AI WebGPU Lab Benchmark

Blackhole Render Shootout

`bench-blackhole-render-shootout` compares deterministic blackhole renderer profiles on the same lensing fixture.

The benchmark records winner selection, frame pacing, ray step budget, TAA state, adaptive quality, and fallback metadata before real renderers are integrated.

Benchmark Controls

Run the benchmark to compare renderer profiles under the selected execution mode.

Renderer Lensing Panels

Checks

  • Use one fixed blackhole fixture across three deterministic renderer profiles.
  • Record WebGPU primary and fallback primary captures with the same output schema.
  • Keep result fields stable for later real three.js, raw WebGPU, and WebGL/WebAssembly implementations.

Metrics

Environment

Activity Log

    Schema-Aligned Result Draft

    {
      "status": "pending"
    }

    What This Unlocks

    • First blackhole renderer benchmark raw JSON and comparison summary
    • Shared metrics for `exp-blackhole-three-singularity` and future raw WebGPU/Kerr engines
    • Stable capture target for renderer shootout automation