fjamespriceandClaude Opus 5.5 6025c504cf perf(camera-raw): develop Dehaze, Clarity and Texture on the GPU in the preview (#6916) ...

Step 3 of #6916. Dehaze, Clarity and Texture run as fragment passes
(box, minimum and guided filters) around step 2's Shadows and
Highlights, so every setting but Detail develops on the GPU with one
readback.

- algo: camera_raw::spatial resolves Dehaze, Clarity and Texture per
  frame; develop builds them the same way (130 output hashes unchanged,
  same speed). PerPixel gains the three stages, per_pixel() refuses only
  Detail, and Whites/Blacks are measured after any spatial stage.
- gpu: camera_raw_filters.wgsl; the tone frame's pyramid is made on
  demand; plans record four segments around the main shader's passes.
  Dehaze's light uses two exact selections (cutoff value, then cutoff
  index among its ties). Its <=768 px copy comes from the CPU
  (Dehaze::cells, develop's own code), cached per source and white
  balance: on WARP the GPU's copy split CPU ties at the cutoff and moved
  the light by 5.7e-3.
- GPU vs CPU at most 7.6e-5 (Vulkan, DX12, WARP), tolerance 1/4096.
  858x572: Dehaze +40 6.0 ms vs 29 ms, Clarity +50 3.5 vs 16 ms,
  Texture +50 3.4 vs 10 ms, everything 19 ms vs 650-690 ms (Vulkan).

Part of #6916.

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

2026-11-09 02:42:52 -05:00

..

bench_camera_raw.rs

perf(camera-raw): develop Dehaze, Clarity and Texture on the GPU in the preview (#6916)

2026-11-09 02:42:52 -05:00

bench_shape_drag.rs

Reuse filled-shape effect maps while moving styled groups (#8993)

2026-10-23 17:37:56 -04:00

bench_stack_reuse.rs

test: GPU tests honour WGPU_BACKEND, so they run on Windows with an Intel GPU (#4)

2026-10-23 17:43:05 -05:00

compositor_build_probe.rs

Build the GPU compositor off the UI thread (#9969)

2026-11-05 22:49:12 +01:00

zero_block_probe.rs

zero_block_probe: pick the adapter by name with adapter=<name>

2026-10-20 20:50:26 -04:00