M13.29: GPU effects part D (particles and remaining fallbacks)

GPU render passes for CC Particle World / Systems II, Particle Playground,
CC Ball Action, CC Pixel Polly, CC Scatterize and Curl Noise (fx_particles,
chunked multi-pass raster); Shatter wireframe views, Foam extras, Timewarp
with a Matte Layer, strong Warp Fisheye/Twist (CPU f64 inverse map) and
custom .ocio configs no longer fall back to the CPU. CPU refactors pinned by
particle_golden / timewarp_golden / sim_golden hashes.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

docs/architecture.md

@@ -209,7 +209,7 @@ Preserve Transparency / blend kernels (a 3D track matte renders solo through the
 Environment Light Background layers draw their sky in a kernel (`adv_sky`, from
 `Renderer::sky_draw`) before the rest of the run, in Classic and Advanced 3D comps. Wireframe
 outlines draw on the GPU from the CPU's pixel list (`Renderer::wireframe_pixels`). GPU effects
-(`effects::GPU_EFFECTS`, 236 of them: blurs, colour correction, keying incl. Key Light, mattes,
+(`effects::GPU_EFFECTS`, 243 of them: blurs, colour correction, keying incl. Key Light, mattes,
 channel, stylize, distortion and warps (Warp, Bezier Warp, Smear, Reshape, CC Bend It, CC Page
 Turn, CC Bender, CC Blobbylize), Cartoon, bevels, shapes, the transitions (CC light family, CC
 transitions, Block Dissolve), perspective (Radial Shadow, CC Cylinder / Sphere / Spotlight /
@@ -217,7 +217,10 @@ Environment, 3D Glasses), generators, noise, grain, text (Numbers, Timecode), ti
 extra texture), Immersive Video, colour management (colour programs from
 `effects::color_program`, LUTs in a storage buffer) and the simulations' render passes (the
 simulation stays on the CPU and hands its plan — sprites, pieces, blobs, grids — to a tiled
-rasteriser); see
+rasteriser; the particle effects too, `gpu::fx_particles`: CC Particle World / Systems II,
+Particle Playground with its Layer Map frames in one atlas, CC Ball Action, CC Pixel Polly, CC
+Scatterize; plans too large for one item table draw in several passes over the previous one);
+see
 [effects.md](effects.md)) repeat the CPU effect's steps (padding, box radii, parameters, hashes)
 as kernels, in one module per family (`gpu::fx_*` with `shaders/fx_*.wgsl`); consecutive GPU
 effects run as one chain with one upload and one readback. Statistics that need the whole frame
@@ -229,10 +232,12 @@ combine them on the GPU; Time Displacement's per-pixel times and Timewarp's moti
 (`time_fx::timewarp_plan`) are computed on the CPU. Numbers and Timecode rasterise their glyph
 coverage on the CPU (`textfx::text_layer`) and composite it on the GPU. Other layers an effect
 reads (gradients, reveals, environments) are fitted on the CPU (`util::fit_layer`) and uploaded.
-Settings a kernel cannot match render on the CPU (`effects::catalog::gpu_supported`; e.g. Warp's
-Fisheye and Twist past 50 % or with a distortion, whose Newton inverse wanders chaotically near
-the crease, and Timewarp with a Matte Layer); Card Wipe (the 3D card renderer of Card Dance and
-Shatter) stays on the CPU. Tests render
+Where f32 cannot follow the CPU's f64, the CPU solves the part that needs it and the GPU uses
+the result: Warp's Fisheye and Twist past 50 % or with a distortion (a Newton inverse that
+wanders chaotically near the crease) sample the CPU's inverse map (`warp_inverse_map`). Settings
+a kernel cannot match still render on the CPU (`effects::catalog::gpu_supported`: Cell
+Pattern's HQ variants, Turbulent Displace's locked pinning and the controls in
+`CPU_ONLY_CONTROLS`). Tests render
 scenes on both paths and compare them (≤ 1/255 at 8 bpc, ≤ 1e-3 at 32 bpc); they skip without an
 adapter. Working textures (premultiplied RGBA f32) come from a pool: a released texture is reused
 once every encoder that could still read it has been submitted, transparent inputs share one

docs/effects.md

@@ -5,7 +5,7 @@
 
 EffectCraft ships 306 effects with 3009 parameters in total, grouped into the same categories as the Effects & Presets panel. Every effect is our own implementation, written from public behaviour descriptions and standard image-processing literature. Parameter names, order, popup options, units, defaults and ranges follow the reference application so that projects, expressions and muscle memory carry over.
 
-- **GPU**: 236 effects also run on the GPU compositor with identical results.
+- **GPU**: 243 effects also run on the GPU compositor with identical results.
 - **32**: 306 effects process 32-bit float (HDR, overbright) pixels without clamping.
 - **Status**: 306 implemented in full, 0 partial (what is missing is listed).
 
@@ -265,7 +265,7 @@ Effects are addressed by id (`ec.<category>.<name>`) or display name in commands
 | Effect | Id | Params | GPU | 32 | Status |
 |---|---|---:|:-:|:-:|---|
 | Add Grain | `ec.noise.addgrain` | 32 | GPU | 32 | Implemented |
-| Curl Noise | `ec.noise.curlnoise` | 10 |  | 32 | Implemented |
+| Curl Noise | `ec.noise.curlnoise` | 10 | GPU | 32 | Implemented |
 | Dust & Scratches | `ec.noise.dustscratches` | 3 | GPU | 32 | Implemented |
 | Fractal Noise | `ec.noise.fractal` | 26 | GPU | 32 | Implemented |
 | Match Grain | `ec.noise.matchgrain` | 37 |  | 32 | Implemented |
@@ -317,22 +317,22 @@ Effects are addressed by id (`ec.<category>.<name>`) or display name in commands
 
 | Effect | Id | Params | GPU | 32 | Status |
 |---|---|---:|:-:|:-:|---|
-| CC Ball Action | `ec.sim.ccballaction` | 8 |  | 32 | Implemented |
+| CC Ball Action | `ec.sim.ccballaction` | 8 | GPU | 32 | Implemented |
 | CC Bubbles | `ec.sim.ccbubbles` | 7 | GPU | 32 | Implemented |
 | CC Drizzle | `ec.sim.ccdrizzle` | 14 | GPU | 32 | Implemented |
 | CC Hair | `ec.sim.cchair` | 20 | GPU | 32 | Implemented |
 | CC Mr. Mercury | `ec.sim.ccmrmercury` | 23 | GPU | 32 | Implemented |
-| CC Particle Systems II | `ec.sim.ccparticlesystems2` | 21 |  | 32 | Implemented |
-| CC Particle World | `ec.sim.ccparticleworld` | 30 |  | 32 | Implemented |
-| CC Pixel Polly | `ec.sim.ccpixelpolly` | 10 |  | 32 | Implemented |
+| CC Particle Systems II | `ec.sim.ccparticlesystems2` | 21 | GPU | 32 | Implemented |
+| CC Particle World | `ec.sim.ccparticleworld` | 30 | GPU | 32 | Implemented |
+| CC Pixel Polly | `ec.sim.ccpixelpolly` | 10 | GPU | 32 | Implemented |
 | CC Rainfall | `ec.sim.ccrainfall` | 14 | GPU | 32 | Implemented |
-| CC Scatterize | `ec.sim.ccscatterize` | 4 |  | 32 | Implemented |
+| CC Scatterize | `ec.sim.ccscatterize` | 4 | GPU | 32 | Implemented |
 | CC Snowfall | `ec.sim.ccsnowfall` | 19 | GPU | 32 | Implemented |
 | CC Star Burst | `ec.sim.ccstarburst` | 6 | GPU | 32 | Implemented |
 | Card Dance | `ec.sim.carddance` | 53 | GPU | 32 | Implemented |
 | Caustics | `ec.sim.caustics` | 28 | GPU | 32 | Implemented |
 | Foam | `ec.sim.foam` | 36 | GPU | 32 | Implemented |
-| Particle Playground | `ec.sim.particleplayground` | 92 |  | 32 | Implemented |
+| Particle Playground | `ec.sim.particleplayground` | 92 | GPU | 32 | Implemented |
 | Shatter | `ec.sim.shatter` | 60 | GPU | 32 | Implemented |
 | Wave World | `ec.sim.waveworld` | 35 | GPU | 32 | Implemented |
 

docs/parity.md

@@ -10,7 +10,7 @@ How close EffectCraft is to After Effects 2026, feature by feature, and how much
 | Unweighted | ≈ 97% (half-credit) / 99% (fractions) |
 | P0 / P1 / P2 | ≈ 99% / 96% / 90% (half-credit); 99.8% / 99.2% / 97.5% (fractions) |
 | Features done / partial / missing | 87 / 5 / 0 of 92 |
-| **Effects** | **306** effects (all After Effects 2026 effects), every one implemented in full ([effects.md](effects.md)); 236 run on the GPU |
+| **Effects** | **306** effects (all After Effects 2026 effects), every one implemented in full ([effects.md](effects.md)); 243 run on the GPU |
 | Disabled menu entries left | 1, on purpose: Import ▸ Vanishing Point (.vpe), whose format has no public specification |
 | Remaining work | ≈ 9–11 agent-hours at the pace measured so far (≈ 29 on the conservative audit scale); ≈ 6–7 without the learned-model items |
 | **Wall-clock estimate** | **≈ 2–2.5 hours** with five agents in parallel; ≈ 1.5 hours for 100% of P0 + P1 |
@@ -19,7 +19,7 @@ Partial features:
 
 | id | Tier | Done | What is missing |
 |---|---|---|---|
-| EFF-5 GPU effects | P1 | 0.97 | GPU kernels for the audio-driven effects and the stepped particle systems' sprite pass (the CC light / transition families, Numbers, Timecode and Time effects landed in M13.22; 3D Channel, the VR family, OCIO / LUT and the simulations' render passes in M13.23); custom `.ocio` configs, Shatter's wireframe views and Foam's texture / environment / flow-map extras render on the CPU |
+| EFF-5 GPU effects | P1 | 0.98 | GPU kernels for the audio-driven effects (the CC light / transition families, Numbers, Timecode and Time effects landed in M13.22; 3D Channel, the VR family, OCIO / LUT and the simulations' render passes in M13.23; the particle effects' render passes — CC Particle World / Systems II, Particle Playground, CC Ball Action, CC Pixel Polly, CC Scatterize — and Curl Noise in M13.29, which also moved Shatter's wireframe views, Foam's texture / environment / flow-map extras, Timewarp with a Matte Layer, strong Warp Fisheye / Twist bends and custom `.ocio` configs onto the GPU); Cell Pattern's HQ variants, Turbulent Displace's locked pinning and the `CPU_ONLY_CONTROLS` render on the CPU |
 | WEB-1 Web app | P0 | 0.97 | GPU effects in Render Queue / analysis job workers (they render on their CPU); layer buffers in the browser's disk cache (frames only); threads inside one engine instance (the decided design is one engine instance per worker; a threaded build needs nightly `build-std`). M13.24 added GPU effects and the GPU compositor in the frame workers (their own WebGPU devices, deferred readbacks: frames render in passes; Backend Auto per comp), the disk cache in the Origin Private File System (written by the workers with sync access handles, LRU under the settings' limit, served after a reload), the storage manager (Settings ▸ Disk ▸ Browser Storage, `storage.*`) and fixed a WGSL constant Chrome rejected (it disabled every GPU kernel in Chrome) |
 | UI-7 Home screen | P1 | 0.95 | A "start from a template" gallery |
 | MSK-4 Roto Brush | P2 | 0.85 | Segmentation is classical (graph cut + optical flow), not a learned model |
@@ -156,7 +156,7 @@ the disk cache in the formats wave; the 3D Camera Tracker in M12.6.)
 | Text | ≈ 96% | 0.3 | no extruded strokes (M13.12: the Variable Font Axes animator re-spaces the text — advances follow the animated axes; M13.6: variable font axes in the character style — `layer.setText variations`, the Character panel's Variable Font Axes fields — shape with HVAR / gvar advances and draw at that design-space position; OpenType features — stylistic sets, discretionary ligatures, contextual / stylistic alternates, swash, titling, ordinals, fractions, figure styles, true small caps / all small caps and superior / inferior glyphs with faux fallback — per character with the Character panel's OpenType popup and `text.fontFeatures` landed in M13.2; vertical Roman / Tate-Chu-Yoko, forced LTR paragraphs, caret on animated and path text, Variable Font Axes and Lottie style runs landed in M13.5; extruded, bevelled text in M7.6; per-character styles, paragraph settings, on-canvas editing and the `sourceText` style API in M9.9–M9.10) |
 | Web | 97% | 0.2 | GPU effects in Render Queue / analysis workers (job workers render on their CPU), layer buffers in the browser's disk cache (frames only), no shared-memory threads inside one engine instance (the decided design is one engine instance per worker; a threaded build needs nightly `build-std`) (browser storage, Web Audio, Web Worker renders/analyses, WebGPU viewer and offline install landed in M15.2; viewer frames in frame workers fed by project diffs, Roto Brush propagation in a worker, non-blocking `wait: true` jobs and a browser Media Browser (File System Access folders, browser storage) in M13.10; WEB-1 in M13.24: GPU effects and the GPU compositor in the frame workers on their own WebGPU devices with deferred readbacks (frames render in passes; Backend Auto per comp), the disk cache in the Origin Private File System (written by the workers with sync access handles, LRU under the settings' limit, served after a reload), the storage manager (Settings ▸ Disk ▸ Browser Storage, `storage.info` / `storage.persist` / `storage.clear`), and a WGSL constant Chrome rejected (it disabled every GPU kernel in Chrome) fixed) |
 | 3D | 88% | 5.5 | multi-view layouts, the Extended Viewer for Advanced 3D comps (Classic 3D Extended Viewer landed in M13.5 UI completion); collapsed precomps of another size seen through the parent's camera render (fixed in M13.2); stereo rigs, orbit nulls, lights controlled by the camera, cameras/lights from glTF models, environment backgrounds, Advanced 3D motion blur, blend modes and track mattes landed in M7.7; Classic 3D iris-shaped bokeh with highlights, progressive depth of field on tilted layers and the focus-link commands landed in M13.5; Advanced 3D depth of field with the iris and highlight options, collapsed precomps as real Advanced 3D geometry and extruded text/shape strokes landed in M13.8; Advanced 3D (glTF/OBJ models, primitives, extruded text and shapes, PBR, image-based light, shadow maps, GPU rasteriser) in M7.4–M7.6; Advanced 3D end to end on the GPU (motion blur, iris depth of field, compositing) in M13.11 |
-| Effects | ≈ 85% | 3.0 | GPU versions of the remaining effects (236 run on the GPU since M13.23, EFF-5: M13.23 added the 3D Channel and Immersive Video families, Apply Color LUT, the OCIO effects, Color Profile Converter and the simulations' render passes (Card Wipe included); M13.22 added the CC light family, the CC transitions, Block Dissolve, Radial Shadow, CC Bender / Blobbylize / Cylinder / Sphere / Spotlight / Environment, 3D Glasses, Numbers, Timecode and the time effects joined the 166 of M13.13; audio-driven effects and the stepped particle systems still render on the CPU) and the missing controls listed as partial in [effects.md](effects.md) (every After Effects effect exists since M9.11, M12.5 and M12.6; parameter names, order, twirl-downs, popups, units and defaults were aligned in M9.12) |
+| Effects | ≈ 85% | 3.0 | GPU versions of the remaining effects (243 run on the GPU since M13.29, EFF-5: M13.29 added the particle effects (CC Particle World / Systems II, Particle Playground, CC Ball Action, CC Pixel Polly, CC Scatterize, Curl Noise) and removed the fallbacks of Shatter's wireframe views, Foam's extras, Timewarp's Matte Layer, strong Warp Fisheye / Twist bends and custom `.ocio` configs; M13.23 added the 3D Channel and Immersive Video families, Apply Color LUT, the OCIO effects, Color Profile Converter and the simulations' render passes (Card Wipe included); M13.22 added the CC light family, the CC transitions, Block Dissolve, Radial Shadow, CC Bender / Blobbylize / Cylinder / Sphere / Spotlight / Environment, 3D Glasses, Numbers, Timecode and the time effects joined the 166 of M13.13; audio-driven effects still render on the CPU) and the missing controls listed as partial in [effects.md](effects.md) (every After Effects effect exists since M9.11, M12.5 and M12.6; parameter names, order, twirl-downs, popups, units and defaults were aligned in M9.12) |
 | Interface | 75% | 5.0 | more Learn tutorials and pixel-level fidelity of dialogs (the Home ▸ Templates gallery (eight original built-in templates, user templates from File ▸ Save as Template…) and View ▸ Simulate Output ▸ My Custom RGB… landed in M13.25; Timeline layer reordering by drag, a non-snapping viewer pan, a working rename field and twirl arrows, and the full set of property reveal shortcuts — double presses, Alt+Shift keyframes, Ctrl+` — landed in M13.17–M13.20; the Home ▸ Learn tab with interactive tutorials and a UI fidelity pass landed in M13.5 UI completion; visual editors for Lumetri RGB / hue-saturation curves, Colorama's output cycle wheel, Glow's colour map and Reshape's correspondence points (viewer handles), a scrolling Preview panel and AE-style Composition / Timeline tabs (close, label swatch, viewer lock) landed in M13.10; Timeline outline and Project panel columns scroll horizontally, the Layer Style dialog, ROI resize handles, Pan Behind snapping and 3D Reference Axes landed in M13.5; native macOS menu bar, Timeline columns/search/reveal-add, Home screen with recent projects and all AE workspaces landed; viewer rulers/snapping/channels/snapshots landed in M0.13) |
 | Project | ≈ 68% | 5.5 | auto-save, folder moves, OCIO displays beyond the built-in tone map (Color Engine with OCIO/ACES working spaces, HDR compand/tone mapping, Rec. 2100 PQ/HLG output, Feet + Frames, display colour management, Simulate Output and the locked viewer landed in M7.7; proxies and Interpret Footage fields / pixel aspect / alpha guess landed: PRJ-8, PRJ-3) |
 | Masks & roto | 74% | 5.0 | Roto Brush's learned (3.0) segmentation model (variable-width mask feather points with the Mask Feather tool landed in M13.5; mask tracking and Mask Interpolation landed in M6.6; Roto Brush & Refine Edge with graph-cut segmentation, flow propagation, edge matting, decontamination and Freeze in M6.7) |
@@ -460,6 +460,33 @@ composited at 8 and 32 bpc:
   default to the layer). Shatter's wireframe views and Foam's User Defined texture,
   Environment Map and flow-map preview render on the CPU.
 
+## Update: M13.29 GPU effects, part D (EFF-5)
+
+7 more effects run on the GPU (243 in all), and five CPU fallbacks are gone. Each case is
+checked against the CPU oracle (≤ 1/255 at 8 bpc, ≤ 1e-3 at 32 bpc) directly on a buffer (full
+and half resolution) and composited at 8 and 32 bpc; the CPU refactors behind the shared plans
+are pinned bit for bit by golden hashes (`crates/effects/tests/particle_golden.rs`,
+`timewarp_golden.rs`, `sim_golden.rs`).
+
+- **Particle render passes** (`gpu::fx_particles`): CC Particle World and CC Particle Systems II
+  (every particle type and transfer mode; the stepped simulation stays on the CPU or the GPU
+  `ParticleSim`, sorted and shaded as the CPU does), Particle Playground (cannon, grid, text
+  strokes, layer exploder, property mappers, repel, wall; the Layer Map particles' frames are
+  packed into one atlas), CC Ball Action, CC Pixel Polly and CC Scatterize (their plans read the
+  frame back once: the particles take the layer's colours and, for Ball Action's twist, its
+  luminance). Plans too large for one item table draw in several passes, each over the previous
+  one, with the same per-pixel arithmetic.
+- **Curl Noise**: the fBm potential, its curl and the streamline trace as three kernels.
+- **Former fallbacks**: Shatter's wireframe views (the default view) draw their lines as
+  sprites; Foam's User Defined texture, Environment Map and flow-map preview draw over the
+  bubble sprites; Timewarp with a Matte Layer carries the masked frames and the foreground's and
+  background's own vectors in its plan and builds both layers on the GPU; custom `.ocio` configs
+  compile to the colour program (matrix and offset, exponent, log / log-affine, range, CDL, file
+  LUTs, groups, inverses). Warp's Fisheye / Twist past 50 % or with a distortion keeps the CPU's
+  f64 Newton solve (f32, even with more iterations, lands on other roots near the crease): the
+  CPU solves the inverse map from the geometry alone and the GPU samples along it, so the chain
+  stays on the GPU without a readback.
+
 ## Highest-value gaps, in order
 
 1. ~~On-canvas text editing and per-character styles~~ (landed: M9.9–M9.10).