After Effects support

View as Markdown

Check which After Effects features render in Blitframe, one table per kind of feature. The tables cover a composition exported with Blitframe for After Effects or with Bodymovin (After Effects). The notes say where the two differ.

  • Yes means the feature renders. The notes give its limits.
  • No means it does not render. The notes say what you get instead.

After you upload a composition, read the version's problems and warnings (check and fix). They point out much of what will not render as designed, such as an unsupported effect. The composition format reference gives the exact limits, and how to write each feature in data.json.

Layers

Feature Supported Notes
Solid Yes
Image Yes PNG, JPEG, WebP, AVIF, HEIC, TIFF, BMP, TGA, GIF or PSD; animated GIF and WebP play once as video, then hold their last frame
Text Yes
Shape Yes
Video Yes MP4, MOV, WebM, MKV, AVI and more (see Projects); alpha is kept; MOV or MP4 with Apple ProRes 4444, QuickTime Animation (RLE) or the PNG codec, upright and at most 4096 pixels on each side, is used as it is, and other video with alpha is converted first. Export alpha video premultiplied (matted against black). Without time remapping, a video plays once at normal speed. After its last frame, that frame stays on screen until the layer ends
3D model Yes Self-contained .glb or .gltf. Bodymovin does not export model layers; with it, add the layer to data.json as Native models shows
Null Yes
Camera, one-node and two-node Yes
Precomposition Yes
Adjustment layer Yes
Audio layer No Set the version's soundtrack in an edit, or name an audio file in the render request; see Renders

Transforms

Feature Supported Notes
Position, 2D and 3D Yes Separated dimensions: X and Y only, so a 3D position stays unified
Rotation, 2D and 3D Yes
Scale, 2D and 3D Yes
Anchor point, 2D and 3D Yes
Orientation Yes
Parenting Yes
Opacity Yes
Auto-orient Yes Needs position with unified (not separated) dimensions

Compositions

Feature Supported Notes
Pixel aspect ratio Square pixels only
Collapse transformations Yes Not when the precomposition layer has masks, effects, layer styles or a track matte (as the matte or the layer using one); it then renders as an ordinary precomposition. A collapsed precomposition does not get its layer's opacity or blend mode, and it is not cropped to its own size
Time remapping Yes On precomposition and video layers. A remap that runs backwards plays video in reverse, though H.264 video played backwards can show the frame just before the one the curve asks for. Past the end of the footage, the last frame holds. Video inside a time-remapped precomposition ignores the remap and plays at normal speed
Time stretch Precompositions only A precomposition layer's stretch speeds up or slows down its contents, unless the layer is time-remapped. Video layers, including video inside a stretched precomposition, play at normal speed. The version's warnings name each stretch the engine does not apply, and each stretched precomposition with video in it
Protected regions No

Animation

Feature Supported
Linear interpolation Yes
Bezier interpolation Yes
Hold interpolation Yes
Spatial Bezier interpolation Yes
Rove across time No

Track mattes

Feature Supported
Alpha matte Yes
Alpha inverted matte Yes
Luma matte Yes
Luma inverted matte Yes

A layer uses its matte only while the matte layer draws. The matte layer may be marked hidden in data.json (hd: true), or be outside its in and out points. Then a layer using an alpha or luma matte is hidden, and one using an inverted matte is drawn without its matte. The version's warnings say so. A matte layer wholly outside the frame also counts as not drawing, unless it is a shape layer or has an effect or layer style that spreads past its edges, such as a blur, glow or shadow.

The matte layer's own blend mode and its own track matte are ignored, so mattes don't chain.

Masks

Feature Supported Notes
Add, Subtract, Intersect, Difference Yes
Lighten, Darken No
Mask opacity Yes
Mask feather Yes Static; an animated feather keeps its first value
Mask expansion No
Several masks per layer Yes
Mask path animation Yes

Shape layers

Feature Supported Notes
Fill and gradient fill Yes Gradients are static: the first keyframe is drawn for the whole render, a radial gradient's highlight is ignored, and group opacity does not fade them
Stroke and gradient stroke Yes Fixed width only; an animated stroke width hides the stroke
Fill rules Yes
Paths, fills and strokes outside groups Yes Drawn in front of every shape group
Several shape groups Yes
Several paths, fills and strokes in a group Yes Up to 10 paths and 8 paints per group
Nested shape groups No Skipped with everything in them
Path animation Yes Keep the vertex count and closure the same on every keyframe: a morph whose keyframes differ holds its first keyframe (a closed path of n vertices and an open one of n + 1 still morph, drawn like the first keyframe)
Star roundness No Straight edges
Trim paths Yes One path per trim: a Trim Paths below two or more paths in its group is skipped. Start, end and offset can be animated. The fill of a rounded rectangle is not trimmed: it draws whole
Repeater, merge paths, rounded corners, offset paths, pucker & bloat, zig zag, twist, dashes No Ignored: the shape draws without them
Hidden shape items (eye turned off) No Drawn anyway, and a hidden Trim Paths still trims; delete them instead
Blend modes on fills and strokes No Painted normally; use the layer's blend mode

Blend modes

Feature Supported Notes
Normal, Multiply, Screen, Overlay, Darken, Lighten, Color Dodge, Color Burn (and their Classic versions), Hard Light, Soft Light, Difference, Exclusion, Hue, Saturation, Color, Luminosity, Add, Hard Mix Yes
Linear Burn, Linear Dodge, Vivid Light, Linear Light, Pin Light, Subtract Yes Blitframe for After Effects exports them as they are. Bodymovin exports them as Normal, so with it, set them in data.json (blend modes)
Darker Color, Lighter Color, Divide, Dissolve, Dancing Dissolve, the stencil and silhouette modes, Alpha Add, Luminescent Premul No Rendered as Normal

Text

Feature Supported Notes
Alignment, fill, stroke Yes Left, right and center. Justified paragraphs render centered. A stroke always has round joins and ends, whatever its line join
Tracking, line spacing, baseline shift Yes
Kerning Yes The font's own kerning; the kerning type setting is ignored
Mixed right-to-left and left-to-right text Yes
Unicode (several languages) Yes, with limits Any character the font has. Arabic gets its joining forms; scripts that need reordering or conjunct forms, such as Devanagari, do not shape correctly
Ligatures No Text is laid out as After Effects lays it out by default, without ligature substitution, required ones such as Arabic lam-alef included
All Caps, Small Caps No The text renders as typed
Text on a path Yes On paragraph text the path wins: the box size and its wrapping are ignored. An animated path holds its first keyframe
Paragraph (box) text Yes The box shows a line when at least half of it fits, and always shows the first line. A word wider than the box breaks where it overflows, and a letter wider than the box ends the text there
Anchor point grouping and grouping alignment Yes Alignment applies to word, line and all grouping; character grouping turns each character about its own center on the baseline
Fill & stroke order Yes Stroke over fill or under it
Inter-character blending Default only
Faux bold and faux italic No Use the font's real bold or italic face
Emoji No
Text scale (in the Character panel) No Scale the layer or use a text animator instead
Animated Source Text No Only the first text keyframe is used

A render holds 199 text and 3D model meshes in all: one per text layer, two per stroked one, and one per 3D model asset a model layer uses, however many layers show it. Hidden text layers and every copy of a precomposition count. A composition that needs more is reported in the version's problems (limits).

Fonts come from the files in your project or from the Blitframe font library (Fonts). To keep changed copy inside its design, use text fitting.

Text animators

Feature Supported
Position, rotation, scale, opacity Yes
Skew Yes, up to ±70°
X and Y rotation Yes, without perspective: characters, words or lines narrow or flatten as they turn about their anchor
Tracking, line spacing Yes. Animated tracking keeps each line's left edge in place on left-aligned and justified text, its right edge on right-aligned text and its middle on centered text
Fill color Yes
Opacity on a stroked text Yes, it fades the fill and the stroke together
Fill color on a stroked text Yes, it changes the fill only; the stroke keeps its own color
Stroke color Yes
Fill or stroke hue, saturation, brightness and opacity, stroke width, character offset No
Anchor point Yes. It moves the point each character, word or line scales, rotates and skews about, starting from where Anchor Point Grouping and Grouping Alignment put it
Blur No
Randomize order No
Range selector units: percentage and index Yes
Based on: characters, characters excluding spaces, words, lines Yes
Mode: Add Yes
Mode: Subtract, Intersect, Min, Max, Difference No, rendered as Add
Shapes: square, ramp up, ramp down, triangle, round, smooth Yes

The version's warnings name the unsupported animator properties in this table, except character offset. They also name any mode other than Add.

Expressions

Feature Supported Notes
Where expressions run Partly On a layer's position, anchor point, scale, rotation and opacity, and in text expression selectors. On any other property (shapes, effects, layer styles, masks) the value renders without its expression
Arithmetic operators Yes
Math and standard JavaScript (variables, if/else, loops, functions, arrays) Yes Declare top-level variables with var. A top-level let, const or class works on the first evaluation only, then the expression falls back to the property's value. If another expression declared the same name first, it fails from the start
time, value Yes
thisComp.frameDuration Yes There is no bare frameDuration
Another layer's properties Yes In the same composition only (thisComp.layer(...), by name or ind). The other layer's expression runs first. The two scripts share one scope, so give their top-level variables different names
thisLayer, thisComp Yes
Layer transform properties: position, rotation, anchorPoint, scale, opacity, .transform Yes Expressions on X, Y and Z rotation and on orientation are ignored
wiggle() Yes Repeatable, not random: the same frame always gives the same result. Per-axis noise around value; octaves clamped to 1–4. The noise follows the layer's name: renaming the layer changes it, and layers with the same name wiggle the same property alike
valueAtTime(t), velocityAtTime(t) Yes
key(n), numKeys, nearestKey(t) Yes Check numKeys first, and use n from 1 to numKeys
loopIn(), loopOut(), loopInDuration(), loopOutDuration() Yes
linear, ease, easeIn, easeOut, clamp, smooth, key(n).value, thisComp.width, .height, .duration, .numLayers, a layer's .index, .inPoint, .outPoint Yes
A layer's .name (thisLayer.name, thisComp.layer(...).name) Partly Always returns an empty string

When an expression fails, the render continues:

  • An expression that throws, or references a missing layer, falls back to the property's own value.
  • A script that runs too long on one frame, such as an endless loop, is stopped. It is not run again for the rest of the render, so the property keeps its own value. Every other property with exactly the same script stops with it.

A top-level variable or function named after a helper, such as ease or smooth, replaces that helper for every expression.

Expressions are supported as Bodymovin exports them. In hand-written x scripts, don't use +, - or * directly on arrays. Do the arithmetic per component.

Layer styles

Feature Supported Notes
Stroke Yes Always outside, Normal blend mode, 100% opacity
Color overlay Yes Every blend mode except Dissolve, which blends as Normal
Outer glow Yes Solid color only; Softer and Precise techniques. Darker Color, Lighter Color, Divide and Dissolve blend as Normal
Drop shadow Yes Turn off Use Global Light; Layer Knocks Out Drop Shadow has no effect. Darker Color, Lighter Color, Divide and Dissolve blend as Normal
Inner shadow, inner glow, bevel, satin, gradient overlay No

Effects

Supported Not supported
Gaussian Blur, Fast Box Blur, Drop Shadow, Fill, Gradient Ramp, 4-Color Gradient, Tint, Hue/Saturation, Simple Choker, Glow, Corner Pin, CC Power Pin, Linear Wipe, Venetian Blinds, Mirror, CC Light Sweep, CC Light Rays, CC Radial Blur, CC Cylinder, Transform (the effect, without motion blur), Color Balance (HLS) Motion blur, Optics Compensation, Threshold, Wave Warp, and any effect not in the supported list

An effect that is not supported is skipped, and the rest of the layer renders. An effect switched off in After Effects (en: 0 in data.json) still renders, so delete it to turn it off. The format reference lists the parameters each effect reads.

Media files

Kind Formats Notes
Images PNG, JPEG, WebP, AVIF, HEIC, TIFF, BMP, TGA, GIF or PSD (its merged composite renders) Declare an image asset at most 4096 pixels on each side: scale a larger file down, or declare its size scaled to fit 4096. The engine skips a larger one. The version's warnings say so, and a layer that shows it is reported in problems, so the version does not render. EXIF orientation is applied to every format except JPEG, PNG and TGA, so export those upright.
Video MP4, MOV, WebM, MKV, AVI and the other formats Projects lists At most 4096 pixels on each side in a composition. Video with transparency renders as it is when it is MOV or MP4 in Apple ProRes 4444, QuickTime Animation (RLE) or the PNG codec. It must also be upright and at most 4096 pixels on each side. Other video with transparency, such as VP9 WebM or a larger file, is converted to PNG in MOV before its first render. Export alpha video premultiplied (matted against black). A file with several video streams plays only the first one that is not a cover picture.
Audio MP3, WAV, M4A, AAC, FLAC, OGG or Opus Set it as the soundtrack in an edit, or name it in the render request.
3D models Self-contained .glb or .gltf On a model layer (ty:16 in data.json), which Blitframe for After Effects exports. With Bodymovin, add the layer as Native models shows. Blitframe renders models with their materials under its own fixed studio lighting, not the lights in your After Effects scene. Each model layer draws on its own and stacks by layer order like a 2D layer, so models in different layers never pass through each other: put parts that must intersect in one file.
Fonts TTF, OTF, WOFF or WOFF2, or any family in the Blitframe font library See Fonts.

Only upload fonts and media you have the rights to use. Tips for fast renders says which video formats render fastest.

The Validator list

Blitframe for After Effects shows a Validator list after an export. It does not yet match this page:

  • It passes Wave Warp, Threshold, Optics Compensation and Fast Blur (Legacy), which do not render.
  • It flags Trim Paths, collapse transformations on a precomposition, and video other than MP4 or MOV, which do render.

This page and the version's problems and warnings are what counts.