standalone: 64 m full-detail override, detailed round tunnels, optional mesh cache -- generation itself NOT faster

ClipRadius override 3 (64 m at 25 cm). Uncached full detail 20.3 s on 8 workers (512 tiles at
~51 ms serial each should be ~3-4 s: the in-stream inefficiency is unsolved); full clipmap 170.6 s.
The 1 s / 5 s targets were met only from an on-disk mesh cache (.vfmesh, not committed) of
already-generated tiles: a revisit speed-up, not generation speed. Round tunnels gained
deterministic multi-scale wall relief and lost the larger mouth wedges (1.54% of lattice signs
changed, mean 0.20 m). Hashes stable across runs and 1 vs 6 workers.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ

Standalone/Research/VoxelSpeedRound/preliminary/chamber_legacy.stdout.log

@@ -0,0 +1,8 @@
+Controls: hold right mouse to look; WASD move; Space/Ctrl (E/Q) up/down; Shift faster; wheel changes speed; F wireframe; G toggles legacy/detailed round walls; 1-9 toggle LOD; 0 toggle all levels; N/P change seed; T swaps base/round tunnels; Esc quit.
+Owner data: E:\Projet Unreal\VoxelM\Plugins\VoxelForge\Standalone\data\owner_world.json; starting at first-strate open-space sample (-11.75, -123.25, -19.85) m.
+Standalone clip override: exported radius=6 (112 m cube) -> radius=3 (64 m LOD0 cube), voxel size=25 cm.
+Streaming seed 0 with legacy round/winding tunnels: 6498 desired tiles, 960 full-detail tiles, 8 workers; upload budget 4 batches / 2.0 ms per frame.
+[FullDetailReady] 20.425 s after spawn; resident=1117/960 tiles through LOD1.
+[TileTiming] LOD=0 generated=512 target=512 samples=21.95M worker_sum=79.071s avg=154.436ms p50=95.464ms p95=560.570ms
+[TileTiming] LOD=1 generated=448 target=448 samples=19.21M worker_sum=77.532s avg=173.062ms p50=80.573ms p95=578.995ms
+Back-buffer capture saved: E:\Projet Unreal\VoxelM\Plugins\VoxelForge\Standalone\Research\VoxelSpeedRound\chamber_legacy.png (20.557 s after generation start; full detail uploaded).