Resumed round: checkpoint rebuilt and re-verified (parity 3,436/3,436, walk
5,642/5,642, W1/W8 hashes 44/44, far topology 0 diffs vs CPU and WARP, far
survival PASS). Remote GPU holes were missing clipped LOD skirts: tiles with
visible skirt faces now fall back to the CPU mesh path, leaving only 16 far
tiles GPU-rendered until a GPU skirt emitter exists.
My interleaved runs (SkirtSeamVerify, process start -> whole view):
CPU 1.522/1.488/1.430 s; GPU 4.271/1.749/1.659 s. A ~2.8 s first Present
stall hit only my first GPU launch; Luna saw it on every fresh launch, so the
trigger is still unidentified. Headless CPU clipmap 0.789 s.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Luna measured the startup density-parity check at 1,049 ms, nearly all of the
~1.13 s before shape compile; it now runs after first display and still passes
3,436/3,436. CPU hashes 44/44 at 1 and 8 workers, walkability 5,642/5,642,
headless 8-worker clipmap 0.769 s. Stopped by the owner mid-round, while
auditing the remote far holes; first-launch stall and hybrid speed untouched.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Near LOD0/1 (detail band, relief, root refinement, normals) now evaluated and
meshed on the GPU. 960 near tiles: zero sign/topology differences vs CPU and
WARP, vertex max 0.0049 mm (was 15 cm); far survival audit passes; CPU hashes
44/44 unchanged. Far connector-panel outlier (64.56 cm, LOD8) remains a
tessellation difference, and remote captures show missing far geometry on GPU.
Cold readiness (viewer clock, my interleaved runs): CPU 2.51/2.49/2.40 s,
GPU 5.05/2.68/2.80 s. The ~2.1 s Present stall hit only the first GPU launch.
Viewer spends ~1.13 s before shape compile starts (startup checks), which is
the gap between the 0.95 s benchmark and ~2.4 s on-screen.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Whole far field on the RX 7900 XTX: 112 ms total (connector rings 61 ms). Cold hybrid 2.303 s vs
all-CPU 2.314 s (within noise): the far field is no longer on the wall-clock critical path.
0 sign/topology differences vs CPU and WARP incl. 127,401 protected and 353 thin-feature cells;
repeat GPU runs identical; far survival passes on GPU meshes; CPU hashes unchanged.
Open: CPU/GPU vertex distance mean 1.15 mm but max 64.56 cm, 247 tiles over the quantisation bound.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Up to 32 tiles per deferred command list, one polled event per batch, no per-tile waits; CPU
upload/enqueue 723 -> 46 ms. GPU far meshes pass far survival (7,200 strata boundaries, 10,976
connectors, 0 missing/holes). But hybrid cold 4.36 s vs CPU 2.29 s: whole-tile GPU timestamps sum to
646 ms (0.58 ms/tile; step 3's 0.03 ms covered part of the work only), divergent per-thread shape
loops; CPU far-input export 1.6 worker-s. CPU 44/44 hashes unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
1,114 of 1,266 far tiles meshed on the GPU; vs CPU and vs WARP: 0 sign/topology differences,
including the protected/thin exception audit; max symmetric vertex distance 0.125 cm; repeat RX runs
identical; GPU field+mesh 0.029 ms/tile average. Strata survival 7,200/7,200, but connector survival
fails (GPU lacks the connector repair panels). Hybrid cold clipmap 5.74 s vs CPU 2.23 s: dispatch and
readback overhead, not compute (~33 ms of GPU work in total).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Seven LOD2-8 tiles: RX 7900 XTX field+meshing 0.335 ms vs CPU generateTile 15.81 ms; repeat runs
bit-identical; CPU W1/W8 44/44 hashes unchanged. Not yet a real gain: rock/air signs still come from
CPU masks for every sample. Viewer does not draw GPU tiles yet; no whole-view timing.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
- 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
- Relief per wavelength band compared with the owner field; the 4-8 and 8-16 voxel bands were lost.
Restored within the partition-of-unity blend; max detail gradient 0.545 (limit 0.55).
- New owner-band gate: fails all six views on de14f00, passes all six now.
- W1/W8 44/44; path hash both directions; routes 5642/5642; stair-distance, ring, thin-sheet,
coarse-panel, far-survival audits pass; bad-shell control fails.
- Cold 1.127 s vs 1.108 s (within spread); nearest-tile readiness 12.20 vs 12.16 ms.
Caveats: the 143-ray GridStair normal audit is at 62.2% vs its 60% limit (was 76.9%); in the tunnel
capture the floor shows a regular dimple pattern, and the passage shaft close-up is still smooth.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
- Stairs: cave, floor, strata and passage relief were effectively summed in overlaps (owner ray
~0.94 voxel of displacement); now partition-of-unity weights: -0.288 voxel there, max detail
gradient 0.323 voxel/voxel. High-detail patch scan 133 -> 0; exact midpoint edge roots kept
(removing them: 7,993 bad roots, 46 patches).
- Fins: exposed skirt triangle scan 1,776 -> 0; thin sheets 0; coarse off-surface panels 0.
- Path hash both directions equal; W1/W8 equal; routes 5642/5642; passage detail ratio 0.850;
far survival passes, bad-shell fails; 30-crossing flight: nearest tile 15.1 ms median, 124 ms max.
Caveats seen in the captures: the owner-ray wall has visibly less relief than before; the fin
before/after captures do not show the fins (evidence is the scan only). Cold StreamFullClipmap
0.758 s vs 0.741 s (+17 ms, Luna's interleaved run). Owner to judge: bin\VoxelForgeLookCheck4.exe.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Not verified as a whole. In the tree when Luna's session ended ("model at capacity"):
- fins: LOD seam skirts could hang into tunnel air; skirt triangles are now clipped against the solid
half-space and carry a per-face mask so a host can hide skirts between same-LOD neighbours.
- choppy LOD0 stairs: marching cubes' linear edge interpolation misplaced vertices where the relief
bends within one cell; edge crossings corrected with the exact field (82 patches, midpoint-edge
errors 18,818 -> 3,288); normal halo now ignores the +-4.125 fallback samples.
- an explicit unclamped Core distance query for audits.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
- Core compiles exact-tile analytic slabs for entering tiles; a Z-column sampler reuses the coherent
noise context along each tile column; the viewer keeps residents and prioritises nearby tiles.
- 30-crossing straight flight: nearest tile ready 15.8 ms median / 27.6 ms max; full crossing wall
121 ms peak; worker CPU per crossing 545 ms median, 834 ms max.
- W1/W8 hashes equal; routes 5642/5642; look, passage, stair, thin-sheet, coarse-panel and far
audits pass; bad-shell control fails; path-hash audit kept.
Cold whole view, measured by me interleaved (busy machine): 1.242 s vs 1.255 s for 3b37d57, parity.
Still open from the owner (LOD0): fins along a strata seam, choppy stair facets.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
From the owner's fly-through ("very impressive, in some cases slightly better than the original"):
- Streaming: Core compiles world-aligned fine-detail pages for tiles entering the view and releases
them when they leave; the viewer keeps resident tiles. One chunk crossing: 144 new tiles, 1,822
kept (was a full rebuild). A tile built as a late page hashes identically (density, mesh, payload,
topology) to the same tile from an initial build.
- LOD>=1 fins: off-surface planar connector panels are now checked against the source surface and
projected onto it; coarse audit 0 off-surface panels; far survival passes, bad-shell still fails.
- LOD0 stair patches: 2,630 sign-changing edges had one endpoint at the fixed +-4.125 fallback
(carve surfaces were excluded from distance storage); distances now completed analytically, 462
packed signs corrected; the audit now fails on any sign edge missing either endpoint (0 of 952,997).
- W1/W8 hashes equal; routes 5642/5642; look, passage and far audits pass.
Verified by me, interleaved (8 workers, cold): 1.014 s vs 1.008 s for aeeedaa, no regression.
Open: a chunk crossing still costs ~1.2 s of page compilation.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
- The copied generator, mesher and Compat stand-ins compile inside vfcore::detail.
- CheckCoreSymbols.ps1 (run by Build.ps1) fails on any unqualified Unreal-style symbol in the
library: 16,255 on 557f6aa, 0 now. A UE 5.7 CoreMinimal.h + VoxelForgeCore.h smoke compile passes.
- Output identical: 11 hash rows at W1/W8, routes 5642/5642, look/passage/far audits pass,
bad-shell control fails, six captures byte-identical.
Cost, measured by me interleaved (8 workers, cold, 4 runs each, first run excluded): 0.963 s vs
0.937 s for 557f6aa, ~27 ms in shape compilation; same math, so codegen/inlining. To be found.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
- Standalone/Core: public header VoxelForgeCore.h (no Unreal types), static library target,
ResolvedWorldSnapshot (from the schema-v1 JSON), World (compile, generateTile, density/sign,
point/segment solid, classifyBox), host-supplied JobScheduler (the core starts no threads),
sequenced, idempotent EditTransaction returning affected tiles per LOD.
- The viewer, benchmarks, edits and audits run on Core; StreamingWorld stays host code.
- Same as round 11: 11 hash rows at W1 and W8, routes 5642/5642, look/passage/far audits pass,
bad-shell control fails, six fixed captures byte-identical; cold W8 1.025 s vs 1.041 s reference.
- SHARED_CPP_LIBRARY_PLAN.md rewritten for a UBT module compiling the same sources via shims.
Open issue before Unreal: Core's implementation still compiles the Compat stand-ins named like
Unreal types (TArray, FVector...); inside a UE binary that is an ODR clash. Must be namespaced.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
- AddLanding no longer adds a shallow positive ellipsoid (floor plate check 71.3% -> 0%).
- Round 10 had removed all relief on passage walls to pass the ring check (smooth tube; rejected).
Cause found instead: the owner roughness's 8-voxel Z anchors with linear interpolation. Passage-
controlled walls now use 2-voxel anchors, blended smoothly into the 8-voxel ones.
- Ring gate (lag-20 correlation) 0.376 -> 0.038 AND shaft relief RMS 1.294 (smooth-tube control
0.276 fails); passage/cave relief ratio 1.152. Faint rings remain deep in the shaft.
- Whole view 0.984 s cold @8w (interleaved, full payload); W1/W8 44/44 hash fields equal; routes
5642/5642; look and far-survival audits pass; bad-shell control fails.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
- Eight-sample X batching of fine density rows, reuse of solved Z intervals in the candidate mask,
AVX2 marching-cubes cell classification. LOD0/1 density-grid time -61/-62%; interval/sign stage
140 -> 97 ms; candidate-axis allocation -18 MiB. Shape stage ~413 ms, detail band ~219 ms.
- W1/W8 hashes equal across 44 fields and equal to round 8; routes 5642/5642; look and far-survival
audits pass; bad-shell control still fails; six fixed captures pixel-identical.
Remote-view recapture pending (camera 500,0,0).
Verified by me, interleaved (8 workers, cold, BelowNormal 0xFF00, 2,226 tiles, 110.9 MiB payload):
1.012 / 0.927 / 0.926 s vs round 8 1.100 / 1.099 / 1.088 s.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
- Octave noise evaluated in AVX2 batches (one canonical path), fBM and ridged share samples, lattice
hash products factored; the outputs are preserved: W1/W8 and pre-change hashes equal across 44
fields, six fixed captures pixel-identical, remote views differ by at most one pixel.
- Routes 5642/5642, shape and far-survival audits pass, bad-shell control still fails.
Verified by me, interleaved head to head (8 workers, cold, BelowNormal 0xFF00): 1.092 / 1.098 s
vs 1.408 / 1.396 s for round 7.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
- Far repairs are source-aware (only where the sampled mesh lost a shell) and redundant radius probes
trimmed; ~175k two-triangle repair panels remain (+18 MiB over the no-shell 92.6 MiB).
- Pipelined start builds the shape world once; the 27-tile neighbourhood is ready ~0.73 s from
process start; pipelined whole view is no slower than direct.
- Far audit passes, bad-shell control still fails (16 missing / 16 holes); look audits pass; routes
5642/5642; 60 W1/W8 + direct/pipeline hash fields equal; six fixed captures pixel-identical.
Verified by me head to head (8 workers, cold, BelowNormal 0xFF00, quiet machine, two runs each):
1.40 s vs 1.80 s for the previous default.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
- FarSurvivalAudit: checks every stratum top/bottom boundary (4x4 columns per far tile) and every
in-view far connector (sections <=128 voxels, 4 radial rays) against the emitted mesh. Falsified:
removing one tile's shell triangles -> 16 missing, 16 holes, exit 14. Passes at LOD6/7/8.
- Far connectors follow source radial crossings; strata patches emitted at LOD7/8. Default LOD8.
- Viewer: separate near ShapeWorld for the 27 player tiles, full world built on its own thread;
27/27 tiles hash-identical to the direct build. W1/W8 60 hash fields equal; routes 5642/5642.
Cost, measured by me head to head (8 workers, cold, BelowNormal 0xFF00): LOD8 without the shells
1.17-1.25 s, LOD8 with the shells 1.80 s; payload 92.6 -> 276.7 MiB. The shells are far too
expensive, and the pipelined benchmark whole view is 2.92 s (shape compiled twice): both to fix.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
No source change kept. LOD8 changes a remote view (29k of 1.02 M pixels) and no check proves distant
strata survive; LOD7 renders the same remote view as LOD8. Default stays LOD6.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
- Profiled the band: detail noise ~70% of worker time, workers balanced (2.5% spread).
- Local amplitude envelope skips detail on 34% of candidates; owner cave roughness evaluated on
global 8-voxel Z anchors and interpolated. Detail evaluations 7.07 M -> 3.53 M.
- The field changes slightly (mean channel delta ~1.7-5.4 /255 in the six captures); all look
audits pass, routes 5642/5642, W1/W8 hashes equal on all nine outputs.
- Whole view 1.953 s @8w full payload: no end-to-end gain shown (stream timing noise ~10%).
- Rejected: detail band radius 5 (fails GridStair).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
- Per-vertex biome lookup was the payload cost (~365 ms); far tiles now take the biome palette from a
per-tile grid. Same channels (palette, slope, border blend, UV) and ceiling/ground sections.
- Opt-in --max-clip-level=8: 6,498 -> 2,226 tiles, stream 0.876 -> 0.428 s, 1.451 s process wall.
Kept opt-in: six fixed views cannot prove every distant thin stratum survives.
- W1/W8 hashes equal across 11 outputs; routes 5642/5642; five captures pixel-identical, passage
382 px differ (max delta 16).
- Rejected with measurements: LOD6 empty-block proof (-21% samples, no wall gain), far supertiles
(no complete groups exist).
Remaining long pole: shape compile 0.854 s.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Takeover after the PC reset. Verified at 1 and 8 workers: density, geometry, full payload and
topology hashes equal for all nine outputs; routes 5642/5642; crater transaction 17 ms; player
neighbourhood mesh-ready 23 ms after streaming starts.
- The plugin's materials consume UV, the packed biome/slope/blend colours and the ceiling/ground
sections, so the viewer-only skips stay off by default (diagnostic only: 1.53 s without them).
- UVs staged as FVector2f (matches the game stream; -20 MB, no speed change). Far-bin reserve 24x.
- LOD6 is the long pole: 32.7 M density samples, 1.83 M vertices, ~1.1 s.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Verified before the reset (density + full mesh hashes equal at 1/8/16 workers, routes 5642/5642,
look audits green):
- merged field lookups, parallel mask merge, compact inline spatial bins, line-local relief hash
cache, lattice-specific mesher entry, fused sign+magnitude evaluation, far connectors as one
capsule each (181,836 -> 16,491 far primitives).
- crater edit (sphere+box+capsule, one transaction across LOD0/1): 15-27 ms; player 3x3x3
neighbourhood meshes ready ~18 ms after streaming starts.
NOT yet verified (in the tree, unhashed): standalone-only skips of UV/colour arrays and of the
ceiling/ground material grouping (best cold whole view 0.958 s @16 workers) and a far-bin reserve
tweak. Those skips drop data the plugin's materials may need: not a game-path win as-is.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
- Removed the thin automatic room-support slabs that left floating caps (in-view detached thin
sheets 9 -> 0; the audit now also fails on isolated small components).
- Bounded 16-voxel floor/wall blend (18 rejected: it created a sub-voxel sheet), 3-voxel
strata/cave blend, C1 tunnel radius profile (0 radius/slope steps at joins).
- Surface relief: broad octave with amplitude modulation instead of uniform bumps.
- Cold, no cache: 64 m full detail 0.924 s (8 workers) / 0.715 s (16); whole view 2.016 / 1.605 s.
- 1-worker and 6-worker hashes match; 5,642/5,642 passage samples open.
Not solved: regular bands in the narrow passage shafts, a few slivers on tile boundaries, faint
wall/floor plates; close-view mean surface distance to the owner field got worse (see REPORT.md).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Luna's continuation stopped when Windows could not start new processes (0xC0000142, low memory).
Fixed-view join step: chamber 0.041, tunnel 0.434 (bar 0.450, barely). Routes 5642/5642 open.
Shape stages 0.963 s / 0.870 s. Floor join plates still visible; final checks not run.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
The look round was stopped by the system for low memory (not a code fault) while
checking worker-independent hashes and refreshed owner-distance comparisons.
Saved state, not yet reviewed:
- passage floors carry rock detail (flat-floor check passes)
- tunnel radius step between joined segments 0.0000 voxels across 240 joins (was 2.9156)
- 3-octave surface relief
- cold 64 m full detail 0.982 s (0.703 shapes + 0.279 mesh), whole view 2.490 s, 8 workers
Pending: determinism 1 vs 6 workers, owner-distance numbers, final captures, report.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Owner's rooms, tunnels, passages, landings and strate shells compiled as shapes, solved along
lattice lines with packed signs; rock detail only in a band sized by its supremum; far levels from
the same shapes; multithreaded meshing. 8 workers, BelowNormal, affinity 0xFF00, zero cache hits:
full detail (960 tiles): 669 ms shapes + 290 ms mesh = 0.959 s (was 20.3 s)
full clipmap (6,498 tiles): 870 ms + 1.398 s = 2.268 s (was 170.6 s)
Checks: 5,642 passage centreline samples open; smoothed and per-voxel-noise variants both fail the
detail audit; same hash across runs and 1 vs 6 workers. V toggles owner field / shape world.
Look: 6-12% of lattice signs changed; captures show flat planes without detail (passage floor, a
wall band), telescoping tunnel segments, blocky stair patches. Not accepted as the owner's world yet.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
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
Cause found by profiling: room/SDF graph-cache churn (exact XY bounds + sample order rebuilt it
constantly). Region grouping (one LOD3 tile: 1,084 -> 229 builds), taller-tile traversal for LOD6
(36,480 -> 12,160), extent-dependent cache slots, lambdas instead of recursive std::function.
Serial ms/tile, 64 nearest tiles per level (before -> after):
L0 50.4 -> 50.6, L1 112 -> 110, L2 24 -> 22, L3 223 -> 73, L4 269 -> 134, L5 393 -> 127, L6 283 -> 145.
Hashes of all levels unchanged; parity 3,436/3,436 bit-identical.
Full clipmap, 16 workers BelowNormal: full detail 66.6 s, all 12,054 tiles 175 s. In-stream L0 tail is
high (median 83 ms, mean 198 ms, p95 669 ms) vs 51 ms serial.
Streaming: recentre keeps the intersection, generates only missing tiles nearest/finest first;
one crossing = exactly 441 new tiles in 6.1 s, no stale results.
Lighting: normals already come from the density gradient; the striping was the squared headlamp;
broad linear fill + warmer palette. Daytime CPU rule documented in Standalone/README.md.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Hashed TMap/TSet and real TInlineAllocator storage did not move level 0 (72.1 -> 74.1 s, within
noise): the slowness is elsewhere. StreamingWorld.h started (first build hit an int64/uint64
mismatch), library plan drafted (SHARED_CPP_LIBRARY_PLAN.md), before-normal captures saved.
Uncompiled/unmeasured parts: streaming. Resume by forking session 01a0fd2e-3c3b-7050-b4e9-628fdcc4d553.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Standalone/ (C++20, MSVC /O2 /arch:AVX2 /fp:precise, Build.ps1): copied density pipeline, op stack,
cave morphology, passages, strate manager/composer, noise, DetSinCos, mesher, clipmap selection,
with a UE compatibility layer (Compat/). Owner data exported once to Standalone/data/owner_world.json
by a default-off explore-commandlet exporter (the only plugin-source change). 25 cm voxels.
Parity: seed 0, 3,436 samples (strate edges, negative/far coords, rooms, tunnels, passages,
landings): bit-identical. Viewer: full 7-level clipmap, per-level toggles 1-7, T = round winding
tunnels (standalone option), N/P seed, F wireframe.
Timing NOT yet a baseline: 515 s for 12,054 tiles (L0 70 s, L6 296 s); the compat TMap/TSet are
vector-backed with linear lookup, which the plugin's caches do not expect.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Bounded smooth blends at junctions (3.1% of cells), QEF dual contouring on the exact crossings and
analytic normals, band-limited surface relief (bound 0.60 m, observed 0.40-0.44 m, R toggles it),
sparser strata. Exact signs vs the blended reference at every node, thin features kept, same hash
across worker counts. Mesh has 0 boundary edges but 3-7 overused edge groups: not collision-ready.
Near CPU total, 6 workers: 241 -> 836 ms. The new stages all run on ONE thread: blend index 186 ms,
blend evaluation 201 ms, dual contouring 322 ms, relief 61 ms.
Look: slab stack gone, joins rounder; chambers read as broad crumpled facets, junction seams still
flaky. Viewer built as interval_viewer2.exe (the owner had the first one open).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Viewer generates the world at startup with the spike's code (timing shown), free-fly camera,
wireframe, near/far toggles, seed cycling. Benchmark hashes unchanged (e5ad62303de5f641).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Standalone C++20 (MSVC /O2 /arch:AVX2), no Unreal. A 372-shape corpus (rooms at the hardest
profile's scale, winding tunnels, pillars, stalactites, shelves, floating masses, thin strata,
rotated shapes, crowded junctions, craters), built two ways: point-by-point reference vs column
intervals on 3 axes + 64-bit sign words + bit-scanned active cells + meshing from analytic crossings.
64 m near region at 25 cm (17.4 M nodes), 5 seeds, fresh processes:
workers near CPU total (median/p95/worst) interval stage mesh stage (single-threaded)
1 349 / 358 / 358 ms 154 ms 188 ms
6 241 / 280 / 288 ms 48 ms 191 ms
8 240 / 271 / 277 ms 43 ms 193 ms
Near mesh ~563k triangles, 13.5 MB; total output 19.9 MB; busiest line 58 candidates (< 64).
Correctness: 0 sign mismatches vs reference at every node, 0 missing crossings/open contours, thin
strata with no lattice node inside get explicit triangles (dropping them is caught), same hash on
1/6/8 workers and a repeated run. Far field: 47k nodes / 92k patches in ~8.6 ms, but fixed-grid
ellipsoid sheets, not a validated cave hierarchy: unproven. No noise detail in this corpus.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Stopped by the owner's decision (in-game lit captures were eating the round); he judges the look in
game instead.
1. Fall-through with the research field: fixed. Deep drop, gate on: 2 holds (0.50 s), min foot
clearance +43 cm; gate off: penetrates (-278 cm). Stress fixture offsets converted to metres.
2. VOXEL_SIZE 25 -> 12.5 cm. Physical sizes kept in metres (player, steps, gate, insets, research
field sizes, dig radii, test thresholds); class defaults that count chunks/voxels doubled
(ClipRadius 6, RenderDistanceChunks 1536, ViewDistance*, StrateViewMargin 6,
PawnCollisionRadius 2, SkirtCells 4, EdgeSealThickness 128, decoration spacings, density volume
res 256). Asset values override these defaults: the owner must double his own.
Ring scale, six workers, idle, step-2f field before the look changes: 64 m cube (519^3) 4.55 s;
32 m cube (259^3) 0.95 s.
3. In game: tiles go through the brick evaluator (proved by a log marker); profile switch rebuilds
the loaded world (~5,800 tiles); full-detail fill 12.4 s after spawn in the visible-window test.
4. Look: round winding tunnels, per-column ellipsoid/lattice caches. All step-2f geometry gates pass
at 12.5 cm (0 crumbs, 1 solid component per profile, solid share 69-87%). 259^3 ring, six workers:
1.08-2.02 s (worst HyperComplex) -- the 600 ms gate FAILS. Lit captures failed (hidden/off-screen
window, then blank frames); not pursued.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Stopped by low memory (another app at 14 GB; processes failed with 0xC0000142).
r6 (compiled and measured, owner playing during the run), step 2e -> step 2f r6:
crumbs < 1 m^3 (max per profile) 2,898 -> 3 FAIL (0,0,0,0,2,1,3,2)
flakes (shard share / p95 angle) 40.7% / 84.2 deg -> 2.1% / 33.1 deg pass
lattice-scale detail energy 0.832 -> 0.016 pass
grid signature (autocorrelation) 0.666 -> 0.349 pass
min solid share / roof openings 2.7%, 26,657 -> 79.2%, 0 pass
worst profile ring time 642.8 -> 636.3 ms FAIL (6 workers, 259^3)
Step 2e measured 642.8 ms here vs 570 ms idle earlier: the game was costing ~12% during this run.
Brick proofs + audit, seams + injected seam, 1 vs 6 worker determinism pass. After r6, feature
probabilities were lowered to remove the last crumbs: NOT compiled, NOT measured. TraversalSeeds,
in-game captures and the LOD-rejection figure for 2e are still to do. Reject screenshots included.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Editor-only, default off. Seeded chamber graph (jittered XY grid, spanning tree, optional loops,
tunnels, shafts, pillars, ledges, spires, floating masses, basins), 8 profiles, four detail bands
at every surface-affecting 25 cm sample, sound brick proofs. 259^3 ring, six workers, medians of 3:
404-570 ms (worst TerracedBasin 570.0 ms). Detail, seam, bound-audit (0 violations, undersized
control caught), determinism (1 vs 6 workers) and TraversalSeeds all pass; live tests untouched.
Not accepted yet: the profiles are 81-97% open air. In the slices a "TightNetwork" is a large void
with scattered solid blobs (rooms of 0.5-0.76 x pitch overlap and merge), and the finest detail
band shows as salt-and-pepper speckle with 22-109 separate 1 m components per ring: fast partly
because air bricks skip, and the look needs the owner's judgement. Also missing vs live: maze,
flat/plain, crystal-chamber, surface-world and underwater grammars; water geometry.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Stopped by low memory (another app held 12.5 GB; processes failed with 0xC0000142).
Editor-only, default off: a seeded connected room graph (varied chambers on a spanning network,
optional loops, shafts, rock features, ledges, basin floors), eight parameter profiles, the
step-2d brick driver and detail bands, a new test VoxelForge.ParameterizedGeneration.TraversalSeeds
(PASS in step2e_functional_01, which also proves the tree compiles). Existing live tests untouched.
Not yet run: the eight-profile ring benchmark, variety metrics, slices, detail/artefact/determinism
checks.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Editor-only, default off (voxel.ResearchRedefinedDensityField). The redefined field is a seeded,
periodic cross-corridor pattern per strate plus the authored passages, landings and a shaft tree,
with four value-noise detail bands at every near-surface 25 cm sample and sound 8^3 brick proofs.
259^3 ring, six workers: 308 ms median (467 ms without skips) vs 8.9 s for the live field.
Detail, boundary-artefact, determinism (1 vs 6 workers, two runs) and bound-audit checks pass.
Why it is not a pass: the slices show a mostly-solid field with a few straight corridors, where the
owner's world is mostly open space; rooms, the tunnel network, composer recipes and archetypes are
gone. Much of the speed comes from 60% of bricks being proven solid in that simpler world.
What it does show: detail at full resolution plus brick skipping costs ~80-130 ns/sample; the
cost of the live field is how rooms/tunnels/passages are represented, not the detail or the lattice.
The two walkability tests were redirected to the research field in the working tree; that change
is NOT committed (saved as scratchpad step2d-test-redirect.patch) so they keep testing the live world.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Editor-only, default off. 259^3 ring split into 1,000 aligned 32^3 chunks; cache/index builds
bounded by chunks (948 room/SDF caches, 800 active-op lists, 8,532 index tables); 800 block setups
+ 201,243 scalar fallback points (1.16%); no per-point GetDensityAt on supported routes.
Six workers: live 7.07 s wall / 1.678 us per sample; prepared 7.92 s / 1.595 us.
Candidate visits 21.2 -> 15.8 per sample, evaluations 6.31 -> 5.75; room / tunnel / tunnel-core
evaluations unchanged (48.1 M / 5.0 M / 38.4 M for 17.4 M samples). Passage evals 18.1 M -> 8.4 M.
Exact: 0 Q16/sign disagreements, five seeds, same hash on 1 and 6 workers, reach shrink caught.
Luna's verdict: 600 ms is not reachable while every lattice point evaluates the exact current
density; 27.7 worker-CPU seconds need >= 7.7x less work.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Editor-only brick API (<= 8^3 points per call, TunnelNetwork/Underwater op-stack routes, 1.16%
scalar fallback). Exact: 0 Q16/sign disagreements on the full 259^3 ring and the five-seed probe
set, same hash on 1 and 6 workers, reach shrink still caught (18,107 Q16 / 735 sign).
Speed: 69.19 s wall, 17.6 us/sample vs live 7.04 s, 1.65 us/sample. Stack sampler (inclusive):
BuildChunkCache 93.5%, PropagateSdfOverBox 92.5%, BuildActiveOpList 92.6% -- each brick rebuilds
the cache windows and active-op lists; EvalBlock itself is 3.6%.
Live path attribution (inclusive, 68,934 stacks): op-stack RoomGraph EvalBlock 54.4%, SDF and
spatial candidate scans ~28%, tunnel core 18.4%, passage floor segment 15.0%, structural posts
14.6%, noise octaves 2.6%, disturbances 0.6%, diff layer 0.3%.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Editor-only ablation (explore commandlet, -regioneval, default off): 8^3 bins for rooms,
room-floor joins and tunnel chains, but every point still goes through GetDensityAt.
259^3 ring, seed 1337, six workers:
live : 6.776 s wall, 29.16 s worker CPU, 1.679 us/sample, 21.2 candidate visits/sample
binned: 19.54 s wall, 50.02 s worker CPU, 2.879 us/sample, 16.4 candidate visits/sample
Correctness: zero Q16 (x16, ties away from zero) disagreements on the full ring and ~284k probes
on 4 more seeds (composer recipes, warped rooms, passages, landings, strate edges, edits);
shrinking room reach to 0.5 -> 18,107 Q16 / 735 sign differences (the check bites).
Not the planned region program: the step-2 question is still open. Baseline for the next attempt:
1.68 us CPU per sample on the live path.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Rebuild plan step 1 (RESEARCH-SPEED-2026-09.md), all behind default-off switches.
A. Direct cook on a dedicated thread, 27-tile support cohort: core cook p50 1.50 ms, p95 2.28 ms;
a 3,949-tri tile 1.67 ms extract+cook. PASSES the 10 ms gate. Through RMC's route the same
support bodies became ready 8.46 s p50 / 10.80 s p95 after their meshes existed: FAILS 150 ms.
Exact-target capsule sweep confirmed (0.75 ms after body update).
B. 100 rapid underfoot digs on CMC with a field override: 883 false-support frames, 0 false-air,
0 invented velocity. CMC's floor comes from collision-scene traces -> step 4 = field-aware Mover.
C. 341 prequeued tiles, render on: cap 4 = 238 applies/s, cap 8 = 476, cap 16 = 639 (drain p95
2.22 ms vs 2 ms gate); first visible 1 frame after release; no RT/RHI backlog. Tune drain,
no grouping needed yet.
D. Same 128^3 region, 64/64 tile hashes equal in export and game. The "263 ns/sample" was export
WALL time with 15 workers sharing a lattice (~3.4 us CPU/sample); game tile time includes
meshing and duplicated borders (+22%); contention cap 1 vs 14: 74 vs 112 ms per tile.
Profiling is not the cause.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Compiled and run (cap 14 / 12 / 8), then stopped: another app held 12.5 GB, free memory fell
under the 6 GB launch floor and processes failed to start (STATUS_DLL_INIT_FAILED).
Measured so far (owner world, 120 s windows):
- 343 L0 tiles, request->collision-ready p50/p95: cap 14 = 262/440 ms, cap 8 = 196/238 ms;
scheduler queue wait p95 141 ms -> 0.9 ms. Initial support tile at cap 14: RMC submit->body
7.47 s, first check->ready 9.73 s; at cap 12: 0.33 s / 2.54 s; at cap 8 a 3,949-tri tile
reached body update 2.8 ms after RMC submission. Waiting, not cooking, dominates.
- NOT confirmed: a capsule sweep hitting the exact target body (the 0.75 ms hit was another
component); the 10 ms core-cook and 150 ms/27-tile thresholds are unscored.
- CMC cannot take a same-frame density floor: FindFloor/ComputeFloorDist trace the collision
scene. Underfoot digs (100-dig test), upload throughput (render-enabled caps 4/8/16) and the
8x per-sample question are not run. Per-sample profiling was fully disabled in these runs, so
it does not explain the 8x.
Switches (all default off): voxel.Step1GenerationCap, Step1PauseGenerationAfterSeconds,
Step1MaxMeshApplies, Step1PrequeueVisibleLOD0, Step1UploadDisableCollision, Step1CollisionProbe,
Step1CoreCookBenchmark; harness -EnableRender. Also: plan step 3b, AVX2 need not match scalar.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Round three (Sol fork): keep precise float procedural math, restructure evaluation (region
programs, feature binning, brick certificates, shared lattice), a versioned 16-bit lattice as the
common mesh/edit/collision cache, integer edit composition on GPU for craters. Seven ordered steps,
each with a numeric pass/fail gate; day one instruments collision and upload clocks.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Fork of the Sol research session with the owner's answers (player-only 1 s collision; bounded
digging but fast crater spells). AVX2 integer kernel 4.25-5.86x over its scalar reference with
exact parity; radius-10 m crater = 714 bricks, cached-field edit < 1 ms (render not measured).
Cost estimate revised to 6-12 engineer-months; kill test gate <= 0.35 s worker wall on real plans.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Sol 6 (high, web search) research round: recommendation, three architectures, determinism
contract, migration with a 5-day standalone kill test. Standalone probe in
Saved/Research/brick_kernel.cpp (not VoxelForge code): 16.8M integer samples in ~0.36-0.48 s,
brick-classified ~0.1 s, 0 sign mismatches.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
1. Collision cook breakdown (request -> ready), critical support tile:
spawn (0,0,-1): generate 107 ms, apply < 1 ms, RMC submit->body 479 ms, total 587 ms;
deep floor (0,0,-7): generate 82 ms, RMC 1,441 ms, total 1,548 ms.
Almost all of the wait is inside RMC (end-of-frame update + async pool + Chaos cook + body
install, not separable from our side). Critical priority does not shorten it. Collision is
async, complex-as-simple, fast-cook off; skirts are ~16-18% of collision triangles.
No cook change made. Startup outlier: first support tile waited 2.09 s for its game-thread apply.
Underfoot dig: a carving brush that intersects the pawn's capsule and support tile now
disables the old body and withdraws readiness on the carve frame: 0 frames standing on removed
ground (replacement 91.8 ms later). 8 neighbouring digs: 0 gate frames. Gap audit 0.
2. Stack sampler, 49.4 s game path: top resolved leaf was the profiler's own
VoxelDensityProfile::BucketFromName (4.86%). Linear label scan -> CRC-keyed lookup with exact
compare: 59.26 -> 57.87 ms/tile (-2.35%), 980/980 tile hashes identical.
3. Measured only: F (classifier mode 1, no exact check) +48% worker CPU, 0/980 tiles differ ->
not worth it. H (handoff component scans) p95 0.07 ms/frame -> not worth it.
E (FullResClipLevels=1): worker CPU 56.7 -> 35.7 s, settle 9.6 -> 7.0 s, crossing samples
-30%; changes the look, owner's decision.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
- Dig near the pawn: a same-tile collision recook withdrew readiness although the old body was
still installed, so the gate held a standing pawn (957 of 1,931 frames over 8 near digs).
A tile now stays ready during a recook while collision is still wanted and its existing body is
valid. After: 0 gate frames, 0 falling frames over 2,604 frames. Dig directly under the pawn
still falls (~127 cm, max ~502 cm/s). The reported "accelerating toward the ground" was not
reproduced; the false hold is the measured part.
- See-through: the retire logic only waited for replacements that overlapped the old tile in
volume and missed desired neighbours touching at a face/edge/corner. Handoff now includes
contact. New surface-witness gap audit (24 surface points per tile, skirts excluded, catches an
injected early hide). 6000 cm/s walk, 75 s: 3 gaps / 51 frames / longest 603 ms -> 0.
Dig session: 0 -> 0. Overlap audit and edit ghost stay at 0.
- bEnableDensityVolume defaults to false; when off: no volume, no fill thread, uploads or tick
work, no volume-dependent proof bypass; terrain MIDs get VolP5.z = 0 (enable off), other VolP*
black, VolTex* null, so M_Test1's volume nodes return their neutral result.
Gate: deep gate-on passes (79.9 m, 3 holds, -0.101 cm), gate-off penetrates (-5.1 cm).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
From the outside performance review (items 1-4; item 5 prototyped and reverted).
1. Collision gate: a support request outside the stamped desired set was cancelled by the rebuild
it forced, and the gate held the pawn with gravity 0. Baseline deep-drop (floor at (0,0,-7),
outside the aligned L0 ring): 1 hold of 30.0 s, never released. Fix: the target and 3 tiles
below are retained through rebuilds at CollisionCritical; hold only when the swept capsule is
about to enter pending support; release restores movement mode, gravity and downward speed.
After: 3 holds, 2.21 s total, 112.8 m horizontal progress; gate off still penetrates (-7.4 cm).
2. Suspected cleanup regression (out-of-layout stacks): not real. 44 paired boundary tiles,
30.9 ms vs 32.6 ms at b4ac996, hashes 44/44 identical. No change.
3. PawnCollisionRadiusChunks (new, default 1): L0 tiles outside the collision set stay visible
without Chaos cooks; anchors and the gate's support column are unioned in. Cook requests
~217/min -> ~30/min (-86%). Critical support ready time did NOT improve (9.5 s -> 10.6 s in the
deep-drop fixture; the wait is the cook itself, ~7.4 s cook-to-ready there).
4. Dig: exact pre-edit L0 density cached (one 35^3 grid, 171.5 KB), carve replays current mods on
it. Sync remesh 99.4 ms -> 4.7 ms. 16/16 random carves hash-identical to full regeneration.
5. Parallel grid fill for the critical tile: 16/16 hash parity, but the waited-on tile was already
meshed; the wait was collision cooking. Reverted.
Review G answered: M_Test1 still ray-marches VolTex0 (custom HLSL mini-sun shadow), so the density
volume is live. Known and left alone: outside the strate layout the fallback density can be solid
while the gate treats those tiles as air.
Tests: VoxelForgeCollisionDesiredTilesTest, VoxelForgePreEditDensityCacheTest; no-overlap, edit
ghost, descent corridor pass.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Stopped mid-round: the PC ran out of memory (another app held 11 GB), new processes failed to start
(0xC0000142), and the round ended without verifying anything after the edit.
Confirmed in code (performance review finding 1): a support request outside the stamped desired set
forces a rebuild that cancels it, the gate asks for one unresolved tile at a time, and it holds
before the pending tile is anywhere near the pawn's next step.
Uncompiled edits: a short collision-only support column retained through desired rebuilds and
prioritised; hold only when the swept capsule is about to enter a pending tile; downward speed
restored on release; opt-in deep-drop fixture (carved shaft) in the stress harness.
Evidence so far: shallow 10-voxel fall, gate on, before the edit: 1 hold of 1.51 s, clearance
+120.6 cm. The 80/160-voxel attempts spawned OUTSIDE the strate layout (support unavailable,
zero holds) and failed the penetration check at -11.1 cm: not the reviewed loop, but worth a look.
Compile risks: GetGravityZ(), TSet pruning in the rebuild, the fixture's carve/log arguments.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
- Passage proof bounds now enclose everything the passage code can write: separated switchback
legs, outward door approaches, native floor slabs with a derived relief supremum, and the final
body-envelope carve (segment XY extent + half-width, floor range + player height + step).
- ClassifyTile: the cave-only AllSolid shortcut could restore a solid verdict after a passage
carve candidate was found; it now keeps the tile mixed.
- Descent corridor test asserts no corridor stencil tile gets a uniform verdict.
Evidence: SealedSolidProofSoundness and ClassifyTileSoundness pass; 64 owner seeds, 0 body
blockers, 0 uniform stencil tiles, CRCs match. Negative control (passage reach shrunk to 4 voxels):
7 sealed-solid proofs among 620 corridor tiles, i.e. the check bites.
KNOWN COST, not fixed: tile profile, owner world, seed 0, 15 s window: passage-candidate tiles
83.6 ms (de05e29) -> 107.2 ms (+28%), all tiles +5.4%. Two segment-rejection filters did not help
and were removed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
What the pillars were: at switchbacks the native-floor D-section clip in
UVoxelStrateManager::EvaluateModifierSDF used ONE floor branch (the nearest by projection) for
neighbouring capsule segments, so one leg's floor clipped the other leg's carve and left rock
standing in the corridor (seed 0 witness: density -29.45, carve SDF +3.28; with
voxel.TunnelAblateNativeFloor=1 the same point is air). Separately, the lower ramp was routed
through the lower landing room (sign error in the lower "outward" score), where room ownership
correctly erased the tunnel floor: up to 12.5-voxel drops.
Fix: each segment clips with its own floor branch before the union; switchback legs separated
and door approaches moved outward so ramps leave their landing rooms; a final carve clears the
player body envelope (1.36 + 2x1 voxel wide, 1.8..8.8 voxels above the route floor) after all
writers, with a room-SDF guard. No floor slab, no fill, no blockade on the through edge.
Owner world, 64-seed sweep, stations every 0.5 voxel x 5 lateral x 14 heights:
before 89,603 body blockers in 60/64 seeds; after 0 in 64/64, 0 floor/width gaps, worst step
0.958 voxels (limit 1.8). Seed 0 built twice: identical density CRC.
Crossing at default speed still stops on capsule_overlaps_pending_floor (also before); 500 cm/s
reaches B.
NOT YET CHECKED: sealed-solid / classifier soundness against the moved passage geometry, and the
generation cost of the per-segment projection.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
The old criteria (travel distance) could not tell a working gate from a broken one: every run,
gate on or off, ended at (455.7, 0, -58.1). That spot is measured air ~1.2 m above a real floor
(density zero-crossing at world Z ~ -275.7 cm), so it is a terrain trap, not a fall-through.
The stress test now drops the pawn fast onto an unready support tile (0,0,-1) and samples the
density at the capsule feet every frame, sweeping at <= half a voxel between positions. PASS =
no measured clearance below -5 cm.
Owner config (radius 3, max level 6, distance 768, sheets off):
- gate on: 10/10 PASS, min clearance +120.5..+120.9 cm, one hold of 0.43-0.47 s each,
support submitted in ~0.7 ms, collision ready in 0.43-0.47 s
- gate off: 3/3 FAIL terrain_penetration, clearance -5.7..-8.1 cm
Edit-ghost, clipmap no-overlap sweep: pass. Test-harness code only; no production path changed.
This validates a49725d.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
- RequestCollisionGateSupportTile submits the level-0 support tile immediately, even before the
first desired set; the desired rebuild adopts the same pending key (1 early request, 1 adoption,
1 ready callback in all 10 runs). First check -> submit went from ~2.2 s to ~1 us.
- The 2 s gate release timeout is removed: at the owner's startup the game thread applies nothing
for ~2.2 s, so the timeout always fired and dropped the pawn. The gate holds downward Z only;
horizontal movement stays free.
- Controls (previous round): legacy selector+reconcile failed the stress repro 0/5, aligned 1/5,
so 85a803b did not regress the gate.
NOT VALIDATED: the stress repro's pass criteria were rewritten (support submitted/ready, a hold,
progress during a hold). 10/10 "pass" under them, but every one of those runs, and the gate-
disabled run, ends stuck at (455.7, 0, -58.1), ~1.55 m below the Z~97-100 at which the old
passing runs travelled 113 m. Whether that spot is a real pit or the pawn below the surface is
unresolved; until it is, a pass under these criteria proves nothing about falling through.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
- Clipmap: finer rings snap to child-pair boundaries, so no two desired tiles of different
levels overlap. Owner config 635 overlapping pairs -> 0; LOD0 ring 343 -> 512 tiles.
New test sweeps 6,174 cases; fails on 7486c29 as intended, passes here.
- Shared room-graph cache: Entry->Cache read/written only under the mutex (race fix).
- ReconcileReadyTransitionVisibility indexed: moving p95 32.0 ms -> 0.31 ms.
- Block-vs-scalar density equivalence test: 5.1M samples bit-identical; catches an injected
optimistic EffectOverBox (5,612 mismatches).
UNVALIDATED for merge: collision-gate stress passed 1 of 4 runs on this code (fails at 4.56 m,
support collision unready at start); it passed on 7486c29. No control run yet.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
The owner reported a wall staying visible after he carved through it, and overlapping LODs when crossing slopes at speed. Two causes: an edit only re-meshed the level-0 tiles of the touched chunks, so coarser tiles kept pre-edit geometry until something else re-queued them; and superseded tiles stayed drawn until the budgeted teardown reached them.
Now: edits fence late pre-edit async results and re-mesh every affected loaded/pending level; a replaced tile is hidden as soon as its replacement is fully covered, while destruction and collision removal stay budgeted (the anti-spike rule that made teardown budgeted in the first place). Load-before-unload is unchanged, so no holes.
Measured: edit ghost 40.095 s (unresolved at test exit) -> 0.172 s; 6000 cm/s slope overlap 19.952 s with 153 overlapping pairs -> 0.000 s with zero events. Gate stress PASS (113.68 m, 2 holds, 0 full stops). Apply-frame cost peaked at 0.979 ms against a 2 ms budget.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Scheduling rule: CollisionCritical -> Edited -> LOD0 -> Horizon, with coarse work capped so 4 task slots stay reserved for near work; result queues follow the same order; epoch and cancellation handling unchanged.
Measured at MaxClipLevel=5 / RenderDistanceChunks=768: spawn ring request-to-ready 53.632 s -> 1.171 s; spawn ring from trace start 68.252 -> 62.549 s; moving first complete ring 66.808 -> 63.917 s, warm ring p50 0.267 s, per-tile moving LOD0 p50 69.4 -> 62.0 ms; full settle of 10,841 tiles 110.7 -> 106.9 s. Gate stress PASS (139.99 m, 0 full stops); static perf PASS.
⚠️ NOT SAFE TO MERGE YET: the strate-crossing test fails in both patched runs with capsule_overlaps_pending_floor -- the exact guarantee the gate exists for. Codex suspects the owner's current DA_Settings (seal 50, clip level 5) rather than this change, since older passing captures used different settings. Control needed: same test, same assets, on 370c39e.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
The sheet system models terrain as one ground and one cap height per column, which cannot express this world's overhangs, floating islands and open sky; three attempts to fix the MC/sheet seam produced holes, then caps over the void, then both. Measured alternative: cover the render distance with real marching-cubes tiles and switch the far-sheet ring off.
Benchmarks at RenderDistanceChunks=768, all tiles loaded at steady state: L4 57,473 tiles / 30.1M tris / 837 MiB / 2,514 s worker / 296 s to settle; L5 10,841 / 8.1M / 226 MiB / 375 s / 108 s; L6 3,281 / 2.9M / 82 MiB / 127 s / 77 s. Defaults now MaxClipLevel=5, bFarSheetRing=false. Invariants: strate crossing PASS, gate stress PASS (140.06 m, 0 full stops), determinism PASS with 10,841 tile hashes identical vs -voxel.CrtFma3=0. Peak memory ~4.2 GiB in all three.
Owner action required: his DA_Settings still serializes MaxClipLevel=3 (asset edits are out of scope for Codex), so his world needs that value raised to 5 to use the new far field. At twice the distance the estimate is ~39.5k tiles / 29-30M tris / ~0.8 GiB / ~1,500 worker-seconds, which would want L6 rather than L5.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
ParameterRoll's 5 false box verdicts are fixed: fine room-graph intervals are now checked against exact lattice samples before a uniform verdict is accepted, while the SDF bound keeps its conservative lower bound. Lattice sweep 600 tiles / 0 violations; ParameterRoll passes with 9,577 exact lattice leaves and 774 interval disagreements caught.
The owner's ceiling holes are NOT explained by this: in his recorded session outer_classifier_mode=0, solid-proof skips were 0, and air-proof skips were outside the strate. The trace points at the far-field sheet system instead: sheet tile L5 (0,1,0) is deliberately empty inside the sheet hole while its bridge tile L4 (0,1,-1) meshed 662 ceiling triangles. Continuous ceiling is NOT verified: every render launch failed (D3D12 driver error, stalls before map load, the 6 GB cap), so the round produced no matched after-render.
Cost note: this round used 3.46M tokens on Luna 6 at max, most of it burned on failed render launches. Verification by render is unreliable in this harness; verify ceiling continuity from tile/geometry data instead.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
The owner marked missing tiles around him in a 64-chunk surface strate with RenderDistanceChunks=768. Cause: the desired-set vertical clamp used the player's small window (bClampViewToStrate on, ViewDistanceUp/Down=5, margin 3), so at chunk Z=0 only [-5..3] was ever requested; the rest of his strate was never asked for. Second cause: with MaxClipLevel=3 and a two-level sheet span the outer ring is level 5, while the marching-cubes loop stopped at 3, leaving level 4 empty.
The band now follows the player's whole strate plus margin ([-66..3] in the pinned case) and keeps the small window only when there are no strate bounds (inter-strate gap), so the hidden-rock column the clamp exists to avoid does not return. Bridge shells are emitted up to the level below the sheet ring, filling level 4.
Pinned case desired tiles per level, before -> after: l0 343->343, l1 218->218, l2 129->178, l3 80->178, l4 0->178, l5 4801->9586; total 5,571 -> 10,681 (+91.7%). New focused fixture VoxelForgeVerticalStrateReachTest fails before and passes after. Stress repro passes (140.089 m, 4 holds, 0 full stops); strate crossing passes (0 holds, 0 floor overlap).
⚠️ Cost not established: the post-fix perf sample was incomplete and showed a 55.7 s p50 queue wait, which may be the fixture's extreme settings (768 render distance -> 9,586 sheet tiles) or a real backlog. Measure in the owner's world before merging.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
The owner could not do speed runs: the gate stopped the whole pawn whenever streaming fell behind. Measured at 6000 cm/s on 0dd45ca: one 18.9 ms hold followed by terminal non-resume, 4.522 m of a 120 m route. The gate now cancels only downward motion into unresolved ground, keeps horizontal velocity and player control, ignores air and out-of-layout tiles, and requests a blocking support tile at high priority. voxel.CollisionGate (default 1) disables the safeguard for stress runs; it is snapshotted at start-up.
After: stress run PASS -- 4 holds, 1.881924 s total, full_stops=0, 139.947 m of 120 m. Strate crossing PASS with pending_floor_overlap=0 and the lower strate reached. Static perf sanity PASS. Verification budget kept small on purpose (no suite, no exports): the owner's complaint was that a one-line request triggered a thousand tests.
Known, NOT fixed here: the surface-fall repro fails as surface_ground_timeout -- the pawn now lands, but on top of the authored (0,0) entry at Z=799.5 cm instead of dropping to its floor at -150 cm. The gate has released by then; this looks like the entry opening being capped, i.e. the owner's 'surface generation is bugged'. Next round.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
The round was stopped when Windows began refusing to start processes for it (0xC0000142) under memory pressure, before any after-change test run. Treat as a draft: the baseline run happened, the verification did not.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
The owner fell through a surface-only world and hovered in mid-air: log showed 'ready=1 predicted_wait=1' for 50 s. Two causes, both fixed: (1) engaging the gate zeroes velocity, so the predicted-support probe never re-ran and the gate kept waiting on the tile remembered from when the pawn was still moving -- the prediction is now re-derived every tick from the pawn's current pose; (2) a tile the streamer never requests (out of layout, or outside the desired set) could never become ready, so the wait was unbounded -- such tiles are now logged and treated as nonblocking, and any wait is bounded to 2 s with a warning.
Repro added to the harness (surface-fall scenario): before, the pawn stops at Z=1315.9 with collision_gate_prediction_deadlock; after, it reaches the ground in 2.25 s at Z=-45.6. Strate crossing still protected: 145.49 s route, max gate wait 0.109 s (limit 10 s). Perf S/M/S/M 24.265/54.262 and 23.947/53.057 s, within repeat noise. Owner and canonical geometry unchanged (44E7C813 / 94B47A1C).
Known red, pre-existing and unrelated: VoxelForge.Composer.ParameterRoll still reports false box verdicts in the owner's current single-strate surface layout (5 violations, identical in the baseline). Next round.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
The harness's first full suite run showed 16 of 44 tests failing. Two were real defects in the owner's world:
1. ParameterRoll: the tile classifier could certify a box AllAir while the final field is solid there (3 false proofs found, e.g. density -2.092319 at (-192,16,-224)). Cause: the box diagnostic narrowed 'tail reach' to 'air reach', so the swept tunnel FLOOR -- which writes solid -- did not count as reaching the box. A false AllAir skips meshing, i.e. missing floor in game. The classifier now uses the same reach predicate as the final writer (swept chain, floor band, relief, room-mouth envelope) and propagates it through Mixed refinement nodes. Re-run: 1,288,143 lattice samples, 0 violations.
2. VerticalShaftEquivalence: LedgeSpacing and LedgeDepth were missing from the production TLS cache key, so a shaft could reuse stale ledge parameters. Both are now in the key.
The other 14: broken fixtures (missing staged corpus assets, a sampler leaking the shaft TLS marker between queries, stale candidate IDs, a window entirely inside the common tail) or stale expectations after this week's owner decisions (no authored tunnel floor, slope in degrees, rooms owning their interior). No test was deleted or weakened into vacuity; retired legacy comparisons stay as telemetry and the current stack is asserted instead. PassageLandsInOpenSpace now requires a real player-capsule fit in the final density at each authored landing (14/14, worst gradient 0.115723).
Suite: 42 tests green in one run, the two long op-stack sweeps green in serialized filtered runs (the 1800 s launch guard); a harness timeout-recording bug was fixed. Bounded probe exports unchanged (canonical 94B47A1C, owner 44E7C813). ⚠️ Harness perf scenario moved 50.74 s -> 53.58 s worker (+5.6%): the wider reach predicate skips fewer boxes. Measure on the game path and decide before merging.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Tools/VoxelForgeTest.ps1 runs scenarios (canonical, owner, probe, perf, tests, parity) with the known-good flag sets, isolated user/config dirs, serialized launches, an owner-process guard, and a compact result.json + summary.txt instead of 300 MB logs that Codex has to stream. Answers the owner's question 'can the commandlet reflect in-game behaviour': -gameconfig=1 makes the commandlet use the authored settings, and the parity scenario compares per-tile geometry hashes between the commandlet export and the game's streaming path. Parity PASS: 64 commandlet tiles vs 810 game records, no differing tile.
First run of the tests scenario: 44 tests, 28 pass, 16 fail (8 composer, 6 op-stack equivalence, PassageLandsInOpenSpace, StrateParamBlendCoverage). Triage is the next round: stale expectations from the floor removal and the degrees change, versus real defects, especially the legacy-vs-op-stack equivalence ones.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
The room-wins rule cost +55% static / +36% moving worker CPU because every sample re-answered the room-mouth question. A bounded exact memo around BuildRoomMouth (4096 sets x 4 ways, 7.49 MiB, keyed on every geometry-affecting room/tunnel input) removes that: hit rate 99.91% static / 99.89% moving, and VF_FindRoomFloorCrossingWorld (19.5%/17.1% inclusive), VF_FindRoomMouthBoundary (16.0%/14.0%) and VF_EvaluateRoomLandingDensity leave the sampler's top 30 entirely.
vs the tip DLL (bf9384a runtime), same staged owner map: worker sum 29.003 s vs 30.749 s static (-5.7%), 57.232 s vs 60.923 s moving (-6.1%), request p95 -3.3% / -8.4%. Geometry unchanged by construction: canonical hash 86A99075, 64/64 tile hashes, OBJ SHA 7E0D8109..C543, identical across a repeat run and with -voxel.CrtFma3=0. Walk Connected 42,870 / 17,413; tunnel-attributed solid inside rooms 0; mouth amplitude 1.4903 cm; TilePostReachProof and DetSinCos pass. The crossing test was not rerun (the memo returns the same values; its PASS stands at 248.43 m / 26.584 s).
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Crossing PASS (248 m, 26.6 s), TilePostReachProof PASS, DetSinCos PASS, canonical export identical with -voxel.CrtFma3=0 (86A99075), walk Connected 42,870 / 17,413. FAILED: one S/M/S/M pair on the owner's staged map shows worker generation 47.7 s vs 30.7 s static (+55%) and 83.1 s vs 60.9 s moving (+36%), request p95 +19-28%. The room-ownership containment test now runs per sample in the generic tunnel SDF, the core evaluator and both native fallbacks. Next round: same geometry (export hash must stay 86A99075), cost back to the tip's level.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Stopped by the owner (he wanted his PC back to play) after most acceptance was in. Measured with the owner's assets: the shelf across rooms came from the tunnel core reasserting solid below its floor inside room volumes, plus the landing-floor post. Room ownership now suppresses the tunnel bottom in the generic tunnel SDF, the core evaluator and both native fallbacks: tunnel-attributed solid inside rooms 469 -> 0, and a matched-camera re-render shows the slab gone. Trimmed mouth endpoints now take their height from the room's own vertical zero crossing (join offsets -2.08/+0.40 -> -1.13/-0.71 voxels). Mouth surface periodic amplitude 1.49 cm floor / 0 cm wall (degree-six trend removal; raw local-median audit 6.19 cm kept). Canonical walk Connected, 42,870 fit / 17,413 walk-reachable (was 41,184 / 13,879); export 141,054 triangles, geometry hash 86A99075. Launches use an isolated engine saved/config directory so the owner's editor settings are not rewritten.
NOT DONE: the -voxel.CrtFma3=0 export comparison, TilePostReachProof / DetSinCos / strate crossing, the perf pair vs the experimental tip, the EditorSettings.ini diff. Unvalidated in game.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Records the owner-playtest findings (stale editor build, owner assets vs harness defaults, the 0.1 slope freeze, ribs with every terrace setting at 0), the proven rib cause and its fix on fix/tunnel-shape, the reverted Catmull-Rom attempt and why its evidence was wrong, the in-editor ablation procedure (cvar read once at first Play), current canonical numbers, and the queue awaiting the owner.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
The owner proved in game that the tunnel core caused the ribs (voxel.TunnelAblateTunnelCore=1 removes them). Root cause: the core's final reassertion replaced any sufficiently interior sample with a fixed +16 air density, destroying the distance gradient, so marching cubes snapped the surface to the voxel grid (all 385 centre samples saturated). The core now reasserts the continuous -Core.SDF, with bottom polarity from the swept shape. The Catmull-Rom sweep from 0231072 is reverted (it cost 18-30% and did not help); the linear tapered-capsule sweep is restored.
Owner decisions implemented: bTunnelFloorEnabled has no runtime effect (reflected only for asset compatibility, ignored on PostLoad), and the authored slab / graph support-floor paths are detached from generation, so the tunnel's own carved shape is the floor. Winding and dramatic ledges no longer depend on the floor switch. Degrees authoring, its migration, and the 3840-voxel route cap stay. Mouth trim stays behind voxel.TunnelMouthTrim (default off).
Evidence: along owner tunnel 15, 385 samples at 0.25 voxel, periodic surface amplitude floor 3.55 -> 0.50 cm, wall 16.9 -> 1.10 cm, ceiling 8.80 -> 0.69 cm (acceptance < 2 cm); 0/385 samples saturated. Matched-camera render: ribs absent. Canonical walk: Connected, player-fit 41,184 (was 20,830), walk-only 13,879 (was 10,909), mouth gap 0; mesh identical with -voxel.CrtFma3=0. TilePostReachProof, DetSinCos, strate connectivity 16/16, strate crossing: pass. Perf, one S/M/S/M pair vs a rebuilt bf9384a DLL: static 26.10 vs 25.41 s (+2.7%), moving 55.34 vs 53.08 s (+4.2%), request p95 about equal. Treat as a possible small cost to re-measure, not as noise.
Not merged into experimental: waiting for the owner's in-game look.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Tunnel round from the owner's playtest. Luna's ablations traced the ribs to the world-space tunnel-core stage: disabling it removes them, while terracing, detail ops, roughness, the support floor, lateral warp, radius jitter, winding and LOD do not. The per-span capsule union is replaced by a deterministic Catmull-Rom sweep. Winding and graph expansion are now gated on bTunnelFloorEnabled (they ran even with the floor off). TunnelFloorGentleSlopeThreshold is authored in degrees (0-90, min 1 deg, DetSinCos conversion, legacy tangent values migrated: 0.3 -> 16.7 deg). The route length is capped at 6 waves x 640 = 3840 voxels, so 0.1 deg loads instead of freezing (2048 collapsed canonical reach; 24576 used ~2 GB). Room-wall trimming is behind voxel.TunnelMouthTrim (default OFF): with it on, a 36 m gap disconnects the walk.
Canonical walk: 20,907 / 10,981, connected (was 20,830 / 10,909); exports identical with -voxel.CrtFma3=0; TilePostReachProof, DetSinCos and strate crossing pass. NOT merged because: one S/M pair vs the tip DLL shows worker CPU +17.8% static / +30.3% moving and worse p95; and the orchestrator's look at the before/after renders is inconclusive (camera not matched; regular grooves still visible in the after interior). Next: owner checks the look in game; if smooth, make the sweep cheap before merging; if not, drop the sweep and keep the degrees/bound/wind-gate parts.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Bet A: the per-sample lookup it would remove is 5.1-6.1% exclusive, and feature-major visiting would touch 5.2-6.6x more lattice points than the current code evaluates; the swept kernel is branch-divergent. Killed on its own numbers. Bet B: 20.04% of static LOD0 density calls are exact duplicates with 80.6% donors already finished (~2.96 s of 18.3 s), moving 20.70% / 63.3%; re-derived independently from the raw trace. Its rung-2 prototype is committed unvalidated on sol/tile-sample-reuse after the Codex usage limit ended the round mid-rebuild; resume brief in the scratchpad.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TaFdXhTdYMkW1xrV6ApT6Y
- WORK-NEXT: state after 2026-09-16 -- Sol's review, the sol/deterministic-math result, new
canonical export/worms/field hashes, and the next branches in order.
- CODEMAP 3.2: VoxelMath::DetSinCos / IsFiniteFast rows.
- ARCHITECTURE: passage landings have had no connector to the (0,0) room since the owner's
d97373c playtest correction; the doc still claimed one. Records the open gap Sol found (the
random-reach fallback when the footing query declines is unproven) and the proposed fix.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TaFdXhTdYMkW1xrV6ApT6Y
On Win64, FMath::Sin/Cos resolve to CRT sinf/cosf (GenericPlatformMath.h:521/530; no platform
override), and the UCRT picks FMA3 or non-FMA3 implementations at startup by CPU. /fp:precise
closed only the compiler half, so the same build and seed could build a slightly different world
on two Windows PCs, and on Linux.
- VoxelMath::DetSin/DetCos/DetSinCos (VoxelTypes.h): fixed double-precision period reduction to
a quarter period, then fixed float minimax polynomials; no CRT, FMath or CPU-dependent
intrinsic. Measured max error vs sinf 5.96e-8 over the used range (|arg| <= 6.19).
- Every world-deciding FMath::Sin/Cos replaced (generator, op stacks, cave morphology, strate
manager, content placement, strate measure); none remain in runtime code.
- Pinned test VoxelForge.Determinism.DetSinCos (golden bit patterns + error bound).
- Diagnostic start-up switch -voxel.CrtFma3=0 (Windows) forces the CRT's non-FMA3 path.
- Parse-once console switches now initialise in StartupModule on the game thread; workers only
read them (was an unsynchronised first parse inside GetDensityAt).
- Corrected the CRC comment that claimed a CRC32 cannot give a false positive.
Proof (Codex round; orchestrator re-checked):
- Before the fix, FMA3 on vs off was identical on this Ryzen 7 5800X3D: the hole exists but was
not observed in this world. After the fix, the canonical exports are byte-identical with
FMA3 on and off.
- The world changed once, as expected. New canonical export SHA-256 CBB8AE18...906F (geometry
B9EF84AC; was B3E5F4C3...377); worms region 9242C093 (was 311CA42C), walk OBJ 20CCF7BD....
Triangle counts unchanged in every tile. Order-independent nearest-vertex distance, worms
region: max 8.7 micrometres (the reported "11.4 m" was a line-order artefact of a line-by-line
OBJ diff); export region max 3e-5 m.
- Capability exact: player-fit 20,830, walk-reachable 10,909, connected. TilePostReachProof,
sealed-proof, cross-platform and strate-crossing tests pass.
- Performance, headless game path, S/M/S/M vs the committed-tip DLL: static worker 27.41 s vs
28.44 s (-3.6%), moving 59.59 s vs 61.74 s (-3.5%); LOD0 p50/p95 not worse.
From Sol's code review (DESIGN-SOL-2026-09-15-CODE.md, section 4.1), branch sol/deterministic-math.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TaFdXhTdYMkW1xrV6ApT6Y
Round two of the owner's "build it every better way" question: Sol's round-one session was forked
read-only and given the repository, to say what could reasonably change and how each change would
prove its worth. Advice only; nothing was changed, built or run.
Sol's verdict: don't rebuild. The representation, mesher, clipmap and operator stack are already
close to its clean-slate design. Three things are worth testing, each with a cheap
instrumentation-only first step and explicit kill criteria:
- tile-wide, feature-major tunnel evaluation (inferred 8-18% of worker CPU);
- exact density reuse across adjacent gameplay tiles (theoretical ceiling ~23%);
- a canonical numeric contract for world-deciding maths (correctness, ~neutral on speed).
The orchestrator verified against the code and the UE 5.7 engine source:
- FMath::Sin/Cos resolve to the CRT sinf/cosf on Win64 (FMA3 variant chosen at runtime);
- per-sample lookup is ~5% exclusive, so round one's "the loop direction is the largest gap"
overstated it;
- worker-thread parse-once flags, raw-memory CRC cache keys and a docs-vs-code contradiction on
landing connectors are real.
Round one's note gets a pointer to the corrections.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TaFdXhTdYMkW1xrV6ApT6Y
The owner's reframing idea: after local "find and fix the hot spot" rounds hit diminishing
returns, we asked Sol High, with no code and no file access, how it would build this project in
every better way. It recommends feature-rasterised sparse bricks (each feature stamps into the
bricks it touches instead of every sample searching for features), noise only in surface bands,
coarse LOD from finer data, a fixed-point determinism contract and deadline-driven collision-first
streaming.
The orchestrator's check against our code: 34 FMath::Sin/Cos calls sit on the generation path
(per-sample strata line/rib detail and passage/tunnel shape building). The MSVC C runtime switches
those to FMA3 implementations at runtime, so cross-machine bit-identity is not guaranteed. Not
changed yet; recorded for a deterministic replacement.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TaFdXhTdYMkW1xrV6ApT6Y
GetDensityAt asked the strate manager on every sample whether the chunk is a gap chunk, which
generator type it uses, and whether it uses the operator stack. All three are constant per chunk.
voxel.ChunkRoutingCache (default 1) memoises them per worker thread (Codex; reviewed by me):
- thread_local storage only, so no cross-thread sharing;
- key = generator owner id + manager pointer + manager lifetime id + layout version + full chunk
coord, the same identity pattern as the existing CP_* refetch;
- the editor composer override is still queried in the CP_* refetch, not cached here.
Attribution for the re-rank's ~28% shared cost (offline-symbolised sampler, exclusive): GetDensityWithParams
16.4%, GetDensityAt 6.2%, unknown 5.8%, FChunkSDFSpatialIndex::GetRange 3.8%, FScopedTimer::End 1.3%,
VF_ForEachSpatialCandidate 1.2%, UsesOperatorStackForChunk 1.0%. The tunnel-cache lookup was already
near-perfect (48 misses in 2,248,091 samples). A thread-local timer cache was tried and rejected as
slower (+0.4% static, +1.2% moving). Spatial-iteration and mesher-loop changes found no isolated
exact win.
Routing A/B: static worker 28.50 -> 27.78 s (-2.5%), moving 63.01 -> 59.81 s (-5.1%).
Final DEFAULT build vs the committed-tip DLL (two interleaved runs each, same command line):
static worker 28.79 -> 27.58 s (-4.2%) LOD0 p50/p95 52.3/94.8 -> 50.1/92.6 ms
moving worker 62.25 -> 59.97 s (-3.7%) LOD0 p50/p95 51.4/95.8 -> 48.8/92.3 ms
Field exact: OBJ SHA-256 b3e5f4c3..9b377 on both builds, 0 per-tile differences, capability
20,830 / 10,909; TilePostReachProof, sealed-proof and strate-crossing tests pass.
Ledger: every launch was serialized with 0 editors before it. Two attribution attempts were
rejected on harness bookkeeping/output-path bugs before a clean run; one crossing run was invalid
(profiling left on, exit -1) and its retry passed. No crash folders.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TaFdXhTdYMkW1xrV6ApT6Y
Records why the merge must not be re-proposed without a new idea. WorldControlPoints keep the
authored room-floor anchors and walkable routes before cave warp; replacing them lost 1,146 fit /
975 walk-reachable cells, and the exact shared-segment variant was slower. The remaining large
costs are the geometry itself, so the exact-skip perf push has reached diminishing returns; the
reach proofs are now guarded by an automation test (8ef1a3b).
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TaFdXhTdYMkW1xrV6ApT6Y
VoxelForge.Correctness.TilePostReachProof (VoxelForgeTilePostReachTest.cpp) runs the real mesher
twice in one process: at full reach it compares the skipped posts' FINAL field (356,946
comparisons, 0 differences), and at a deliberately shrunk reach it must find differences (774
found). Native/fused sample counts are identical (137,180). So a future tunnel or passage geometry
change that breaks a reach proof (ce4d878, a668833) fails loudly instead of silently making holes.
VoxelGenLOD::ResetTilePostReachDiagnostics resets every per-kind counter between the two runs, so
the proof cannot pass vacuously.
Test fixture fix: the strate connectivity test's density-sign/coarse-lie controls use boxes up to
256 voxels, which the DiffLayer budget added in 8879045 rejected. The fixture now uses an unlimited
budget; the test measures the field, not the gameplay brush.
The tunnel SDF/core merge was attempted and NOT shipped (Codex; reviewed by me):
- The world-space core is load-bearing. WorldControlPoints keep the authored room-floor anchors and
walkable tunnel routes before cave warp (8e42d6d, against warped floors becoming vertical
breaks), and cf2b2ad made the world chain authoritative for option B winding/ledge walkability.
EvaluateTunnelCoreWorld also supplies the support/room-floor ownership the tail uses.
- Replacing the world shape: player-fit 20,830 -> 19,684, walk 10,909 -> 9,934, components
261 -> 220. Raw world evaluation without floor handling also failed (20,348 / 10,713): a
coordinate-frame / junction mismatch, not only floor projection.
- The exact shared-segment variant was SLOWER than the committed tip: static worker +5.6%, moving
+12.6%, LOD0 p95 +6% / +13%.
- A 0.0001-0.01% world-chain blend that merely passed the gate was rejected as cosmetic.
All merge source was reverted. This build is bit-identical to the committed tip (311CA42C, 75,482
triangles, capability 20,830 / 10,909); connectivity 6/6 + 16/16, passage open-space 14/14 and
sealed proof 22/22 pass. Every launch was serialized.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TaFdXhTdYMkW1xrV6ApT6Y
Records the playtest-driven fixes (strate crossing, sealed-rock proof, option B floors, gap
D-floors), the cold-audit correctness round (including the tile-cache determinism bug), the perf
gains measured against committed-tip DLLs (static worker ~184 s -> ~28 s over the whole push, same
world), the current remove-one cost map, and the parked exact-cull branch.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TaFdXhTdYMkW1xrV6ApT6Y
Applies ce4d878's decide-once-then-prove pattern to the two largest exact targets from the re-rank
(tunnel core 20.8%, passage carving 9.1%). Codex; reviewed by me.
Measured first, where the time is:
core world: far 223,860 vs near 35,515,979 cycles -> tile gate only; the core is
near-dominated (~99%)
core tail: far 363,440 vs near 20,110,219
passage, tile: far 7,315,426 vs near 30,168,796 -> refined to an 8^3 block gate
passage, 8^3: far 11,233,384 vs near 26,250,838
The core already uses the tunnel spatial index. An exact skip cannot remove its cost: that cost is
in tiles near tunnels. The remaining lever there is the field-changing SDF/core merge (next).
- Core reach per tile, from lattice plus halo: swept capsule reach, floor band, floor-profile
deviation, relief bound, room-mouth ownership. The GetDensityAt tail is identity unless
bSupportFloor, bRoomFloor or CoreSDF < -inset.
- Passage carving per 8^3 block: capsule-chain and landing-room domains, the carve threshold and
the SmoothMin K/6 margin. Malformed descriptors stay candidates (conservative).
- voxel.TileReachCostDiagnostics (default 0) for the far/near cost split.
Soundness: the final-field debug comparison ran 130,285,889 comparisons with 0 differences; shrunk
reach (0.5) produced 140,035 differences (60,098 in the final field) and failed as required.
Field exact: canonical 07C14005 / b3e5f4c3..9b377, 64/64 per-tile ledger identical, capability
20,830 / 10,909, strate-crossing and sealed-proof tests pass.
Final DEFAULT build vs the committed-tip DLL (same command line, two interleaved runs each):
static worker 28.84 -> 28.10 s (-2.6%) LOD0 p50 53.3 -> 51.5 ms p95 94.4 -> 93.1 ms
moving worker 61.39 -> 60.61 s (-1.3%) LOD0 p50 50.9 -> 49.7 ms p95 92.21 -> 92.27 ms (flat,
+0.06%, within noise)
Maintenance note: these reach proofs must track tunnel/passage geometry. The next round adds an
automation test for the final-field comparison so a geometry change cannot silently break them.
Ledger: 15 builds (one C2065 fixed, one access-denied path retry); rejected TileOnly/BlockLookup
variants were slower; one hung diagnostic (a non-owner PID) was stopped; wrapper quoting failures
launched no editor. No crash folders.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TaFdXhTdYMkW1xrV6ApT6Y
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
The remove-one ablation, re-confirmed on the current build, puts the cave warp at 11.0% static /
11.8% moving of worker time: three independent Perlin3D calls per sample. VoxelNoise::Perlin3D_x3
evaluates the three warp channels together through the existing SIMD evaluator (Codex; reviewed
by me). It is wired into the generator, op-stack, classifier and room/player-fit unwarp paths.
Exact: the field is byte-identical. 0/841 tiles changed, 0 triangle delta, 0 voxels of maximum
surface displacement; canonical export 07C14005 / b3e5f4c3..9b377; capability 20,830 / 10,909;
strate-crossing and sealed-proof tests pass. The seam test (VoxelForgeMazeSeamTest, extended) has
48 neighbouring-tile and 24 cross-LOD comparisons, 0 mismatches. Before/after renders from a
camera aimed at the real arrival tunnel are byte-identical.
Cross-machine determinism: Perlin3D_x3 adds no new intrinsics, no AVX, no FMA and no runtime CPU
dispatch. It reuses the evaluator behind the compile-time gate
`#if PLATFORM_ENABLE_VECTORINTRINSICS && PLATFORM_CPU_X86_FAMILY`, so every x64 machine runs the same
instruction path; its output matches the scalar path bit for bit.
Final DEFAULT build vs the committed-tip DLL (two interleaved runs each):
static worker 30.83 -> 29.47 s (-4.4%) LOD0 generation p50/p95 56.3/99.3 -> 53.3/95.3 ms
moving worker 67.28 -> 64.56 s (-4.0%) LOD0 generation p50/p95 55.9/99.0 -> 53.4/96.4 ms
Ledger: every launch passed; two harness false negatives (the seam test on the game-only
validator, and a trace ended by intentional forced termination) were rerun. No crash folders.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TaFdXhTdYMkW1xrV6ApT6Y
The remove-one ablation put landing posts at ~10% and passage structural posts at ~6% of worker
time, though landings exist only near passage ends and the origin. The parked exact-cull round's
per-SAMPLE post bounds had been a net loss. This round decides reach once per TILE (Codex;
reviewed by me).
- VoxelTilePostReach.h: one per-tile reach mask (thread-local byte), set when the tile is set
up. The origin landing, passage landing SDF/floors, native floors, tunnel air and passage
structural posts return their input unchanged on tiles no instance can reach. Per sample this
costs an inline flag read.
- Reach proofs come from each post's geometry (origin rooms, landing SDF and floors, native floor
band, structural support spheres), not from comments.
- voxel.TilePostReachDebug=1 evaluates the skipped posts anyway and compares the FINAL field;
voxel.TilePostReachScale=0.5 deliberately shrinks the reach and must fail.
Default build: 694 tiles, landing reachable on 4 (skipped on 690), structural posts skipped on 169.
Soundness: 123,958,673 comparisons, 0 differences; shrunk reach: 13,929 differences, FAIL as
required.
Final DEFAULT build vs the committed-tip DLL (same command line, two interleaved runs each):
static worker 34.80 -> 30.52 s (-12%) LOD0 generation p50/p95 67.0/107.6 -> 55.8/95.8 ms
moving worker 75.96 -> 66.66 s (-12%) LOD0 generation p50/p95 66.3/106.0 -> 56.0/95.8 ms
Field exact: 841/841 keyed tiles vs a fresh committed-tip trace (669,902 triangles, which includes
9260775's gap floors), export b3e5f4c3..9b377, capability 20,830 / 10,909; strate-crossing and
sealed-proof tests pass.
Ledger: all launches passed or failed where designed to; early raw-SDF debug attempts failed
until the comparator was corrected to compare the final field. No crash folders.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TaFdXhTdYMkW1xrV6ApT6Y
Investigation of the two tests that had been failing for days (Codex; the fix reviewed and its
field claims re-run by me).
Real bug, introduced by adb34df: gap chunks take the compact bedrock path in GetDensityAt, which
applied the passage modifier but never the build-authored native D-floor. So the walkable passage
floor vanished exactly at the gap/strate seam, in the passages the player uses to cross strates.
Now gap chunks set Result = -Density, then ApplyPassageNativeFloorMC, then VF_ApplyXYEdgeSealMC.
The seal runs on the MC value, in the same order as before.
Stale tests, not code bugs:
- PassageLandsInOpenSpace: since d97373c production passes the cached world/warp seed to
VF_SuggestLandingPoint, while the test still queried with the default seed. Both queries are
fixed; the test now allows only exact graph-owned support-floor samples (the production
bProtectAuthoredTunnelFloor contract), and anything else still fails. 14/14 landing ends,
5,774/5,774 floor and capsule samples.
- ClassifyTileSoundness: since d38d4ca ClassifyTile covers the cell vertices g=0..CPA, but the test
still sampled the mesher's outer normal halo. Both "holes" were halo-only (g = CPA+1). The test now
asserts the current contract. TryProveSealedSolidTile already uses the exact core lattice, so it
is not affected.
Verified: crossing test PASS (max gate 0.129 s); sealed-proof test 22/22; both classifier variants
pass (600 tiles each). Re-run by the orchestrator against this build: export OBJ SHA-256
b3e5f4c3..9b377 unchanged; capability 20,830 / 10,909 / walk-only 10,909.
Ledger: one DDC startup fatal (crash folder 43101DC8, 04:38) and a set of expected diagnostic
failures while locating the gap-branch omission; one malformed ablation argument (not a valid
ablation).
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TaFdXhTdYMkW1xrV6ApT6Y
Owner's playtest showed a staircase (dozens of <= 45 cm risers) on a steep tunnel into a room. He
chose option B: "smooth passage and 'dramatic ledge' on tunnels that crossed the slope limit, so we
both have serpenting tunnels and sometimes dramatic drops that needs to be escalated".
Implementation (Codex; the edge rule reviewed by me):
- VF_TunnelFloorNeedsTerrace and all tunnel-floor NumSteps quantisation are removed; new profiles
always have NumSteps = 0. Gentle floors remain continuous ramps with bounded relief.
- The room graph's bridges and a deterministic reserved walk spanning forest decide each steep
edge. bGraphLedgeEligible = leaf || (redundant && !walk-forest edge). Eligible edges may receive a
few large drops at control-point boundaries, never repeated small risers. Every other steep edge
WINDS (a sinusoidal wander lengthens the WorldControlPoints route until its floor is walkable).
Several drops chosen together cannot disconnect walking, because the forest stays walkable.
- The collect margin includes the longer wound-route envelope (window invariance). Classifier
interval bounds include both endpoint floor levels, covering a ledge riser.
Verified:
- default acceptance (slot 1, threshold 0.9657): player-fit 20,830, walk 10,909, arrival ->
departure connected, TunnelFloorMultiStepEdges 0, mesh 311CA42C unchanged. NOTE: this config
has 0 steep edges, so it does not exercise option B at all.
- stress (threshold 0.2): arrival -> departure connected; player-fit 21,305 (winding adds
volume), walk-only 10,787. Fewer walk-only cells are expected behind climbable ledges. Ledges
4-8 voxels: 52,826 observations; 8+: 215,880; below 4: 0.
- determinism: two runs identical (164,126 triangles, 70F987AA, 0 per-tile differences).
- performance vs baseline DLL: moving worker 93.6 -> 85.2 s; LOD0 request p95 static 196 ->
181 ms, moving 188 -> 164 ms.
- the two already-failing tests fail identically before and after (ClassifyTileSoundness: 2
expected-uniform holes; PassageLandsInOpenSpace: upper landing endpoint mismatch).
NOT verified visually: every render reused one fixed camera, never aimed at a changed tunnel. The
arrival case was rendered at the default threshold (8/8 views identical); in the steep/wind cases
the changed views did not show a tunnel clearly. The owner's in-game check is the real acceptance.
Ledger: several intermediate capability gates failed before the final configuration passed; two
malformed launches were aborted; no crash folders.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TaFdXhTdYMkW1xrV6ApT6Y
Correctness round on AUDIT-COLD-2026-09-14.md (Codex; key fixes reviewed by me).
#7 (determinism, introduced by the tile-cache work): the refetch guard in GetDensityAt now
rebuilds on `bTileCacheContext != CP_UsesTileCacheWindow` or when the worker-local stack/core were
moved into a cache entry (`CP_ActiveTunnelDensityCache != nullptr`). A cache miss can no longer
evaluate a moved-from, empty operator stack. Test (VoxelForgeCorrectnessRoundTest): the pre-fix
binary CRASHED in the context-switch case (negative control); the fixed binary passes with 0/5
coarse and fine mismatches, and warm vs fresh origin-tile meshes are identical.
Also fixed, each with a focused test that passes:
- #3/#8: worker diff/surface/biome caches carry owner/layer identity (no cross-world reuse);
#9: the shaft key gains LedgeSpacing/LedgeDepth and the island key BoundarySealThickness;
#14: clearing passages bumps PassagesVersion; #21: HasAnyModInChunkRange is proportional to
the queried range (the sealed-rock proof calls it per candidate).
- #1: the density-volume fill thread is paused/drained with generation; #2: each content manager
joins its own decoration tasks before destruction; #15: non-finite/invalid modification
geometry is rejected; #20: zero-width transition blend is guarded at runtime.
- Also #4-#6, #10-#13, #16-#18, #22. #19 is valid C++20 and unchanged.
Verified: canonical export b3e5f4c3..9b377; capability 20,830 / 10,909; fresh trace 841/841
identical, 0 origin-column mismatches; static mesher calls 18,078,130. Performance: two pairs
disagree in sign, and the means are final -2.5% static / -0.8% moving; no repeatable regression.
Known and NOT caused by this round: ClassifyTileSoundness and PassageLandsInOpenSpace were already
failing (15/17 in the grouped suite). Queued for investigation.
Ledger: first determinism launch hit the DDC fatal (crash folder 57DB6ADC, 23:21); the retry
exited 1 on the two known failures; one malformed multi-test launch was inconclusive; one
malformed trace launch was stopped; one launch with an unquoted project path found no test and its
retry passed. Everything else passed.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TaFdXhTdYMkW1xrV6ApT6Y
Fr0zkaandClaude Opus 5.5