strongly recommended, while with the Point tracker it is
recommended to try tracking without a mask.
8 Track: Click the Go button to return to the reference frame. Press the Track To End button and
wait for the track to complete. Click the Go button to return to the reference frame again. Press the
Track To Start button and wait for the track to complete. Note that the tracks in the viewer are not
selectable or deletable as they are in a Camera Tracker.
9 Check track quality: Visually inspect the track to see how accurate it is. Does it stick to the
surface? Switching to Steady mode can help here.
10 Use the track: At this point, in most cases you will create a Planar Transform node from the
Inspector and use it to apply the tracked perspective distortion onto masked images. If the image
you are applying the tracking data to is full frame unmasked clip, you can use the Steady, Corner
Pin, and Stabilize operation modes in in the Planar Tracker.

Inspector

The Planar Tracker Controls tab

Controls Tab
The Controls tab contains controls for determining how the Planar Tracker will be used, setting the
reference frame and initiating the track.

Operation Mode
The Operation Mode menu selects the purpose of the Planar Tracker node. The Planar Tracker has
four modes of operation:
– Track: Used to isolate a planar surface and track its movement over time. Then, you can create a
Planar Transform node that uses this data to match move another clip in various ways.
– Steady: After analyzing a planar surface, this mode removes all motion and distortions from the
planar surface, usually in preparation for some kind of paint or roto task, prior to “unsteadying” the
clip to restore the motion.
– Corner Pin: After analyzing a planar surface, this mode computes and applies a matching
perspective distortion to a foreground image you connect to the foreground input of the Planar
Tracker node, and merges it on top of the tracked footage.
– Stabilize: After analyzing a planar surface, this mode allows smoothing of a clip’s translation,
rotation, and scale over time. This is good for getting unwanted vibrations out of a clip while
retaining the overall camera motion that was intended.

Chapter 57 Tracker Nodes 1378

The last three modes (Steady, Corner Pin, and Stabilize) use the tracking data produced in Track mode.

NOTE: None of the operations can be combined together. For example, both Corner Pin and
Stabilize cannot be done at the same time, nor can a track be done while in corner
pinning mode.

Reference Time
The Reference Time determines the frame where the pattern is outlined. It is also the time from which
tracking begins. The reference frame cannot be changed once it has been set without destroying all
pre-existing tracking information, so scrub through the footage to be tracked and choose carefully.
The reference frame must be chosen carefully to give the best possible quality track.
You choose a reference frame by moving the playhead to an appropriate frame and then clicking the
Set button to choose that frame.

Pattern Polygon
You specify which region of the image you want to track by drawing a polygon on the reference frame.
Typically, when you first add a Planar Tracker node, you are immediately ready to start drawing a
polygon in the viewer, so it’s best to do this right away. When choosing where to draw a polygon, make
sure the region selected belongs to a physically planar surface in the shot. In a pinch, a region that is
only approximately planar can be used, but the less planar the surface, the poorer the quality of the
resulting track.
As a rule of thumb, the more pixels in the pattern, the better the quality of the track. In particular, this
means the reference frame pattern should be:
– As large as possible.
– As much in frame as possible.
– As unoccluded as possible by any moving foreground objects.
– At its maximal size (e.g., when tracking an approaching road sign, it is good to pick a later frame
where it is 400 x 200 pixels big rather than 80 x 40 pixels).
– Relatively undistorted (e.g., when the camera orbits around a flat stop sign, it is better to pick a
frame where the sign is face-on parallel to the camera rather than a frame where it is at a highly
oblique angle).
If the pattern contains too few pixels or not enough trackable features, this can cause problems with
the resulting track, such as jitter, wobble, and slippage. Sometimes dropping down to a simpler motion
type can help in this situation.
After you’ve drawn a pattern, a set of Pattern parameters lets you transform and invert the resulting
polygon, if necessary.

Track Mode
Track mode is unlike the other three options in the Operation menu in that is the only option that
initiates the planar tracking. The other modes use the tracking data generated by the Track mode.

Tracker
There are two available trackers to pick from:
– Point: Tracks points from frame to frame. Internally, this tracker does not actually track points-per-
se but rather small patterns like Fusion’s Tracker node. The point tracker possesses the ability
to automatically create its internal occlusion mask to detect and reject outlier tracks that do not
belong to the dominant motion. Tracks are colored green or red in the viewer, depending on
whether the point tracker thinks they belong to the dominant motion or they have been rejected.
The user can optionally supply an external occlusion mask to further guide the Point tracker.

Chapter 57 Tracker Nodes 1379

 – Hybrid Point/Area: Uses an Area tracker to track all the pixels in the pattern. Unlike the Point
tracker, the Area tracker does not possess the ability to automatically reject parts of the pattern
that do not belong to the dominant motion, so you must manually provide it with an occlusion
mask. Note that for performance reasons, the Hybrid tracker internally first runs the Point tracker,
which is why the point tracks can still be seen in the viewer.
There is no best tracker. They each have their advantages and disadvantages:
– Artist Effort (occlusion masks): The Point tracker will automatically create its internal occlusion
mask. However, with the Hybrid tracker, you need to spend more time manually creating
occlusion masks.
– Accuracy: The Hybrid tracker is more accurate and less prone to wobble, jitter, and drift since it
tracks all the pixels in the pattern rather than a few salient feature points.
– Speed: The Hybrid tracker is slower than the Point tracker.

In general, it is recommended to first quickly track the shot with the Point tracker and examine the
results. If the results are not good enough, then try the Hybrid tracker.

Motion Type
Determines how the Planar Tracker internally models the distortion of the planar surface being tracked.
The five distortion models are:
– Translation.
– Translation, Rotation (rigid motions).
– Translation, Rotation, Scale (takes squares to squares, scale is uniform in x and y).
– Affine includes translation, rotation, scale, skew (maps squares to parallelograms).
– Perspective (maps squares to generic quadrilaterals).

Each successive model is more general and includes all previous models as a special case.
When in doubt, choose Perspective for the initial track attempt. If the footage being tracked has
perspective distortions in it, and the Planar Tracker is forced to work with a simpler motion type, this
can end up causing the track to slide and wobble.
Sometimes with troublesome shots, it can help to drop down to a simpler motion model—for example,
when many track points are clustered on one side of the tracked region or when tracking a small
region where there are not many trackable pixels.

Output
Controls what is output from the Planar Tracker node while in the Track operation mode.
– Background: Outputs the input image unchanged.
– Background - Preprocessed: The Planar Tracker does various types of preprocessing on the input
image (e.g., converting it to luma) before tracking. It can be useful to see this when deciding which
track channel to choose.
– Mask: Outputs the pattern as a black and white mask.
– Mask Over Background: Outputs the pattern mask merged over the background.

Track Channel
Determines which image channel in the background image is tracked. It is good to pick a channel with
high contrast, lots of trackable features, and low noise. Allowed values are red, green, blue, and
luminance.

Chapter 57 Tracker Nodes 1380

Tracking Controls
These controls are used to control the Tracker. Note that while tracking, only track to a new frame if
the current frame is already tracked or it is the reference frame.
– Track to start: Tracks from the current frame backward in time to the start (as determined by the
current render range).
– Step tracker to previous frame: Tracks from the current frame to the previous frame.
– Stop tracking: Stops any ongoing tracking operations.
– Step tracker to next frame: Tracks from the current frame to the next frame.
– Track to end: Tracks from the current frame forward in time to the end (as determined by the
current render range).
– Trim to start: Removes all tracking data before the current frame.
– Delete: Deletes all tracking data at all times. Use this to destroy all current results and start
tracking from scratch.
– Trim to end: Removes all tracking data after the current frame. This can be useful, for example, to
trim the end of a track that has become inaccurate when the pattern starts to move off frame.

Show Splines
This button to the right of the “Trim to end” button opens the Spline Editor and shows the splines
associated with the Planar Tracker node. This can be useful for manually deleting points from the Track
and Stable Track splines.

Right-Click Here for Track Spline

While tracking, a spline containing 4 x 4 matrices at each keypoint is created. This is known as the
“Track spline” or just “Track” for short. These matrices completely describe the distortions of the
tracked pattern.

Create Planar Transform

After tracking footage, this button can be pressed to create a Planar Transform node on the Node
Editor. The current tracking data is embedded in the Planar Transform node so that it can replicate the
planar distortions tracked by the Planar Tracker node. Unless you are compositing a full frame
foreground that matches the same dimensions as the raster, it is best to create a Planar Transform and
use it to apply motion to the foreground.

Steady Mode
In Steady mode, the Planar Tracker transforms the background plate to keep the pattern as motionless
as possible. Any leftover motion is because the Planar Tracker failed to follow the pattern accurately or
because the pattern did not belong to a physically planar surface.
Steady mode is not very useful for actual stabilization, but is useful for checking the quality of a track. If
the track is good, during playback the pattern should not move at all while the rest of the background
plate distorts around it. It can be helpful to zoom in on parts of the pattern and place the mouse cursor
over a feature and see how far that feature drifts away from the mouse cursor over time.

Steady Time
This is the time where the pattern’s position is snapshotted and frozen in place. It is most common to
set this to the reference frame.

Invert Steady Transform

This causes the Planar Tracker node to reverse the effects of the steady transform. This means two
Planar Tracker nodes connected back to back with the second set to invert should give back the
original image. If you place an effects node in between the two, then the effect will be locked in place.

Chapter 57 Tracker Nodes 1381

This should only be used to accomplish effects that cannot be done through corner pinning, since it
involves two resamplings, causing softening of the background image.

Clipping Mode
Determines what happens to the parts of the background image that get moved off frame by the
steady transform:
– Domain: The off frame parts are kept.
– Frame: The off frames parts are thrown away.

Domain mode is useful when Steady mode is being used to “lock” an effect to the pattern.
As an example, consider painting on the license plate of a moving car. One way to do this is to use a
Planar Tracker node to steady the license plate, then a Paint node to paint on the license plate, and
then a second Planar Tracker to undo the steady transform. If the Clipping mode is set to Domain, the
off frame parts generated by the first Planar Tracker are preserved so that the second Planar Tracker
can, in turn, map them back into the frame.

Corner Pin Mode

In Corner Pin mode, one or more textures can be attached to a previously tracked planar surface and
undergo the same perspective distortions as the surface.
The corner pin workflow with Planar Tracker is:
1 Track: select a planar surface in the shot that you wish to attach a texture to or replace the texture
on. Track the shot (see the tracking workflow in the Track section).
2 Switch the Operation Mode to Corner Pin: When Corner Pin mode is entered from Track mode,
the pattern polygon is hidden and a corner pin control is shown in the viewer.
3 Connect in the texture: In the Node Editor, connect the output of the MediaIn node containing the
texture to the Corner Pin 1 input on the Planar Tracker node.
4 Adjust corner pin: Drag the corners of the corner pin in the viewer until the texture is positioned
correctly. Sometimes the Show Grid option is useful when positioning the texture. Additionally, if
it helps to position it more accurately, scrub to other times and make adjustments to the corner pin.
5 Review: Play back the footage and make sure the texture “sticks” to the planar surface.

Merge Mode
Controls how the foreground (the corner pinned texture) is merged over the background (the tracked
footage). If there are multiple corner pins, this option is shared by all of them. There are four options to
pick from:
– BG only
– FG only
– FG over BG
– BG over FG

Number of Corner Pins

Use the + and - buttons to increase or decrease the number of corner pins. Each time an additional
corner pin is created, a corresponding input appears on the node in the Node Editor.

Corner Pin 1 Input Group

Each corner pin has a group of related inputs:
– Enable: Controls the visibility of the corner pin in the viewer.
– Show Grid: Shows a grid over the corner pin. This can be useful when positioning the corners.

Chapter 57 Tracker Nodes 1382

 – Merge Options: Controls merging of corner pin texture over the background. (See the
documentation for the Merge node.)
– Reference Time Positions: The positions of the four corners at the reference time. If the track was
not perfect, these positions can be animated to make adjustments on top of the track.

Stabilize Mode
Stabilize mode is used to smooth out shakiness in the camera by applying a transform that partially
counteracts the camera shake. This stabilizing transform (contained in the Stable Track spline) is
computed by comparing neighboring frames.

NOTE: Stabilize mode only smooths out motions, while Steady mode tries to completely “lock
off” all motion.

Be aware that the Planar Tracker stabilizes based on the motion of the pattern, so it is essential to
choose the pattern carefully. If the motion of the pattern does not represent the motion of the camera,
then there may be unexpected results. For example, if tracking the side of a moving truck and the
camera is moving alongside it, the Planar Tracker smooths the combined motion of both the truck and
the mounted camera. In some cases, this is not the desired effect. It may be better to choose the
pattern to be on some fixed object like the road or the side of a building, which would result in
smoothing only the motion of the camera.
One unavoidable side effect of the stabilization process is that transparent edges appear along the
edges of the image. These edges appear because the stabilizer does not have any information about
what lies off frame, so it cannot fill in the missing bits. The Planar Tracker node offers the option to
either crop or zoom away these edges. When filming, if the need for post-production stabilization is
anticipated, it can sometimes be useful to film at a higher resolution (or lower zoom).

The Planar Tracker Stabilization Workflow

1 Track: Select a roughly planar region that represents the motion that you want to stabilize. Track
the shot (see the tracking workflow in the Track section).
2 Switch the Operation Mode to Stabilize: Until a stabilization is computed, the Planar Tracker will
just output the input footage.
3 Adjust stabilization options: Frequently changed options are Parameters to Smooth and
Smoothing Radius.
4 Compute stabilization: Press the Compute Stabilization button and wait for the stabilization
computations to finish. Play back the output of the Planar Tracker node to see the effects of the
stabilization. Notice that transparent edges have been introduced around the edges of the image
by the stabilization transform.
5 Refine: Adjust the stabilization options and recompute the stabilization as many times as desired.
6 Handle transparent edges (optional): Set the Frame Mode to either Zoom or Crop as desired
and then click the Auto Zoom or Auto Crop button. Play back the footage to observe the effects.
If there is too much zoom or the image has been cropped too small, try reducing the amount of
smoothing.

Parameters to Smooth
Specify which of the following parameters to smooth:
– X Translation
– Y Translation
– Rotation
– Scale

Chapter 57 Tracker Nodes 1383

Smoothing Window
When stabilizing a particular frame, this determines how the contributions of neighboring frames are
weighted. Available choices are Box and Gaussian.

Smoothing Radius (Frames)

Determines the number of frames whose transforms are averaged together to compute the
stabilization. A larger Smoothing Radius results in more smoothing but introduces more
transparent edges.

Compute Stabilization
Clicking this button runs the stabilizer, overwriting the results of any previous stabilization. As soon as
the stabilization is finished, the output of the Planar Tracker node will be immediately updated with the
stabilization applied.

NOTE: The stabilizer uses the Track spline (created by the tracker) to produce the Stable
Track spline. Both of these splines’ keyframes contain 4 x 4 matrices, and the keyframes are
editable in the Spline Editor.

Clipping Mode
Determines what happens to the parts of the background image that get moved off frame by the
stabilization:
– Domain: The off frame parts are kept.
– Frame: The off frames parts are thrown away.

Frame Mode
This controls how transparent edges are handled. The available options include:
– Full: Do nothing. Leaves the transparent edges as is.
– Crop: Crops away the transparent edges. When this option is selected, the size of Planar Tracker’s
output image is smaller than the input image. No image resamplings occur. In Crop mode, use
the Auto Crop button or manually adjust the crop window by changing the X Offset, Y Offset, and
Scale sliders.
– Auto Crop Button: When this button is clicked, the Planar Tracker will examine all the frames
and pick the largest possible crop window that removes all the transparent edges. The
computed crop window will always be centered in frame and pixel aligned. When clicked, Auto
Crop updates the X/Y Offset and Scale sliders.

– Zoom: Scales the image bigger until the transparent edges are off frame. Choosing this option
causes an image resampling to occur. The downside of this approach is that it reduces the quality
(slightly softens) of the output image. In Zoom mode, use the Auto Zoom button or manually adjust
the zoom window by changing the X Offset, Y Offset, and Scale sliders.
– Auto Zoom: When this button is clicked, the Planar Tracker will examine all the frames and pick
the smallest possible zoom factor that removes all the transparent edges. The computed zoom
window will always be centered in frame. When clicked, Auto Zoom updates the X/Y Offset and
Scale sliders.

Chapter 57 Tracker Nodes 1384

Right-Click Here for Stable Track Spline
Right-clicking over this label provides access to a spline whose keyframes contain 4 x 4 matrices
which in turn represent the stabilization transforms. This is mostly here for completeness and for
advanced users.

Options Tab
These controls affect the look of onscreen controls in the viewer.

The Planar Tracker Options tab

Darken Image
Darkens the image while in Track mode in order to better see the controls and tracks in the viewer.
The Shift+D keyboard shortcut toggles this.

Show Track Markers

Toggles the display of the dots marking the location of trackers at the current time.

Show Trails
Toggles the display of the trails following the location of trackers.

Trail Length
Allows changing the length of tracker trails. If the pattern is moving very slowly, increasing the length
can sometimes make the trails easier to follow in the viewer. If the pattern is moving very fast, the
tracks can look like spaghetti in the viewer. Decreasing the length can help.

Inlier/Outlier Colors
When tracking, the tracker analyzes the frame and detects which of the multitudinous tracks belong to
the dominant motion and which ones represent anomalous, unexplainable motion. By default, tracks
belonging to the dominant motion are colored green (and are called inliers) and those that do not
belong are colored red (and are called outliers). Only the inlier tracks are used when computing the
final resulting track.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Tracking nodes. These common controls
are described in detail at the end of this chapter in “The Common Controls” section.

Chapter 57 Tracker Nodes 1385

Planar Transform Node [PXF]

The Planar Transform node

The Planar Transform node applies perspective distortions generated by a Planar Tracker node onto
any input mask or masked image. The Planar Transform node can be used to reduce the amount of
time spent on rotoscoping objects. The workflow here centers around the notion that the Planar
Tracker node can be used to track objects that are only roughly planar. After an object is tracked, a
Planar Transform node can then be used to warp a rotospline, making it approximately follow the
object over time. Fine-level cleanup work on the rotospline then must be done.
Depending on how well the Planar Tracker followed the object, this can result in substantial time
savings in the amount of tedious rotoscoping. The key to using this technique is recognizing situations
where the Planar Tracker performs well on an object that needs to be rotoscoped.
A rough outline of the workflow involved is:
1 Track: Using a Planar Tracker node, select a pattern that represents the object to be rotoscoped.
Track the shot (see the tracking workflow in the Track section for the Planar Tracker node).
2 Create a Planar Transform node: Press the Create Planar Transform button on the Planar Tracker
node to do this. The newly created Planar Transform node can be freely cut and pasted into
another composition as desired.
3 Rotoscope the object: Move to any frame that was tracked by the Planar Tracker. When unsure if a
frame was tracked, look in the Spline Editor for a tracking keyframe on the Planar Transform node.
Connect a Polygon node into the Planar Transform node. While viewing the Planar Transform
node, rotoscope the object.
4 Refine: Scrub the timeline to see how well the polygon follows the object. Adjust the polyline on
frames where it is off. It is possible to add new points to further refine the polygon.

Inputs
The Planar Transform has only two inputs:
– Image Input: The orange image input accepts a 2D image on which the transform will be
applied.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the output of
the Planar Transform to certain areas.

Basic Node Setup

The example below uses a Planar Transform node between a masked MediaIn2 node and the
foreground input to a Merge node. The background MediaIn1 node connects to the Planar Tracker,
which was used to generate the Planar Transform. Once the Planar Transform is created, the Planar
Tracker is no longer needed in the node tree.

Chapter 57 Tracker Nodes 1386

A Planar Transform creating a match move

Inspector

The Planar Transform Controls tab

Controls Tab
The Planar Transform node has very few controls, and they are all located in the Controls tab. It’s
designed to apply the analyzed Planar Tracking data as a match move,

Reference Time
This is the reference time that the pattern was taken from in the Planar Tracker node used to produce
the Planar Transform.

Right-Click Here for Track Spline

The Track spline contains information about the perspective distortions stored in 4 x 4 matrices. When
a Planar Transform node is exported from a Planar Tracker node, the Track spline produced by the
Planar Tracker is shared by connecting it with the Planar Transform node. A consequence of this
sharing of the Track spline is that if the track is changed in the Planar Tracker node, the Planar
Transform will be automatically updated. Note that this spline can be examined in the Spline Editor,
which is useful for seeing the extent of tracked frames.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Tracking nodes. These common controls
are described in detail at the end of this chapter in “The Common Controls” section.

Chapter 57 Tracker Nodes 1387

Camera Tracker [CTra]

The Camera Tracker node

Camera Tracker Introduction

Camera tracking is match moving, and a vital link between 2D and 3D, allowing compositors to
integrate 3D renders into live-action scenes. The Camera Tracker node is used to calculate the path of
a live-action camera and generate a virtual camera in 3D space. This virtual camera’s motion is
intended to be identical to the motion of the actual camera that shot the scene. Using the calculated
position and movement of the virtual camera provides the flexibility to add 3D elements to a live-
action scene. Also, the Camera Tracker creates a point cloud in 3D space that can be used to align
objects and other 3D models to the live-action scene.
For more information about how to use the Camera Tracker, see Chapter 26, “3D Camera Tracking” in
the Fusion Reference Manual or Chapter 87, in the DaVinci Resolve Reference Manual.

The original video clip

3D objects composited on to the video clip that use the camera tracker to remain
aligned with the objects in the frame as the image moves

Chapter 57 Tracker Nodes 1388

Inputs
The Camera Tracker has two inputs:
– Background: The orange image input accepts a 2D image you want tracked.
– Occlusion Mask: The white occlusion mask input is used to mask out regions that do not need
to be tracked. Regions where this mask is white will not be tracked. For example, a person
moving in front of and occluding bits of the scene may be confusing to the tracker, and a
quickly-created rough mask around the person can be used to tell the tracker to ignore the
masked-out bits.

Basic Node Setup

The Camera Tracker background input is used to connect the image you want tracked. Polygon masks
can be connected into the occlusion mask input to identify areas the tracker should ignore.

The Camera Tracker with occlusion masks

Inspector

The Camera Tracker tab

Track Tab
The Track tab contains the controls you need to set up an initial analysis of the scene.

Auto Track
Automatically detects trackable features and tracks them through the source footage. Tracks will be
automatically terminated when the track error becomes too high, and new tracks are created as
needed. The values of the Detection Threshold and Minimum Feature Separation sliders can be used
to control the number and distribution of auto tracks.

Chapter 57 Tracker Nodes 1389

Reset
Deletes all the data internal to the Camera Tracker node, including the tracking data and the solve
data (camera motion path and point cloud). To delete only the solve data, use the Delete button on the
Solve tab.

Preview AutoTrack Locations

Turning this checkbox on will show where the auto tracks will be distributed within the shot. This is
helpful for determining if the Detection Threshold and Minimum Feature Separation need to be
adjusted to get an even spread of trackers.

Detection Threshold
Determines the sensitivity to detect features. Automatically generated tracks will be assigned to the
shot and the Detection Threshold will force them to be either in locations of high contrast or
low contrast.

Minimum Feature Separation

Determines the spacing between the automatically generated tracking points. Decreasing this slider
causes more auto tracks to be generated. Keep in mind that a large number of tracking points will also
result in a lengthier solve.

Track Channel
Used to nominate a color channel to track: red, green, blue, or luminance. When nominating a channel,
choose one that has a high level of contrast and detail.

Track Range
Used to determine which frames are tracked:
– Global: The global range, which is the full duration of the Timeline.
– Render: The render duration set on the Timeline.
– Valid: The valid range is the duration of the source media.
– Custom: A user determined range. When this is selected, a separate range slider appears to set
the start and end of the track range.

Bidirectional Tracking
Enabling this will force the tracker to track backward after the initial forward tracking. When tracking
backward, new tracks are not started but rather existing tracks are extended backward in time. It is
recommended to leave this option on, as long tracks help give better solved cameras and
point clouds.

Gutter Size
Trackers can become unstable when they get close to the edge of the image and either drift or jitter or
completely lose their pattern. The Camera Tracker will automatically terminate any tracks that enter the
gutter region. Gutter size is given as a percentage of pattern size. By default, it’s 100% of pattern size,
so a 0.04 pattern means a 0.04 gutter.

New Track Defaults

There are three methods in which the Camera Tracker node can analyze the scene, and each has its
own strengths when dealing with certain types of camera movement.
– Tracker: Internally, all the Trackers use the Optical Flow Tracker to follow features over time and
then further refine the tracks with the trusted Fusion Tracker or Planar Tracker. The Planar Tracker
method allows the pattern to warp over time by various types of transforms to find the best fit.
These transforms are:
– Translation
– Translation and Rotation Translation, Rotation, and Scale Affine
– Perspective
Chapter 57 Tracker Nodes 1390
 It is recommended to use the default TRS setting when using the Planar Tracker. The Affine and
Perspective settings need large patterns in order to track accurately.
– Close Tracks When Track Error Exceeds: Tracks will be automatically terminated when the
tracking error gets too high. When tracking a feature, a snapshot of the pixels around a feature are
taken at the reference time of the track. This is called a pattern, and that same pattern of pixels is
searched for at future times. The difference between the current time pattern and the reference
time pattern is called the track error. Setting this option higher produces longer but increasingly
less accurate tracks.
– Solve Weight: By default, each track is weighted evenly in the solve process. Increasing a track’s
weight means it has a stronger effect on the solved camera path. This is an advanced option that
should be rarely changed.

Auto Track Defaults

Set a custom prefix name and/or color for the automatically generated tracks. This custom color will be
visible when Track Colors in the Options tab is set to User Assigned.

Camera Tab

The Camera Tracker Camera tab

The controls of the Camera tab let you specify the physical aspects of the live-action camera, which
will be used as a starting point when searching for solve parameters that match the real-world camera.
The more accurate the information provided in this section, the more accurate the camera solve.
The Camera tab includes controls relating to the lens and gate aspects of the camera being solved for.

Focal Length
Specify the known constant focal length used to shoot the scene or provide a guess if the Refine
Focal Length option is activated in the Solve tab.

Film Gate
Choose a film gate preset from the drop-down menu or manually enter the film back size in the
Aperture Width and Aperture Height inputs. Note that these values are in inches.

Aperture Width
In the event that the camera used to shoot the scene is not in the preset drop-down menu, manually
enter the aperture width (inches).

Aperture Height
In the event that the camera used to shoot the scene is not in the preset drop-down menu, manually
enter the aperture height (inches).

Chapter 57 Tracker Nodes 1391

Resolution Gate Fit
This defines how the image fits the sensor size. Often, film sensors are sized to cover a number of
formats, and only a portion of the sensor area is recorded into an image.
For example, a 16:9 image is saved out of a full aperture-sized sensor.
Typically, fit to Width or Height is the best setting. The other fit modes are Inside, Outside, or
Stretched.

Center Point
This is where the camera lens is aligned to the camera. The default is (0.5, 0.5), which is the middle of
the sensor.

Use Source Pixel Aspect

This will use the squeeze aspect of the pixels that is loaded in the image. HD is square pixels, but
NTSC has a pixel aspect ratio of 0.9:1, and Anamorphic CinemaScope is 2:1 aspect. Disabling this
option exposes Pixel X and Y number fields where you can customize the source pixel aspect.

Auto Camera Planes

When enabled, the camera’s image plane and far plane are automatically moved to enclose the point
cloud whenever a solve completes. Sometimes, though, the solver can atypically fling points off really
deep into the scene, consequently pushing the image plane very far out. This makes the resulting
scene unwieldy to work with in the 3D views. In these cases, disable this option to override this default
behavior (or delete the offending tracks).

Solve Tab

The Camera Tracker Solve tab

Chapter 57 Tracker Nodes 1392

SolveTab
The Solve tab is where the tracking data is used to reconstruct the camera’s motion path along with
the point cloud. It is also where cleanup of bad or false tracks is done, and other operations on the
tracks can be performed, such as defining which marks are exported in the Point Cloud 3D. The
markers can also have their weight set to affect the solve calculations.
For example, a good camera solve may have already been generated, but there are not enough
locators in the point cloud in an area where an object needs to be placed, so adding more tracks and
setting their Solve Weight to zero will not affect the solved camera but will give more points in the
point cloud.

Solve
Pressing Solve will launch the solver, which uses the tracking information and the camera
specifications to generate a virtual camera path and point cloud, approximating the motion of the
physical camera in the live-action footage. The console will automatically open, displaying the
progress of the solver.

Delete
Delete will remove any solved information, such as the camera and the point cloud, but will keep all
the tracking data.

Average Solve Error

Once the camera has been solved, a summary of the solve calculation is displayed at the top of the
Inspector. Chief among those details is the Average Solve Error. This number is a good indicator of
whether the camera solve was successful. It can be thought of as the difference (measured in pixels)
between tracks in the 2D image and the reconstructed 3D locators reprojected back onto the image
through the reconstructed camera. Ultimately, in trying to achieve a low solve error, any value less than
1.0 pixels will generally result in a good track. A value between 0.6 and 0.8 pixels is considered
excellent.

Clean Tracks by Filter

Clicking this button selects tracks based on the Track Filtering options. If the Auto Delete Tracks
By Filter checkbox is activated, the selected tracks will be deleted as well.

Clean Foreground Tracks

Clicking this button makes a selection of the tracks on fast-moving objects that would otherwise cause
a high solve error. The selection is determined by the Foreground Threshold slider.

Foreground Threshold
This slider sets the detection threshold for finding the tracks on fast-moving objects. The higher the
value, the more forgiving.

Auto Delete Tracks by Filter

With this checkbox enabled, tracks that are selected by the Clean Tracks By Filter button will be
deleted. Enable the checkbox, and then press Clean Tracks By Filter. Any track that meets the filtering
options is then selected and deleted.

Auto Delete Foreground Tracks

With this checkbox enabled, tracks that are selected by the Clean Foreground Tracks button will be
deleted. Enable the checkbox, and then press Clean Foreground Tracks. Any track that meets the
foreground threshold criteria is deleted.

Accept Solve Error

This slider sets an acceptable maximum threshold level for the solve error. If the solve error is greater
than this value, the Camera Tracker will sweep the focal length setting in an attempt to bring the solve
error under the Accept Solve Error value. If the solver cannot find a solution, the Camera Tracker will

Chapter 57 Tracker Nodes 1393

display a message in the console that the solver failed. If a solution cannot be found, ideally you
should try to input the correct focal length or alternatively manually clean some noisy tracks
then re-solve.

Auto Select Seed Frames

With this enabled, the Camera Tracker nominates two frames that will be used as a reference for
initiating the solve. These two frames are initially solved for and a camera is reconstructed, and then
gradually more frames are added in, and the solution is “grown” outward from the seed frames. The
choice of seed frames strongly affects the entire solve and can easily cause the solve to fail. Seed
frames can be found automatically or defined manually.
Disabling this will allow the user to select their own two frames. Manual choice of seed frames is an
option for advanced users. When choosing seed frames, it is important to satisfy two conflicting
desires: the seed frames should have lots of tracks in common yet be far apart in perspective (i.e., the
baseline distance between the two associated cameras is long).

Refine Focal Length

Enabling this will allow the solver to adjust the focal length of the lens to match the tracking points. You
can prevent the focal length being adjusted by setting the Focal Length parameter in the Camera Tab.

Enable Lens Parameter

When enabled, lens distortion parameters are exposed to help in correcting lens distortion
when solving.
– Refine Center Point: Normally disabled, camera lenses are normally centered in the middle
of the film gate but this may differ on some cameras For example, a cine camera may be set
up for Academy 1.85, which has a sound stripe on the left, and shooting super35, the lens is
offset to the right.
– Refine Lens Parameters: This will refine the lens distortion or curvature of the lens. There tends to
be larger distortion on wide angle cameras

NOTE: When solving for the camera’s motion path, a simulated lens is internally created to
model lens distortion in the source footage. This simulated lens model is much simpler than
real-world lenses but captures the lens distortion characteristics important for getting an
accurate camera solve. Two types of distortion are modeled by Camera Tracker:
Radial Distortion: The strength of this type of distortion varies depending on the distance
from the center of the lens. Examples of this include pincushion, barrel, and mustache
distortion. Larger values correspond to larger lens curvatures. Modeling radial distortion is
especially important for wide angle lenses and fisheye lenses (which will have a lot of
distortion because they capture 180 degrees of an environment and then optically squeeze it
onto a flat rectangular sensor).
Tangential Distortion: This kind of distortion is produced when the camera’s imaging sensor
and physical lens are not parallel to each other. It tends to produce skew distortions in the
footage similar to distortions that can be produced by dragging the corners of a corner pin
within Fusion. This kind of distortion occurs in very cheap consumer cameras and is
practically non-existent in film cameras, DSLRs, and pretty much any kind of camera used in
film or broadcast. It is recommended that it be left disabled.

Enable Lens Parameters

When disabled, the Camera Tracker does not do any lens curvature simulations. This is the default
setting and should remain disabled if there is a very low distortion lens or the lens distortion has
already been removed from the source clip. Activating the Enable Lens Parameters checkbox

Chapter 57 Tracker Nodes 1394

determines which lens parameters will be modeled and solved for. Parameters that are not enabled
will be left at their default values. The following options are available:
– Radial Quadratic: Model only Quadratic radial lens curvature, which is either barrel or pincushion
distortion. This is the most common type of distortion. Selecting this option causes the low and
high order distortion values to be solved for.
– Radial Quartic: Model only Quartic radial lens curvature, which combines barrel and pincushion
distortion. This causes the low and high order distortion values to be solved for.
– Radial & Tangential: Model and solve for both radial and tangential distortion. Tangential relates to
misaligned elements in a lens.
– Division Quadratic: Provides a more accurate simulation of Quadratic radial lens curvature. This
causes the low and high order distortion values to be solved for.
– Division Quartic: Provides a more accurate simulation of Quartic radial lens curvature. This causes
the low and high order distortion values to be solved for.
– Lower Order Radial Distortion: This slider is available for all simulations. It determines the
quadratic lens curvature.
– Higher Order Radial Distortion: This slider is available only for Quartic simulations. Determines
the quartic lens curvature.
– Tangential Distortion X/Y: These sliders are available only for Tangential simulations.
Determines skew distortion.

Track Filtering
The Camera Tracker can produce a large number of automatically generated tracks. Rather than
spending a lot of time individually examining the quality of each track, it is useful to have some less
time-intensive ways to filter out large swaths of potentially bad tracks. The following input sliders are
useful for selecting large amounts of tracks based on certain quality metrics, and then a number of
different possible operations can be made on them. For example, weaker tracks can selected and
deleted, yielding a stronger set of tracks to solve from. Each filter can be individually enabled
or disabled.

Minimum Track Length (Number of Markers)

Selects tracks that have a duration shorter than the slider’s value. Short tracks usually don’t get a
chance to move very far and thus provide less perspective information to the solver than a longer
track, yet both short and long tracks are weighted evenly in the solve process, making long tracks
more valuable to the solver. Locators corresponding to shorter tracks are also less accurately
positioned in 3D space than those corresponding to longer tracks. If the shot has a lot of long tracks, it
can be helpful to delete the short tracks. For typical shots, using a value in the range of 5 to 10 is
suggested. If there are not a lot of long tracks (e.g., the camera is quickly rotating, causing tracks to
start and move off frame quickly), using a value closer to 3 is recommended.

Maximum Track Error

Selects tracks that have an average track error greater than the slider’s value. When tracking, tracks
are automatically terminated when their track error exceeds some threshold. This auto termination
controls the maximum track error, while this slider controls the average track error. For example, tracks
following the foliage in a tree tend to be inaccurate and sometimes may be detected by their high
average error.

Maximum Solve Error

Selects tracks that have a solve error greater than the slider’s value. One of the easiest ways to
increase the accuracy of a camera solve is to select the 20% of the tracks with the highest solve error
and simply delete them (although this can sometimes make things worse).

Chapter 57 Tracker Nodes 1395

Select Tracks Satisfying Filters
Selects the tracks within the scene that meet the above Track Filtering values. Note that when this
button is pressed, the tracks that satisfy the filter values are displayed in the Selected Tracks area of
the Solve tab and are colored in the viewer. This button is useful when Auto Select Tracks While
Dragging Sliders is turned off or if the selection, for example, was accidentally lost by mis-clicking in
the viewer.

Auto Select Tracks While Dragging Sliders

When this is ticked on, dragging the above sliders (Minimum Track Length, Maximum Track Error,
Maximum Solve Error) will cause the corresponding tracks to be interactively selected in the viewer.

Operations on Selected Tracks

Tracks selected directly in the viewer with the mouse or selected via track filtering can have the
following operations applied:

Delete Will remove the tracks from the set. When there are bad tracks, the simplest and
easiest option is to simply delete them.

Trim Previous Will cut the tracked frames from the current frame to the start of the track.
Sometimes it can be more useful to trim a track than deleting it. For example,
high quality long tracks that become inaccurate when the feature they are
tracking starts to become occluded or when the tracked feature moves too close
to the edge of the image.

Trim Next Will cut the tracked frames from the current frame to the end of the track.

Rename Will replace the current auto generated name with a new name.

Set Color Will allow for user assigned color of the tracking points.

Export Flag This controls whether the locators corresponding to the selected tracks will be
exported in the point cloud. By default all locators flagged as exportable.

Solve Weight By default, all the tracks are used and equally weighted when solving for the
camera’s motion path. The most common use of this option is to set a track’s
weight to zero so it does not influence the camera’s motion path but is still
has a reconstructed 3D locator. Setting a tracks’ weight to values other than
1.0 or 0.0 should only be done by advanced users.
Onscreen display of track names and values are controlled by
these functions:”

None Will clear/hide the selected tracks.

Toggle Will swap the selected tracks and unselect sets.

All Will select all tracks.

Show Names Will display the track name, by default these are a number.

Show Frame Range Will display the start and end frame of a track.

Show Solve Error Will display the amount of solve error each selected track has.

Chapter 57 Tracker Nodes 1396

Selected Tracks
This area displays the properties of a track point or group of points. It has options to:
– Clear: Deselects all tracks and clears this area.
– Invert: Deselects the current selected tracks and selects the other tracks.
– Visible: Selects all the trackers at the current frame.
– All: Selects all trackers on all frames.
– Search: Selects tracks whose names contain a substring.

TIP: Also select tracks directly in the 2D viewer using the mouse or in the 3D viewer by
selecting their corresponding locators in the point cloud.

Export Tab
The Export tab lets you turn the tracked and solved data this node has generated into a form that can
be used for compositing.

The Camera Tracker Export tab

Chapter 57 Tracker Nodes 1397

Export
The Export button will create a basic setup that can be used for 3D match moving:
– A Camera 3D with animated translation and rotation that matches the motion of the live-action
camera and an attached image plane.
– A Point Cloud 3D containing the reconstructed 3D positions of the tracks.
– A Shape 3D set to generate a ground plane.
– A Merge 3D merging together the camera, point cloud, and ground plane. When the Merge 3D is
viewed through the camera in a 3D viewer, the 3D locators should follow the tracked footage.
– A Renderer 3D set to match the input footage.

The export of individual nodes can be enabled/disabled in the Export Options tab.

Update Previous Export

When this button is clicked, the previously exported nodes are updated with any new data generated.
These previously exported nodes are remembered in the Previous Export section at the bottom of this
section. Here’s an example of how this is handy:
1 Solve the camera and export.
2 Construct a complex Node Editor based around the exported nodes for use in set extension.
3 The camera is not as accurate as preferred or perhaps the solver is rerun to add additional tracks
to generate a denser point cloud. Rather than re-exporting the Camera 3D and Point Cloud 3D
nodes and connecting them back in, just press the Update Previous Export button to “overwrite”
the existing nodes in place.

Automatically Update Previous Export After Solves

This will cause the already exported nodes (Camera 3D, Point Cloud 3D, Lens Distort, Renderer 3D,
and the ground plane) to auto update on each solve.

3D Scene Transform
Although the camera is solved, it has no idea where the ground plane or center of the scene is
located. By default, the solver will always place the camera in Fusion’s 3D virtual environment so that
on the first frame it is located at the origin (0, 0, 0) and is looking down the -Z axis. You have the choice
to export this raw scene without giving the Camera Tracker any more information, or you can set the
ground plane and origin to simplify your job when you begin working in the 3D scene. The 3D Scene
Transform controls provide a mechanism to correctly orient the physical ground plane in the footage
with the virtual ground plane in the 3D viewer. Adjusting the 3D Scene Transform does not modify the
camera solve but simply repositions the 3D scene to best represent the position of the live-
action camera.

NOTE: If you export the scene and then make changes in the 3D Scene Transform, it is
important to manually click Update Previous Export to see the results in the exported nodes.

Aligned/Unaligned
The Aligned/Unaligned menu locks or unlocks the origin and ground plane settings. When set to
Unaligned, you can select the ground plane and origin either manually or by selecting locators in the
viewer. When in unaligned mode, a 3D Transform control in the 3D viewer can be manually
manipulated to adjust the origin.
Once alignment of the ground plane and origin has been completed, the section is locked by
switching the menu to Aligned.

Chapter 57 Tracker Nodes 1398

Set from Selection
When set to unaligned, buttons labeled Set from Selection are displayed under the Origin, Orientation,
and Scale sections. Clicking these buttons takes the selecting locators in the viewer and aligns the
ground plane or origin based on the selection.
For instance, to set the ground plane, do the following:
1 After solving, set the 3D Scene Transform menu to Unaligned.
2 Find a frame where the ground plane is at its largest and clearest point.
3 In the viewer, drag a selection rectangle around all the ground plane locators.
4 Hold Shift and drag again to add to the selection.
5 In the Orientation section, make sure the Selection Is menu correctly matches the orientation of
the selected locators.
6 Click the Set from Selection button located under the Orientation parameters.
7 Set the 3D Scene Transform menu back to Aligned.

To get the best result when setting the ground plane, try to select as many points as possible
belonging to the ground and having a wide separation.

TIP: When selecting points for the ground plane, it is helpful to have the Camera Tracker
node viewed in side-by-side 2D and 3D views. It may be easier to select tracks belonging to
the ground by selecting tracks from multiple frames in the 2D viewer rather than trying to box
select locators in the 3D viewer.

Setting the origin can help you place 3D objects in the scene with more precision. To set the origin,
you can follow similar steps, but only one locator is required for the origin to be set. When selecting a
locator for the origin, select one that has a very low solve error.

Ground Plane Options

These controls let you adjust the ground plane for the scene, which is a crucial step in making sure the
composite looks correct.

Color Will set the color of the ground plane.

Size Controls how big the ground plane can be set.

Subdivision Level Shows how many polygons are in the ground plane.

Wireframe Sets whether the ground plane is set as wireframe or solid surface when
displayed in 3D.

Line Thickness Adjusts how wide the lines will draw in the view.

Offset By default, the center of the ground plane is placed at the origin (0, 0, 0).
This can be used to shift the ground plane up and down along the y-axis.

Export Options
Provides a checkbox list of what will be exported as nodes when the Export button is pressed. These
options are Camera, Point Cloud, Ground Plane, Renderer, Lens Distortion, and Enable Image Plane in
the camera.
The Animation menu allows you to choose between animating the camera and animating the point
cloud. Animating the camera leaves the point cloud in a locked position while the camera is keyframed
to match the live-action shot. Animating the point cloud does the opposite. The camera is locked in
position while the entire point cloud is keyframed to match the live-action shot.

Chapter 57 Tracker Nodes 1399

Previous Export
When the Update Previous Export button is clicked, the previously exported nodes listed here are
updated with any new data generated (this includes the camera path and attributes, the point cloud,
and the renderer).

Options Tab
The Options tab lets you customize the Camera Tracker’s onscreen controls so you can work most
effectively with the scene material you have.

The Camera Tracker Options tab

Trail Length
Displays trail lines of the tracks overlaid on the viewer. The amount of frames forward and back from
the current frame is set by length.

Location Size
In the 3D viewer, the point cloud locators can be sized by this control.
Track Colors, Locator Colors, and Export Colors each have options for setting their color to
one of the following:
– User Assigned
– Solve Error
– Take From Image
– White

Track Colors Onscreen tracks in the 2D view.

Locator Colors Point Cloud locators in the 3D view.

Export Colors Colors of the locators that get exported within the Point Cloud node.

Chapter 57 Tracker Nodes 1400

Darken Image
Dims the brightness of the image in viewers to better see the overlaid tracks. This affects both the 2D
and 3D viewers.

Visibility
Toggles which overlays will be displayed in the 2D and 3D viewers. The options are Tracker Markers,
Trails, Tooltips in the 2D Viewer, Tooltips in the 3D viewer, Reprojected Locators, and Tracker Patterns.

Colors
Sets the color of the overlays.
– Selection Color: Controls the color of selected tracks/locators.
– Preview New Tracks Color: Controls the color of the points displayed in the viewer when the
Preview AutoTrack Locations option is enabled.
– Solve Error Gradient: By default, tracks and locators are colored by a green-yellow-red gradient
to indicate their solve error. This gradient is completely user adjustable.

Reporting
Outputs various parameters and information to the Console.

Understanding Camera Tracking

On large productions, camera tracking or 3D match moving is often handed over to experts who have
experience with the process of tracking and solving difficult shots. There is rarely a shot where you
can press a couple of buttons and have it work perfectly. It does take an understanding of the whole
process and what is essential to get a good solved track.
The Camera Tracker must solve for hundreds of thousands of unknown variables, which is a complex
task. For the process to work, it is essential to get good tracking data that exists in the shot for a long
time. False or bad tracks will skew the result. This section explains how to clean up false tracks and
other techniques to get a good solve.

Workflow
Getting a good solve is a repeated process.
Track > Solve > Refine Filters > Solve > Cleanup tracks > Solve > Cleanup from point cloud >
Solve > Repeat.
Initially, there are numerous tracks, and not all are good, so a process of filtering and cleaning up
unwanted tracks to get to the best set is required. At the end of each cleanup stage, pressing Solve
ideally gives you a progressively lower solve error. This needs to be below 1.0 for it to be good for use
with full HD content, and even lower for higher resolutions. Refining the tracks often but not always
results in a better solve.

False Tracks
False tracks are caused by a number of conditions, such as moving objects in a shot, or reflections
and highlights from a car. There are other types of false tracks like parallax errors where two objects
are at different depths, and the intersection gets tracked. These moiré effects can cause the track to
creep. Recognizing these False tracks and eliminating them is the most important step in the
solve process.

Chapter 57 Tracker Nodes 1401

Track Lengths
Getting a good set of long tracks is essential; the longer the tracks are, the better the solve. The
Bi-Directional tracking option in the Tracker tab is used to extend the beginning of tracks in time. The
longer in time a track exists and the more tracks that overlap in time of a shot, the more consistent and
accurate the solve.

Seed Frames
Two seed frames are used in the solve process. The algorithm chooses two frames that are as far
apart in time yet share the same tracks. That is why longer tracks make a more significant difference in
the selection of seed frames.
The two Seed frames are used as the reference frames, which should be from different angles of the
same scene. The solve process will use these as a master starting point to fit the rest of the tracks in
the sequence.
There is an option in the Solve tab to Auto Detect Seed Frames, which is the default setting and most
often a good idea. However, auto detecting seed frames can make for a longer solve. When refining
the Trackers and re-solving, disable the checkbox and use the Seed 1 and Seed 2 sliders to enter the
previous solve’s seed frames. These seed frames can be found in the Solve Summary at the top of the
Inspector after the initial solve.

Refine Filters
After the first solve, all the Trackers will have extra data generated. These are solve errors and
tracking errors.
Use the refine filters to reduce unwanted tracks, like setting minimum tracker length to eight frames.
As the value for each filter is adjusted, the Solve dialog will indicate how many tracks are affected by
the filter. Then Solve again.

Onscreen Culling
Under the Options tab, set the track to 20. This will display each track on footage with +-20 frame trail.
When scrubbing/playing through the footage, false tracks can be seen and selected onscreen, and
deleted by pressing the Delete key. This process takes an experienced eye to spot tracks that go bad.
Then Solve again.
You can view the exported scene in a 3D perspective viewer. The point cloud will be visible. Move and
pan around the point cloud, select and delete points that seem to have no inline with the image and
the scene space. Then Solve again.
Repeat the process until the solve error is below 1.0 before exporting.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Tracking nodes. These common controls
are described in detail in the following “The Common Controls” section.

Chapter 57 Tracker Nodes 1402

The Common Controls
Nodes that handle tracking operations share several identical controls in the Inspector. This section
describes controls that are common among Tracking nodes.

Inspector

The Common Tracking Controls Settings tab

Settings Tab
The Settings tab in the Inspector can be found on every tool in the Tracking category. The controls are
consistent and work the same way for each tool.

Blend
The Blend control is used to blend between the tool’s original image input and the tool’s final modified
output image. When the blend value is 0.0, the outgoing image is identical to the incoming image.
Normally, this will cause the tool to skip processing entirely, copying the input straight to the output.

Process When Blend Is 0.0

The tool is processed even when the input value is zero. This can be useful if the node is scripted to
trigger a task, but the node’s value is set to 0.0.

Red/Green/Blue/Alpha Channel Selector

These four buttons are used to limit the effect of the tool to specified color channels. This filter is often
applied after the tool has been processed.

Chapter 57 Tracker Nodes 1403

For example, if the Red button on a Blur tool is deselected, the blur will first be applied to the image,
and then the red channel from the original input will be copied back over the red channel of the result.
There are some exceptions, such as tools for which deselecting these channels causes the tool to skip
processing that channel entirely. Tools that do this will generally possess a set of identical RGBA
buttons on the Controls tab in the tool. In this case, the buttons in the Settings and the Controls tabs
are identical.

Apply Mask Inverted

Enabling the Apply Mask Inverted option inverts the complete mask channel for the tool. The mask
channel is the combined result of all masks connected to or generated in a node.

Multiply by Mask
Selecting this option will cause the RGB values of the masked image to be multiplied by the mask
channel’s values. This will cause all pixels of the image not included in the mask (i.e., set to 0) to
become black/transparent.

Use Object/Use Material (Checkboxes)

Some 3D software can render to file formats that support additional channels. Notably, the EXR file
format supports Object ID and Material ID channels, which can be used as a mask for the effect. These
checkboxes determine whether the channels will be used, if present. The specific Material ID or
Object ID affected is chosen using the next set of controls.

Correct Edges
This checkbox appears only when the Use Object or Use Material checkboxes are selected. It toggles
the method used to deal with overlapping edges of objects in a multi-object image. When enabled,
the Coverage and Background Color channels are used to separate and improve the effect around the
edge of the object. If this option is disabled (or no Coverage or Background Color channels are
available), aliasing may occur on the edge of the mask.
For more information on the Coverage and Background Color channels, see Chapter 18,
“Understanding Image Channels” in the Fusion Studio Reference Manual or Chapter 79 in the
DaVinci Resolve Reference Manual.

Object ID/Material ID (Sliders)

Use these sliders to select which ID will be used to create a mask from the object or material channels
of an image. Use the Sample button in the same way as the Color Picker: to grab IDs from the image
displayed in the viewer. The image or sequence must have been rendered from a 3D software
package with those channels included.

Motion Blur
– Motion Blur: This toggles the rendering of Motion Blur on the tool. When this control is toggled
on, the tool’s predicted motion is used to produce the motion blur caused by the virtual camera’s
shutter. When the control is toggled off, no motion blur is created.
– Quality: Quality determines the number of samples used to create the blur. A quality setting of 2
will cause Fusion to create two samples to either side of an object’s actual motion. Larger values
produce smoother results but increase the render time.
– Shutter Angle: Shutter Angle controls the angle of the virtual shutter used to produce the motion
blur effect. Larger angles create more blur but increase the render times. A value of 360 is the
equivalent of having the shutter open for one full frame exposure. Higher values are possible and
can be used to create interesting effects.
– Center Bias: Center Bias modifies the position of the center of the motion blur. This allows for the
creation of motion trail effects.
– Sample Spread: Adjusting this control modifies the weighting given to each sample. This affects
the brightness of the samples.

Chapter 57 Tracker Nodes 1404

Use GPU
The Use GPU menu has three settings. Setting the menu to Disable turns off GPU hardware-
accelerated rendering. Enabled uses the GPU hardware for rendering the node. Auto uses a capable
GPU if one is available and falls back to software rendering when a capable GPU is not available.

Hide Incoming Connections

Enabling this checkbox can hide connection lines from incoming nodes, making a node tree appear
cleaner and easier to read. When enabled, empty fields for each input on a node will be displayed in
the Inspector. Dragging a connected node from the node tree into the field will hide that incoming
connection line as long as the node is not selected in the node tree. When the node is selected in the
node tree, the line will reappear.

Comments
The Comments field is used to add notes to a tool. Click in the empty field and type the text. When a
note is added to a tool, a small red square appears in the lower-left corner of the node when the full
tile is displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the
note in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 57 Tracker Nodes 1405

Chapter 58

Transform Nodes
This chapter details the Transform nodes available in Fusion.

The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Camera Shake [CSH]  1407
Crop [CRP]  1410
DVE [DVE]  1413
Letterbox [LBX]  1415
Resize [RSZ]  1418
Scale [SCL]  1421
Transform [XF]  1424
The Common Controls  1428

Chapter 58 Transform Nodes 1406

Camera Shake [CSH]

The Camera Shake node

Camera Shake Node Introduction

This node can simulate a variety of camera shake-style motions from organic to mechanical. It is not
the same as the Shake Modifier, which generates random number values for parameters.
For more information on the Shake modifier, see Chapter 61, “Modifiers” in the Fusion Reference
Manual or Chapter 122 in the DaVinci Resolve Reference Manual.
The Camera Shake node concatenates its result with adjacent transformation nodes for higher-quality
processing.

Inputs
The two inputs on the Camera Shake node are used to connect a 2D image and an effect mask, which
can be used to limit the camera shake area.
– Input: The orange input is used for the primary 2D image that shakes.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the camera
shake area to only those pixels within the mask. An effects mask is applied to the tool after the
tool is processed.

Basic Node Setup

The Camera Shake background input is used to connect the image you want to transform. Polygon
masks can be connected into the occlusion mask input to identify areas the camera shake
should ignore.

The Camera Shake node can be connected directly after

a MediaIn node or any node providing a 2D output.

Chapter 58 Transform Nodes 1407

Inspector

The Camera Shake Controls tab

Controls Tab
The Controls tab includes parameters for adjusting the offsets, strength, speed, and frequency of the
simulated camera shake movement.

Deviation X and Y
These controls determine the amount of shake applied to the image along the horizontal (X) and
vertical (Y) axes. Values between 0.0 and 1.0 are permitted. A value of 1.0 generates shake positions
anywhere within the boundaries of the image.

Rotation Deviation
This determines the amount of shake that is applied to the rotational axis. Values between 0.0 and 1.0
are permitted.

Randomness
Higher values in this control cause the movement of the shake to be more irregular or random. Smaller
values cause the movement to be more predictable.

Overall Strength
This adjusts the general amplitude of all the parameters and blends that affect in and out. A value of
1.0 applies the effect as described by the remainder of the controls.

Speed
Speed controls the frequency, or rate, of the shake.

Frequency Method
This selects the overall shape of the shake. Available frequencies are Sine, Rectified Sine, and Square
Wave. A Square Wave generates a much more mechanical-looking motion than a Sine.

Filter Method
When rescaling a pixel, surrounding pixels are often used to give a more realistic result. There are
various algorithms for combining these pixels, called filters. More complex filters can give better
results but are usually slower to calculate.

Chapter 58 Transform Nodes 1408

The best filter for the job often depends on the amount of scaling and on the contents of the
image itself.
– Box: This is a simple interpolation resize of the image.
– Linear: This uses a simplistic filter, which produces relatively clean and fast results.
– Quadratic: This filter produces a nominal result. It offers a good compromise between speed and
quality.
– Cubic: This produces better results with continuous-tone images. If the images have fine detail in
them, the results may be blurrier than desired.
– Catmull-Rom: This produces good results with continuous-tone images that are resized down.
This produces sharp results with finely detailed images.
– Gaussian: This is very similar in speed and quality to Bi-Cubic.
– Mitchell: This is similar to Catmull-Rom but produces better results with finely detailed images. It is
slower than Catmull-Rom.
– Lanczos: This is very similar to Mitchell and Catmull-Rom but is a little cleaner and also slower.
– Sinc: This is an advanced filter that produces very sharp, detailed results; however, it may produce
visible “ringing” in some situations.
– Bessel: This is similar to the Sinc filter but may be slightly faster.

Window Method (Sinc and Bessel Only)

Some filters, such as Sinc and Bessel, require an infinite number of pixels to calculate exactly. To
speed up this operation, a windowing function is used to approximate the filter and limit the number of
pixels required. This control appears when a filter that requires windowing is selected.
– Hanning: This is a simple tapered window.
– Hamming: Hamming is a slightly tweaked version of Hanning that does not taper all the way down
to zero.
– Blackman: A window with a more sharply tapered falloff.
– Kaiser: A more complex window with results between Hamming and Blackman.

Most of these filters are useful only when making an image larger. When shrinking images, it is
common to use the Bi-Linear filter, however, the Catmull-Rom filter will apply some sharpening to the
results and may be useful for preserving detail when scaling down an image.

Example

Resize filters. From left to right: Nearest Neighbor, Box, Linear, Quadratic, Cubic, Catmull-Rom,
Gaussian, Mitchell, Lanczos, Sinc, and Bessel.

Edges
This menu determines how the Edges of the image are treated.
– Canvas: This causes the edges that are revealed by the shake to be the canvas color—usually
transparent or black.
– Wrap: This causes the edges to wrap around (the top is wrapped to the bottom, the left is
wrapped to the right, and so on).

Chapter 58 Transform Nodes 1409

 – Duplicate: This causes the Edges to be duplicated, causing a slight smearing effect at the edges.
– Mirror: Image pixels are mirrored to fill to the edge of the frame.

Invert Transform
Select this control to Invert any position, rotation, or scaling transformation. This option might be useful
for exactly removing the motion produced in an upstream Camera Shake.

Flatten Transform
The Flatten Transform option prevents this node from concatenating its transformation with adjacent
nodes. The node may still concatenate transforms from its input, but it will not concatenate its
transformation with the node at its output.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Transform nodes. These common controls
are described in detail at the end of this chapter in “The Common Controls” section.

Crop [CRP]

The Crop node

Crop Node Introduction

The Crop node can be used to cut out a portion of an image or to offset the image into a larger image
area. However, unlike using a mask, this node actually changes the resolution of the image.

TIP: You can crop an image in the viewer by activating the Allow Box Selection in the upper-
left corner of the viewer while the Crop node is selected and viewed. Then, drag a crop
rectangle around the area of interest to perform the operation.

NOTE: Because this node changes the physical resolution of the image, animating the
parameters is not advised.

Inputs
The single input on the Crop node is used to connect a 2D image for cropping.
– Input: The orange input is used for the primary 2D image you want to crop.

Chapter 58 Transform Nodes 1410

Basic Node Setup
Below, the Crop node is inserted between the MediaIn1 node and the background input of the Merge.
Unlike using a mask tool, cropping the MediaIn1 changes the resolution of the clip. The cropped
MediaIn1 node connected to the orange background input also sets the resolution of the
Merge output.

The Crop node can be used to cut out a portion of an image.

Inspector

The Crop Controls tab

Controls Tab
The Controls tab provides XY Offset and XY Size methods for cropping the image.

Offset X and Y
These controls position the image off the screen by pushing it left/right or up/down. The cropped
image disappears off the edges of the output image. The values of these controls are measured
in pixels.

Size X and Y
Use these controls to set the vertical and horizontal resolution of the image output by the Crop node.
The values of these controls are measured in pixels.

Keep Aspect
When toggled on, the Crop node maintains the aspect of the input image.

Keep Centered
When toggled on, the Crop node automatically adjusts the X and Y Offset controls to keep the image
centered. The XY Offset sliders are automatically adjusted, and control over the cropping is done with
the Size sliders or the Allow Box Selection button in the viewer.

Chapter 58 Transform Nodes 1411

Reset Size
This resets the image dimensions to the size of the input image.

Reset Offset
This resets the X and Y Offsets to their defaults.

Change Pixel Aspect

Enable this checkbox to reveal a Pixel Aspect control that can be used to change the image’s
pixel aspect.

Clipping Mode
This option sets the mode used to handle the edges of the image when performing domain of
definition (DoD) rendering. This is profoundly important for nodes like Blur, which may require samples
from portions of the image outside the current domain.
– Frame: The default option is Frame, which automatically sets the node’s domain of definition
to use the full frame of the image, effectively ignoring the current domain of definition. If the
upstream DoD is smaller than the frame, the remaining area in the frame will be treated as black/
transparent.
– Domain: Setting this option to Domain will respect the upstream DoD when applying the node’s
effect. This can have adverse clipping effects in situations where the node employs a large filter.
– None: Setting this option to None does not perform any source image clipping at all. This means
that any data required to process the node’s effect that would normally be outside the upstream
DoD is treated as black/transparent.

Auto Crop Tab

Auto Crop tab analyzes the selected channel and crops the image based on that channel’s
boundaries. The adjustments from auto crop are seen in the Crop tab parameters.

The Auto Crop tab

RGBA Color Channels

Select which channels are examined for an Auto Crop. This is useful for auto cropping images with
non-solid backgrounds in a specific color channel, like a blue color gradient. Toggling the channel off
causes Auto Crop to ignore it when evaluating the image.

Auto Crop
This evaluates the image and attempts to determine the background color. It then crops each side of
the image to the first pixel that is not that color.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Transform nodes. These common controls
are described in detail at the end of this chapter in “The Common Controls” section.

Chapter 58 Transform Nodes 1412

DVE [DVE]

The DVE node

DVE Node Introduction

The DVE (Digital Video Effects) node is a 3D-image transformation similar to nodes found in old,
tape-based online editing suites. The node encompasses image rotations, perspective changes, and
Z moves. The axis can be defined for all transformations.

Inputs
The three inputs on the DVE node are used to connect a 2D image, DVE mask, and an effect mask,
which can be used to limit the DVE area.
– Input: The orange input is used for the primary 2D image that is transformed by the DVE.
– DVE Mask: The white DVE mask input is used to mask the image prior to the DVE transform
being applied. This has the effect of modifying both the image and the mask.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input causes the DVE to
modify only the image within the mask. An effects mask is applied to the tool after the tool is
processed.

Basic Node Setup

In the example below, the DVE node is inserted between the MediaIn2 node and the foreground input
of the Merge. The MediaIn1 node is manipulated in the DVE node and composited over the top of the
MediaIn1 node.

The DVE node modifying the foreground input of a Merge node

Chapter 58 Transform Nodes 1413

Inspector

The DVE Controls tab

Controls Tab
The Controls tab includes all the transform parameters for the DVE.

Pivot X, Y, and Z
Positions the axis of rotation and scaling. The default is 0.5, 0.5 for X and Y, which is in the center of
the image, and 0 for Z, which is at the center of Z space.

Rotation Order
Use these buttons to determine in what order rotations are applied to the image.

XYZ Rotation
These controls are used to rotate the image around the pivot along the X-, Y- and Z-axis.

Center X and Y
This positions the center of the DVE image onscreen. The default is 0.5, 0.5, which positions the DVE
in the center of the image.

Z Move
This zooms the image in and out along the Z-axis. Visually, when this control is animated, the effect is
similar to watching an object approach from a distance.

Perspective
This adds additional perspective to an image rotated along the X- or Y-axis, similar to changing the
Field of View and zoom of a camera.

Masking Tab
The DVE node allows pre-masking of its input image. This offers the ability to create transformations
from the masked area of the image while leaving the remainder of the image unaffected.

The DVE Masking tab

Chapter 58 Transform Nodes 1414

Unlike regular effect masks, the masking process occurs before the transformation. All the usual mask
types can be applied to the DVE mask.

Black Background
Toggle this on to erase the area outside the mask from the transformed image.

Fill Black
Toggle this on to erase the area within the mask (before transformation) from the DVE’s input,
effectively cutting the masked area out of the image. Enabling both Black Background and Fill Black
will show only the masked, transformed area.

Alpha Mode
This determines how the DVE will handle the alpha channel of the image when merging the
transformed image areas over the untransformed image.
– Ignore Alpha: This causes the input image’s alpha channel to be ignored, so all masked areas will
be opaque.
– Subtractive/Additive: These cause the internal merge of the pre-masked DVE image over the
input image to be either Subtractive or Additive.
– An Additive setting is necessary when the foreground DVE image is premultiplied, meaning
that the pixels in the color channels have been multiplied by the pixels in the alpha channel.
The result is that transparent pixels are always black, since any number multiplied by 0 always
equals 0. This obscures the background (by multiplying with the inverse of the foreground
alpha), and then simply adds the pixels from the foreground.
– A Subtractive setting is necessary if the foreground DVE image is not premultiplied. The
compositing method is similar to an Additive merge, but the foreground DVE image is first
multiplied by its own alpha, to eliminate any background pixels outside the alpha area.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Transform nodes. These common controls
are described in detail at the end of this chapter in “The Common Controls” section.

Letterbox [LBX]

The Letterbox node

Letterbox Node Introduction

Use the Letterbox node to adapt existing images to the frame size and aspect ratios of any other
format. The most common use of this node is to convert film resolution images to HD-sized frames for
viewing on an external television monitor. Horizontal or vertical black edges are automatically added
where necessary to compensate for aspect ratio differences. This node actually changes the
resolution of the image.

Chapter 58 Transform Nodes 1415

 NOTE: Because this node changes the physical resolution of the image, animating the
controls is not recommended.

Inputs
The single input on the Letterbox node is used to connect a 2D image for letterbox/cropping.
– Input: The orange input is used for the primary 2D image you want to letterbox/crop.

Basic Node Setup

The Letterbox node is used in the example below to change the resolution of the Merge node’s
output. Depending on how the resolution is modified, side pillars, or a horizontal letterbox mask, is
applied to “fill in” the frame area, which the Merge node output does not cover.

The Letterbox node converting the Merge output resolution and adding letterbox masking where needed.

Inspector

The Letterbox Controls tab

Controls Tab
The Controls tab includes parameters for adjusting the resolution and pixel aspect of the image. It also
has the option of letterboxing or pan-and-scan formatting.

Width and Height

The values of these controls determine the size of the output image as measured in pixels.

TIP: You can use the formatting contextual menu to quickly select a resolution from a list.
Place the pointer over the Width or Height controls, and then right-click to display the
contextual menu. The bottom of the menu displays a Select Frame Format submenu with
available frame formats. Select any one of the choices from the menu to set the Height,
Width, and Aspect controls automatically.

Chapter 58 Transform Nodes 1416

Auto Resolution
Activating this checkbox automatically sets the Width and Height sliders to the Frame Format settings
found in the Preferences window for Fusion Studio or to the resolution of the DaVinci Resolve Timeline.

Pixel Aspect X and Y

These controls determine the pixel aspect ratio of the output image.

Center X and Y
This Center control repositions the image window when used in conjunction with Pan-and-Scan mode.
It has no effect on the image when the node is set to Letterbox mode.

Mode
This control is used to determine the Letterbox node’s mode of operation.
– Letterbox/Envelope: This corrects the aspect of the input image and resizes it to match the
specified width.
– Pan-and-Scan: This corrects the aspect of the input image and resizes it to match the specified
height. If the resized input image is wider than the specified width, the Center control can be used
to animate the visible portion of the resized input.

Filter Method
When rescaling a pixel, surrounding pixels are often used to give a more realistic result. There are
various algorithms for combining these pixels, called filters. More complex filters can give better
results but are usually slower to calculate. The best filter for the job often depends on the amount of
scaling and on the contents of the image itself.
– Box: This is a simple interpolation resize of the image.
– Linear: This uses a simplistic filter, which produces relatively clean and fast results.
– Quadratic: This filter produces a nominal result. It offers a good compromise between
speed and quality.
– Cubic: This produces better results with continuous-tone images. If the images have fine detail in
them, the results may be blurrier than desired.
– Catmull-Rom: This produces good results with continuous-tone images that are resized down.
This produces sharp results with finely detailed images.
– Gaussian: This is very similar in speed and quality to Bi-Cubic.
– Mitchell: This is similar to Catmull-Rom but produces better results with finely detailed images. It is
slower than Catmull-Rom.
– Lanczos: This is very similar to Mitchell and Catmull-Rom but is a little cleaner and also slower.
– Sinc: This is an advanced filter that produces very sharp, detailed results; however, it may produce
visible “ringing” in some situations.
– Bessel: This is similar to the Sinc filter but may be slightly faster.

Window Method (Sinc and Bessel Only)

Some filters, such as Sinc and Bessel, require an infinite number of pixels to calculate exactly. To
speed up this operation, a windowing function is used to approximate the filter and limit the number of
pixels required. This control appears when a filter that requires windowing is selected.
– Hanning: This is a simple tapered window.
– Hamming: Hamming is a slightly tweaked version of Hanning that does not taper all the way down
to zero.

Chapter 58 Transform Nodes 1417

 – Blackman: A window with a more sharply tapered falloff.
– Kaiser: A more complex window with results between Hamming and Blackman.

Most of these filters are useful only when making an image larger. When shrinking images, it is
common to use the Bi-Linear filter; however, the Catmull-Rom filter will apply some sharpening to the
results and may be useful for preserving detail when scaling down an image.

Example

Different resize filters. From left to right: Nearest Neighbor, Box, Linear, Quadratic, Cubic,
Catmull-Rom, Gaussian, Mitchell, Lanczos, Sinc, and Bessel.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Transform nodes. These common controls
are described in detail at the end of this chapter in “The Common Controls” section.

Resize [RSZ]

The Resize node

Resize Node Introduction

Use the Resize node to increase or decrease the resolution of an input image. This is useful for
converting images from one format to another (for example, from film to video resolution). This node
actually changes the resolution of the image.

NOTE: Because this node changes the physical resolution of the image, animating the
controls is not advised.

Inputs
The single input on the Resize node is used to connect a 2D image for resizing.
– Input: The orange input is used for the primary 2D image you want to resize.

Chapter 58 Transform Nodes 1418

Basic Node Setup
Below, the Resize node is inserted between the MediaIn1 node and the background input of the
Merge. Unlike using a Transform tool, resizing the MediaIn1 changes the resolution of the clip. The
resized MediaIn1 node connected to the orange background input also sets the resolution of the
Merge output.

The Resize node can be used to scale an image and change its resolution.

Inspector

The Resize Controls tab

Controls Tab
The Controls tab includes parameters for changing the resolution of the image. It uses pixel values in
the Width and Height controls.

Width
This controls the new resolution for the image along the X-axis.

Height
This controls the new resolution for the image along the Y-axis.

TIP: You can use the formatting contextual menu to quickly select a resolution from a list.
Place the mouse pointer over the Width or Height controls, and then right-click to display the
contextual menu. The bottom of the menu displays a Select Frame Format submenu with
available frame formats. Select any one of the choices from the menu to set the Height and
Width controls automatically.

Auto Resolution
Activating this checkbox automatically sets the Width and Height sliders to the Frame Format settings
found in the Preferences window for Fusion Studio or the resolution in the DaVinci Resolve Timeline.

Chapter 58 Transform Nodes 1419

Reset Size
Resets the image dimensions to the original size of the image.

Keep Frame Aspect

When toggled on, the Resize node maintains the aspect of the original image, preserving the original
ratio between width and height.

Only Use Filter in HiQ

The Resize node will normally use the fast Nearest Neighbor filter for any non-HiQ renders, where
speed is more important than full accuracy. Disable this checkbox to force Resize to always use the
selected filter for all renders.

Change Pixel Aspect

Enable this checkbox to reveal a Pixel Aspect control that can be used to change the image’s
pixel aspect.

Filter Method
When rescaling a pixel, surrounding pixels are often used to give a more realistic result. There are
various algorithms for combining these pixels, called filters. More complex filters can give better
results but are usually slower to calculate. The best filter for the job often depends on the amount of
scaling and on the contents of the image itself.
– Box: This is a simple interpolation resize of the image.
– Linear: This uses a simplistic filter, which produces relatively clean and fast results.
– Quadratic: This filter produces a nominal result. It offers a good compromise between speed and
quality.
– Cubic: This produces better results with continuous-tone images. If the images have fine detail in
them, the results may be blurrier than desired.
– Catmull-Rom: This produces good results with continuous-tone images that are resized down.
This produces sharp results with finely detailed images.
– Gaussian: This is very similar in speed and quality to Bi-Cubic.
– Mitchell: This is similar to Catmull-Rom but produces better results with finely detailed images. It is
slower than Catmull-Rom.
– Lanczos: This is very similar to Mitchell and Catmull-Rom but is a little cleaner and also slower.
– Sinc: This is an advanced filter that produces very sharp, detailed results; however, it may produce
visible “ringing” in some situations.
– Bessel: This is similar to the Sinc filter but may be slightly faster.

Window Method (Sinc and Bessel Only)

Some filters, such as Sinc and Bessel, require an infinite number of pixels to calculate exactly. To
speed up this operation, a windowing function is used to approximate the filter and limit the number of
pixels required. This control appears when a filter that requires windowing is selected.
– Hanning: This is a simple tapered window.
– Hamming: Hamming is a slightly tweaked version of Hanning that does not taper all the way down
to zero.
– Blackman: A window with a more sharply tapered falloff.
– Kaiser: A more complex window with results between Hamming and Blackman.

Chapter 58 Transform Nodes 1420

Most of these filters are useful only when making an image larger. When shrinking images, it is
common to use the Bi-Linear filter; however, the Catmull-Rom filter will apply some sharpening to the
results and may be useful for preserving detail when scaling down an image.

Example

Different resize filters. From left to right: Nearest Neighbor, Box, Linear, Quadratic, Cubic,
Catmull-Rom, Gaussian, Mitchell, Lanczos, Sinc, and Bessel.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Transform nodes. These common controls
are described in detail at the end of this chapter in “The Common Controls” section.

Scale [SCL]

The Scale node

Scale Node Introduction

The Scale node is almost identical to the Resize node, except that Resize uses exact dimensions,
whereas the Scale node uses relative dimensions to describe the change to the source image’s
resolution. This node actually changes the resolution of the image.

NOTE: Because this node changes the physical resolution of the image, animating the
controls is not advised.

Inputs
The single input on the Scale node is used to connect a 2D image for scaling.
– Input: The orange input is used for the primary 2D image you want to scale.

Basic Node Setup

Below, the Scale node is inserted between the MediaIn1 node and the background input of the Merge.
Unlike using a Transform tool, scaling the MediaIn1 changes the resolution of the clip. The resized
MediaIn1 node connected to the orange background input also sets the resolution of the
Merge output.

Chapter 58 Transform Nodes 1421

The Scale node can be used to scale an image and change its resolution.

Inspector

The Scale Controls tab

Controls Tab
The Controls tab includes parameters for changing the resolution of the image. It uses a multiplier of
size to set the new resolution. An Edges menu allows you to determine how the edges of the frame
are handled if the scaling decreases.

Lock X/Y
When selected, only a Size control is shown, and changes to the image’s scale are applied to both
axes equally. If the checkbox is cleared, individual Size controls appear for both X and Y Size.

Size
The Size control is used to set the scale used to adjust the resolution of the source image. A value of
1.0 would have no affect on the image, while 2.0 would scale the image to twice its current resolution.
A value of 0.5 would halve the image’s resolution.

Only Use Filter in HiQ

The Scale node will normally use the fast Nearest Neighbor filter for any non-HiQ renders, where
speed is more important than full accuracy. Disable this checkbox to force Scale to always use the
selected filter for all renders.

Change Pixel Aspect

Enable this checkbox to reveal a Pixel Aspect control that can be used to change the image’s
pixel aspect.

Filter Method
When rescaling a pixel, surrounding pixels are often used to give a more realistic result. There are
various algorithms for combining these pixels, called filters. More complex filters can give better

Chapter 58 Transform Nodes 1422

results but are usually slower to calculate. The best filter for the job often depends on the amount of
scaling and on the contents of the image itself.
– Box: This is a simple interpolation resize of the image.
– Linear: This uses a simplistic filter, which produces relatively clean and fast results.
– Quadratic: This filter produces a nominal result. It offers a good compromise between
speed and quality.
– Cubic: This produces better results with continuous-tone images. If the images have fine detail in
them, the results may be blurrier than desired.
– Catmull-Rom: This produces good results with continuous-tone images that are resized down.
This produces sharp results with finely detailed images.
– Gaussian: This is very similar in speed and quality to Bi-Cubic.
– Mitchell: This is similar to Catmull-Rom but produces better results with finely detailed images. It is
slower than Catmull-Rom.
– Lanczos: This is very similar to Mitchell and Catmull-Rom but is a little cleaner and also slower.
– Sinc: This is an advanced filter that produces very sharp, detailed results; however,
it may produce visible “ringing” in some situations.
– Bessel: This is similar to the Sinc filter but may be slightly faster.

Window Method (Sinc and Bessel Only)

Some filters, such as Sinc and Bessel, require an infinite number of pixels to calculate exactly. To
speed up this operation, a windowing function is used to approximate the filter and limit the number of
pixels required. This control appears when a filter that requires windowing is selected.
– Hanning: This is a simple tapered window.
– Hamming: Hamming is a slightly tweaked version of Hanning that does not taper all the
way down to zero.
– Blackman: A window with a more sharply tapered falloff.
– Kaiser: A more complex window with results between Hamming and Blackman.

Most of these filters are useful only when making an image larger. When shrinking images, it is
common to use the Bi-Linear filter; however, the Catmull-Rom filter will apply some sharpening to the
results and may be useful for preserving detail when scaling down an image.

NOTE: Because this node changes the physical resolution of the image, animating the
controls is not advised.

Example

Different resize filters. From left to right: Nearest Neighbor, Box, Linear, Quadratic, Cubic,
Catmull-Rom, Gaussian, Mitchell, Lanczos, Sinc, and Bessel.

Chapter 58 Transform Nodes 1423

Transform [XF]

The Transform node

Transform Node Introduction

The Transform node can be used for simple 2D transformations of the image, such as moving, rotating,
and scaling. The image’s aspect can also be modified using the Transform node.
The Transform node concatenates its result with adjacent Transformation nodes. The Transform node
does not change the image’s resolution.

Inputs
The two inputs on the Transform node are used to connect a 2D image and an effect mask, which can
be used to limit the transformed area.
– Input: The orange input is used for the primary 2D image that gets transformed.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the transform
area to only those pixels within the mask. An effects mask is applied to the tool after the
tool is processed.

Basic Node Setup

The Transform node in the example below is inserted between the MediaIn2 node and the foreground
input of the Merge. Unlike using a Scale or Resize tool, transforming the MediaIn2 does not change
the resolution of the clip. For that reason, it is the tool most often used to scale, move, and rotate a clip.

The Transform node can be used to scale an image without changing its resolution.

Chapter 58 Transform Nodes 1424

Inspector

The Transform Controls tab

Controls Tab
The Controls tab presents multiple ways to transform, flip (vertical), flop (horizontal), scale, and rotate
an image. It also includes reference size controls that can reinterpret the coordinates used for width
and height from relative values of 0-1 into pixel values based on the image’s resolution.

Center X and Y
This sets the position of the image on the screen. The default is 0.5, 0.5, which places the image in the
center of the screen. The value shown is always the actual position multiplied by the reference size.
See below for a description of the reference size.

Pivot X and Y
This positions the axis of rotation and scaling. The default is 0.5, 0.5, which is the center of the image.

Use Size and Aspect

This checkbox determines whether the Transform node provides independent Size controls for the X
and Y scale or if Size and Aspect controls are used instead.

Size
This modifies the scale of the image. Values range from 0 to 5, but any value greater than zero can be
entered into the edit box. If the Use Size and Aspect checkbox is selected, this control will scale the
image equally along both axes. If the Use Size and Aspect option is off, independent control is
provided for X and Y.

Aspect
This control changes the aspect ratio of an image. Setting the value above 1.0 stretches the image
along the X-axis. Values between 0.0 and 1.0 stretch the image along the Y-axis. This control is
available only when the Use Size and Aspect checkbox is enabled.

Angle
This control rotates the image around the axis. Increasing the Angle rotates the image in a
counterclockwise direction. Decreasing the Angle rotates the image in a clockwise direction.

Chapter 58 Transform Nodes 1425

Flip Horizontally and Vertically
Toggle this control on to flip the image along the X- or Y-axis.

Edges
This menu determines how the edges of the image are treated when the edge of the raster
is exposed.
– Canvas: This causes the edges of the image that are revealed to show the current Canvas Color.
This defaults to black with no Alpha and can be set using the Set Canvas Color node.
– Wrap: This wraps the edges of the image around the borders of the image. This is useful for
seamless images to be panned, creating an endless moving background image.
– Duplicate: This causes the edges of the image to be duplicated as best as possible, continuing
the image beyond its original size.
– Mirror: Image pixels are mirrored to fill to the edge of the frame.

Filter Method
When rescaling a pixel, surrounding pixels are often used to give a more realistic result. There are
various algorithms for combining these pixels, called filters. More complex filters can give better
results but are usually slower to calculate. The best filter for the job often depends on the amount of
scaling and on the contents of the image itself.
– Box: This is a simple interpolation resize of the image.
– Linear: This uses a simplistic filter, which produces relatively clean and fast results.
– Quadratic: This filter produces a nominal result. It offers a good compromise
between speed and quality.
– Cubic: This produces better results with continuous-tone images. If the images have fine detail in
them, the results may be blurrier than desired.
– Catmull-Rom: This produces good results with continuous-tone images that are resized down.
This produces sharp results with finely detailed images.
– Gaussian: This is very similar in speed and quality to Bi-Cubic.
– Mitchell: This is similar to Catmull-Rom but produces better results with finely detailed images. It is
slower than Catmull-Rom.
– Lanczos: This is very similar to Mitchell and Catmull-Rom but is a little cleaner and also slower.
– Sinc: This is an advanced filter that produces very sharp, detailed results; however, it may produce
visible “ringing” in some situations.
– Bessel: This is similar to the Sinc filter but may be slightly faster.

Window Method (Sinc and Bessel Only)

Some filters, such as Sinc and Bessel, require an infinite number of pixels to calculate exactly. To
speed up this operation, a windowing function is used to approximate the filter and limit the number of
pixels required. This control appears when a filter that requires windowing is selected.
– Hanning: This is a simple tapered window.
– Hamming: Hamming is a slightly tweaked version of Hanning that does not taper all the way down
to zero.
– Blackman: A window with a more sharply tapered falloff.
– Kaiser: A more complex window with results between Hamming and Blackman.

Chapter 58 Transform Nodes 1426

Most of these filters are useful only when making an image larger. When shrinking images, it is
common to use the Bi-Linear filter; however, the Catmull-Rom filter will apply some sharpening to the
results and may be useful for preserving detail when scaling down an image.

Example

Different resize filters. From left to right: Nearest Neighbor, Box, Linear, Quadratic, Cubic,
Catmull-Rom, Gaussian, Mitchell, Lanczos, Sinc, and Bessel.

Invert Transform
Select this control to invert any position, rotation, or scaling transformation. This option is useful when
connecting the Transform to the position of a tracker for the purpose of reintroducing motion back into
a stabilized image.

Flatten Transform
The Flatten Transform option prevents this node from concatenating its transformation with adjacent
nodes. The node may still concatenate transforms from its input, but it will not concatenate its
transformation with the node at its output.

Reference Size
The controls under the Reference Size menu do not directly affect the image. Instead they allow you to
control how Fusion represents the position of the Transform node’s center.
Normally, coordinates are represented as values between 0 and 1, where 1 is a distance equal to the
full width or height of the image. This allows for resolution independence, because you can change
the size of the image without having to change the value of the center.
One disadvantage to this approach is that it complicates making pixel-accurate adjustments to an
image. To demonstrate, imagine an image that is 100 x 100 pixels in size. To move the center of the
image to the right by 5 pixels, we would change the X value of the transform center from 0.5, 0.5 to
0.55, 0.5. We know the change must be 0.05 because 5/100 = 0.05.
The Reference Size controls allow you to specify the dimensions of the image. This changes the way
the control values are displayed, so that the Center shows the actual pixel positions in the X and Y
number fields of the Center control. Extending our example, if you set the Width and Height to 100
each, the Center would now be shown as 50, 50, and we would move it 5 pixels toward the right by
entering 55, 50.
Internally, the Transform node still stores this value as a number between 0 to 1, and if you were to
query the Center controls value via scripting, or publish the Center control for use by other nodes,
then you would retrieve the original normalized value. The change is visible only in the value shown for
Transform Center in the node control.

Reference Width and Height Sliders

Set these to the width and height of the image to change the way that Fusion displays the values of
the Transform node’s Center control.

Auto Resolution
Enable this checkbox to use the current frame format settings in Fusion Studio or the timeline
resolution in DaVinci Resolve to set the Reference Width and Reference Height values.

Chapter 58 Transform Nodes 1427

The Common Controls
Nodes that handle Transform operations share several identical controls in the Inspector. This section
describes controls that are common among Transform nodes.

Inspector

The Transform Common controls

Settings Tab
The Settings tab in the Inspector can be found on every tool in the Transform category. The Settings
controls are even found on third-party Transform-type plug-in tools. The controls are consistent and
work the same way for each tool.

Blend
The Blend control is used to blend between the tool’s original image input and the tool’s final modified
output image. When the blend value is 0.0, the outgoing image is identical to the incoming image.
Normally, this will cause the tool to skip processing entirely, copying the input straight to the output.

Process When Blend Is 0.0

The tool is processed even when the input value is zero. This can be useful if the node is scripted to
trigger a task, but the node’s value is set to 0.0.

Chapter 58 Transform Nodes 1428

Red/Green/Blue/Alpha Channel Selector
These four buttons are used to limit the effect of the tool to specified color channels. This filter is often
applied after the tool has been processed.
For example, if the Red button on a Blur tool is deselected, the blur will first be applied to the image,
and then the red channel from the original input will be copied back over the red channel of the result.
There are some exceptions, such as tools for which deselecting these channels causes the tool to skip
processing that channel entirely. Tools that do this will generally possess a set of identical RGBA
buttons on the Controls tab in the tool. In this case, the buttons in the Settings and the Controls tabs
are identical.

Apply Mask Inverted

Enabling the Apply Mask Inverted option inverts the complete mask channel for the tool. The mask
channel is the combined result of all masks connected to or generated in a node.

Multiply by Mask
Selecting this option will cause the RGB values of the masked image to be multiplied by the mask
channel’s values. This will cause all pixels of the image not included in the mask (i.e., set to 0) to
become black/transparent.

Use Object/Use Material (Checkboxes)

Some 3D software can render to file formats that support additional channels. Notably, the EXR file
format supports Object ID and Material ID channels, which can be used as a mask for the effect. These
checkboxes determine whether the channels will be used, if present. The specific Material ID or
Object ID affected is chosen using the next set of controls.

Correct Edges
This checkbox appears only when the Use Object or Use Material checkboxes are selected. It toggles
the method used to deal with overlapping edges of objects in a multi-object image. When enabled,
the Coverage and Background Color channels are used to separate and improve the effect around the
edge of the object. If this option is disabled (or no Coverage or Background Color channels are
available), aliasing may occur on the edge of the mask.
For more information on the Coverage and Background Color channels, see Chapter 18,
“Understanding Image Channels” in the Fusion Reference Manual or Chapter 79 in the
DaVinci Resolve Reference Manual.

Object ID/Material ID (Sliders)

Use these sliders to select which ID will be used to create a mask from the object or material channels
of an image. Use the Sample button in the same way as the Color Picker: to grab IDs from the image
displayed in the viewer. The image or sequence must have been rendered from a 3D software
package with those channels included.

Motion Blur
– Motion Blur: This toggles the rendering of Motion Blur on the tool. When this control is toggled
on, the tool’s predicted motion is used to produce the motion blur caused by the virtual camera’s
shutter. When the control is toggled off, no motion blur is created.
– Quality: Quality determines the number of samples used to create the blur. A quality setting of 2
will cause Fusion to create two samples to either side of an object’s actual motion. Larger values
produce smoother results but increase the render time.
– Shutter Angle: Shutter Angle controls the angle of the virtual shutter used to produce the motion
blur effect. Larger angles create more blur but increase the render times. A value of 360 is the
equivalent of having the shutter open for one full frame exposure. Higher values are possible and
can be used to create interesting effects.

Chapter 58 Transform Nodes 1429

 – Center Bias: Center Bias modifies the position of the center of the motion blur. This allows for the
creation of motion trail effects.
– Sample Spread: Adjusting this control modifies the weighting given to each sample. This affects
the brightness of the samples.

Use GPU
The Use GPU menu has three settings. Setting the menu to Disable turns off GPU hardware-
accelerated rendering. Enabled uses the GPU hardware for rendering the node. Auto uses a capable
GPU if one is available and falls back to software rendering when a capable GPU is not available.

Hide Incoming Connections

Enabling this checkbox can hide connection lines from incoming nodes, making a node tree appear
cleaner and easier to read. When enabled, empty fields for each input on a node will be displayed in
the Inspector. Dragging a connected node from the node tree into the field will hide that incoming
connection line as long as the node is not selected in the node tree. When the node is selected in the
node tree, the line will reappear.

Comments
The Comments field is used to add notes to a tool. Click in the empty field and type the text. When a
note is added to a tool, a small red square appears in the lower-left corner of the node when the full
tile is displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the
note in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 58 Transform Nodes 1430

Chapter 59

VR Nodes
This chapter details the Virtual Reality (VR) nodes available in Fusion.

The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
VR Nodes  1432
Lat Long Patcher [LLP]  1433
Pano Map [PaM]  1434
Spherical Camera [3SC]  1436
Spherical Stabilizer  1437
The Common Controls  1439

Chapter 59 VR Nodes 1431

VR Nodes
You can create and fix spherical (360°) video, often described as Virtual Reality, or VR, using Fusion’s
set of VR nodes. Dome productions, planetariums, and other special-venue theaters have utilized the
flexibility of Fusion and its 3D system to produce and deliver special content for years.
The equirectangular (lat-long) format often used for 360° video is similar to how a globe is represented
by a flat world map, with the poles at the top and bottom edges of the image and the forward
viewpoint at the center.

TIP: You can create stereo VR using two stacked Lat Long images, one for each eye.

A VR Sphere displayed in the Node Editor and Fusion’s 3D viewer

Fusion supports several common spherical image formats and can easily convert between them.
– VCross and HCross: VCross and HCross are the six square faces of a cube laid out in a cross,
vertically or horizontally, with the forward view in the center of the cross in a 3:4 or 4:3 image.
– VStrip and HStrip: VStrip and HStrip are the six square faces of a cube laid vertically or
horizontally in a line, ordered as Left, Right, Up, Down, Back, Front (+X, -X, +Y, -Y, +Z, -Z) in a 1:6 or
6:1 image.
– LatLong: LatLong is a single 2:1 image in an equirectangular mapping.

You can display both spherical video and live 3D scenes from the comp directly to headsets, including
those from Oculus Rift and HTC Vive.
Fusion’s “Fix it in post” tools for VR make it easy to perform several important tasks that are common in
these types of productions.

Chapter 59 VR Nodes 1432

Lat Long Patcher [LLP]

The Lat Long Patcher node

NOTE: The VR category and Lat Long node are available only in Fusion Studio and
DaVinci Resolve Studio.

Lat Long Patcher Node Introduction

Equirectangular stitched images often need patches, paintwork, or other VFX applied. The Lat Long
Patcher extracts and de-warps a section of a lat-long (equirectangular) image to be treated, and can
warp and merge fixes back over the original. You can quickly pick a section of the spherical image to
patch or paint, and then apply it back to the original image. Note that matching rotations are used in
both Extract and Apply modes, allowing a node’s operation to be easily reversed by a copy or
instance with the same rotation settings.

Input
The Lat Long Patcher node includes two inputs. The orange input accepts a 2D image in an
equirectangular format, where the X-axis represents 0–360 degrees longitude, and the Y-axis
represents –90 to +90 degrees latitude. The effect mask input is provided, although rarely used
on VR nodes.
– Image Input: The orange image input accepts a equirectangular (lat-long) 2D RGBA image.
– Effect Mask: The effect mask input is provided, although rarely used on VR nodes.

Basic Node Setup

The Loader node connects to the input on a Lat Long Patcher node. The output of a Lat Long Patcher
node is set to Extract. It is then connected to whatever image-processing operation is required. A
second Lat Long Patcher node set to Apply takes an input from the processed extraction and merges
it over the top of the original source.

Two Lat Long Patchers used to repair a section

Chapter 59 VR Nodes 1433

Inspector

The Lat Long Patcher Controls tab

Controls Tab
The Controls tab is used to extract and later reapply a section from an equirectangular image. Rotation
controls allow you to select the exact portion you need to repair.

Mode
– Extract: Pulls a de-warped 90-degree square image from the equirectangular image.
– Apply: Warps and merges a 90-degree square image over the equirectangular image. Because
the square image’s alpha is used, this allows, for example, paint strokes or text drawn over a
transparent black background to be applied to the original equirectangular image, avoiding any
double-filtering from de-warping and re-warping the original.

Rotation Order
These buttons choose the ordering of the rotations around each axis. For example, XYZ rotates first
around the X axis (pitch/tilt), then around the Y axis (pan/yaw), and then around the Z axis (roll). Any of
the six possible orderings can be chosen.

Rotation
These dials rotate the spherical image around each of the X, Y, and Z axes, offering independent
control over pitch/tilt, pan/yaw, and roll, respectively.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other VR nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Pano Map [PaM]

The Pano Map node

NOTE: The VR category and Pano Map node are available only in Fusion Studio and
DaVinci Resolve Studio.

Chapter 59 VR Nodes 1434

Pano Map Node Introduction
The Pano Map node converts images from one spherical layout to another, such as from a cube map
to an equirectangular format. The node can also perform rotations of the spherical images
when converting.

Input
The Pano Map node includes two inputs. The orange input accepts a 2D image in an equirectangular,
cube map or other spherical formats. The effect mask input is provided, although rarely used
on VR nodes.
– Image Input: The orange Image input accepts a spherical formatted 2D RGBA image that gets
converted to another spherical format.
– Effect Mask: The effect mask input is provided, although rarely used on VR nodes.

Basic Node Setup

In the example below, a Loader node containing a Lat Long image connects to the input on a Pano
Map node. The Pano Map node is used to convert the image to an H-Cross format. It is then
connected to whatever image-processing operation is required.

A Pano Map used to convert from one VR format to another

Inspector

The Pano Map Controls tab

Controls Tab
The Controls tab is used to determine the format of the input image and the desired output format.

From/To
– Auto: Auto detects the incoming image layout from the metadata and image frame aspect.
– VCross and HCross: VCross and HCross are the six square faces of a cube laid out in a cross,
vertically or horizontally, with the forward view in the center of the cross in a 3:4 or 4:3 image.

Chapter 59 VR Nodes 1435

 – VStrip and HStrip: VStrip and HStrip are the six square faces of a cube laid vertically or
horizontally in a line, ordered as Left, Right, Up, Down, Back, Front (+X, -X, +Y, -Y, +Z, -Z)
in a 1:6 or 6:1 image.
– LatLong: LatLong is a single 2:1 image in equirectangular mapping.

Rotation Order
These buttons choose the ordering of the rotations around each axis. For example, XYZ rotates first
around the X axis (pitch/tilt), then around the Y axis (pan/yaw), and then around the Z axis (roll). Any of
the six possible orderings can be chosen.

Rotation
These dials rotate the spherical image around each of the X, Y, and Z axes, offering independent
control over pitch/tilt, pan/yaw, and roll, respectively.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other VR nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Spherical Camera [3SC]

The Spherical Camera node

Spherical Camera Node Introduction

The Spherical Camera is not located in the VR category of the Effects Library but in the 3D category.
However, it is commonly used in creating and fixing VR content, so it is referenced here. The Spherical
Camera allows the 3D Renderer node to output an image covering all viewing angles, laid out in
several different formats. This image may be used, for example, as a skybox texture or reflection map
or viewed in a VR headset. The Image Width setting in the 3D Renderer sets the size of each square
cube face so that the resulting image may be a multiple of this size horizontally and vertically.
For more detail on the Spherical Camera node, refer to Chapter 29, “3D Nodes” in the Fusion
Reference Manual or Chapter 90 in the DaVinci Resolve Reference Manual.

Chapter 59 VR Nodes 1436

Spherical Stabilizer

The Spherical Stabilizer node

NOTE: The VR category and Spherical Stabilizer node are available only in Fusion Studio and
DaVinci Resolve Studio.

Spherical Stabilizer Node Introduction

VR live action often uses handheld cameras, causing shaky footage to be a common problem. The
Spherical Stabilizer node automatically identifies and tracks visible features in the footage, and then
analyzes their movement to identify pan, tilt, and roll rotations. After tracking, it is then possible to
smooth out or stabilize the rotation of the footage.

Inputs
The Spherical Stabilizer node has a single orange input.
– Image: This orange image input node requires an image in a spherical layout, which can be any
of Lat Long (2:1 equirectangular), Horizontal/Vertical Cross, or Horizontal/Vertical Strip.

Basic Node Setup

In the example below, a 2:1 Lat Long image is connected to the input of the Spherical Stabilizer node.
Once the image is stabilized, the output of the Spherical Stabilizer node is a steadied clip.

Spherical Stabilizer set up to steady a 2:1 Lat Long clip

Chapter 59 VR Nodes 1437

Inspector

The Spherical Stabilizer Controls tab

Controls Tab
The Controls tab contains parameters to initiate the tracking and modify the results for
stabilization or smoothing.

Reject Dominant Motion Outliers While Tracking

With this control activated (the default setting), features that move contrary to the majority of other
features are ignored. This helps ignore the movement of subjects in the shot, preferring stable and
consistent markers from the surrounding environment.

Track Controls
These buttons initiate tracking and analysis of the shot. Be aware that the reference frame used for
stabilization is set to the first frame tracked.
– Track Backward from End Frame starts tracking backward from the end of the
current render range.
– Track Backward from Current Time starts tracking backward from the current frame.
– Stop ceases tracking, preserving all results so far.
– Track Forward from Current Time starts tracking forward from the start of the current render range.
– Track Forward from Start Frame starts tracking forward from the current time.

Append to Track
– Replace causes the Track Controls to discard any previous tracking results and replace them with
the newly-created track.
– Append adds the new tracking results to any earlier tracks.

Stabilization Strength
This control varies the amount of smoothing or stabilization applied, from 0.0 (no change) to 1.0
(maximum).

Smoothing
The Spherical Stabilizer node can eliminate all rotation from a shot, fixing the forward viewpoint (Still
mode, 0.0) or gently smooth out any panning, rolling, or tilting to increase viewer comfort (Smooth
mode, 1.0). This slider allows either option or anything in between.

Chapter 59 VR Nodes 1438

Offset Rotation
Often a shot is not entirely level and needs the horizon to be realigned, or perhaps a desired pan
should be reintroduced after fully stabilizing the shot. The Offset Rotation controls allow additional
manual control of the Spherical Stabilizer’s rotation of the footage, for pitch/tilt (X), pan/yaw (Y),
and roll (Z), respectively. Rotation is always performed in the order X, Y, Z.

Common Controls
Settings Tab
The Settings tab in the Inspector is duplicated in other VR nodes. These common controls are
described in detail in the following “The Common Controls” section.

The Common Controls

Nodes that handle VR operations share a number of identical controls in the Inspector. This section
describes controls that are common among VR nodes.

Inspector

The VR Common controls

Settings Tab
The Settings tab in the Inspector can be found on every tool in the VR category. The controls are
consistent and work the same way for each tool.

Apply Mask Inverted

Enabling the Apply Mask Inverted option inverts the complete mask channel for the tool. The mask
channel is the combined result of all masks connected to or generated in a node.

Chapter 59 VR Nodes 1439

Use GPU
The Use GPU menu has three settings. Setting the menu to Disable turns off hardware-accelerated
rendering using the graphics card in your computer. Auto uses a capable GPU if one is available and
falls back to software rendering when a capable GPU is not available.

Hide Incoming Connections

Enabling this checkbox can hide connection lines from incoming nodes, making a node tree appear
cleaner and easier to read. When enabled, empty fields for each input on a node are displayed in the
Inspector. Dragging a connected node from the node tree into the empty field hides that incoming
connection line as long as the node is not selected in the node tree. When the node is selected in the
node tree, the line reappears.

Comments
The Comments field is used to add notes to a tool. Click in the empty field and type the text. When a
note is added to a tool, a small red square appears in the lower-left corner of the node when the full
tile is displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the
note in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 59 VR Nodes 1440

Chapter 60

Warp Nodes
This chapter details the Warp nodes available in Fusion.

The abbreviations next to each node name can be used in the Select Tool dialog
when searching for tools and in scripting references.

For purposes of this document, node trees showing MediaIn nodes in

DaVinci Resolve are interchangeable with Loader nodes in Fusion Studio, unless
otherwise noted.

Contents
Coordinate Space [CDS]  1442
Corner Positioner [CPN]  1443
Dent [DNT]  1445
Displace [DSP]  1447
Drip [DRP]  1449
Grid Warp [GRD]  1451
Lens Distort [LENS]  1458
Perspective Positioner [PPN]  1460
Vector Distortion [DST]  1462
Vortex [VTX]  1464
The Common Controls  1466

Chapter 60 Warp Nodes 1441

Coordinate Space [CDS]

The Coordinate Space node

Coordinate Space Node Introduction

The Coordinate Space node changes the coordinate space of the image from rectangular to polar or
from polar to rectangular.

Inputs
The two inputs on the Coordinate Space node are used to connect a 2D image and an effect mask,
which can be used to limit the distorted area.
– Input: The orange input is used for the primary 2D image that is distorted.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive
shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the
distortion to only those pixels within the mask. An effects mask is applied to the tool after the
tool is processed.

Basic Node Setup

The Coordinate Space node is used below to make a circular pattern based on a Fast Noise, Mosaic
Blur (DaVinci Resolve Resolve FX only), and a Transform node. The Crop node at the end is used to set
the desired resolution.

The Coordinate Space node can help create motion graphics backgrounds

Example
To demonstrate a basic tunnel effect that can be achieved with this node:
1 Add a Text+ node with some text, and then animate it to move along a path from the top of
the frame to the bottom.
2 Connect the output of the Text+ node to a Coordinate Space node.
3 Select Polar to Rectangular from the Shape menu.

Chapter 60 Warp Nodes 1442

 As the text moves from top to bottom along the original path, it appears to move from an
infinite distance in the Coordinate Space node. It may be necessary to flip the text using a
Transform node to make it appear the correct way in the Coordinate Space node. Another
common use for the Coordinate Space node is to use it in pairs: two of them set to different
Shape settings with a Drip or Transform node in between. When used in this way, the effect
gets modified while the image remains the same.

Inspector

The Coordinate Space Controls tab

Controls Tab
The Controls tab Shape menu switches between Rectangular to Polar and Polar to Rectangular.
Consider the following example to demonstrate the two coordinate spaces.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Warp nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Corner Positioner [CPN]

The Corner Positioner node

Corner Positioner Node Introduction

The Corner Positioner can be used to position the four corners of an image interactively. This would
typically be used to replace a sign or other rectangular portion of a scene. Connect all corners to
Paths or Trackers for animation purposes.

Inputs
The two inputs on the Corner Positioner node are used to connect a 2D image and an effect mask,
which can be used to limit the warped area.

Chapter 60 Warp Nodes 1443

 – Input: The orange input is used for the primary 2D image that is warped.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive shapes,
paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the Corner
Positioner to only those pixels within the mask. An effects mask is applied to the tool after the
tool is processed.

Basic Node Setup

Below, the Corner Positioner is used to position the rectangular corners of the MediaIn2 to fit within a
specific area of the MediaIn1 node. A Planar Tracker tracks the background, and then a Planar
Transform is used to keep the Corner Positioner in place as the background clip moves. Once the
planar tracking is completed, and the Planar Transform is created, the Planar Tracker node is no longer
needed, and you can delete it.

The Corner Positioner corner pins a clip in place and uses

a Planar Transform to match move it to the background.

Inspector

The Corner Positioner Controls tab

Chapter 60 Warp Nodes 1444

Controls Tab
The Controls tab includes transform and offset adjustments for the four corners of the image

Mapping Type
This determines the method used to project the image caused by the Corner Positioner. In Bi-Linear
mode, a straight 2D warping takes place. In Perspective mode, the image is calculated with the offsets
in 2D space and then mapped into a 3D perspective.

Corners X and Y
There are four points in the Corner Positioner. Drag these around to position each corner of the image
interactively. Attach these control points to any of the usual modifiers.
The image input is deformed and perspective corrected to match the position of the four corners.

Offset X and Y
These controls can be used to offset the position of the corners slightly. This is useful when the
corners are attached to Trackers with patterns that may not be positioned exactly where they
are needed.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Warp nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Dent [DNT]

The Dent node

Dent Node Introduction

The Dent node creates a circular deformation of an image similar to a Fish Eye Lens effect, with the
choice of six different Dent filters.

Inputs
The two inputs on the Dent node are used to connect a 2D image and an effect mask, which can be
used to limit the warped area.
– Input: The orange input is used for the primary 2D image that is warped.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive
shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the
Dent to only those pixels within the mask. An effects mask is applied to the tool after the
tool is processed.

Chapter 60 Warp Nodes 1445

Basic Node Setup
The Dent node is used below to make a circular pattern based on a Fast Noise and a Mosaic Blur
(DaVinci Resolve Resolve FX only). The Crop node at the end is used to set the desired resolution.

The Dent node can help create lens distortion effects or a motion graphics background.

Inspector

The Dent Controls tab

Controls Tab
The adjustments in the Controls tab are used to change the Dent style, position, size, and strength.

Type
Select the type of Dent filter to use from this menu. All parameters for the Dent can be keyframed.

Dent 1
This creates a bulge dent.

Kaleidoscope
This creates a dent, mirrors it, and inverts it.

Dent 2
This creates a displacement dent.

Dent 3
This creates a deform dent.

Cosine Dent
This creates a fracture to a center point.

Sine Dent
This creates a smooth rounded dent.

Center X and Y
This positions the Center of the Dent effect on the image. The default values are 0.5, 0.5, which center
the effect in the image.

Size
This changes the size of the area affected by the dent. Animate this slider to make the dent grow.

Chapter 60 Warp Nodes 1446

Strength
This changes the overall strength of the dent.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Warp nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Displace [DSP]

The Displace node

Displace Node Introduction

The Displace node uses a map image to displace or refract another image. This is useful for creating a
vast range of effects from bevels and heat distortion to glass and water effects.

Inputs
There are three inputs on the Displace node: The primary image, the displacement map foreground
image, and an effect mask.
– Input: The orange image input is a required connection for the primary image you wish to
displace.
– Foreground Image: The green input is also required as the image used to displace the
background. Once connected, you can choose red, green, blue, alpha, or luminance channel to
create the displacement.
– Effect Mask: The optional blue effect mask input expects a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the displacement to only those pixels within the mask. An effects mask is applied to
the tool after it is processed.

Basic Node Setup

Below, the Displace node uses a Fast Noise to generate the Displace map. Increasing the seethe rate
can produce heat distortion or flag waving effects.

The Displace node using a Fast Noise node for the Displace map

Chapter 60 Warp Nodes 1447

Inspector

The Displace Controls tab

Controls Tab
The Controls tab is used to change the style, position, size , strength, and lighting (embossing) of the
displacement.

Type
The Type menu is used to choose in what mode the Displace node operates. The Radial mode uses
the map image that refracts each pixel out from the center, while X/Y mode provides control over the
amount of displacement along each axis individually.

NOTE: There is one set of Refraction controls while in Radial mode, and two sets in XY
mode—one for each of the X and Y channels.

Center (Radial Only)

The Center control defines the point from which pixels are displaced toward or away.

Refraction Channel
This drop-down menu controls which channel from the foreground image is used to displace the
image. Select from Red, Green, Blue, Alpha, or Luminance channels. In XY mode, this control appears
twice, once for the X displacement and once for the Y displacement.

Refraction Strength (Radial)

Controls the strength of the refraction. Higher values cause stronger or more pronounced refraction.

X and Y Refraction (X/Y)

Two separate sliders appear to control the Refraction strength along the X- and Y-axis separately.
Otherwise, this is exactly like Refraction Strength.

Light Power
This controls the intensity, or strength, of the simulated light, causing bright and dim areas to form
according to the contour of the refraction image. Higher values cause the bright and dim areas to be
more pronounced.

Light Angle
This sets the angle of the simulated light source.

Chapter 60 Warp Nodes 1448

Spread
This widens the Displacement effect and takes the edge off the Refraction map. Higher values cause
the ridges or edges to spread out.

Light Channel
Select the channel from the refraction image to use as the simulated light source. Select from Color,
Red, Green, Blue, Alpha, or Luminance channels.

NOTE: The Radial mode pushes pixels inward or outward from a center point, based on pixel
values from the Displacement map. The XY mode uses two different channels from the map
to displace pixels horizontally and vertically, allowing more precise results. Using the XY
mode, the Displace node can even accomplish simple morphing effects. The Light controls
allow directional highlighting of refracted pixels for simulating a beveled look.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Warp nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Drip [DRP]

The Drip node

Drip Node Introduction

The Drip node creates a ripple effect over the entire image, which has the potential to animate
outward from a central source. There are a variety of different Drip effects from which to choose.

Inputs
The two inputs on the Drip node are used to connect a 2D image and an effect mask, which can be
used to limit the warped area.
– Input: The orange input is used for the primary 2D image that is warped.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive
shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the
warping to only those pixels within the mask. An effects mask is applied to the tool after the
tool is processed.

Basic Node Setup

Below, the Drip node is used to make rippling water-style effects using a MediaIn node.

Chapter 60 Warp Nodes 1449

The Drip node can be connected directly after a
MediaIn node or any node providing a 2D output.

Inspector

The Drip Controls tab

Controls Tab
The Controls tab is used to change the style, position, size , strength, and phase for animating the
“ripples” of the Drip.

Shape
Use this control to select the shape of the Drip.

Circular
This creates circular ripples. This is the default Drip mode.

Square
This creates even-sided quadrilateral drips.

Random
This creates a randomly dispersed noise that distorts your image and is similar to a particle effect.

Horizontal
This creates horizontal waves that move in one direction.

Vertical
This creates vertical waves that move in one direction.

Exponential
This creates a Drip effect that looks like a diamond shape with inverted, curved sides (an exponential
curve flipped and mirrored).

Star
This creates an eight-way symmetrical star-shaped ripple that acts as a kaleidoscope when the phase
is animated.

Chapter 60 Warp Nodes 1450

Radial
This creates a star-shaped ripple that emits from a fixed pattern.

Center X and Y
Use this control to position the center of the Drip effect in the image. The default is 0.5, 0.5, which
centers the effect in the image.

Aspect
Control the aspect ratio of the various Drip shapes. A value of 1.0 causes the shapes to be
symmetrical. Smaller values cause the shape to be taller and narrower, while larger values cause
shorter and wider shapes.

Amplitude
The Amplitude of the Drip effect refers to the peak height of each ripple. Use the slider to change the
amount of distortion the Drip applies to the image. A value of 0.0 gives all ripples no height and
therefore makes the effect transparent. A maximum Amplitude of 10 makes each ripple extremely
visible and completely distorts the image. Higher numbers can be entered via the text entry boxes.

Dampening
Controls the Dampening, or falloff, of the amplitude as it moves away from the center of the effect. It
can be used to limit the size or area affected by the Drip.

Frequency
This changes the number of ripples emanating from the center of the Drip effect. A value of 0.0
indicates no ripples. Move the slider up to a value of 100 to correspond with the density of
desired ripples.

Phase
This controls the offset of the frequencies from the center. Animate the Phase value to make the ripple
emanate from the center of the effect.

Common Controls
Settings Tab
The Settings Tab in the Inspector is also duplicated in other Warp nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Grid Warp [GRD]

The Grid Warp node

Grid Warp Node Introduction

The Grid Warp node is a 2D deformation grid with flexible vertices. The image is deformed so that the
source grid matches the destination grid.

Chapter 60 Warp Nodes 1451

Inputs
The two inputs on the Grid Warp node are used to connect a 2D image and an effect mask, which can
be used to limit the warped area.
– Input: The orange input is used for the primary 2D image that is warped.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive
shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this input limits
the warping to only those pixels within the mask. An effects mask is applied to the tool after
the tool is processed.

Basic Node Setup

Below, two Grid Warp nodes are used to warp different areas of a frame. This can be useful for
chaining the framing in a shot or adding slight movement to a still. Using the Copy Src to Dest button,
only the area modified by Grid Warp 2 is pasted as the foreground in the Merge.

The Grid Warp can be used to shift areas within a

shot for reframing or adding animation to a still.

Inspector

The Grid Warp Controls tab

Controls Tab
The Controls tab contains parameters that configure the onscreen grid as well the type of distortion
applied when a control point on the grid is moved.

Chapter 60 Warp Nodes 1452

Source and Destination
The Source and Destination buttons determine whether the source grid or destination grid is currently
active. Only one grid can be displayed or manipulated at a time. The selected button is highlighted to
indicate that it is the currently active grid.
All other controls in this tab affect the grid selected by this control.

Selection Type
These three buttons determine the selection types used for manipulating the points. There are three
options available.

Selected
When in Selected mode, adjustments to the grid are applied only to the currently selected points.
This mode is identical to normal polyline operation.

Region
In Region mode, all points within the area around the mouse pointer move when the mouse button is
clicked. New points that enter the region during the move are ignored. Choosing this option exposes
Magnet Distance and Magnet Strength controls to determine the size and falloff of the area.

Magnetic
In Magnetic mode, all points within the area around the mouse pointer move when the mouse button
is clicked. New points that enter the region during the move are affected as well. Choosing this option
exposes Magnet Distance and Magnet Strength controls to determine the size and falloff of the area.

Magnet Distance
The default node for selecting and manipulating the grid is a Magnet node. The magnet is represented
in the viewer by a circle around the mouse pointer. The Magnet Distance slider controls how large the
region of effect for the magnet is, as in the size of the circle. Drag on the grid and any vertex within the
range of the slider moves.
To increase the size of the magnet, increase the value of this slider. Alternately, adjust the size of the
magnet by holding down the D key while dragging the mouse.

Magnet Strength
The Magnet Strength slider increases or decreases the falloff of the magnet cursor’s effect. At a
setting of 0.0, the magnetic cursor has no effect, and vertices do not move at all. As the values
increase, the magnet causes a greater range of motion in the selected vertices. Use smaller values for
a more sensitive adjustment and larger values for broad-sweeping changes to the grid.

X and Y Grid Size

The X and Y Grid Size sliders control the number of divisions in the grid. Where the X and Y divisions
intersect, a control vertex is created.
Be aware that changing either of these controls after applying changes in the grid resets the entire
grid. Set the X and Y grid sizes to the appropriate resolution before making detailed adjustments
to the grid.

Subdivision Level
The Subdivision Level determines how many subdivisions there are between each set of divisions.
Subdivisions do not generate vertices at intersections. The more subdivisions, the smoother the
deformation is likely to be, but the slower it is to render.

Center
The Center coordinates determine the exact center of the grid. The onscreen Center control is
invisible while editing the grid. Select the Edit Rect mode, and the grid center becomes visible and
available for editing.

Chapter 60 Warp Nodes 1453

Use the Center control to move the grid through a scene without affecting the animation applied to the
individual vertices. For example, while deforming lips, track the motion of the face with a Tracker, and
connect the grid center to the Tracker. This matches the grid with slight movements of the head while
focusing on the deformation of the lips.

Angle
This Angle control rotates the entire grid.

Size
The Size control increases or decreases the scale of the grid.

Edit Buttons
There are four edit modes available, each of which can be selected by clicking on the
appropriate button.

Edit None
Set the grid to Edit None mode to disable the display of all onscreen controls.

Edit Grid
The Edit Grid mode is the default mode. While this mode is enabled, the grid is drawn in the viewer,
and the control vertices of the grid can be manipulated directly.

Edit Rectangle
When the grid is in Edit Rectangle mode, the onscreen controls display a rectangle that determines
the dimensions of the grid. The sides of the rectangle can be adjusted to increase or decrease the
grid’s dimension. This mode also reveals the onscreen Center control for the grid.

Edit Line
The Edit Line mode is beneficial for creating grids around organic shapes. When this mode is enabled,
all onscreen controls disappear, and a spline can be drawn around the shape or object to be
deformed. While drawing the spline, a grid is automatically created that best represents that object.
Additional controls for Tolerance, Over Size, and Snap Distance appear when this mode is enabled.
These controls are documented below.

Set Mesh to Entire Image

The Set Mesh to Entire Image button automatically resets the size of the grid to the exact dimensions
of the image. Any adjustments to vertices within the grid are reset.

Copy Buttons
These two buttons provide a technique for copying the exact shape and dimensions of the source grid
to the destination, or the destination grid to the source. This is particularly useful after setting the
source grid to ensure that the destination grid’s initial state matches the source grid before beginning
a deformation.

Point Tolerance
This control is visible only when the Edit Line mode is enabled. The Point Tolerance slider determines
how much tessellation the grid applies to match the density of points in the spline closely. The lower
this value, the fewer vertices there are in the resulting grid, and the more uniform the grid appears.
Higher values start applying denser grids with variations to account for regions in the spline that
require more detail.

Oversize Amount
This control is visible only when the Edit Line mode is enabled. The Oversize Amount slider is used to
set how large an area around the spline should be included in the grid. Higher values create a larger
border, which can be useful when blending a deformation back into the source image.

Chapter 60 Warp Nodes 1454

Snap Distance
This control is visible only when the Edit Line mode is enabled. The Snap Distance slider dictates how
strongly the drawn spline attracts surrounding vertices. If a vertex is close enough to a spline’s edge,
the vertex moves to line up with the spline. The higher the value, the farther the reach of the spline.

Right-Click Here for Mesh Animation

The grids are static by default. Right-clicking on the Right-Click Here for Mesh Animation label
provides a contextual menu with options for animating the grid or connecting it to another grid in the
composition.
The grid uses a Polychange spline. Any adjustment to the control points adds or modifies the
keyframe for all points on that spline.

Right-Click Here for Shape Animation

This label appears only in the Edit Line mode. Right-clicking on the Right-Click Here for Shape
Animation label reveals a pop-up menu used to animate the shaping polyline or to connect it to other
polylines.

The Grid Warp Render tab

Render Tab
The Render tab controls the final rendered quality and appearance of the warping.

Render Method
The Render Method drop-down menu is used to select the rendering technique and quality applied to
the mesh. The three settings are arranged in order of quality, with the first, Wireframe, as the fastest
and lowest of quality. The default mode is Render, which produces final resolution, full-quality results.

Anti-Aliasing
The Anti-Aliasing control appears only as a checkbox when in Wireframe Render mode.
In other modes, it is a drop-down menu with three levels of quality. Higher degrees of anti-aliasing
improve image quality dramatically but vastly increase render times. The Low setting may be an
appropriate option while setting up a large dense grid or previewing a node tree, but rarely for a
final render.

Filter Type
When the Render Method is set to something other than Wireframe mode, the Filter Type menu is
visible and set to Area Sample. This setting prevents the grid from calculating area samples for each
vertex in the grid, providing good render quality. Super Sample can provide even better results but
requires much greater render times.

Chapter 60 Warp Nodes 1455

Wireframe Width
This slider appears only when the Render Method is set to Wireframe. It determines the width of the
lines that make up the wireframe.

Anti-Aliased
This checkbox appears only when the Render Method is set to Wireframe. Use this checkbox to
enable/disable anti-aliasing for the lines that make up the wireframe.

Black Background
The Black Background checkbox determines whether pixels outside of the grid in the source image
are set to black or if they are preserved.

Object ID and Material ID

Enable the Object ID or Material ID checkboxes to have the grid output the proper ID channel in the
final render.

Set Image Coordinates at Subdivision Level

This checkbox defaults to enabled and sets the image coordinates at the subdivision level.

Force Destination Render

This checkbox defaults to enabled and forces a destination render.

Contextual Menu Options

The Grid Warp node places a submenu for both source and destination grids in the viewer’s contextual
menu. Both menus have the exact same name, where only the menu for the active grid is populated
with options. The other menu is empty. The contextual menu options are also available from the
toolbar that appears in the viewer.

Grid Warp Contextual Menu options

Chapter 60 Warp Nodes 1456

Modify Only/Done
These two options set the mesh to Modify Only and Done modes, respectively. Select Modify Only to
edit the mesh or Modify Done to prevent any further changes to a mesh.

Smooth/Linear
Use Smooth and Linear options to apply or remove smoothing from selected vertices.

Auto Smooth Points

When Auto Smooth Points is enabled, the vertices in the grid are automatically smoothed whenever
they are moved. This is generally enabled by default.

Z Under/Z Same/Z Over

When two vertices in a grid overlap, one ends up getting clipped by the other. Z Under, Z Same, and Z
Over are used to select which vertices are rendered on top and which are rendered behind.

Select All
This option selects all points in the mesh.

Show Key Points, Handles, Grid, and Subdivisions

Use these four options to enable or disable the display of the grid, key points (vertices), Bézier
handles, and subdivisions in the viewers.

Reset Selected Points

This resets Selected Points (vertices) to their default positions.

Reset All Points

This resets all points (vertices) in the mesh to their default positions.

Stop Rendering
This option stops rendering, which disables all rendering of the Grid Warp node until the mode is
turned off. This is frequently useful when making a series of fine adjustments to a complex grid.

Grid Warp Toolbar

Whenever the Grid Warp node is selected and is in Edit Grid mode, the Grid Warp toolbar is displayed
in the views. This toolbar provides a variety of options for manipulating and adjusting the grid. The
toolbar buttons in this toolbar are described in the preceding “Contextual Menu Options” section.

The Grid Warp viewer toolbar

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Warp nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Chapter 60 Warp Nodes 1457

Lens Distort [LENS]

The Lens Distort node

Lens Distort Node Introduction

This node can be used to remove or add lens distortion in an image. The lens distortion in an image
depends on the lens, quality, number of elements, and many other factors.
One reason to remove lens distortion is to composite with an undistorted layer. For example,
compositing a 3D element over a distorted live-action layer will cause unwanted effects like straight
lines not matching up on the foreground and background. The resulting composite will not look
believable.

Inputs
The two inputs on the Lens Distort node are used to connect a 2D image and an effect mask, which
can be used to limit the distorted area.
– Input: The orange input is used for the primary 2D image that is distorted.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive
shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this input limits
the distortion to only those pixels within the mask. An effects mask is applied to the tool
after the tool is processed.

Basic Node Setup

A simplified example below applies the Lens Distort in Undistort mode to the MediaIn1 live-action
layer, composites the 3D elements, and finally applies the Lens Distort at the end with the same
settings, but this time in Distort mode to get the original look and distortion back into the image.

Lens Distort applied on the live-action media at the beginning of the node tree, and once again at the end

Chapter 60 Warp Nodes 1458

Inspector

The Lens Distort Controls tab

Controls Tab
The Controls tab presents various ways to customize or build the lens distortion model you want.
Camera Settings allow you to specify the camera used to capture the content.

Mode
Undistort removes the lens distortion to create a flattened image. Distort brings the original lens
distortion back into the image.

Edges
Determines how samples that fall outside the frame are treated.
– Canvas: Pixels outside the frame are set to the default canvas color. In most cases, this is black
with no alpha.
– Duplicate: Pixels outside the frame are duplicated. This results in “smeared” edges but is useful
when, for example, applying a blur because in that case black pixels would result in the unwanted
blurring between the actual image and the black canvas.

Clipping Mode
– Domain: Retains all pixels that might be moved out of the frame for later re-distorting.
– Frame: Pixels moved outside the frame are discarded.

Chapter 60 Warp Nodes 1459

Output Distortion Map
Outputs the location of pixels as a warped screen-coordinate map.

Camera Settings
The options known from the Camera 3D are duplicated here. They can either be set manually or
connected to an already existing Camera 3D.

Lens Distortion Model

Select the appropriate 3D Equalizer Lens Distortion model here: 3DE Classic Model, 3DE4
Anamorphic, 3DE4 Radial Fisheye, or 3DE4 Radial. Please consult the 3D Equalizer manual for further
explanation. The sliders in the 3DE Classic LD Model are most likely best suited for manually applying
(un)distortion, without having imported lens data.

Supersampling [HiQ]
Sets the number of samples used to determine each destination pixel. As always, higher
supersampling leads to higher render times. 1×1 bilinear is usually of sufficient quality, but with high
lens distortion near the edges of the lens, there are noticeable differences to higher settings.

Supersampling Mode [HiQ]

The type of sample done for each supersample. Nearest leads to a crisper but more aliased image.
Bi-Linear gives a blurrier result.

Load Distortion Data

Allows the user to load a Lens Distortion profile created, for example, by the 3D Equalizer.

How to Manually Determine Lens Distortion

In an ideal world, one would have exact lens parameters from each lens that was used during the
shoot, and one could use those values to undistort the image. However, in the real world, those
parameters have not been taken on set or don’t match. Another approach is to use software like 3D
Equalizer, which analyzes the footage and delivers a dataset that can be imported into the Lens Distort
node right away.
And finally, one could try to manually eyeball the amount of lens distortion using the control sliders.
To do that, one could either look for horizontal or vertical lines in the footage that are supposed to be
straight and straighten them out using the controls, or shoot a full-frame checkerboard pattern on set
as a reference.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Warp nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Perspective Positioner [PPN]

The Perspective Positioner node

Chapter 60 Warp Nodes 1460

Perspective Positioner Node Introduction
The Perspective Positioner is the complementary node to the Corner Positioner node. It “unpins” an
image by positioning corner points on a perspective distorted area, thereby removing the perspective
from the image. This function can also be used to wobble and warp the image by animating the points
over time.

Inputs
The two inputs on the Perspective Positioner node are used to connect a 2D image and an effect
mask, which can be used to limit the transformed area.
– Input: The orange input is used for the primary 2D image that is transformed.
– Effect Mask: The blue input is for a mask shape created by polylines, basic primitive
shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this input limits the
transform to only those pixels within the mask. An effects mask is applied to the tool after
the tool is processed.

Basic Node Setup

In the example below, the Perspective Positioner is used to unpin a perspective distorted area of the
MedianIn2 in order to paint on the flat texture. The MediaIn2 is then corner pinned back into place.
The Perspective Positioner and Corner Positioner do not concatenate, so some softness is introduced
with these nodes.

The Perspective Positioner unpins an image to paint on a texture map.

Inspector

The Perspective Positioner Controls tab

Chapter 60 Warp Nodes 1461

Controls Tab
The Controls tab contains parameters for selecting vector channels and controlling how much
distortion they apply to an image.

Mapping Type
The Mapping Type menu is used to select the type of transform used to distort the image. Bi-Linear is
available for support of older projects. Leaving this on Perspective is strongly suggested since the
Perspective setting maps the real world more accurately.

Corners X and Y
There are the four control points of the Perspective Positioner. Interactively drag these in the viewers
to position each corner of the image. You can refine their position using the Top, Bottom, Left, and
Right controls in the Inspector.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Warp nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Vector Distortion [DST]

The Vector Distortion node

Vector Distortion Node Introduction

The Vector Distortion node distorts the main source image along the X- and Y-axis separately, based
on the vector channel data in the source image or vector channels from a second reference image.

Inputs
There are three inputs on the Vector Distort node for the primary 2D image, the distort image with
vector channels and an effect mask.
– Input: The orange image input is a required connection for the primary image you wish to
distort. If this image has vector channels, they are used in the distortion.
– Distort: The green input is an optional distort image input used to distort the background image
based on vector channels. Once connected, it overrides vector channels in the input image.
– Effect Mask: The optional blue effect mask input expects a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the displacement to only those pixels within the mask. An effects mask is applied to
the tool after it is processed.

Basic Node Setup

In the example below, the MediaIn1 node uses an Optical Flow node to generate the vector channels,
which are then passed on to the Vector Distort node. MediaIn2 is distorted and composited over the
background.

Chapter 60 Warp Nodes 1462

The Vector Distort node relies on the image having vector
channels or an Optical Flow node to generate them.

Inspector

The Vector Distortion Controls tab

Controls Tab
The Controls tab contains parameters for selecting vector channels and controlling how much
distortion they apply to an image.

X Channel and Y Channel

These two menus are used to select which channel of the (green) distort image input is used to distort
the X and Y channels. If no distort reference image is connected, then channels from the main orange
input are used instead.

Flip Channel X and Flip Channel Y

Use these checkboxes to flip the direction of the distortion along the specified axis.

Lock Scale X/Y

Select this checkbox to separate the Scale slider into separate Scale X and Scale Y sliders.

Scale
Use the Scale slider to apply a multiplier to the values of the distortion reference image.

Lock Bias X/Y

Select this checkbox to separate the Bias slider into separate Bias X and Bias Y sliders.

Center Bias
Use the Center Bias slider to shift or nudge the distortion along a given axis.

Chapter 60 Warp Nodes 1463

Edges
This menu determines how the edges of the image are treated.
– Canvas: This causes the edges that are revealed by the shake to be the canvas color—usually
transparent or black.
– Duplicate: This causes the edges to be duplicated, causing a slight smearing effect at the edges.

Glow
Use this slider to add a glow to the result of the vector distortion.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Warp nodes. These common controls are
described in detail at the end of this chapter in “The Common Controls” section.

Vortex [VTX]

The Vortex node

Vortex Node Introduction

The Vortex effect appears as a swirling whirlpool in specified regions of the image. The Vortex can be
made to move and grow by animating the various controls.

Inputs
There are two inputs on the Vortex node for the primary 2D image and the effect mask.
– Input: The orange image input is a required connection for the primary image you wish to swirl.
– Effect Mask: The optional blue effect mask input expects a mask shape created by polylines,
basic primitive shapes, paint strokes, or bitmaps from other tools. Connecting a mask to this
input limits the swirling vortex to only those pixels within the mask. An effects mask is applied
to the tool after it is processed.

Basic Node Setup

Below, the Vortex is applied to text for creating motion graphics. Since the Vortex will cause the text to
swirl outside the text boundary, a Set Domain node is used to expand the text boundary, ensuring that
the text is not cropped when the Vortex is applied.

Chapter 60 Warp Nodes 1464

The Vortex is used to create swirling whirlpool effects.

Inspector

The Vector Distortion Controls tab

Controls Tab
The Controls tab contains parameters for adjusting the position, size, and strength of the Vortex effect.

Center X and Y
This control is used to position the center of the Vortex effect on the image. The default is 0.5, 0.5,
which positions the effect in the center of the image.

Size
Size changes the area affected by the Vortex. You can drag the circumference of the effect in the
viewer or use the Size slider.

Angle
Drag the rotation handle in the viewer or use the thumbwheel control to change the amount of rotation
in the Vortex. The higher the angle value, the greater the swirling effect.

Power
Increasing the Power slider makes the Vortex smaller but tighter. It effectively concentrates it inside
the given image area.

Common Controls
Settings Tab
The Settings tab in the Inspector is also duplicated in other Warp nodes. These common controls are
described in detail in the following “The Common Controls” section.

Chapter 60 Warp Nodes 1465

The Common Controls
Nodes that handle Warp operations share a number of identical controls in the Inspector. This section
describes controls that are common among Warp nodes.

Inspector

Warp Common Controls tab

Settings Tab
The Settings tab in the Inspector can be found on every tool in the Warp category. The Settings
controls are even found on third-party Warp-type plug-in tools. The controls are consistent and work
the same way for each tool.

Blend
The Blend control is used to blend between the tool’s original image input and the tool’s final modified
output image. When the blend value is 0.0, the outgoing image is identical to the incoming image.
Normally, this will cause the tool to skip processing entirely, copying the input straight to the output.

Process When Blend Is 0.0

The tool is processed even when the input value is zero. This can be useful if the node is scripted to
trigger a task, but the node’s value is set to 0.0.

Chapter 60 Warp Nodes 1466

Red/Green/Blue/Alpha Channel Selector
These four buttons are used to limit the effect of the tool to specified color channels. This filter is often
applied after the tool has been processed.
For example, if the Red button on a Blur tool is deselected, the blur will first be applied to the image,
and then the red channel from the original input will be copied back over the red channel of the result.
There are some exceptions, such as tools for which deselecting these channels causes the tool to skip
processing that channel entirely. Tools that do this will generally possess a set of identical RGBA
buttons on the Controls tab in the tool. In this case, the buttons in the Settings and the Controls tabs
are identical.

Apply Mask Inverted

Enabling the Apply Mask Inverted option inverts the complete mask channel for the tool. The mask
channel is the combined result of all masks connected to or generated in a node.

Multiply by Mask
Selecting this option will cause the RGB values of the masked image to be multiplied by the mask
channel’s values. This will cause all pixels of the image not included in the mask (i.e., set to 0) to
become black/transparent.

Use Object/Use Material (Checkboxes)

Some 3D software can render to file formats that support additional channels. Notably, the EXR file
format supports Object ID and Material ID channels, which can be used as a mask for the effect. These
checkboxes determine whether the channels will be used, if present. The specific Material ID or
Object ID affected is chosen using the Object and Material ID sliders explained below.

Correct Edges
This checkbox appears only when the Use Object or Use Material checkboxes are selected. It toggles
the method used to deal with overlapping edges of objects in a multi-object image. When enabled,
the Coverage and Background Color channels are used to separate and improve the effect around the
edge of the object. If this option is disabled (or no Coverage or Background Color channels are
available), aliasing may occur on the edge of the mask.
For more information on the Coverage and Background Color channels, see Chapter 18,
“Understanding Image Channels” in the Fusion Reference Manual or Chapter 79 in the
DaVinci Resolve Reference Manual.

Object ID/Material ID (Sliders)

Use these sliders to select which ID will be used to create a mask from the object or material channels
of an image. Use the Sample button in the same way as the Color Picker: to grab IDs from the image
displayed in the viewer. The image or sequence must have been rendered from a 3D software
package with those channels included.

Motion Blur
– Motion Blur: This toggles the rendering of Motion Blur on the tool. When this control is toggled
on, the tool’s predicted motion is used to produce the motion blur caused by the virtual camera’s
shutter. When the control is toggled off, no motion blur is created.
– Quality: Quality determines the number of samples used to create the blur. A quality setting of 2
will cause Fusion to create two samples to either side of an object’s actual motion. Larger values
produce smoother results but increase the render time.
– Shutter Angle:L Shutter Angle controls the angle of the virtual shutter used to produce the motion
blur effect. Larger angles create more blur but increase the render times. A value of 360 is the
equivalent of having the shutter open for one full frame exposure. Higher values are possible and
can be used to create interesting effects.

Chapter 60 Warp Nodes 1467

 – Center Bias: Center Bias modifies the position of the center of the motion blur. This allows for the
creation of motion trail effects.
– Sample Spread: Adjusting this control modifies the weighting given to each sample. This affects
the brightness of the samples.

Use GPU
The Use GPU menu has three settings. Setting the menu to Disable turns off GPU hardware-
accelerated rendering. Enabled uses the GPU hardware for rendering the node. Auto uses a capable
GPU if one is available and falls back to software rendering when a capable GPU is not available.

Hide Incoming Connections

Enabling this checkbox can hide connection lines from incoming nodes, making a node tree appear
cleaner and easier to read. When enabled, empty fields for each input on a node will be displayed in
the Inspector. Dragging a connected node from the node tree into the field will hide that incoming
connection line as long as the node is not selected in the node tree. When the node is selected in the
node tree, the line will reappear.

Comments
The Comments field is used to add notes to a tool. Click in the empty field and type the text. When a
note is added to a tool, a small red square appears in the lower-left corner of the node when the full
tile is displayed, or a small text bubble icon appears on the right when nodes are collapsed. To see the
note in the Node Editor, hold the mouse pointer over the node to display the tooltip.

Scripts
Three Scripting fields are available on every tool in Fusion from the Settings tab. They each contain
edit boxes used to add scripts that process when the tool is rendering. For more details on scripting
nodes, please consult the Fusion scripting documentation.

Chapter 60 Warp Nodes 1468

Chapter 61

Modifiers
This chapter details the modifiers available in Fusion.

Contents
Modifiers  1470
Anim Curves  1470
Bézier Spline  1473
B-Spline  1474
Calculation  1475
CoordTransform Position  1477
Cubic Spline  1478
Expression  1478
From Image  1483
Gradient Color  1484
Key Stretcher Modifier  1486
MIDI Extractor  1486
Natural Cubic Spline  1490
Offset (Angle, Distance, Position)  1490
Path  1493
Perturb  1494
Probe  1496
Publish  1498
Resolve Parameter  1498
Shake  1499
Track  1501
Vector Result  1502
XY Path  1503

Chapter 61 Modifiers 1469

Modifiers
Modifiers are extensions to a node’s standard set of parameters found in the Inspector; in fact,
modifiers are designed to control other parameters. They can be as simple as a motion path or linking
two parameters. However, they can also be elaborate expressions, procedural functions, external
data, third-party plug-ins, or scripted Fuses. You can add modifiers by right-clicking over a parameter
in the Inspector and choosing a modifier from the menu. Alternatively, you can right-click a control in
the viewer. Not all modifiers are displayed in the right-click menu for all parameters. Some modifiers
work only on specific parameter types.

NOTE: Text3D and Text+ have additional text-specific modifiers, which are covered in their
nodes’ sections.

Modifiers right-click menu

Anim Curves
The Animation Curves modifier (Anim Curves) is used to dynamically adjust the timing, values, and
acceleration of an animation, even if you decide to change the duration of a comp. Using this modifier
makes it infinitely easier to stretch or squish animations, create smooth motion, add bouncing
properties, or mirror animations curves without the complexity of manually adjusting splines.

The Inspector for the Anim Curves modifier

Chapter 61 Modifiers 1470

When creating Fusion templates for the Edit and Cut page in DaVinci Resolve, the Anim Curves
modifier allows the keyframed animation you’ve created in Fusion to stretch and squish appropriately
as the transition, title, or effect’s duration changes on the Edit and Cut page Timelines.

Curve Shape Controls

The controls for the Anim Curves modifier appear in the modifier’s tab of the Inspector. The Curve
Shape controls determine the acceleration or shape of the animation curve.
– Source: This drop-down menu has three options based on how the comp is created from
DaVinci Resolve’s Edit page.
– Transition: This setting is automatically selected when the comp is created from an Edit page
transition effect. If the duration of the transition is updated in the Edit page, the timing of the
animation updates as well.
– Duration: Use this setting when the comp is created from a clip on the Edit page. The animation
timing will update if the clip’s duration changes by trimming.
– Custom: Displays an Input dial to manually control the timing.

– Input: This dial is only visible when Source is set to Custom. It is used to change the input
keyframe value.
– Curve: The Curve drop-down menu selects the interpolation method used between keyframes.
The three choices are: linear, easing, or custom.
– Linear: The default Linear interpolation method maintains a fixed, consistent acceleration
between keyframes.
– Easing: Displays interpolation menus for both the start of the curve (In) and the end of
the curve (Out).
– Custom: Opens a mini Spline Editor to customize the interpolation from the start of the
animation to the end.

– Mirror: Plays the animation forward, and after reaching the end, it returns to the starting value. This
causes the initial animation to be twice as fast, since the second half of the comp is used for the
reverse animation.
– Invert: Flips the animation curve upside-down so that the values start high and end low.

Scaling
The Scale parameters modify the animation values using relative adjustments.
– Scale: This number is a multiplier applied to the value of the keyframes. If the Scale value is 2 and
a keyframe has a value of 0, it remains 0. If the Scale value is 2 and a keyframe has a value of
10, the result is as if the keyframe is set to 20. This can be thought of as the ending value for the
animation. It is best to set this while viewing the last frame in the comp.
– Offset: The offset is added to the keyframe values and can be thought of as the starting value for
the animation. It is best to set this while viewing the first frame in the comp.
– Clip Low: Ensures the output value never dips below 0.0.
– Clip High: Ensures the output value never exceeds 1.0.

Chapter 61 Modifiers 1471

Timing
The Timing parameters adjust the animation timing using relative values.
– Time Scale: Stretches or squishes the animation, causing it to run faster or slower. A value of 1.0
keeps the animation running for the comp’s duration (unless you have customized the animation
using other controls in the Modifier).
– Time Offset: This value delays the animation as a fraction of its total duration. A value of
0.0 applies no delay. A value of 0.5 delays the animation starting point midway into the
comp’s duration.

Using the Anim Curves Modifier to Create a Custom Transition

To understand how to use the Anim Curves modifier for a transition, let’s create a simple
scaling dissolve.
1 Add a standard cross dissolve in the Edit page Timeline.
2 Right-click over the transition and choose Convert to Fusion Cross Dissolve.
3 Right-click over the transition again and choose Open in Fusion page.
4 Add a Transform node to MediaIn1 and to MediaIn2.
5 Select the Transform node attached to the MediaIn2.
6 In the Inspector, right-click the Size control, then choose Modify With > Anim Curves from the
contextual menu. Adding this modifier to the Size control will cause the slider to animate from 0 to
1 for the Cross Dissolve’s duration.
7 Select the Transform node attached to the MediaIn1.
8 In the Inspector, right-click the Size control, then choose Modify With > Anim Curves from the
contextual menu.
9 At the top of the Inspector, click the Modifier tab and click the Invert button. Inverting the
animation curves causes MediaIn1 to scale opposite of MediaIn2.
10 In the Modifiers tab, set the Curve drop-down menu to Easing, and experiment with the different
ease-in ease-out curve types from the In/Out drop-down menus.

Once you create a Macro from this node tree and save it as a Transition template, you can apply it in
the Edit page Timeline. If you change the transition duration in the Edit page, the animation timing will
update appropriately.

Using the Anim Curves Modifier with Paths

To understand how to use the Anim Curves modifier’s with a Path modifier, let’s use the premise that
you want to create text that falls from the top of the frame and bounces as it reaches the bottom of
the frame.
1 In Fusion, create two keyframes that cause text to start at the top of the frame and drop to the
bottom. This automatically creates a Path modifier.
2 In the Inspector’s Modifier tab, right-click over the Displacement parameter and choose Insert >
Anim Curves. The animation is normalized to the duration of the comp.
3 Set the Source menu to Duration, since this is not a transition and we are not customizing
the duration.
4 From the Curve menu, choose Easing, then for the Out menu, choose Bounce.
5 Play the animation to see the Bounce animation.
6 To make the bounce occur halfway down the frame, change the Scale to .05.
7 To make the animation run twice as fast, enter 2.0 in the Time Scale parameter.

Chapter 61 Modifiers 1472

Once you create a macro from this node tree and save it as a Title template, you can apply it in the Edit
page Timeline. If you change the title’s duration in the Edit page, the animation timing will update
appropriately.

TIP: To view the resulting animation curve in the Spline Editor, select the parameter name in
the Spline Editor’s header. The spline is updated as you change the controls.

Bézier Spline
The Bézier Spline is one of the animation modifiers in Fusion and is typically applied to numerical
values rather than point values. It is automatically applied when you keyframe a parameter or each
time you right-click a number field and select Animate.

Bézier Spline modifier menu

Usage
You can add the Bézier Spline to the Spline Editor by right-clicking a number field and selecting
BezierSpline. Since this is the most common choice for animation splines, it is separated from the
Modify With menu for quicker access. Selecting BezierSpline from the menu adds a keyframe at the
current location and displays a Bézier Spline in the Spline Editor.

Bézier Spline Editor

Unlike most modifiers, this modifier has no actual Controls tab in the Inspector. However, the Spline
Editor displays the Bézier Spline, and it can be controlled there. The Bézier Spline offers individual
control over each control point’s smoothness using Bézier handles. The smoothness is applied in
multiple ways:
– To make the control points smooth, select them, and press Shift-S. The handles can be used to
modify the smoothness further.

Chapter 61 Modifiers 1473

 – To make the control points linear, select them, and press Shift-L. These operations can also be
performed using the contextual menu.
– Select the control point(s), right-click, and select Smooth or Linear. The menu also allows the user
to smooth a spline using a convolution analysis called a Savitzky-Golay filter. Select the control
point(s), right-click, and select Smooth Points -Y Dialog.

Ease In/Out
Traditional Ease In/Out can also be modified by using the number field virtual sliders in the Spline
Editor. Select the control points you want to modify, right-click, and select Ease In/Out... from the
contextual menu. Then use the number field virtual sliders to control the Ease In/Out numerically.

Spline Ease In/Out modifier

B-Spline
An alternative to the Bézier Spline, B-spline is another animation modifier in Fusion and is typically
applied to numerical values rather than point values. It is applied by right-clicking a parameter and
selecting Modify With > B-Spline.

Usage

B-Spline Editor

– This animation spline modifier has no actual Controls tab. However, the Spline Editor displays
the B-spline, and it can be controlled there. Notice that, though the actual value of the second
keyframe is 0, the value of the resulting spline is 0.33 due to the unique smoothing and weighing
algorithms of a B-spline.
– The weight can be modified by clicking the control point to select it, holding the W key, and
moving the mouse left and right to lower or increase the tension. This is also done with multiple
selected control points simultaneously.

Chapter 61 Modifiers 1474

Calculation
Calculations are used to create indirect connections between parameters. A Calculation can perform a
mathematical expression based on two operands, where each operand can be connected to another
parameter or set manually.
Additionally, using Time offsets and Scale controls in the Time tab, the Calculation control can access
values of a parameter at times other than the current time.
The Calculation’s most common use is for connecting two parameters when one value range or scope
is inappropriate for the other parameter.

NOTE: The Expression modifier is essentially a more flexible version of the Calculation
modifier, with a single exception. It is far easier to manipulate the timing of the operands
provided in the Calculation modifier than it is to do so with an Expression.

Inspector

Calculation Calc tab

Calc Tab
The Calc tab includes two dials used for the connected parameter and value that gets mathematically
combined. The Operator menu selects how the Second Operand value combines with the
parameter’s value.

First and Second Operand

These sliders are connected to published or animated parameters or manually set to the desired
values for the calculation.

Operator
Select from the mathematical operations listed in this menu to determine how the two operands are
combined. Clicking the drop-down arrow opens the menu with the following options:

– Add – Divide (Second / First) – Average

– Subtract (First - Second) – Subtract (Second - First) – First only
– Multiply – Minimum
– Divide (First / Second) – Maximum

Calculation Time tab

Chapter 61 Modifiers 1475
Time Tab
The Time tab is used to modify the time of the Calculation modifier. The controls here retime the
speed of the effect or offset it in time.

First and Second Operand Time Scale

These sliders multiply the frame number and return the value of the operands at the multiplied frame
number. A value of 1 returns the value of the operand at frame x when the composition is on frame x.
For example, if the first operand is animated with a value of 1 to 50 from frame 0 to 10, then a scale of
0.5 would cause the calculation to return a value of 25 at frame 10 (effectively slowing the animation by
half for the purposes of the calculation).

First Operand and Second Operand Time Offset

These sliders return the value of the operand at the Time Offset specified. A value of 10 would return
the value of the operand 10 frames forward in time, and -10 would return the value of the operand 10
frames back in time. See the example below for a practical example.

Example
The following example uses a calculation to apply blur to a Text node in inverse proportion to
the size of the text.
1 Open a new composition that starts on frame 0 and ends on frame 100.
2 At frame 0, add a Text+ node to the composition.
3 Enter a small amount of text and set the size to 0.05
4 Click the Keyframe button to the right of the Size slider to add a keyframe.
5 Move to frame 100 and set the Size value to 0.50.
6 Connect a Blur node after the Text+ node.
7 View the Blur node in one of the viewers.
To have the blur decrease in strength as the text gets bigger, a simple “pick whip”-style
parameter linking does not work. The controls cannot be directly connected together
because the values of the Text Size control are getting bigger instead of smaller.
8 Right-click the Blur Size and select Modify With > Calculation from the contextual menu.
This adds a Calculation modifier to the Blur node. At the top of the Inspector, a new set of
controls appears in the Modifiers tab while the Blur node is selected.
9 At the top of the Inspector, select the Modifiers tab (F11).
10 Right-click the First Operand slider and select Connect To > Text 1 > Size from the
contextual menu.
Although the Blur Size is now connected to the Text Size parameter, this connection isn’t
very useful. The maximum value of the Blur Size control is 0.5, which is hardly noticeable
as a blur.
11 Set the Operator drop-down menu to Multiply.
12 Set the Second Operand slider to 100.
Now the first operand is multiplied by 100, and adjusting the dial gives you a much
blurrier blur.

Chapter 61 Modifiers 1476

 13 Switch to the Time tab of the modifier and set the First Operand Time Scale to -1.0.
Normally, the first operand gets the value of the control it is connected to from the same
frame as the current time. So at frame 10, the first operand is set to the same value as the
Text size at frame 10. By setting this value to -1, the value is read from one frame back in
time whenever the current time of the composition advances by 1 frame.
However, this means that the Calculation would be reading the value of the Text size at
frame -10 when we are at frame 10 in the composition.
14 To correct for this, set the First Operand Time Offset slider to 100.
15 Return to the Tools tab at the top of the Inspector and press Play (Spacebar) to watch how
the value of the Blur Size relates to the value of the Text Size.

CoordTransform Position
Because of the hierarchical nature of 3D in Fusion, the original position of an object in a 3D scene
often fails to indicate the current position of the object. For example, an image plane might initially
have a position at 1, 2, 1, but then be scaled, offset, and rotated by other tools further downstream in
the node tree, ending up with an absolute location of 10, 20, 5. This can complicate connecting an
object further downstream in the composition directly to the position of an upstream object. The
Coordinate Transform modifier can be added to any set of the XYZ coordinate controls to calculate the
current position of a given object at any point in the scene hierarchy. To add a Coordinate Transform
modifier, right-click the numeric input on any node, and select Modify With > CoordTransform Position
from the contextual menu.

Inspector

The Coordinate Transform modifier Controls tab

Controls Tab
The Controls tab has two fields for the target and scene input. The target is for the node continuing the
original coordinates, while the scene input is used for the scene with the new coordinates.

Target Object
This control is connected to the 3D tool that produces the original coordinates to be transformed. To
connect a tool, drag the node from the Node Editor into the text edit control, or right-click the control
and select the tool from the contextual menu. It is also possible to type the tool’s name directly into
the control.

SubID
The SubID slider can be used to target an individual sub-element of certain types of geometry, such as
an individual character produced by a Text 3D tool or a specific copy created by a Duplicate 3D tool.

Chapter 61 Modifiers 1477

Scene Input
This control should be connected to the 3D tool, which outputs the scene containing the object at the
new location. To connect a tool, drag and drop a tool tile from the Node Editor into the text edit
control, or right-click the control and select an object from the Connect To pop-up menu.

Cubic Spline
The Cubic Spline is another animation modifier in Fusion that is normally applied to numerical values
rather than point values. It can be applied by right-clicking a numerical control and selecting Modify
With > Natural Cubic Spline.

Usage
Being an animation spline, this modifier has no actual Controls tab. However, its effect can be seen
and influenced in the Spline Editor.

Cubic Spline Editor

Expression
An expression is a variable or a mathematical calculation added to a parameter, rather than a straight
numeric value. You can add an expression to any parameter in Fusion, or you can add the Expression
modifier, which adds several tabs to the modifier Inspector. Adding this modifier to a parameter adds
the ability to manipulate that parameter based on any number of controls, either positional or value-
based. This modifier offers exceptional flexibility compared to the more limited Calculation or Offset
modifiers, but it is unable to access values from frames other than the current time.
The Expression modifier accepts nine value inputs and nine position inputs that are used as part of a
user-defined mathematical expression to output a value.
To add the Expression modifier to a parameter, right-click the parameter in the Inspector and choose
Modify With > Expression from the contextual menu. The type of value returned by the Expression
depends entirely on the type of control it is modifying.
When used with a value control (like a slider), the Expression in the Number Out tab is evaluated to
create the result. When used to modify a positional control (like Center), the Point Out tab controls
the result.
The Inspector’s Modifiers tab contains the controls for the Expression modifier, described below.

Chapter 61 Modifiers 1478

Inspector

Expressions modifier controls

Controls Tab
This tab provides nine number controls and nine point controls. The values of the number controls can
be referred to in an expression as n1 through n9. The X-coordinate of each point control can be
referred to as p1x through p9x, while the Y-coordinate is p1y through p9y.
These values can be set manually, connected to other parameters, animated, and even connected to
other Expressions or Calculations.

The Number Out tab

Number Out Tab

Mathematical formulas are entered using the Number In and Point In values from the Controls tab.
The output modifies the control in which the Expression was applied. See below for the syntax to use
in this field.

The Point Out tab

Chapter 61 Modifiers 1479

Point Out Tab
The two text boxes in this tab use mathematical formulas, accessing the Number In and Point In values
from the Controls tab. The output value modifies the control in which the Expression was applied. The
Expression in the top text box control is used to calculate the X-axis value, and the bottom text box is
used to calculate the Y-axis control. See below for the syntax to use in this field.

The Expressions modifier Config tab

Config Tab
A good expression is reused over and over again. As a result, it can be useful to provide more
descriptive names for each parameter or control and to hide the unused ones. The Config Tab of the
Expressions modifier can customize the visibility and name for each of the nine point and
number controls.

Random Seed
The Random Seed control sets the starting number for the Rand() function. The rand(x, y) function
produces a random value between X and Y, producing a new value for every frame. As long as the
setting of this Random Seed slider remains the same, the values produced at frame x are always the
same. Adjust the Seed slider to a new value to get a different value for that frame.

Show Number or Point X

There are eighteen of these checkbox controls, one for each of the nine Number and nine Point
inputs. Enable this checkbox to display the control for Number x or Point x in the Controls tab.

Name for Number or Point X

There are eighteen of these edit controls, one for each of the nine Number and nine Point inputs. Type
a new name for the input into this edit control to assign a new name for the Input’s label in the
Controls tab.

Chapter 61 Modifiers 1480

Expression Syntax Formulas
Formulas are entered into the Number Out or Point Out tabs as part of an expression. They can be
made up of the following functions:

n1...n9 The value of Number Input 1..9.

p1x...p9x The X of Positional Control 1..9.

p1y...p9y The Y of Positional Control 1..9.

time The current time (frame number).

pi The value of pi.

e The value of e.

log(x) The base-10 log of x.

ln(x) The natural (base-e) log of x.

sin(x) The sine of x (x is degrees).

cos(x) The cosine of x (x is degrees).

tan(x) The tangent of x (x is degrees).

asin(x) The arcsine of x, in degrees.

acos(x) The arccosine of x, in degrees.

atan(x) The arctangent of x, in degrees.

atan2(x, y) The arctangent of x,y, in degrees.

abs(x) The absolute (positive) value of x.

int(x) The integer (whole) value of x.

frac(x) The fractional value of x.

sqrt(x) The Square Root of x.

rand(x, y) A random value between x and y.

rands(x, y, s) A random value between x and y, based on seed s.

min(x, y) The minimum (lowest) of x and y.

max(x, y) The maximum (highest) of x and y.

dist(x1, y1, x2, y2) The distance between point x1,y2 and x2,y2.

dist3d(x1,y1,z1,x2,y2,z2) The distance between 3D points x1,y2,z1 and x2,y2,z2

noise(x) A smoothly varying Perlin noise value based on x

noise2(x, y) A smoothly varying Perlin noise value based on x and y

noise3(x, y, z) A smoothly varying Perlin noise value based on x, y and z

if(c, x, y) Returns x if c not 0, otherwise y.

Chapter 61 Modifiers 1481

Expression Syntax Operators
Operators are used to evaluate statements. They are combined with functions to perform logical and
mathematical calculations in the Number Out and Point Out tabs.

x+y x plus y.

x-y x minus y.

x>=y 1.0 if x is less than y, otherwise 0.0.

x>y 1.0 if x is greater than y, otherwise 0.0.

!x 1.0 if x = 0, otherwise 0.0.

-x (0.0 - x).

+x (0.0 + x) (effectively does nothing).

x^y x raised to the power of y.

xy x multiplied by y.

xy x divided by y.

x%y x modulo y, (remainder of (x divided by y)).

x>=y 1.0 if x is less than or equal to y, otherwise 0.0.

x>=y 1.0 if x is greater than or equal to y, otherwise 0.0.

x=y 1.0 if x is exactly equal to y, otherwise 0.0.

x == y 1.0 if x is exactly equal to y, otherwise 0.0 (identical to above).

x<>y 1.0 if x is not equal to y, otherwise 0.0.

x != y 1.0 if x is not equal to y, otherwise 0.0 (identical to above).

x&y 1.0 if both x and y are not 0.0, otherwise 0.0.

x && y 1.0 if both x and y are not 0.0, otherwise 0.0 (identical to above).

x|y 1.0 if either x or y (or both) are not 0.0, otherwise 0.0.

x || y 1.0 if either x or y (or both) are not 0.0, otherwise 0.0 (identical to above).

Example 1
To make a numeric control equal to the Y value of a motion path, add an expression to the
desired target control and connect the Path to Point In 1. Enter the formula:

p1y

into the Number Out field.

Example 2
To make the result of the Expression’s Number Out be the largest of Number In 1 and Number
In 2, multiplied by the cosine of Number In 3, plus the X coordinate of Point In 1, enter
the formula:

max(n1, n2) * cos(n3) + p1x

into the Number Out field.

Chapter 61 Modifiers 1482

 Example 3
Add a Background node set to solid black and a Hotspot node. Set the Hotspot size to 0.08
and set the Strength to maximum. Modify the Hotspot center with an expression. Change the
current frame to 0.
Set n1 to 0.0 and add a Bézier Spline. At frame 29, set the value of n1 to 1.0. Select both points
and loop the spline using the Spline Editor. Now enter the following equations into the Point
Out tab of the expression.

X-Axis Expression

n1

Y-Axis Expression

0.5 + sin(time*50) 4

Render out a preview and look at the results. (Try this one with motion blur.)

From Image
The From Image modifier only works on gradients, like the gradient on a Background node. It takes
samples of an image along a user-definable line and creates a gradient from those samples.
Unlike other modifiers, From Image is not located in the Modify With menu. This modifier can be
applied by right-clicking a Gradient bar in the Inspector and selecting From Image.

Inspector

From Image modifier

Controls Tab
The From Image controls tab in the Inspector is used to select the node that contains the image you
want to sample. It allows you to define the starting and ending point in the image as well as how many
color samples to use in creating the gradient.

Image to Scan
Drop into this box the node from the Node Editor that you want to be color sampled.

Start X/Y, End X/Y

These two point controls define the Start and End points of the line along which the samples are taken
from the image defined in the Image to Scan box.
The points can also be moved directly in the viewer.

Chapter 61 Modifiers 1483

Number of Sample Steps
This defines how many individual color samples are taken along the line. You can also see the result of
this setting when you look at the actual node’s Gradient control. The more sample steps you define
here, the more individual color stops appear on the Gradient control. It is also possible to first create a
gradient using the From Image modifier and then remove that modifier from the Gradient control again.
The created gradient stays intact and can then be fine tuned by hand.

Edges
Edges determines how the edges of the image are treated when the sample line extends over the
actual frame of the image to be sampled.

Black
This outputs black for every point on the sample line outside of the image bounds.

Wrap
This wraps the edges of the line around the borders of the image.

Duplicate
This causes the edges of the image to be duplicated as best as possible, continuing the image beyond
its original size.

Color
This outputs a user-definable color instead of black for every point on the sample line outside of the
image bounds.

Example

The source image on the left shows the color selection line in red. The image on the right
shows the resulting gradient from that selection.

Gradient Color
The Gradient Color modifier allows you to create a customized gradient and map it into a specific time
range to control a value. Use the Start and End time controls to set the frames for the animation. If the
Start and End time values are set to 0, then the modifier returns the value at the starting point of the
gradient. You can use the Offset control to animate the gradient manually.
It can be applied by right-clicking a parameter and selecting Modify With > Gradient Color.

Chapter 61 Modifiers 1484

Inspector

The Gradient modifier Controls tab

Controls Tab
The Controls tab consists of a Gradient bar where you add and adjust points of the gradient. Start
Time and End Time thumbwheels at the bottom of the Inspector determine the time range the gradient
is mapped into.

Gradient
The Gradient control consists of a bar where it is possible to add, modify, and remove points of the
gradient. Each point has its color. It is possible to animate the color as well as the position of the point.
Furthermore, a From Image modifier can be applied to the gradient to evaluate it from an image.

Gradient Interpolation Method

The gradient is linearly interpolated from point to point in RGB color space by default. This can
sometimes result in unwanted colors. Choosing another color space may provide a better result.

Repeat
Defines how the left and right borders of the gradient are treated.

Gradients set to Once, Repeat, and Ping Pong from top to

bottom, respectively, and shifting the gradient to the left.

Chapter 61 Modifiers 1485

 – Once: When using the Gradient Offset control to shift the gradient, the border colors keep their
values. Shifting the default gradient to the left results in a white border on the left; shifting it to the
right results in a black border on the right.
– Repeat: When using the Gradient Offset control to shift the gradient, the border colors are
wrapped around. Shifting the default gradient to the left results in a sharp jump from white to
black; shifting it to the right results in a sharp jump from black to white.
– Ping Pong: When using the Gradient Offset control to shift the gradient, the border colors ping-
pong back and forth. Shifting the default gradient to the left results in the edge fading from white
back to black; shifting it to the right results in the edge fading from black back to white.

Gradient Offset
Allows you to pan through the gradient.

Time Controls
The Start Time and End Time thumbwheels determine the time range the gradient is mapped into. This
is set in frames. The same effect can be achieved by setting the Gradient to Once and animating the
offset thumbwheel.

Key Stretcher Modifier

The Keyframe Stretcher modifier is primarily used when creating title templates in Fusion for use in
DaVinci Resolve’s Edit page or Cut page. The Keyframe Stretcher modifier is added to an animated
parameter, so the keyframes in the animated parameter stretch when the template is trimmed in the
Timeline. It can be applied by right-clicking a parameter and selecting Modify with > KeyStretcher.
For more information on the Keyframe Stretcher Modifier controls, see the Keyframe Stretcher Node in
Chapter 49, “Miscellaneous Nodes” in the Fusion Reference Manual or Chapter 110 in the
DaVinci Resolve Reference Manual.

MIDI Extractor
The MIDI Extractor modifier provides the ability to modify the value of a control using the values stored
in a MIDI file. This modifier relies on some knowledge of MIDI, which is beyond the scope of
this manual.
The value produced by the modifier is extracted from the MIDI event selected in the Mode menu. Each
mode can be trimmed so that only specific messages for that event are processed—for example, only
some notes are processed, while others are ignored. The value of the event can be further scaled or
modified by additional factors, such as Scale, Velocity, Attack, and Decay.
It can be applied by right-clicking a parameter and selecting Modify With > MIDI Extractor.

Chapter 61 Modifiers 1486

Inspector

The MIDI Extractor modifier Controls tab

Controls Tab
The Controls tab is used to load the MIDI file, modify its timing, and determine which MIDI messages
and events trigger changes in the Fusion parameter.

MIDI File
This browser control is used to specify the MIDI file that is used as the input for the modifier.

Time Scale
Time Scale is used to specify the relationship between time as the MIDI file defines it and time as
Fusion defines it. A value of 1.0 plays the MIDI events at normal speed, 2.0 plays at double speed,
and so on.

Time Offset
Time Offset adjusts the sync between the MIDI file’s timing and Fusion’s timing. If there is an
unexpected delay, or if the MIDI file should start partway into or before some animation in Fusion, this
control can be used to offset the MIDI data as required.

Result Offset, Result Scale

These sliders adjust the range of values produced by the modifier. By default, values between 0 and 1
(or -1 and 1 for PitchBend mode) are generated. This does not always suit the node/parameter, and
scale can be used to make this range larger (such as * 0.0 - 2.0). Offset is used to provide some
constant value as a base.

Chapter 61 Modifiers 1487

Result Curve
The Result Curve can also be used to adjust the output. However, this adjusts the curve of the result.
By default, for any input MIDI data, the results fall linearly between 0.1 and 1.0 (for example, a velocity
127 note generates 1.0, whereas 63 generates approximately 0.5).
The Result Curve applies a gamma-like curve so that middle values can produce higher or lower
results while still maintaining the full scale.

Mode
This menu provides Beat, Note, Control Change, Poly AfterTouch, Channel AfterTouch, or Pitch Bend,
indicating from which MIDI event the values are being read. Beat mode is slightly different in that it
produces regular pulses based on the tempo of the MIDI file (including any tempo maps).
The Beat mode does not use any specific messages; it bases its event timing on the tempo map
contained in the MIDI file.

Combine Events
This menu selects what happens when multiple events occur at the same time. In Notes mode, this
can happen easily. For other events, this can happen if Multiple Channels are selected.
Use this to take the result from the most recent event to occur, the oldest event still happening, the
highest or lowest valued event, the average, sum, or the median of all events currently occurring.

Beat (Quarters) (Beat Mode Only)

This defines how often a beat occurs when in Beat mode. This is in quarter notes, so a value of 1.0
gives a beat every quarter.

Note Range (Note and Poly Aftertouch Modes Only)

This defines what range of notes causes a value to be generated. For example, use this to pick out the
kick drum from a GM drum track by setting the note range between 35–36.

Pitch Scale (Note Mode Only)

Pitch Scale defines how much the result changes with pitch. A value of 1.0 causes the result to vary
from 0.0 to 1.0 over the entire range.

Velocity Scale (Note Mode Only)

This defines how much the result changes with velocity. A value of 1.0 causes the result to vary from
0.0 to 1.0 over the entire range. This is added to the result from Pitch Scale for the final result.

Control Number (Control Change Mode Only)

This specifies the MIDI controller number from which to extract events.

Envelope Controls (Note and Beat Modes Only)

These define an Envelope to follow for values before, during, and after the note or beat. Pre-Attack
Time defines how long before the event it starts ramping up to the pre-attack level. Attack is the Time/
Level to ramp to once the event has occurred, followed by the Decay ramp and Sustain, until the event
stops. This stage is for Notes only. Beats have an instantaneous duration, so it goes straight to
Release. Release is the ramp-down time after the event finishes. When trying to do a Beat, set Release
to some value, or there likely will not be much of a beat.
These values can be used to follow actual sounds in the MIDI sequence or just to create interesting
effects. All time values used in the MIDI Extractor are in seconds.

Chapter 61 Modifiers 1488

The MIDI Extractor modifier Channels tab

Channels Tab
The Channels tab is used to select the Channels used in the modifier.

Channels
Channels checkboxes select which of the 16 channels in the MIDI file are actually considered for
events. This is a good way to single out a specific instrument from an arrangement.

About MIDI
A single MIDI interface allows 16 channels. Typically, these are assigned to different
instruments within a device or different devices. Usually, MIDI data is 7 bits, ranging from
0–127. In Fusion, this is represented as a value between 0–1 to be more consistent with how
data is handled in Fusion.
There are numerous different MIDI messages and events, but the ones that are particularly
useful with this modifier are detailed below.
MIDI Messages
– Note On: This indicates that a note (on a specific channel) is being turned on, has a pitch
(0–127, with middle C being 60) and a Velocity (0–127, representing how fast the key or
strings or whatever was hit).
– Note Off: This indicates that a note (on a specific channel) is being turned off, has a pitch
(0–127, with middle C being 60) and a Velocity (0–127, representing how fast the key or
strings or whatever was released).
– Control Change: This message indicates that some controller has changed. There are
128 controllers (0–127), each of which has data from 0–127. Controllers are used to set
parameters such as Volume, Pan, amount of Reverb or Chorus, and generic things like foot
controllers or breath controllers.
MIDI Events
– Channel Aftertouch: This event defines that pressure is being applied to the keys (or
strings or whatever) during a note. This represents general, overall pressure for this
channel, so it simply uses a pressure value (0–127).
– Poly Aftertouch: This event defines that pressure is being applied to the keys (or strings or
whatever) during a note. It is specific to each particular note and therefore contains a note
number as well as a pressure value (0–127).

Pitch Bend
The Pitch Bend controller generally specifies the degree of pitch bending or variation applied
to the note. Because pitch bend values are transmitted as a 14-bit values, this control has a
range between -1 and 1 and a correspondingly finer degree of resolution.

Chapter 61 Modifiers 1489

Natural Cubic Spline
The Natural Cubic Spline is one of the animation modifiers in Fusion and normally is applied to
numerical values rather than point values. It can be applied by right-clicking a numerical control and
selecting Modify With > Natural Cubic Spline.

NOTE: Unlike other spline types, Cubic splines have no control handles. They attempt to
automatically provide a smooth curve through the control points.

Usage
Being an animation spline, this modifier has no actual Controls tab. However, its effect can be seen
and influenced in the Spline Editor.

Offset (Angle, Distance, Position)

There are three Offset modifiers used to create variances between values. Depending on the modifier,
these values relate to controls, paths, and points. The three types of Offset modifiers available in
Fusion are:
– Offset Angle
– Offset Distance
– Offset Position

Offset Angle
The Offset Angle modifier outputs a value between 0 and 360 that is based on the angle between
two positional controls. The Position and Offset parameters may be static, connected to other
positional parameters, or connected to paths of their own. All offsets use the same set of controls,
which behave differently depending on the offset type used. These controls are described below.

Offset Distance
The Offset Distance modifier outputs a value that is based on the distance between two positional
controls. This modifier is capable of outputting a value based on a mathematical expression applied to
a position.

Chapter 61 Modifiers 1490

Offset Position
The Offset Position modifier generates a position (X and Y coordinates) that is based on the
relationship between positional controls. This modifier is the equivalent of a calculation control except
that it outputs X and Y coordinates instead of a value.
It can be applied by right-clicking a control and selecting Modify With > Offset.

Inspector

The Offset Position modifier controls

Offset Tab
The Inspector for all three Offset modifiers is identical. The Offset tab includes Position and Offset
values as well as a Mode menu for selecting the mathematical operation performed by the
offset control.

Position X and Y
The X and Y values are used by the Position to generate the calculation.

Offset X and Y
The X and Y values are used by the Offset to generate the calculation.

Flip Position Horizontal and Vertical

When these controls are selected, the Position is mirrored along the vertical or horizontal axis of
the image.

Flip Offset Horizontal and Vertical

When these controls are selected, the Offset position is mirrored along the vertical or horizontal axis
of the image.

Mode
The Mode menu includes mathematical operations performed by the Offset control. Available
options include:

– Offset – Use Position Only – Invert Position

– Difference (Position - Offset) – Use Offset Only – Invert Offset
– Difference (Offset - Position) – Maximum – Invert Sugar
– Average – Minimum – Random Offset

Chapter 61 Modifiers 1491

Image Aspect
Adjust the modifier’s output to compensate for the image aspect (not pixel aspect) of the project. A
square image of 500 x 500 would use an Image Aspect value of 1, and a rectangular image of 500 x
1000 would use an Aspect Value of 2. The default value is always based on the current frame format
selected in the preferences. To calculate image aspect, divide the width by the height. This control can
also be used to create the illusion of aspect.

The Offset Time tab

Time Tab
Position Time Scale
This returns the value of the Position at the Time Scale specified (for example, 0.5 is the value at half
the current frame time).

Position Time Offset

This returns the value of the Position at the Time Offset specified (for example, 10 is 10 frames back).

Offset Time Scale

This returns the value of the Offset at the Time Scale specified.

Offset Time Offset

This returns the value of the Offset at the Time Offset specified.

Example
This is a simple comp to illustrate one potential use of offsets.
1 Create a new Comp 100 frames long.
2 Create a node tree consisting of a black background and a Text node foreground
connected to a Merge.
3 In the Text Layout tab, use the Center X control to animate the text from the left side of the
screen to the right.
4 Move to frame 0.
5 In the Text tab in the Inspector, right-click the Size control and select Modify With > Offset
Distance from the contextual menu.
This adds two onscreen controls: a crosshair for the position and an X control for the
offset. These onscreen controls represent the Position and Offset controls displayed in
the Modifiers tab.
The size of the text is now determined by the distance, or offset, between the two
onscreen controls.
6 Drag the X onscreen control in the viewer to see how the distance from the crosshair
changes the size of the merge and by association the text.

Chapter 61 Modifiers 1492

 Both the crosshair and the X onscreen controls are animatable and can be connected to
other controls.
7 Position the X centered at the bottom of the viewer.
8 In the Inspector, select the Modifiers tab.
9 In the Offset on Text size section, right-click over Position and choose Connect To >
PathConnect the position value of the Offset to the existing path by right-clicking the
Position control and selecting Connect To > Path1 Position.
Play the comp to view the animation.
10 Now, the text shrinks near the center of the path (when the distance between the offset
and the path is at its minimum) and grows at its ends (where the distance between the
offset and the path is at its maximum).

Path
The Path modifier uses two splines to control the animation of points: an onscreen motion path
(spacial) and a Time spline visible in the Spline Editor (temporal). To animate an object’s position
control using a Path, right-click the Position control either in the Inspector or in the viewer and select
Path from the contextual menu. This adds a keyframe at the current position. You can begin creating a
path by moving the playhead and dragging the center position control in the viewer. The Spline Editor
shows a displacement spline for editing the temporal value, or “acceleration,” of the path.

Controls

The Path modifier Controls tab

Controls Tab
The Controls tab for the path allows you to scale, reposition, and rotate the path. It also provides the
Displacement parameter, allowing you to control the acceleration of an object attached to the path.

Center
The actual center of the path. This can be modified and animated as well to move the entire
path around.

Size
The size of the path. Again, this allows for later modification of the animation.

Chapter 61 Modifiers 1493

X Y Z Rotation
The path can be rotated in all three dimensions to allow for sophisticated controls.

Displacement
Every motion path has an associated Displacement spline in the Spline Editor. The Displacement
spline represents the position of the animated element along its path, represented as a value between
0.0 and 1.0. Displacement splines are used to control the speed or acceleration of an object’s
movement along the path.
To slow down, speed up, stop, or even reverse the motion of the control along the path, adjust the
values of the points for the path’s displacement in the Spline Editor or in the Inspector.
– A Displacement value of 0.0 in the Spline Editor indicates that the control is at the very beginning
of a path.
– A value of 1.0 indicates that the control is positioned at the end of the path.
– Each locked point on the motion path in the viewer has an associated point on the
Displacement spline.
– Unlocked points have a control point in the viewer but do not have a corresponding point on the
Displacement spline.

Heading Offset
Connecting to the Heading adjusts the auto orientation of the object along the path. For instance, if a
mask’s angle is connected to the path’s heading, the mask’s angle will adjust to follow the angle
of the path.

Right-Click Here for Shape Animation

It’s possible to animate the shape of the path as well or to connect it to other path controls like Polyline
Masks or Paint Strokes.

TIP: Switching Default Paths

You can change the default path type used when animating a position or center control to a
path (if this is the preferred type of animation). Open the Preferences window and select the
Global Settings. In the Default category, select the Point With menu and choose Path. The
next time Animate is selected from a Position or Center control’s contextual menu, a
path is used.

Perturb
The Perturb modifier generates smoothly varying random values for a given parameter based on
Perlin noise. It can be used to add jitter, shake, or wobble to any animatable parameter, even if the
parameter is already animated. Its results are similar to those of the Shake modifier, although it uses a
different set of controls that may be more intuitive. Unlike other random modifiers, you can apply the
Perturb modifier to polylines, shapes, grid meshes, and even color gradients.
For example, to add camera shake to an existing path, right-click the crosshair and choose Insert >
Perturb, and then adjust the Strength down to suit. Alternatively, right-clicking the path’s “Right-click
here for shape animation” label at the bottom of the Inspector lets you apply perturb to the path’s
polyline. This works best if the polyline has many points—for example, if it has been tracked or
hand-drawn with the Draw Append pencil tool. A third usage option is to use the Insert contextual
menu to insert the modifier onto the Displacement control. This causes the motion along the path to
jitter back and forth without actually leaving the path.

Chapter 61 Modifiers 1494

 NOTE: Perturb can only add jitter; it cannot smooth out existing animation curves.

Inspector

The Perturb modifier Controls tab

Controls Tab
The Controls tab for Perturb is mainly used for controlling the Strength, Wobble, and Speed
parameters of the random jitter.

Value
The content of this control depends on what type of control the modifier was applied to. If the Perturb
modifier was added to a basic Slider control, the Value is a slider. If it was added to a Gradient control,
then a Gradient control is displayed here. Use the control to set the default, or center value, for the
Perturb modifier to work on.

The Perturb modifier Gradient controls

Chapter 61 Modifiers 1495

Jaggedness
(Polylines and meshes only) This allows you to increase the amount of variation along the length of the
polyline or mesh, rather than over time. Increasing Jaggedness gives a squigglier polyline or more
tangled mesh, independent of its movement.

Phase
(Polylines and meshes only) Animating this can be used to move the ripple of a polyline or mesh along
itself, from end to end. The effect can be most clearly seen when Speed is set to 0.0.

Random Seed Randomize

The Random Seed is used to “seed” the amount of jitter applied by the modifier. Two Perturb modifiers
with identical settings, but different random seeds, produce two completely different results. Click the
Randomize button to assign a random seed value.

Strength
Use this control to adjust the strength of the Perturb modifier’s output, or its maximum variation from
the primary value specified above.

Wobble
Use the Wobble control to determine how smooth the resulting values are. Less wobble implies a
smoother transition between values, while more wobble produces less predictable results.

Speed
Increasing the Speed slider value speeds up the rate at which the value changes. This can increase
the apparent wobbliness in a more predictable fashion than the Wobble control and make the jitter
more frantic or languorous in nature.

Probe
The Probe modifier is one of the most versatile modifiers in Fusion. It allows you to control any numeric
parameter by the color or luminosity of a specific pixel or rectangular region of an image. Think of
driving the Brightness node by probing the pixel values of flickering lights in a shot, or measuring
graded LUTs to compare values.
It can be applied by right-clicking a parameter and selecting Modify With > Probe.

Inspector

The Probe modifier Controls tab

Controls Tab
The Controls tab for the Probe modifier allows you to select the node to probe, define the channel
used to drive the parameter, and control the size of the probed area.

Image to Probe
Drag a node from the Node Editor to populate this field and identify the image to probe.

Chapter 61 Modifiers 1496

Channel
Select the channel you want to probe. The usual options are:
– Red
– Green
– Blue
– Alpha

Luma
Once a Probe modifier is present somewhere in your comp, you can connect other node’s values to its
outputs as well. The Probe allows you to connect to its values individually:
– Result
– Red
– Green
– Blue
– Alpha

Position X Y
The position in the image from where the probe samples the values.

Probe Rectangle
By default, the Probe samples only the value of a single pixel at its position. By using the Probe
Rectangle mode, you can sample from a larger area of pixels based on the Evaluation method.

Width Height Controls

These determine the size of the area to be probed.

Evaluation
Sets how the pixels inside the rectangle are computed to generate the output value.
Options include:
– Average: All pixel values inside the rectangle are averaged.
– Minimum: The smallest value of all pixels inside the rectangle is used.
– Maximum: The highest value of all pixels inside the rectangle is used.

The Probe modifier Value tab

Value Tab
The Value tab controls the range or scale of the modifier adjustment, thereby adjusting the sensitivity
of the Probe.

Chapter 61 Modifiers 1497

Scale Input
By default, the Probe generates the Black Value when the probed area results in a value of 0 (i.e.,
black), and it generates its White Value when the probed area results in a value of 1 (i.e., white). By
using this range control, you can modify the sensitivity of the Probe.

Black Value
The value that is generated by the Probe if the probed area delivers the result set in Scale Input Black.

White Value
The value that is generated by the Probe if the probed area delivers the result set in Scale Input White.

Out of Image Value

The value that is generated by the Probe if the probed area is outside the frame boundaries of the
probed image. If probing a rectangle, this value does not generate before the entire rectangle is
outside the frame boundaries of the image to be probed.

Publish
Only parameters that are animated will be available from the Connect To menu. To connect to non-
animated parameters, you must Publish them first. Animated controls are automatically published,
whereas static controls must be published manually.
To publish a static control, right-click the control and select Publish from the contextual menu.

The Publish modifier Controls tab

Controls Tab
The Controls tab shows the published control available for linking to other controls.

Published Value
The display of the published control is obviously dependent on which control is published from
which node.

Resolve Parameter
The Resolve Parameter Modifier is used when creating a transition template in Fusion for use in
DaVinci Resolve’s Edit page or Cut page. When building a transition in Fusion, the Resolve Parameter
modifier is added to any control you want to animate. The Resolve Parameter modifier automatically
animates the parameter for the duration of the transition, allowing you to trim the transition in the Edit
page or Cut page.

Chapter 61 Modifiers 1498

For example, to create a cross dissolve, do the following:
1 From the Effects Library, add a Fusion Composition to the Edit page Timeline.
2 In Fusion, add a Dissolve node to the Node Editor.
3 In the Inspector, right-click the Background/Foreground parameter, and then choose Resolve
Parameter from the Modifier contextual menu. Adding the modifier to the Background/Foreground
parameter automatically updates the slider if the transition is modified back on the Edit/Cut page.
4 In the Node Editor, right-click the Dissolve node and choose Macro > Create Macro.
5 When creating a macro that’s to be used as a Fusion transition it’s important that two inputs and
one output are selected in the Macro Editor. In this example, under the Dissolve heading, enable
the Output, Background and Foreground check boxes.
6 Give the transition a name, then save the macro from the top File menu.
7 Save the macro in the following folder so it appears the Edit page or Cut page Effects Library:
– For MacOS: $TEMPLATE_MAC_OS_PATH/Transitions or $TEMPLATE_MAC_USER_PATH/
Transitions
– For Windows: $TEMPLATE_WIN_OS_PATH\Transitions or $TEMPLATE_WIN_USER_PATH\
Transitions

8 Quit and reopen DaVinci Resolve to update the list of transitions in the Effects Library.
9 On the Edit page, open the Effects Library. Navigate to Video Transitions Fusion Transitions, and
the custom Fusion transition will be listed.

Shake
The Shake modifier is used to randomize a Position or Value control to create semi-random numeric
inputs. The resulting shake can be entirely random. The motion can also be smoothed for a more
gentle, organic feel.
To add the Shake modifier to a parameter, select Modify With > Shake from the parameter’s contextual
menu. The Shake modifier uses the following controls to achieve its effect. It can be applied by
right-clicking a parameter and selecting Modify With > Shake.

Inspector

The Shake modifier Controls tab

Chapter 61 Modifiers 1499

Controls Tab
Random Seed
The Random Seed control contains the value used to seed the random number generator. Given the
same seed, a random number generator always produces the same results. Change the seed if the
results from the randomizer are not satisfying.

Smoothness
This control is used to smooth the overall randomness of the Shake. The higher the value, the
smoother the motion appears. A value of zero generates completely random results, with no
smoothing applied.

Lock X/Y
This checkbox is used to unlock the X- and Y-axis, revealing independent slider controls for each axis.

Minimum and Maximum

This control is used to determine the overall strength of the shake. The low values represent the
lowest value that can be generated by the randomizer, and the high values represent the highest
values. To create a Shake that moves a center crosshair anywhere within the image, set the Minimum
to 0.0 and the Maximum to 1.0. To restrict the motion to a smaller shake in the bottom-right corner of
the image, set the Minimum to 0.70 and the Maximum to 0.90.

Example
1 Create a new comp, and then add and view a Text node.
2 Type some text in the Text node.
3 In the viewer, right-click over the Center control of the text and choose Modify With >
Shake Position.
4 In the Inspector, select the Modifiers tab and set the smoothing to 5.0.
5 Set the Minimum to 0.1 and the Maximum to 0.9.
This adds some chaotic movement to the text. However, we can change this over Go to
frame 0 and in the Inspector click the Keyframe button to the right of both the Minimum
and the Maximum controls.
6 Go to frame 0 and in the Inspector click the Keyframe button to the right of both the
Minimum and the Maximum controls.
7 Go to frame 90 and adjust the Minimum to 0.45 and the Maximum to 0.55.
8 View the results.
Now, the text starts out by flying all over the screen and tightens in toward the center of
the screen as the comp plays.

Chapter 61 Modifiers 1500

Track
Although there is a standard Tracker node, you can also use a Tracker modifier to add a tracker
directly to a parameter. To apply the Tracker modifier, in the viewer right-click the Center control of of
any transform, text, mask, or other positionable element. From the contextual menu, choose Object x
Center > Modify With > Tracker. Then choose one of three options:
– Tracker Position: Tracks a point from the source.
– Steady Position: Stabilizes based on a single point in the source.
– Unsteady Position: Adds original motion back after stabilizing.
This adds a modifier in the Inspector with a set of controls almost identical to those found in the
Tracker node itself.

Inspector

The Tracker modifier

For an in-depth explanation of this node, see Chapter 57, “Tracker Nodes” in the Fusion Reference
Manual or Chapter 118 in the DaVinci Resolve Reference Manual.

Tracker Modifier vs. Tracker Node

The differences between a Tracker modifier and a Tracker node are as follows:
– The Tracker modifier can only track a single pattern.
– The Tracker modifier can only output a single value and cannot be used for complex stabilization
or match-moving procedures.
– The default source image for the modifier is the node immediately upstream of the node that
contains the modifier (i.e., when adding a Tracker modifier to a Glow node with a Loader as its
input, the Tracker Source input defaults to the output of the Loader). Set a different source image
for the Tracker modifier by typing in the name of the node. Alternatively, drag the source node
from the Node Editor into the Text Box control.

Chapter 61 Modifiers 1501

 Example
Imagine that you need to track an actor’s eyes so that an unearthly, alien glow can be applied
to the eyes.
1 Select the Loader node.
2 Add a Glow node.
3 Add an Ellipse mask to the Glow in the shape of one of the eyes.
4 Right-click the center of that mask and select Modify With > Tracker > Position.
Since the track is on the mask, the tracker takes the glow as the image for tracking. This
could cause problems since the eye might be very obscured by the glow. A cleaner
source will be the Loader that feeds the glow.
5 Drag the Loader into the modifier Inspector’s Track Source field.
6 Track the actor’s eye.
7 Perform the same steps on the other eye.

Vector Result
The Vector Result modifier is used to offset positional controls, such as crosshairs, by distance and
angle. These can be static or animated values.
It can be applied by right-clicking a control and selecting Modify With > Vector.

Inspector

The Vector Result modifier Controls tab

Controls Tab
Origin
This control is used to represent the position from which the vector’s distance and angle values
originate.

Distance
This slider control is used to determine the distance of the vector from the origin.

Angle
This thumbwheel control is used to determine the angle of the vector relative to the origin.

Chapter 61 Modifiers 1502

Image Aspect
This slider control is used primarily to compensate for image aspect differences. A square image of
500 x 500 would use an Image Aspect value of 1, while a rectangular image of 500 x 1000 would use
an Image Aspect value of 2. The default for this value is taken from the current Frame Format
preferences using width/height. It may be necessary to modify this control to match the current image.

Example
1 Create a 100-frame comp.
2 Create a simple node tree consisting of a black background and a Text node foreground
connected to a Merge.
3 Enter some text in the Text node.
4 Select the Merge node.
5 In the viewer, right-click the Center control of the Merge and choose Modify With >
Vector Result.
This adds a crosshair onscreen control for the Vector distance and angle. The onscreen
control represents the Distance and Angle controls displayed in the Modifiers tab.
6 In the Modifiers tab of the Inspector, drag the Distance control to distance the text from
the Vector origin.
7 Drag the Angle thumbwheel to rotate the text around the Vector origin.
This is different from changing a pivot point, since the text itself is not rotating.
These points are animatable and can be connected to other controls.
8 In the Inspector, right-click the Origin control and choose a path to add a motion path
modifier to the Origin control.
9 Verify that the current frame is set to frame 0 (zero) and use the Origin controls in the
Inspector or drag the Vector Origin crosshair to the bottom-left corner of the screen.
10 On the Vector Angle thumbwheel, click the Keyframe button to animate this control.
11 Set the Angle thumbwheel to a value of 10.
12 Go to frame 100 and click at the top-left corner of the screen to move the Vector Origin
crosshair.
13 Set the Vector Angle thumbwheel to a value of 1000.
14 Play the comp to see the results.
This causes the text to orbit around the path just created.

XY Path
The XY Path type uses two separate splines for the position along the X-axis and for the position
along the Y-axis.
To animate a coordinate control using an XY path, right-click the control and select Modify With > XY
Path from the contextual menu.
At first glance, XY paths work like Displacement paths. To describe the path, change frames and
position the control where it should be on that frame, and then change frames again and move the
control to its new position. Fusion automatically interpolates between the points. The difference is that
no keyframes are created on the onscreen path.

Chapter 61 Modifiers 1503

Look in the Spline Editor to find the X and Y channel splines. Changes to the control’s values are
keyframed on these splines. The advantage to the XY path is that it becomes very easy to work with
motion along an individual axis.

Inspector

XY Path modifier controls

X Y Z Values
These reflect the position of the animated control using X, Y, and Z values.

Center
The actual center of the path. This can be modified and animated as well to move the entire
path around.

Size
The size of the path. Again, this allows for later modification of the animation.

Angle
The angle of the path. Again, this allows for later modification of the animation.

Heading Offset
If another control (for example, a mask’s Angle) is connected to the path’s heading, this control allows
for adding or subtracting from the calculated angle.

Plot Path in View

Toggles whether or not the actual path is displayed in the views.

Switching Default Paths

You can change the Default Path type to XY path (if this is the preferred type of animation). Open the
Default category in the Global Preferences and locate the Point With drop-down menu. Change this
from the current value to XY Path. The next time Animate is selected from a Coordinate control’s
contextual menu, an XY path is used instead of a Displacement path.

Chapter 61 Modifiers 1504

 PART 6

Other Information
DaVinci Resolve 17

Regulatory Notices,
Safety Information
and Warranty

Contents
Regulatory Notices  1507
Safety Information  1509
Warranty  1510

DaVinci Resolve 17 Regulatory Notices, Safety Information and Warranty 1506

Regulatory Notices

Disposal of Waste of Electrical and Electronic Equipment Within the European Union.
The symbol on the product indicates that this equipment must not be disposed of with other waste
materials. In order to dispose of your waste equipment, it must be handed over to a designated
collection point for recycling. The separate collection and recycling of your waste equipment at the
time of disposal will help conserve natural resources and ensure that it is recycled in a manner that
protects human health and the environment. For more information about where you can drop off your
waste equipment for recycling, please contact your local city recycling office or the dealer from whom
you purchased the product.

This equipment has been tested and found to comply with the limits for a Class A digital device,
pursuant to Part 15 of the FCC rules. These limits are designed to provide reasonable protection
against harmful interference when the equipment is operated in a commercial environment.
This equipment generates, uses, and can radiate radio frequency energy and, if not installed and used
in accordance with the instructions, may cause harmful interference to radio communications.
Operation of this product in a residential area is likely to cause harmful interference, in which case the
user will be required to correct the interference at personal expense.
Operation is subject to the following two conditions:
1 This device may not cause harmful interference.
2 This device must accept any interference received,
including interference that may cause undesired operation.

Davinci Resolve Advanced Panel KCC-REM-BMD-DaVinciResolve

Davinci Resolve Mini Panel MSIP-REM-BMD-201708001
Davinci Resolve Micro Panel MSIP-REM-BMD-201703002
Davinci Resolve Studio USB Keylock MSIP-REM-BMD-201705001
DaVinci Resolve Editor Keyboard R-R-BMD-201907001
DaVinci Resolve Speed Editor R-R-BMD-20200211001

Fairlight Desktop Audio Editor R-R-BMD-2020103002

Fairlight Console Audio Editor R-R-BMD-2020103002
Fairlight Console LCD Monitor R-R-BMD-2020103003
Fairlight Console Channel Fader R-R-BMD-2020103004
Fairlight Console Channel Control R-R-BMD-2020103005
Fairlight PCIe Audio Accelerator R-R-BMD-2020103006
Fairlight Audio Interface R-R-BMD-2020103007
Fairlight PCIe Audio MADI Upgrade R-R-BMD-2020103008
Fairlight Desktop Console R-R-BMD-20200728001
Fairlight HDMI Monitor Interface R-R-BMD-20200729001

DaVinci Resolve 17 Regulatory Notices 1507

ISED Canada Statement
This device complies with Canadian standards for Class A digital apparatus.
Any modifications or use of this product outside its intended use could void
compliance to these standards.
Connection to HDMI interfaces must be made with high quality shielded HDMI cables.
This equipment has been tested for compliance with the intended use in a commercial environment.
If the equipment is used in a domestic environment, it may cause radio interference.

Bluetooth®
The DaVinci Resolve Speed Editor is a Bluetooth wireless technology enabled product.
Contains transmitter module FCC ID: QOQBGM113
This equipment complies with FCC radiation exposure limits set forth for an uncontrolled environment.
Contains transmitter module IC: 5123A-BGM113
This device complies with Industry Canada’s license-exempt RSS standards and exception from routine
SAR evaluation limits given in RSS-102 Issue 5.
Certified for Japan, certificate number: 209-J00204. This equipment contains specified radio equipment
that has been certified to the technical regulation conformity certification under the radio law.
This module has certification in South Korea, KC certification number: MSIP‑CRM-BGT-BGM113

Pending Certification for South Africa by ICASA, approval number TA-2021/XXXX

Cerified for Mexico (NOM), for Bluetooth module manufactured by Silicon Labs, model number BGM113A
Includes transmitter module certified in Mexico IFT: RCBSIBG20-2560

Hereby, Blackmagic Design declares that the product (DaVinci Resolve Speed Editor) is using wideband
transmission systems in 2.4 GHz ISM band is in compliance with directive 2014/53/EU.
The full text of the EU declaration of conformity is available from [email protected]

DaVinci Resolve 17 Regulatory Notices 1508

Safety Information
Weight Warning
The Fairlight Console has considerable weight even when empty. For example, a 3 Bay console
weighs up to 110 kg empty, and 157 kg fully assembled. You should always move a Fairlight console
with at least 4 people using safe lifting procedures, such as keeping the back straight, bending the
knees and lifting with careful, controlled movements.

Electrical Warning Notice and Disclaimer

For installations involving the fitting of more than five Fairlight modules, additional earthing
requirements must be fitted before connecting the supply. This requirement does not apply if
each group of five Fairlight modules can be connected to separate wall or floor socket outlets.
Earth posts are welded internally at both ends of the console frame for connecting earth wires
from the console frame to the building earth point. Either of these posts can be used and they
are marked with the following label.

Blackmagic Design recommends appointing a qualified and licenced electrician to install, test
and commission this wiring system.
Blackmagic Design does not accept responsibility for the safety, reliability, damage or personal
injury caused to, or by, any third-party equipment fitted into the console.

For protection against electric shock, the equipment must be connected to a mains socket outlet
with a protective earth connection. In case of doubt contact a qualified electrician.
To reduce the risk of electric shock, do not expose this equipment to dripping or splashing.
Product is suitable for use in tropical locations with an ambient temperature of up to 40°C.
Ensure that adequate ventilation is provided around the product and that it is not restricted.
When rack mounting, ensure that the ventilation is not restricted by adjacent equipment.
No operator serviceable parts inside product. Refer servicing to your local Blackmagic Design
service center.
The DaVinci Resolve Speed Editor contains a single cell Lithium battery. Keep lithium batteries away
from all sources of heat, do not use the product in temperatures greater than 40°C.

Use only at altitudes not more than 2000m above sea level.

State of California statement

This product can expose you to chemicals such as trace amounts of polybrominated biphenyls
within plastic parts, which is known to the state of California to cause cancer and birth defects
or other reproductive harm.
For more information go to www.P65Warnings.ca.gov.

DaVinci Resolve 17 Safety Information 1509

Warranty
12 Months Limited Warranty
Blackmagic Design warrants that DaVinci Resolve color grading control panels, editing keyboards and
audio consoles will be free from defects in materials and workmanship for a period of 12 months from the
date of purchase. If a product proves to be defective during this warranty period, Blackmagic Design, at
its option, either will repair the defective product without charge for parts and labor, or will provide a
replacement in exchange for the defective product. Periodical updates to the operational software are
not included under this warranty.
In order to obtain service under this warranty, you the Customer, must notify Blackmagic Design of the
defect before the expiration of the warranty period and make suitable arrangements for the performance
of service. The Customer shall be responsible for packaging and shipping the defective product to a
designated service center nominated by Blackmagic Design, with shipping charges pre paid. Customer
shall be responsible for paying all shipping changes, insurance, duties, taxes, and any other charges for
products returned to us for any reason.
This warranty shall not apply to any defect, failure or damage caused by improper use or improper or
inadequate maintenance and care. Blackmagic Design shall not be obligated to furnish service under
this warranty: a) to repair damage resulting from attempts by personnel other than Blackmagic Design
representatives to install, repair or service the product, b) to repair damage resulting from improper use
or connection to incompatible equipment, c) to repair any damage or malfunction caused by the use of
non Blackmagic Design parts or supplies, or d) to service a product that has been modified or integrated
with other products when the effect of such a modification or integration increases the time or difficulty
of servicing the product.
THIS WARRANTY IS GIVEN BY BLACKMAGIC DESIGN IN LIEU OF ANY OTHER WARRANTIES, EXPRESS
OR IMPLIED. BLACKMAGIC DESIGN AND ITS VENDORS DISCLAIM ANY IMPLIED WARRANTIES OF
MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. BLACKMAGIC DESIGN’S RESPONSIBILITY
TO REPAIR OR REPLACE DEFECTIVE PRODUCTS IS THE WHOLE AND EXCLUSIVE REMEDY PROVIDED
TO THE CUSTOMER FOR ANY INDIRECT, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES
IRRESPECTIVE OF WHETHER BLACKMAGIC DESIGN OR THE VENDOR HAS ADVANCE NOTICE OF
THE POSSIBILITY OF SUCH DAMAGES. BLACKMAGIC DESIGN IS NOT LIABLE FOR ANY ILLEGAL USE
OF EQUIPMENT BY CUSTOMER. BLACKMAGIC IS NOT LIABLE FOR ANY DAMAGES RESULTING FROM
USE OF THIS PRODUCT. USER OPERATES THIS PRODUCT AT OWN RISK.
© Copyright 2021 Blackmagic Design. All rights reserved. ‘Blackmagic Design’, “DaVinci’, ‘Resolve’, ‘DeckLink’, ‘HDLink’, ‘Videohub’,
‘DeckLink’, and ‘Leading the creative video revolution’ are registered trademarks in the US and other countries. All other company and
product names may be trademarks of their respective companies with which they are associated. Thunderbolt and the Thunderbolt logo are
trademarks of Intel Corporation in the U.S. and/or other countries. Dolby, Dolby Vision, and the double-D symbol are registered trademarks
of Dolby Laboratories Licensing Corporation.

DaVinci Resolve 17 Warranty 1510

blackmagicdesign.com