demo.git/src/tests, branch main

style: apply clang-format

2026-05-21T06:10:47Z

fix: audio & effects cleanup — dead code removal, simplifications

2026-05-20T21:21:59Z

ans: order-0 rANS coder + WGSL asset compression

2026-05-14T17:11:28Z

Adds src/util/ans.{h,cc}, a per-chunk-adaptive order-0 rANS entropy coder. Decoder is always built; encoder is gated on ANS_ENABLE_ENCODER (tools only). Both sides take an optional 256-entry initial_counts table to seed the adaptive model. The per-chunk initial state is (1 << kBits). Higher initial states (e.g. with a signature packed into the upper bits) force a renorm-emit at iter 0 that the decoder never consumes, corrupting multi-chunk streams once stats become skewed. Asset pipeline: - AssetRecord gains 'compression' and 'uncompressed_size' fields. - asset_packer scans every WGSL file to build a corpus-wide byte histogram, then ANS-encodes each shader using that histogram as the seed. Histogram and accessor are emitted alongside the asset table. Round-trip verification runs at pack time for every compressed asset; failures fall back to uncompressed storage. - asset_manager decompresses on first GetAsset(), caches the heap-allocated buffer, and DropAsset / ReloadAssetsFromFile free it along with the procedural cache. - Disk-load (dev) builds are unchanged: WGSL paths stay as filenames. Tests: - src/tests/util/test_ans.cc: roundtrip variants (empty, single byte, single-symbol run, all-zeros, random uniform/skewed, repeated ASCII), seeded-vs-uniform compression, rejection of mismatched counts / corruption / truncation, PeekUncompressedSize. - 37/37 dev, 36/36 STRIP_ALL. Compression observed: WGSL shaders shrink to ~0.62-0.71x in the main workspace (81 of 105 assets qualify). Docs: - doc/ANS.md (new): algorithm, bitstream, API, asset pipeline integration, compression numbers, limitations, tests. - doc/ASSET_SYSTEM.md: new Compression section + updated technical guarantees for compressed assets. - doc/COMPLETED.md: May 2026 entry. - PROJECT_CONTEXT.md: Build status line mentions WGSL ANS compression. - CLAUDE.md, GEMINI.md: tier-3 build doc list includes ANS.md.

fix(3d): restore correct orientation in test_3d_render (direct-to-surface)

2026-03-29T00:50:47Z

Add Renderer3D::set_direct_render(bool) flag (must be set before init()). When true, uses standard CCW winding and un-negates perspective Y, restoring the pre-ba7ea27 orientation for direct-to-surface rendering where the post-process Y-flip is absent. handoff(Gemini): set_direct_render() is the escape hatch for any future renderer usage that bypasses the post-process chain.

feat(procedural): add plasma, voronoi, normalmap generators

2026-03-29T00:42:28Z

Add three new procedural texture generators: - gen_plasma: classic sine-sum color texture (RGB output) - gen_voronoi: Worley cellular noise (F1/F2/F2-F1 modes) - gen_normalmap: post-process grayscale→RGB normal map Remove gen_noise_256 (was an alias for gen_noise). Register new generators in asset_manager and asset_packer. Add unit tests for all three, and use them in test_3d_render (plasma sky, voronoi noise, fBm normal map). handoff(Gemini): plasma/voronoi/normalmap procedural generators added; gen_noise_256 removed; tests + 3d_render usage wired up.

fix: reorder WGPURequestAdapterOptions fields to match declaration order

2026-03-29T00:01:14Z

Fixes -Wreorder-init-list warning in webgpu_test_fixture.cc.

feat(cnn_v3): upgrade architecture to enc_channels=[8,16]

2026-03-26T06:03:01Z

Double encoder capacity: enc0 4→8ch, enc1 8→16ch, bottleneck 16→16ch, dec1 32→8ch, dec0 16→4ch. Total weights 2476→7828 f16 (~15.3 KB). FiLM MLP output 40→72 params (L1: 16×40→16×72). 16-ch textures split into _lo/_hi rgba32uint pairs (enc1, bottleneck). enc0 and dec1 textures changed from rgba16float to rgba32uint (8ch). GBUF_RGBA32UINT node gains CopySrc for parity test readback. - WGSL shaders: all 5 passes rewritten for new channel counts - C++ CNNv3Effect: new weight offsets/sizes, 8ch uniform structs - Web tool (shaders.js + tester.js): matching texture formats and bindings - Parity test: readback_rgba32uint_8ch helper, updated vector counts - Training scripts: default enc_channels=[8,16], updated docstrings - Docs + architecture PNG regenerated handoff(Gemini): CNN v3 [8,16] upgrade complete. All code, tests, web tool, training scripts, and docs updated. Next: run training pass.

feat(cnn_v3): 3×3 dilated bottleneck + Sobel loss + FiLM warmup + architecture PNG

2026-03-25T09:05:42Z

- Replace 1×1 pointwise bottleneck with Conv(8→8, 3×3, dilation=2): effective RF grows from ~13px to ~29px at ¼res (~+1 KB weights) - Add Sobel edge loss in training (--edge-loss-weight, default 0.1) - Add FiLM 2-phase training: freeze MLP for warmup epochs then unfreeze at lr×0.1 (--film-warmup-epochs, default 50) - Update weight layout: BN 72→584 f16, total 1964→2476 f16 (4952 B) - Cascade offsets in C++ effect, JS tool, export/gen_test_vectors scripts - Regenerate test_vectors.h (1238 u32); parity max_err=9.77e-04 - Generate dark-theme U-Net+FiLM architecture PNG (gen_architecture_png.py) - Replace ASCII art in CNN_V3.md and HOW_TO_CNN.md with PNG embed handoff(Gemini): bottleneck dilation + Sobel loss + FiLM warmup landed. Next: run first real training pass (see cnn_v3/docs/HOWTO.md §3).

fix(fft): make bit_reverse_permute static, remove from public API and tests

2026-03-24T06:54:59Z

fix(fft): replace iterative twiddle with direct cosf/sinf, add tests A-E

2026-03-24T06:52:25Z

fft_radix2 now computes wr=cosf(angle*k)/wi=sinf(angle*k) directly per k, eliminating float drift over long iteration runs. Iterative approach documented in comment for reference. Tests A-E added (bit-reverse, small-N DFT, twiddle drift, DCT small/large N). arrays_match tolerance reverted to 5e-3. TODO.md updated. handoff(Gemini): fft twiddle fix complete, 38/38 tests passing.