Fr0zka/VoxelForge - VoxelForge - FrozkompaGit

Fr0zka/VoxelForge

0 Fork 0

T

Fr0zkaandClaude Opus 5 0208e93606 measure(perf): re-rank component costs on the current build -- the tunnel core is now the largest

Measurement only (Codex): the same remove-one ablation method and command line as 6ccf371, all 14
voxel.TunnelAblate* bits static x2 and the top five moving x2, on the committed tip after the
landing-post, cave-warp, gap-floor and option B changes.

Static leaders now: tunnel core 20.84%, tunnel SDF 12.64%, passage carving 9.08%, cave warp 6.68%;
27.62% remains unattributed/shared. The tunnel core could not be measured at 6ccf371 because of the
ablation gate bug fixed in 7a972f5. The two tunnel evaluations together are now about a third of
worker time. Next targets as proposed: tunnel SDF/core merge (field-changing), passage carving
reuse (exact), cave warp result reuse (exact). Full tables and the launch ledger are in
PERF-SAMPLED.md.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TaFdXhTdYMkW1xrV6ApT6Y

2026-09-15 09:30:22 +02:00

perf(noise): batch the three cave-warp channels through the SIMD evaluator -- exact, ~4% faster

2026-09-15 08:41:59 +02:00

.gitignore

chore: track the design docs in git (.gitignore !*.md)

2026-07-27 01:39:21 +02:00

ARCHITECTURE.md

perf(generation): exact worm block skip with a proven 3D Perlin bound -- measured NEUTRAL

2026-09-14 04:29:11 +02:00

AUDIT-2026-07.md

docs(audit): C2 is FULLY closed -- the live-edit half was fixed too

2026-08-16 16:11:54 +02:00

AUDIT-2026-08-CODEX.md

docs: Sol investigation -- refutes my boundary-tile hypothesis, corrects my false accusation

2026-08-16 23:22:06 +02:00

AUDIT-COLD-2026-09-14.md

docs(audit): cold read-only examination with verification notes -- one determinism bug confirmed

2026-09-14 20:14:35 +02:00

AUDIT-SOL-2026-09-11.md

docs: Sol's read-only audit, verbatim

2026-09-12 03:42:31 +02:00

CLAUDE.md

chore: track the design docs in git (.gitignore !*.md)

2026-07-27 01:39:21 +02:00

CODEMAP.md

fix(diagnostics): sampler can no longer crash the run it measures; worm skip default off

2026-09-14 05:17:27 +02:00

CODEX-TASK-1-PROFILER-BATCH-RESEARCH.md

tools(profile): the instrument lied by 11.5x -- now it checks itself, and batching kills startup cost

2026-09-09 05:02:21 +02:00

CODEX-TASK-001-tile-skip-stats.md

docs(codex-001): the lumped AllSolid counter cannot prove T1.d -- add the op-stack site

2026-08-16 15:40:23 +02:00

CODEX-TASK-2-PERF-REPORT.md

perf(mesher): adjacent tiles were re-evaluating 18% of the field -- 2.69s to 1.49s

2026-09-09 05:59:56 +02:00

CODEX-TASK-002-column-memo-thrash.md

docs: close C2, sharpen the column-memo prediction, reassess REVIEW_FINDINGS

2026-08-16 16:14:18 +02:00

CODEX-TASK-003-extrareach-perlin-bound.md

fix(opstack): three ExtraReach box verdicts used sup|FBM| = 1.0; the proved bound is 1.5

2026-08-16 16:36:04 +02:00

CODEX-TASK-004-minmax-radius-envelope.md

fix(opstack): two cell-sweep pads assumed MinRadius <= MaxRadius; a third already didn't

2026-08-16 16:50:01 +02:00

CODEX-TASK-005-tile-scan-sampler-extent.md

test(opstack): three tile scans sampled <1 lattice period; fix the CLASS this time

2026-08-16 17:04:55 +02:00

CODEX-TASK-006-cave-morphology-radius-envelope.md

fix(morphology): BuildChunkCache under-bounds room/tunnel collection when Min > Max

2026-08-16 17:32:09 +02:00

CODEX-TASK-007-generation-pause.md

fix(world): drain workers before mutating strate layout/passages (VF-01)

2026-08-16 17:51:19 +02:00

CODEX-TASK-008-column-memo-and-t1d-diagnosis.md

perf(opstack): direct-indexed column box (measured 14.7% miss); + T1.d bail attribution

2026-08-16 18:28:46 +02:00

CodexGuidance.md

fix(strate): aim at the room's centre, not beside it

2026-08-30 00:52:22 +02:00

COMPOSER-NOTES.md

feat(descent): landing ROOMS and a walkable tunnel -- per Jahni's correction

2026-09-08 02:14:37 +02:00

fable-idea.md

perf(streaming): outer tile classifier off by default; delete mode 2

2026-09-13 02:06:24 +02:00

GDD.md

docs: a game design document, distilled from the transcript

2026-08-29 15:44:35 +02:00

INVESTIGATION-2026-08-SOL.md

docs: Sol investigation -- refutes my boundary-tile hypothesis, corrects my false accusation

2026-08-16 23:22:06 +02:00

OPSTACK-DECOMPOSITION.md

feat: TunnelNetwork stage A — the SDF spine, wrapping BuildChunkCache

2026-07-27 18:28:06 +02:00

OPSTACK-HANDOFF.md

docs: close the operator stack -- handoff rewritten in plain language

2026-08-17 03:23:31 +02:00

OPSTACK-PLAN.md

docs(opstack): record the green build and rewrite the handoff for a fresh context

2026-07-28 14:13:59 +02:00

OPSTACK-PROGRESS.md

docs: close the operator stack -- handoff rewritten in plain language

2026-08-17 03:23:31 +02:00

OPSTACK-PROMPT.md

docs(opstack): experimental is pushed and tracked -- correct the flat "never push" rule

2026-08-16 14:41:00 +02:00

PERF-ATTRIBUTION.md

measure(perf): the remaining 4.3x is ALL density evaluation -- and the classifier costs 3.5x in game

2026-09-12 08:45:26 +02:00

PERF-SAMPLED.md

measure(perf): re-rank component costs on the current build -- the tunnel core is now the largest

2026-09-15 09:30:22 +02:00

REVIEW_FINDINGS.md

feat(editor): walk a composed candidate in PIE + why rooms read angular

2026-09-04 22:03:01 +02:00

VoxelForge.uplugin

feat(gen): Jahni's playtest corrections -- rooms flattened, roads deleted

2026-09-08 14:14:29 +02:00

WORK-NEXT.md

docs(work-next): state after 2026-09-14 -- perf ledger, measured costs, what failed and why

2026-09-14 17:26:44 +02:00