AI WebGPU Lab Blackhole Experiment
Blackhole Singularity Readiness
`exp-blackhole-three-singularity` exposes a deterministic blackhole rendering surface before the real three.js/TSL/WebGPU implementation lands.
The harness records capability state, lensing scene load cost, frame pacing, ray step budget, TAA state, and adaptive quality metadata.
Scene Controls
Probe capability first, then run the deterministic lensing scene to export schema-aligned graphics metrics.
Checks
- Use one fixed camera orbit, ray step budget, and disk profile so captures remain comparable.
- Keep WebGPU/WebGL fallback metadata explicit before swapping in the real three.js renderer.
- Record visual artifact notes for lensing rings, accretion disk aliasing, and adaptive quality.
Schema-Aligned Result Draft
{
"status": "pending"
}
What This Unlocks
- First blackhole-specific raw JSON and screenshot capture path
- Renderer-neutral graphics metrics for later shootout benchmarks
- Stable baseline for TSL/WebGPU/WebGL fallback integration work