Fr0zkaandClaude Opus 5.5
971206c2ea
The L70 "2x SurfaceWorld GenerateTile in Unreal" was the instrument: the trace enabled bCaptureVoxelForgeCoreHashesForValidation, which disables the native heightfield-band shortcut and whole-tile skips (W1 same keys: 9.00 s hash-off vs 28.46 s hash-on). With hashing off, Unreal and headless match at W8 within 1.2-2.5% (Core 14.42 s vs 14.26 s) on identical tile keys. LOD0 p95 tail = real cost of non-empty passage tiles near the spawn mouth (p50/p95 18.8/105 ms); most LOD0 tiles are empty. verify.py reads its baselines from Standalone/Tools/verify/baselines/ instead of untracked Saved/Linux. The 11 .vfmesh control meshes (~78 MB) are not committed yet (to be stored compressed). Fast tier green. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_016dhxM43dnWvUu2B5Ph1iG1
680 B
680 B
Verification baselines
These inputs are byte-for-byte copies of the historical files consumed by the
verification gates. The gates now read this tracked directory instead of
Saved/Linux, which is generated and is not present in a fresh checkout.
owner-hash/: L46, L52 and L53 diagnostics plus the pinned v212/v214 owner hash baselines.historical-controls/: L35, L36 and L38 regression-control logs.fixtures/: Linux25 owner JSON inputs and L33 style-override fixtures.landmarks/: L45–L48 landmark mesh and layout controls.mouth/: L35–L37 mouth-shape controls.ring/: L30/L31 negative ring controls.capture/: L34 reference camera records.