WebGPU Exposure Color PipelineEvidence pending

Canonical Lab · Three.js 0.185.1

WebGPU Exposure Color Pipeline

Interactive Three.js WebGPU/TSL demo for WebGPU Exposure Color Pipeline, with 5 scenarios, 6 mechanisms, and 3 quality tiers and evidence-gated validation.

Evidence status: The implementation is available for technical review, but its native-WebGPU runtime and evidence gates remain incomplete.

Readiness & remaining fixes 4

4 open closure items. Loadable does not mean accepted or performance-valid.

  1. P0Define and measure the target performance contract

    The full-histogram, balanced-log-reduction, minimum-fixed-shot tiers have no accepted frame target. Record named device/browser/GPU, viewport, DPR, sustained CPU/GPU/presentation p50 and p95, deadline misses, memory, and settled quality. Submission time is not GPU timing.

    tiers[].frameTargetMs
  2. P0Current Adapter Timing runtime proof

    No named-adapter timestamp capture accepted; performance remains INSUFFICIENT_EVIDENCE.

    runtimeProof.current-adapter-timing
  3. P1Capture a current-source evidence bundle

    Publish source-hash-bound renderer/backend facts, required diagnostic images, aligned readbacks, resource inventory, timing method, and limitations; directly inspect the important artifacts before promotion.

    evidenceBundle
  4. P2Close tier-specific acceptance

    Validate and promote full-histogram, balanced-log-reduction, minimum-fixed-shot as independent visual and performance contracts without crossing their protected invariants.

    tiers[].acceptanceStatus

Use the linked skill and source to inspect route ownership and evidence classification before treating this presentation as an accepted implementation.

Owning skill
Exposure And Color Grading
Published routes
5 scenarios · 6 mechanisms · 3 quality tiers

Supported mechanisms

  • log-luminance-reduction — Log Luminance Reduction
  • histogram-and-percentiles — Histogram And Percentiles
  • adaptation — Adaptation
  • metering-masks — Metering Masks
  • tone-mapping — Tone Mapping
  • lut-grading — Lut Grading
initializing native WebGPU