VoxelForge

0 Fork 0

Files

T

Fr0zkaandClaude Opus 5.5 d4b4377d30 research(standalone): GPU first port -- marching cubes in D3D11 compute, exact vs WARP; field still CPU

- Compute shaders classify cells from CPU sign bits, scan output offsets, emit vertex/index buffers
  and indirect-draw args. Two 32^3 tiles: 0 topology differences hardware vs WARP; two runs on the
  RX 7900 XTX identical; max CPU/GPU vertex distance 0.000122 cm (LOD0), 0.000244 cm (LOD1).
- GPU kernels 0.049 / 0.071 ms vs CPU tile 2.33 / 1.99 ms, but CPU input preparation dominates
  (7.25 / 2.84 ms wall): no whole-view gain until the field itself runs on the GPU.
- VoxelForgeFieldShared.h: first shared C++/HLSL field math (integer noise hash, ...).
CPU path unchanged.

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

2026-10-06 06:44:43 +02:00

include/VoxelForgeCore

wip(standalone): owner LOD0 defects -- skirt curtains clipped, exact edge roots; interrupted (OpenAI capacity)

2026-10-06 00:17:14 +02:00

src

research(standalone): GPU first port -- marching cubes in D3D11 compute, exact vs WARP; field still CPU

2026-10-06 06:44:43 +02:00

verification

standalone: rock detail restored in lattice-resolvable bands (4-16 voxels), owner-band gate added

2026-10-06 05:43:31 +02:00

API.md

standalone: cheap chunk crossings -- exact-tile slab compile, nearest tiles first (~1.2 s -> 121 ms peak)