Region pages fell back to the legacy GetDensityAt evaluator while the base shape used the Core
manifest field, so a tile's density (and sometimes mesh) depended on streaming history. Pages
are now bound to their world's base FStandaloneShapeWorld evaluator before publication; density,
lattice, detail, classification and solid queries all delegate to it. WorldAssemblyEvaluatorVersion
5 -> 6. Route-parity audit: fails on L12 (5/5); after the fix 0 mismatches over 87 tile pairs
(base/page, base/native-only, edited, two histories), 56 exact-query pairs, 2.3M page samples,
4 seeds/recipes. Page-tile proof audit: 1,274 blocks, 0 violations. The spawn and island worlds
are unchanged; crossing output changed as intended. Reviewer interleaved spawn: +1.1% (noise).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CSFYgsdW8ejZSyiMf4qRSX
SharedColumnCache.h: a world-owned, bounded 4-way second-tier cache of exact column values
(heightfield and floating islands), keyed by recipe/plan identity plus content hash, XY and island
id. Lock-free seqlock reads; set-local spin lock on misses. The FLazy thread-local tier stays
first. 83.6% of repeated column work was cross-worker; formula evaluations -60.9%.
W12 L1 view: Luna 2.165 -> 1.754 s; reviewer interleaved 2.059 -> 1.722 s. 8 m crossing new tiles
82 -> 52 ms. +66 MiB peak RSS. All hashes (whole view, per-LOD, exact queries, islands seed 0)
identical at W1/W8/W12, Clang off and GCC default. MSVC/UBT not compiled here.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CSFYgsdW8ejZSyiMf4qRSX
VoxelForgeCoreThreadLocal::FLazy<T> (new public header ThreadLocalStorage.h) makes every large
thread_local cache a pointer that heap-allocates on that thread's first use (still worker-private,
lock-free). Static TLS 17.33 MiB -> 43 KiB (initialised image 5.8 MiB -> 21 KiB); worker start
5.2 -> 0.12 ms; idle worker RSS 17.7 MiB -> 52 KiB. Linux workers back to 8 MiB stacks.
World unchanged (L1 hashes Clang off/GCC default, W1/W12; FloatingIslands seed 0).
Reviewer interleaved W12: 2.256 -> 2.091 s. MSVC/UBT not compiled here.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ED33VnqepD4YY6gDAAA832
Exclusive per-stage tile timers and probes. Findings: LOD0 is MC-exclusive
(10.4 of 14 ms p95, 243 ns/sample); LOD4's 3x cost is column-cache misses
(8,224 vs ~1,600-1,900 at LOD3/5), cap refine secondary; per-tile feature lists
cost ~0.001 ms; cross-worker column repeats common (277k at LOD0). Global octave
drop and stride-2 interpolation probes fail the 2 cm gate. Worker cap 1 kept
(cap 0 worse 40 m/s tick tail). Report: Standalone/Research/WorldAssemblerM3/Round17.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
voxel.StreamingReadyQueueDepth=8 admits one extra worker-set of prioritized
tasks; voxel.StreamingBackgroundWorkerCap=1 sizes slots to the 8 TaskGraph
background workers actually running tile tasks. Full set 5.96 -> 4.91 s,
LOD0 1.40 -> 1.09 s, ring 0.49 -> 0.51 s, 40 m/s ring 75/75. Tick p95
+0.6 ms spawn / +0.26 ms walk, max flat. No geometry change.
Report: Standalone/Research/WorldAssemblerM3/Round16.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Startup apply mode on by default (12 meshes / 14 ms per frame, ends when the
initial set and walkable ring are ready, 30 s cap); normal play keeps the
authored budget. Backlog peak 252 -> 10, LOD8 apply wait 1.1 s -> 4 ms.
Ring 0.49 s, LOD0 1.40 s, 40 m/s ring 75/75. Remaining floor: frame-paced
admission (~8.3 submits/frame, 8 running workers max). No geometry change.
Report: Standalone/Research/WorldAssemblerM3/Round15.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Planar far horizon samples the LOD8 interior at the MC cell spacing, keeping
128-voxel boundary strips; far recipe v3. LOD8 p95 59.8 -> 13.9 ms, spawn
worker-seconds ~41 -> 26.9, full set 7.8 -> 7.1 s, ring 0.49 s, LOD0 1.43 s,
40 m/s ring 75/75. S3/S4 2 cm agreement, 100% cap coverage, W1/W8/W12 and
W1/W8/W16 determinism pass. Report: Standalone/Research/WorldAssemblerM3/Round14.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
voxel.StreamingCollisionLookahead (default on) predicts the support tile 350 ms
ahead along full 3D velocity and prioritises its 9-tile collision ring. Owner
plan, BuildHost actor: 40 m/s arrival-ready 25/74 -> 74/74 (CVar-off control
15/74); 6/12/25 m/s stay 100%; -Z 50 m/s 93/93. Spawn unchanged (~7.7 s).
Critical LOD0 path at 40 m/s: generation 67 ms p50, RMC body update 38 ms p50.
LOD8 cost counted (native horizon 48 ms of 60), optimisation deferred.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
voxel.CoreRegionConcurrentAdmission (default on): tiles independent of the
entering region keep submitting while a region page compiles; pages are
immutable shared_ptr owners so running readers survive eviction. 6 m/s: 11
whole-pass skips -> 0, walkable ring ready at every centre change.
voxel.StreamingResultBacklogCap=256 bounds finished results (slow-apply worst
case 125 -> 76 MB). Spawn unchanged at ~7.7 s: LOD8 is 46% of worker time.
At 60 m/s + teleport the collision ring is outrun (102/113 centres).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Two admission blockers found by per-frame decline codes: the near reserve held
four slots idle after LOD0 was done (coarse capped at 6), and finished-but-
unapplied results counted against the task cap. Admission now counts queued +
running tasks; near reserve releases once the walkable ring is ready while
stationary (voxel.StreamingDynamicNearReserve, default on). Walkable ring
0.65 -> 0.58 s, full LOD0 2.28 -> 1.44 s, busy workers 2.2 -> 5.6 of 10.
Old behaviour behind voxel.StreamingLegacyPendingResultCap. RMC source reviewed
(read-only): batching/fast-cook ranked, not yet worth it on measured apply cost.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Direct per-tile trace on the BuildHost actor (owner plan, 2,578 keys): full set
17.4 s legacy; gated coarse-first (after the 9-tile walkable ring) 11.7 s with
the ring still at 0.63 s, but full LOD0 slips 2.3 -> 10.2 s. Startup apply
boost changed nothing (0.4%). 37.6 worker-s over 17.4 s on 10 slots = ~3.3
busy: the floor is admission/work supply. S5's 251 ms staircase was the trace
sampler. Fixed CVar parsing in headless BeginPlay and the 60 m/s crossing
instrument. All three knobs stay opt-in.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
S3's far cap sat up to 8.2 m above the near roof and read as floating pieces.
The far render cap now projects onto the owning strate's upper boundary
(density unchanged), on a global XY lattice, LOD4 16-voxel grid + LOD5-8
128-voxel. Cap agreement 0.05 cm (100 cm control detected), 2520/2520 upward
rays hit roof on the owner walk. Far recipe v2; LOD0-3 hashes equal S2/S3.
Tile p95 flat; cold clipmap +4.4%. Far-cliff fluting left open (coarse MC step).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Far-shell repair (81-96% of S2's LOD7/8 tile time) is skipped for certified
SurfaceWorld tiles; ground and cap come from the shared planar heightfield.
Near LOD0-3 hashes equal S2; far recipe version 1 salts LOD5+ identities.
Cold first view 6.6 -> 3.4 s (aim 1.57 s). Not a visual sign-off: distant cap
shows floating pieces, sprint streaming arrives late.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Island-only worker context replaced by a shared worker-local candidate domain
(WorldAssemblyFeatureTileListV1 / ScopedWorldAssemblyFeatureQueriesV1) serving
SurfaceWorld, FloatingIslands and far-shell queries; a per-domain cost hint keeps the
full scan on small plans. Content placement prepares the same queries and skips proven
air: owner island cell p95 7.07 -> 0.24 ms; populated fixture 572 exact placements,
3.27 s -> 29 ms. Full/candidate hashes equal with dropped-feature controls on both
families; S1 default W1/W8/W16 hashes reproduced; Core-off and M2 fixtures pass.
Default contract for future families in ARCHITECTURE 8.23. Run on gpt-6.1-sol medium.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
AVoxelWorld::CoreVoxelSizeCm (default 25, 32 voxels/tile) flows through plan, field,
mesher, content, water, atmosphere, edits and collision. Staged-owner BuildHost passes at
20 cm (SurfaceWorld + islands, W1/W8/W16, anchors, BP edits, collision) with a fixed-25cm
negative control; default-size hashes match round 4; Core off stays legacy 12.5 cm.
Reusable BuildWorldAssemblyFeatureTileListV1; wired into FloatingIslands only so far.
256-island stress: tile p95 112.5 -> 2.0 ms, 0 mismatches, dropped-feature control fails.
Content placement measured (no populated cells): island ground search 7.25 ms/cell vs
legacy 0.29 ms -- open. Outside-review evaluation in Round5/REPORT.md.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
The round-3 'field contact error' measured hit-to-anchor distance; V2 contacts are 0 cm and on the zero
isosurface. Tagged-tile cost 5.7-6.7x -> 1.175-1.270x MC; organic 0.998x. Bound, shelf-thickness and route
audits rerun with negative controls; W1/W8/W16 stable; V1 and MarchingCubesV1 unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Owner saw content floating above native terrain, spiky triangles, reversed carve/fill and late relief.
Causes: shared mesher placed Core tiles on a 12.5 cm grid; content/atmosphere asked strate ranges in
12.5 cm voxels; Core edits add density where the legacy diff subtracts. Horizon LOD8 bridge edge
5.657 -> 2.265 cells. Relief mesh p95 60.8 -> 42.0 ms (LOD0). New BuildHost checks for scale, height,
blueprint carve/fill and per-kind streaming lag.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Manifest imported into UVoxelSettings (hash 8933693d9560ba9f), folded into the season
content hash only when Core is on. BuildHost: W1/W8/W16 stable, 2553 surface tiles match
standalone, 432 LOD transitions, collision on all 36 expected tiles. Matched timing
96.8 s native vs 197.3 s bridge (2.04x). Collision support tiles wait for Core region admission.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Luna, M1 round 4 (resumed after the night stop):
- Cliff: sub-voxel audit (29,929 half-voxel XY columns, min run 3.75 voxels, 0 subvoxel runs;
old taper control exposes 20,963) and a root-line shelf at the two-voxel minimum.
- Horizon slab = aliased LOD8 cap ceiling. The first fix hid the ceiling at LOD8 (challenged:
enclosed, interior-lit strates); now the LOD8 cap is rebuilt deterministically from the
native heightfield, skirts preserved; looking up and out shows the cap present.
- Routes: review manifest with 33 stable-ID route points + capsule/support constraints; capsule
sweep 0 blocked samples / 0 unsupported centres, negative control detected.
- Server path: World::densityAt answers without render meshes. Physics cooking not available
standalone.
- Hashes (evaluator + 2,226-tile stream) identical at W1/W8/W16. LOD8: 4.446 s for 529 tiles,
14.25 M field calls, 165 unknown tiles.
- Captures at 60 fps cap + vsync (owner's GPU coil whine). One cliff probe crossed the ~8 GB
guard (7.69 GiB) and was stopped before capturing.
My review: cap view matches legacy closely, no slab; native cliff is a clean attached ledge
(cleaner than legacy); a few small spikes on the ridge top.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Luna, integration round 3, from the owner's four in-game screenshots:
- Capsule rings at the opening and the shaft continuing through the cave room were Core
world defects: the synthetic bore is removed and the opening gets a finite origin room
and floor (solid below). New world version, W1/W8/W16 identical: density
a8bcb72d41092503, mesh 90f29ba15e24579f, payload 82c6b489e7bb7fd7, topology
7b99448a1f844f2c.
- B-side landing and surface: headless coverage found no missing tiles or LOD-transition
gaps (960 / 640 fine tiles compiled after travel); best explained by Core's missing
SurfaceWorld representation. Null-RHI cannot rule out a presentation-only issue.
- Core mode GetBiomeAtWorldLocation now answers (e.g. Biome_Surface_Plain); biome
terrain itself is not generated yet (proposal: native Core SurfaceWorld evaluator).
Owner checked the Step29 viewer himself: the surface strate type is absent there too.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ
Luna, integration round 1 (BuildHost only; owner's project and assets untouched):
- New runtime module VoxelForgeCore: one shim per Standalone/Core/core_sources.txt row
(sources stay canonical under Standalone/), NoPCH/no unity, manifest check; VoxelForge
links it (no runtime routing yet). UE link-safety audit: 18 shims, 83,843 symbol lines
vs 399 Unreal-style names, 0 clashes.
- Adapter (editor commandlet path) turns staged copies of the owner's DA_* assets into
schema-v1 JSON; with clip_radius=3/max_clip_level=8 it reproduces all four Core hashes
(0xdb2991c73215af4c, 0x8236c5edd7fc8e4f, 0x4c48fa75a543d58e, 0xbea22e76b34a4dfc) and
walkability 5,642/5,642 at worker limits 1, 8 and 16.
- Noted: the owner's settings asset says max clip level 6 (standalone used 8); Core uses
25 cm voxels vs the plugin runtime's 12.5 cm.
Not done yet: AVoxelWorld switch and Core runtime route (round 2).
Docs: CODEMAP, ARCHITECTURE, SHARED_CPP_LIBRARY_PLAN, UNREAL_INTEGRATION_REPORT.
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
- CODEMAP: data-flow diagram, §3.2b/§3.2d, generator and strate-manager
rows, test rows and §4.2/§5 now describe the stack-only path; the
deleted archetype functions, bUseOperatorStack and the Not Op Stack
stats are gone; line hints re-synced for VoxelGenerator.cpp,
VoxelStrateManager.cpp and VoxelStrateDefinition.h.
- ARCHITECTURE §8.1: archetype table lists stack builders; §8.10 surface
column and shaft cache entries name their current owners.
- REVIEW_FINDINGS: tick "Operator stack as the only density path"; add
the no-strate-manager fallback world as its own open item.
- fable-idea: T1.a status names the current column caches.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01URDGLsu7c4nunqNFPJEFPU
The single player-following water plane only uses the strate's
WaterMaterial; UVoxelBiomeDefinition::WaterMaterial is never read. The
design (§8.14 content profile) still lists it, so record the gap as an
open owner decision instead of silently dropping it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01URDGLsu7c4nunqNFPJEFPU
- §8.10 render-distance paragraph: remove the F18 far "sheet" ring
description (GenerateSheetMesh, SheetHole*, sheet re-queue). The ring
is plain level-MaxClipLevel MC tiles; bFarSheetRing/FarSheetSpanLevels
are documented as unused settings the runtime always passes as false.
All other §8.10 invariants are kept; the budgeted-teardown note now says
what UnloadTile does today (park the component), not ClearChunk.
- Fix the claim that GetLODForChunk/LODToStep/IsChunkInRange/
GetStrateChunkZBounds and the ViewDistance settings are dead (the first
three are gone, the last two are used); only the ceiling settings are.
- §8.5: decoration results are BuildId-guarded region builds applied by
ApplyRegion (no ReadyResults/DecoEpoch/ApplyDecoResult); water is one
player-following plane (UpdateWater), not per-tile planes; palette tags
are per-grid EntryBiome; FStrateSetPiece no longer "coming" (folded into
landmarks); drop the dead ContentMaxLevel/MaxLODLevel notes.
- §9.3: anchors fold in after the player clipmap (no sheet ring).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01URDGLsu7c4nunqNFPJEFPU
Move closed audits, task briefs, design rounds, perf reports, the OPSTACK
plan/progress logs and WORK-NEXT.md into Docs/archive/ (git mv, history kept).
Whitelist Docs/ in the strict .gitignore, ignore Tools/**/__pycache__/, and
remove the tracked .pyc. Repoint references in the kept docs and in non-test
source comments to the new paths.
- 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
The sampler put the worm carve's first noise call (VoxelGenerator.cpp:3985, FusedWorm) at 19.4% of
LOD0. The worm only carves where N1 + N2 < T with N2 >= 0, so where N1 >= T it contributes nothing,
bit for bit. A block proven to have N1 >= T everywhere can skip both noise calls exactly.
- The Lipschitz bound for this Perlin3D implementation: L = (15/4) * sqrt(3) * 1.25 = 8.118988160,
in scaled-N1 units, with a 1e-3 float-rounding margin. Adversarial sampling found a worst ratio
of 3.187 (0.39 L). The existing 2D floor-relief bound (8.5) is a component bound of a different
field and was confirmed not to transfer.
- A soundness test that can fail (VoxelForgeWormBlockSkipTest): 6,476 proved blocks, 355,945 exact
lattice samples, 0 violations. A deliberately halved L proves a bad block; the full L rejects it.
- voxel.WormBlockSkip, default 1.
Field exact: export OBJ byte-identical (b3e5f4c3..9b377), 841/841 keyed tiles identical to
NextHeadStaticTrace_20260913, capability 20,830 / 10,909.
RESULT, stated plainly: no measurable speed gain. The skip rate is 4.07% of worm-eligible LOD0
samples and 0% at LOD1-4, because the bound is ~2.5x looser than the worst observed ratio at the
block size used. Paired on-vs-off, two interleaved runs each: static worker +2.15%, generation p50
+1.8%; moving worker -1.42%, request p50 -2.5%. Within noise. The next round turns the default OFF.
The proven bound and its test are kept for a tighter skip later (smaller or hierarchical blocks,
or a per-cell bound).
Run ledger (from Codex's report): the first build failed on an ambiguous test overload and was
fixed; two timing runs are void because the orchestrator stopped the session so the owner could
use the PC; the eight valid timing runs, the sampler run and the diagnostics runs completed. No
new crash folders. PERF-SAMPLED.md now lists all eight earlier-round UECC crash records.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TaFdXhTdYMkW1xrV6ApT6Y
Field-preserving floor round (Codex; reviewed by me against the code):
- exact per-(sample, passage) caching of native/generic floor projections and walkable-air
checks, behind voxel.FloorRound1PassageProjectionCache
- a zero-work disabled path for VoxelDensityProfile::FScopedTimer, unchanged when enabled
- pointer / range-for iteration in the proven-hot tunnel, passage, spatial and candidate loops
(TArray RangeCheck was ~6.5% of LOD0 samples on the Development binaries the owner plays on)
- FVector2D temporaries replaced with scalar doubles, same operations in the same order
- FSurfaceColumn made trivial, so thread-local construction does no work. Verified safe:
ComputeSurfaceColumn writes all five outputs unconditionally before any branch, and every read
of Box.Cols is gated by Box.Computed
Verified: export OBJ byte-identical (b3e5f4c3..9b377); capability 20,830 / 10,909; all 8
interleaved game runs gave identical totals (841 tiles / 669,834 triangles static, 2,348 /
1,985,216 moving, the same as previous rounds). Per-tile identity for the always-on changes still
needs a check against a stored trace; that is queued first in the next round.
Game path (sampler off, two interleaved runs each):
projection-cache switch alone: neutral (static worker +0.41%, moving -0.05%)
vs the previous round (cfad3d1, same harness; cross-round numbers reproduced within ~1% here):
LOD0 generation p50 static 76.6 -> 65.8 ms (-14%), moving 72.4 -> 59.5 ms (-18%)
LOD0 request p50/p95 moving 79.4/158.5 -> 67.6/137.6 ms
worker seconds static 39.2 -> 33.4, moving 113.7 -> 97.9
The always-on changes carry the gain; they have no switch, so this is a cross-round comparison.
The sampler's ~12.9% "unknown" LOD0 leaf is engine code: UnrealEditor-Core.dll + RVA 0xC080,
without a Core PDB. The old FSurfaceColumn label is gone. Next round names it by its callers.
PERF-SAMPLED.md: corrected a sentence that repeated my false "daytime use" explanation.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TaFdXhTdYMkW1xrV6ApT6Y
Owner-approved, on the A/B measured in 83a0e4d: the classifier plus exact validation cost more
than it saved on the game path. With it off, ready p95 is 182 vs 190-195 ms, generation p95 is
125-127 vs 139-141 ms, and density calls drop 18%.
- GVoxelForgeOuterClassifierMode defaults to 0. Mode 1 stays reachable via
voxel.OuterClassifierMode=1 as an A/B instrument; the clamp is now 0..1.
- Mode 2 (validation samples reused as the mesher core grid) is deleted: the reuse grid,
FVoxelMesherCoreDensityGrid, GenerateMesh's ReusableCoreGrid parameter, bValidationCoreReused,
TotalValidationCoreReuseTiles and its log field. It could not fire: it reused the grid only
when validation disproved a uniform verdict, which never happened, and Mixed tiles never
validate. The mode-1 validation loop exits early again on the first disagreeing sample.
- All three GenerateMesh call sites checked; the density-count pointer binds after BandVoxHi.
Verified by Codex, checked by me against the logs:
- build ok; the measured process loaded the staged DLL, SHA-256 1024D8CF..04D
- export OBJ SHA-256 unchanged: b3e5f4c3..9b377
- a run WITHOUT the override prints outer_classifier_mode=0 validation_density_calls=0,
proving the new default took effect
- 841/841 tiles have identical triangle counts between mode 0 and mode 1. The 394 tiles mode 1
skipped as uniform produce 0 triangles in mode 0, so the strate band cut (applied by the
mesher, not by validation) was not hiding geometry.
Docs: ARCHITECTURE 8.10, CODEMAP (GenerateTileResult, ClassifyTile) and fable-idea T1.d record
that streaming no longer classifies by default. WORK-NEXT item 1 records the owner's playtest:
he can no longer fall through the world.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VZS94UH8bSsBeg5sZaxe2Y
Two WORK-NEXT items, one null result and one confirmed removal. Item 4 untouched.
ITEM 3 -- CANCEL OBSOLETE WORK AND PRIORITISE THE FLOOR: NULL RESULT
Clean -nullrhi, diagnostics off, 800 cm/s, 200 m of travel:
before after
request->ready p95 0.369117 s 0.367612 s
queue wait p95 0.066949 s 0.064483 s
generation p95 0.287926 s 0.283642 s
p95 travel distance 2.952938 m 2.940894 m
obsolete aborts 0 12
Flat. Aborting obsolete tiles, sorting from the true pawn position and prioritising the
occupied/next tile changed request-to-ready by 0.4%.
⭐ AND THE NULL RESULT IS ITSELF THE FINDING. Look at the split: queue wait is 65 ms of a
368 ms p95; generation is 284 ms. **The queue was never the problem, so reordering it
cannot help.** Only 12 tiles in a 200 m run were obsolete enough to abort. The floor is
late because generation is slow, not because work is badly ordered -- item 3 establishes
that by exclusion, which is worth more than the 0.4%.
The machinery is kept: it is correct, it costs nothing, and it will matter once
generation is fast enough for ordering to be the binding constraint.
ITEM 5 -- VERDICT CACHE REMOVED. It was pure overhead.
Clean moving-session audit:
outer calls/probes/hits 2,109 / 2,109 / 0
nested calls/probes/hits 19,904 / 19,904 / 0
enclosing hits 0
stores 15,877, evictions 15,365
⇒ It stored 15,877 proofs, evicted 15,365 of them, and hit ZERO times. A moving player
visits unique regions; the cache was filling and thrashing a 512-entry table under a
global lock for no benefit. Same-binary A/B: request p95 0.367645 vs 0.365978,
generation p95 0.285559 vs 0.285992 -- neutral within noise, so its cost was the lock and
the linear scans rather than time we can see.
Removed: the global cache, its lock, the entry table, the invalidation path, the
cache-only profiler fields and the synthetic probe. The proof classifier stays.
⚠️ I asked for this cache two days ago on the strength of a commandlet measurement that
re-visited the same region repeatedly. That access pattern is nothing like a moving
player, and I should have said so when specifying it.
VERIFIED BY ME: build 0 errors; export 0.563 / 0.565 / 0.566 s (35de59b was 0.545-0.563,
flat as expected); geometry SHA-256 b3e5f4c3..9b377 UNCHANGED; capability 20,830 /
10,909, connected.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012aheo2crrZmHuYtPpdY2Cb
TunnelNetwork density evaluation was 45x more expensive than Maze for identical
work: 2,097,152 voxels, 64 tiles, same mesher, same machine. Profiling killed
every hypothesis I brought to it -- box classification skipping 0/64 tiles was
true but irrelevant (export bypasses classification), near-surface early-outs
fired at 49.7%, the room cache built 340 entries for 2.74M samples, and the
per-room overrides resolved once per sample rather than once per modifier.
The cause was the wrapper, not the stack. A mesher tile samples an EXPANDED grid
(-1..Cells+1), so one worker crosses several exact chunk coordinates while
meshing a single tile. The generator's cache held only the last coordinate, so
every boundary crossing rebuilt the resolved parameters, the whole 22-operator
stack and the tunnel-core cache. On one representative tile: 3,675 refetches,
38.0s. The entire operator stack was 18.5s of a 4,272s total -- 0.43%.
Replaced with a 128-slot direct-mapped thread_local cache keyed on owner, chunk,
layout version, params fingerprint and generator type. Collisions force a safe
rebuild, never a stale read. No shared mutable state, no order-dependent
reduction.
TunnelNetwork 344.5s / 164.267 us/vox -> 18.4s / 8.763 us/vox (18.7x)
Maze 7.4s / 3.535 us/vox -> 1.08s / 0.513 us/vox ( 6.9x)
same tile 38.0s -> 0.51s ; refetches 3675 -> 27
Output is BYTE-IDENTICAL, before vs after and across repeats, both archetypes:
TunnelNetwork 50d6b4d0..f23126, Maze 364f7489..2fbe99. The permission to change
the field was not needed.
Also in this commit, from the two preceding tasks:
- Explorer renders from the MESH instead of raymarching. It meshes the region
once through the canonical UVoxelMarchingCubesMesher, then CPU-rasterises via
a BVH. A viewpoint went from ~16 minutes to 0.155s; 8 views now cost 1.25s of
drawing. No render pixel samples the density field.
- One sampled grid and player-fit mask per run, shared by render/walk/export,
instead of the same 368s walk computed twice.
- A wall-clock budget (25 min default, hard 30 max) that stops cleanly and
records truncated/truncated_during/completed_modes rather than overshooting.
- ParallelFor across the measurement sampling; the suite was ~10% of a 16-thread
machine.
- -surfaceroughness= on the commandlet, and a surface-relief metric in export
mode (edge lengths, dihedral distribution, face-vs-smooth normal divergence)
so "does this look like rock" is a number, not a screenshot argument.
- Smooth vertex normals in the rasteriser: FVoxelMeshData::Normals is the
density-gradient normal the engine shades with, and the raster path was
recomputing a face normal over it.
- ScaleDiagnosis: baked candidate indices are now derived from the current
corpus, so a legitimate field change stops breaking the test.
NOT verified, and not claimed:
- TunnelNetworkSpineEquivalence exceeded the 30-minute cap and was stopped.
Inconclusive, not passing. The identical hashes are strong evidence the
op-stack/legacy equivalence holds, but the test has not confirmed it.
- CrossPlatformDigest passes but reports 2 samples within 1e-6 of the
isosurface; FMath::Sin/Cos leaves Windows/Linux equivalence unproven.
Pre-existing, not caused by this change.
- The cache's memory footprint (128 entries x ~16 workers) is unmeasured.
What the roughness sweep settled, for the record: SurfaceRoughness 0 -> 8 moves
cost by 0.3% (163.2 / 163.2 / 163.7 us), so roughness was never the cost. It
does reach the surface (dihedral p50 1.1 -> 2.2 -> 5.9), which kills my claim
that the density-channel operator contributed nothing. TunnelNetwork's hard
seams are present at roughness 0 (4.4% of edges past 40 deg vs Maze's 1.0%), so
they are structural -- in how primitives are unioned -- not a roughness
deficiency. More roughness makes them worse, not better.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012aheo2crrZmHuYtPpdY2Cb
Jahni corrected the design: "the connection between level is a tiny tunnel
connecting both strate, not vertical, it needs to be walkable" and "'the
spine' shouldn't be carved except optionally from the top first strate,
then it's used to have a landing room".
COMPOSER-NOTES §6.8 had said the spine "provides a clean landing space, not
a continuous open shaft" -- and the implementation carved a vertical
cylinder anyway. A vertical cylinder has no walkable floor except at its
bottom, so a carve guaranteeing open AIR guaranteed nothing STANDABLE.
That one omission is behind five of my dead hypotheses this week.
Now:
origin spine a finite rounded ROOM per strate, no vertical bore.
radius 14 -> 8 m wide, 7.5 m tall
descent a walkable tunnel, single-file by design: 3 m mouth,
2.5 m waist, gradient capped at a NAMED 15 deg
exotic styles VerticalShaft / SpiralDescent / CascadingDrops /
CrackCrevice preserved, no longer the default
⚠️ THE GEOMETRY BITES, and it is worth Jahni's eye: at 15 deg a 32 m drop
needs 119.4 m of horizontal run -- about 59.7 m per switchback leg. That is
a long walk between strates. It is the honest consequence of "walkable",
not a tuning artifact, and the alternative is a steeper number.
⭐ AND THE ROOT CAUSE GOT NAMED A THIRD TIME, by Codex mid-run: "the round
tube was hollow air but had no authored support strip, so 'open' still did
not mean 'standable'." Tunnels now carry a floor band, and the root join
terminates in the finite origin room rather than an annular shaft hub.
Measured, commandlet sweep, 8 archetypes x 8 seeds, 64/64 valid:
origin column has player-fit cells 17/64 -> 63/64
origin column reachable 0/64 -> 7/64
arrival landing probes 315/320 fit AND in the arrival component
tunnel audit 5,916/5,916 floor and capsule samples pass; zero gradient,
floor, capsule, geometry or seal failures
narrow-gap events 0 box verdicts 14 checked, 0 violations
⚠️ arrival->departure is 6/64. The landings and tunnels are now provably
walkable; what remains is the INTRA-STRATE vertical connectivity problem
found in 6b91669 -- floors at different levels inside a strate with nothing
joining them. Separate and larger. Not tuned toward.
⚠️ PROCESS: this took 5.7 hours, of which Codex used 4 SECONDS of CPU. The
rest was polling the 73-minute automation suite at ~10% machine load. The
commandlet exists to avoid exactly that. New rule recorded: iterate with the
commandlet, full suite ONCE at the end.
Editor + Shipping builds clean. Automation 32/32, 0 failed.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012aheo2crrZmHuYtPpdY2Cb
Every experiment in this project cost a build + full-suite cycle. The
landing task took 52 of them over ~9 hours. And nobody could SEE the world:
every visual judgement all session was made on density SLICES -- diagrams,
not pictures. That is how a floor heaving 1.4 m every two metres got
described as "roughness visible but modest" while Jahni spotted it in
thirty seconds by walking it.
-run=VoxelForgeExplore, editor-only module, three modes, one build many runs:
render raymarched PNGs, fixed 0.25-voxel step + 10 bisection iterations
(NOT sphere tracing -- the op stack yields density, not a signed
distance, so a long step tunnels through walls). Three views, each
with a 1.76 m human marker in frame, because every scale mistake
here came from numbers with no body beside them.
walk an agent using the EXISTING VF player-fit stencil, never its own
movement model -- two definitions of "can a player stand here"
would drift and we could not tell which was lying.
export OBJ through the canonical UVoxelMarchingCubesMesher. No second
mesher; a forked meshing invariant is the bug this project keeps
paying for. 23.2 MB, 106,762 verts, 200,842 tris for 128^3.
⭐ THE NUMBER THAT JUSTIFIES IT: the full automation run takes 4,396 s
(73m16s). The commandlet does all three modes in 48.1 s wall, 5.9 s of
which is engine startup. ~91x faster. And I had guessed the suite cost
~90 s -- it is 73 MINUTES. The iteration loop was far more expensive than
I knew.
First real walk metrics on the Maze (seed 0, slot 4):
distance 214 m · tortuosity 4.98 · dead ends 0.935/100 m
narrow-gap events 0 · reachable fit fraction 3.01%
result NotConnectedAtThisResolution
Tortuosity ~5 is a genuinely winding route, which is right for a maze.
3% reachable is not.
Editor Development build clean. SHIPPING build clean with the receipt
containing VoxelForge and NOT VoxelForgeEditor -- the instrument cannot
reach a packaged game. Automation 32/32, 0 failed.
⚠️ Codex initially shipped this UNBUILT, and correctly: I omitted the
## Validation section from that spec, so it fell back to CLAUDE.md's
"NEVER build or compile". 100+ compile errors from three causes -- bare
TCondensedJsonPrintPolicy and TJsonWriter needing their template
arguments, and a missing UObject/Package.h. That include is the SECOND
miss this session, so CODEMAP §6 now carries the gotcha.
⚠️ Unrelated, found in passing: VoxelForge.uplugin does not declare
RealtimeMeshComponent although the runtime module uses it.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01K8V9my1PL68pmQb2gutZZV
The last unbuilt Tier 1 item. A passage was a bore through rock with no
floor inside the walkable angle, so its mouth was not a standing position
and neither was anything near it -- the departure mouth measured a FOUR-CELL
player-fit component. You arrived and there was nowhere to stand.
Both ends of all 7 passages now open into a landing, sized from the body:
capsule diameter 2.72 vox floor >= 12 vox (3 m) to stand and turn
capsule height 7.04 vox room >= 12 vox tall, stock mouth gives 18 (4.5 m)
3-voxel floor slab, tube tangent to it -- no lip to climb out of
connector 5-vox radius: 2.5 m clear width, 3 m clear height
ramps over 44 deg get a deterministic level dog-leg
⭐ THE MOUTHS ARE NO LONGER SPECKS:
archetype components largest arrival-departure
Maze 309 -> 340 176 -> 3210 43-81 -> 3210-3210
TunnelNetwork 467 -> 227 1292 -> 3841 2-1 -> 3841-269
Underwater 86 -> 117 35 -> 3838 3-11 -> 3838-267
FlatPlain 45 -> 11 259 -> 2130 1-1 -> 268-273
CrystalChamber 197 -> 90 24 -> 286 1-1 -> 286-274
FloatingIslands 2 -> 1 3 -> 72 1-1 -> 72-72
Law 1/8 -> 2/7 connected (1 refused on MaxCells), and critically ZERO
connections now run through a four-cell pocket.
Every landing floor was verified against the FINAL density, after all posts:
14/14 pass the 44 deg slope bracket, worst room gradient .059, worst
connector .522. 14/14 deterministic root joins. 0 seal failures.
0 box-verdict violations across 82 proved + 174 mixed, 109,142 voxels.
Op-stack bit-identical, order-independence green over 10 slots / 31
passages / 3 pool orders. Cache 0.665 us rebuild, 0.646 us hot.
Post order: origin spine -> vertical seal -> passage tube + landing floor
-> XY edge seal. Landings clamp 4 voxels inside the vertical seal; the XY
seal stays last and wins at world edges.
⚠️ ARCHITECTURAL NOTE, not a detail: landings join a deterministic annular
ROOT HUB at (0,0), built identically for every endpoint -- no nearest-
component search, no passage ordering. That is what makes the join
guaranteed rather than likely, and it extends the (0,0) spine concept from
a landmark into a structural network. It also means routes tend to pass
through the centre, which is a gameplay consequence worth Jahni's eye.
⚠️ AND IT DOES NOT CLOSE THE LAW. Codex is explicit: the landings guarantee
membership in the new root network, NOT in the archetype's pre-existing fit
component. The whole-strate Maze check still shows a 71.9-voxel gap between
1,235 and 2,938-cell components. So the native corridor networks remain
disconnected from the root network in places. Real, separate, next.
52 build-and-measure cycles, ~9 hours, finding real geometry bugs on the
way (a landing floor overlapping the ramp cells and hiding a floor
discontinuity; the disturbance pass refilling connector headroom).
Build clean, 32/32 success, 0 failed. Showcase regenerated.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01K8V9my1PL68pmQb2gutZZV
Jahni: "mostly a cube full of straight pathway along the edges... basically
just a repeating pattern of tunnel connecting to each other, definitely
unwanted". And the measurement: the arrival mouth in a 30-cell pocket while
the network held 1,217. SAME CAUSE -- edge selection was independent hash
gating per +X/+Y/+Z edge, i.e. a random subgraph of the cubic lattice.
Every junction identical, and connectivity left to chance.
Fixed the way VerticalShafts was fixed in 7a20eee: guarantee it by
construction. Each non-origin node hashes ONE parent axis toward
(0,0,0) so Manhattan distance strictly decreases -- a proof of a global
rooted tree, not a probability. Loops are capped at 0.18 x
BranchProbability and 0.10 x Verticality and can never supply connectivity.
⛔ No edge probability was raised; percolation is a phase transition, not a
knob.
64 seeds x 9,261 nodes: 0 disconnected cells, 0 edge-count violations
determinism: 20,000 bit-identical checks
seam freedom: 42 probes, 0 legacy and 0 op-stack mismatches
⭐ AND IT READS AS A MAZE, in numbers rather than adjectives:
degree 1 (dead ends) 15.9% mean degree 2.504
degree 2 36.6% run length median 1, p90 3, max 18
degree 3 31.7%
cubic-grid control: 100% degree-6, 0% dead ends, mean 6.000
Two abreast, derived not nudged: 3 m usable floor / 0.7 = 4.286 m clear
bore, + roughness reach 3.75 voxels => CorridorRadius 12.5 (6.25 m bore),
giving 3.06 m of usable floor. CellSize 64 still leaves a 7.875 m wall
between neighbouring bores.
BranchProbability and Verticality are now LOOP-DETAIL knobs, not
connectivity controls. Documented at their declarations so nobody tunes
them expecting the old meaning.
⚠️ The player-fit law still fails, and Codex named the likely reason in its
caveats: "global tree connectivity does not guarantee connectivity inside
an arbitrary finite cropped measurement window." The tree routes toward the
ORIGIN, and the fitted window is the AABB of the two mouths -- so the
connecting path runs A -> origin -> B and leaves the box. The coarse maze
graph now measures Connected while the fine windowed law does not. That
smells like the sixth measurement artifact of this session, and it is the
next thing to test rather than to assume.
Box audits: existing maze 104 proved / 0 violations; novel fill stack 240
proved / 319,440 voxels / 0 violations. Op-stack bit-identical over 20,000
samples. Perf 1.13 us rebuild, 0.145 us hot call.
Showcase regenerated. Build clean, 32 tests, 30 clean successes, 2 known
deliberate warnings, 0 failed.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01K8V9my1PL68pmQb2gutZZV
Jahni: "can't you try and make the world to scale?" Every length parameter
is now derived from the body and annotated with its metre value, so nobody
has to do the arithmetic to notice something is absurd.
CorridorRadius 4 -> 8 1 m -> 2 m radius (4 m bore)
CellSize 40 -> 64 keeps a real wall between wider corridors
TunnelMinRadius 3 -> 6 0.75 m -> 1.5 m
ShaftMinRadius 5 -> 8 1.25 m -> 2 m
IslandMinRadius 18 -> 24 4.5 m -> 6 m
MouthRadius 6 -> 8 passage mouths 4 m
plus passage reach, spiral, cascade, bridges, ridges, chasms
The reasoning that motivated it still looks right: a 2 m circular bore for
a 1.76 m player clears the head only along the exact centreline, so almost
no cell has a floor with 7.04 voxels of clear air above it. A 4 m bore is
genuinely walkable.
⚠️ BUT IT DID NOT FIX TRAVERSABILITY, and this time it is not the
instrument. The player-fit law is measured at STEP 1 on a mouth-fitted
window -- full resolution, both mouths inside the box -- and across the
eight showcase archetypes it reports:
6 NotConnectedAtThisResolution (real negatives at full resolution)
2 OutOfWindow (unmeasurable, not passing)
1 StartCellNotPlayerFit (the arrival mouth itself is too small)
0 CONNECTED
And every archetype reports hard-gate survivors=0, so the showcase is
displaying the least-bad of an entirely failing set.
⛔ DO NOT READ player_fit_fraction / player_clearance FROM THE SHOWCASE
LINE. Those come from the COARSE step-4 measurement in the same log line,
where the 1.36-voxel player is sub-cell and fit is unmeasurable --
player_clearance reads a uniform "1.00 voxels" for all eight archetypes,
which is one cell, not one voxel. Only restricted_law is at step 1 and
only it is trustworthy. Fixing that mixed-resolution reporting is next.
So scale was necessary and insufficient. My hypothesis was that bore
diameter versus body height was the binding constraint; doubling it moved
nothing measurable. Something else is preventing player-sized volume from
existing at all -- worth finding before tuning any further.
Codex applied these edits and then died mid-poll without validating them
(third time a killed agent left a clean-looking unbuilt diff). Build and
suite run by me: build clean, 30 succeeded / 0 failed.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01K8V9my1PL68pmQb2gutZZV
Jahni: "can you not implement those settings? as in, have them be live in
game?" This is that. The composer stops being a lab tool.
THE ARCHITECTURAL LINE, which is the whole design (COMPOSER-NOTES §3.1):
COMPOSITION stays editor-only, EVALUATION becomes runtime. Rolling,
measuring, preview, selection, promotion, the corpus -- none of it reaches
a player's machine. What crosses into the runtime is exactly two things:
reading a season manifest, and materialising a recipe into an op stack.
UVoxelSeasonAsset a cookable UPrimaryDataAsset holding the manifest,
so a season ships in a packaged game. The JSON
stays the reviewable, git-diffable artefact.
UVoxelSettings::Season optional. Unset => behaviour is exactly as today.
GetEffective*() season seed / spine radius / world radius / season
number override the authored values when assigned.
The test deliberately sets DIFFERENT authored
values, so "the season is authoritative" is proved
rather than coincidental.
⭐ MULTIPLAYER DETERMINISM, PROVED: two independent runtime managers built
the same season and compared 393,216 density samples BIT-IDENTICALLY. This
is the property the whole multiplayer model rests on -- replicate seed +
layout + diff, never geometry.
⭐ AND A STALE SEASON IS DETECTABLE, NOT SILENT. Initialize() now returns
bool and an assigned-but-invalid season NEVER falls back to the authored
pool. Falling back would hand a client a world that looks fine and is not
the season -- silent divergence, the worst failure mode in a shared world.
GetCurrentSeasonContentHash() is exposed BlueprintPure to compare at join.
HasComposerRecipeOverride() is gone. It force-classified every custom recipe
as Mixed, which was correct while box verdicts were only proved for
hand-written stacks -- but 8e346a3 made EffectOverBox correct in ISOLATION,
proved on a stack no builder produces. Keeping the force would have cost
tile skipping on every composed world forever.
VF_RollStrateRegionCount moved behind WITH_EDITOR while the partition-seed
helpers stayed runtime: rolling out, evaluation in. Lateral regions remain
gated off.
⚠️ Sol hit its usage limit mid-task, so the edits arrived unvalidated. I
built and ran them myself rather than assume -- and they did not compile:
VoxelForgeComposerSeasonTest.cpp was missing UObject/Package.h, which every
other test picks up transitively through the shared fixture it does not use.
One include; every NewObject/GetTransientPackage in the file was failing.
Editor build clean. SHIPPING BUILD SUCCEEDS -- the invariant that proves no
composer roller is linked into a packaged game. 30 succeeded / 0 failed.
⚠️ UNVERIFIED: whether it looks right on screen. Nothing headless can tell
us that; it needs Jahni to assign a season asset and press play.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJiTcTSDNkKBtnNdiBtdaU
Jahni's own idea: "archetype could even mix within the same strate at some
point, tunnel network leading to a big chamber". An archetype becomes a
spatial REGION inside a strate, not a label on it.
Built by extending the mechanism that already exists rather than inventing
one: the stack blends VERTICALLY between strates, so regions blend the same
way LATERALLY. Deterministic jittered Voronoi sites on a 256-voxel lattice,
1-3 regions per strate from (seed, strateIndex), 24-voxel symmetric blend
band, per-chunk lookup cache. Parameters are never blended -- a shaft's
ShaftSpacing means nothing to a slab -- only the resulting DENSITY is.
Structural posts stay global and unconditional, so an invented region
cannot switch off the primordial law.
single-region path BIT-IDENTICAL to legacy, 256 probes, native + recipe
cross-boundary boxes 20 of 40, 199,800 voxels brute-forced, 0 VIOLATIONS
(every seam-spanning box correctly returned Mixed)
blend-band cost 663 ns vs 338 ns per voxel, over 16.0% of voxels
~1.15x field-weighted
survival 16/16 valid, non-vacuous, largest share >= .50
⛔ AND IT IS GATED OFF, because the seam breaks the one law that matters:
arrival -> departure across a region boundary passes 9/16 seeds, against
11/16 same-region controls in the same harness. Density blending alone does
not guarantee a player can cross a seam. A strate that mixes archetypes and
strands you at the boundary is strictly worse than one that does not mix.
VF_LateralRegionsAreShippable() returns false and the test ASSERTS THE GATE
IS SHUT while the rate is imperfect -- so flipping it without fixing
connectivity turns the suite red by design. The rate is a warning, not an
assertion: it is the measured state of an off-by-default feature, not a
regression in a live world, and a permanently red suite stops being read,
which is how real failures get missed.
Everything that ships is still hard-guarded: single-region bit-identity,
zero cross-boundary box violations, and every shipping-path law test
untouched.
The fix is an explicit CROSS-SEAM CORRIDOR CONTRACT -- a design decision,
recorded rather than made unilaterally. Codex diagnosed this correctly and
declined to tune the threshold, which was the right call.
Build clean, 29 succeeded / 0 failed, run and verified by me.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJiTcTSDNkKBtnNdiBtdaU
PART B -- the thing Jahni asked for: "any way to test those in game?"
A candidate existed only as measurements and PNGs. Now he can stand in one.
Select the VoxelWorld actor in PIE -> Live Edit|Composer ->
ComposerSeed / ComposerCandidateIndex / ComposerTargetStrateIndex /
bComposerRollStructure -> click Apply Composer Candidate.
The button and the automation tests call the SAME roll wrapper, which was
the load-bearing requirement: a second roll path could drift, and then he
would be judging a different world than the contact sheet describes. A
determinism sample is logged on apply (seed, index, recipe, archetype,
metrics, epoch, eight-point density verdict) so drift is caught the day it
happens rather than in a month.
Hazards handled, since this is the first composer code to touch the live
runtime path: preflight materialisation before any live mutation, generation
and decoration tasks drained, PassagesVersion bumped to invalidate
thread_local caches, reuse of the existing RegenerateAllChunks / Epoch
invalidation rather than a new path, no Initialize() or GeneratePassages()
during apply (the re-entrancy trap), DiffLayer locks untouched, and custom
recipes conservatively force ClassifyTile to Mixed.
Scope is deliberate: a DENSITY-ONLY overlay. Layout height, authored
passages, content, water and biome metadata stay from the existing slot;
RebuildStrates() clears it. A full composed layout replacement needs a safe
rebuild path that does not exist yet.
Editor-only, verified: a Shipping build succeeds with the bridge excluded
(96/96 actions), so §3.1's "offline only, never on a player's machine"
still holds.
⚠️ UNVERIFIED BY DESIGN: a button cannot be tested headlessly and Codex was
told not to claim it works. Jahni presses it and confirms the visual plus
the "determinism=PASS" log line.
PART A -- why rooms read angular, measured not guessed. One of three room
shapes is VoxelSDF::RoundedBox:
ShapeA = (.8 Rxy, .8 Rxy, .8 Rz) ShapeR = .25 Rxy
so rounding is 31% of the horizontal half-extent but 78% of the vertical.
Corners are rounded; the broad planar FACES survive. SmoothMin (blend 4)
softens room/tunnel junctions not isolated corners; the cave warp bends
geometry but preserves flat regions; roughness is surface-scale only.
The name is accurate -- "rounded box" is not "blob".
NOT ACTED ON. Jahni corrected his own read: the angular archetype is Maze
(lattice source, rectilinear BY DESIGN and correct), and TunnelNetwork is
"alright-ish". Codex's recommendation is recorded for if he ever wants it:
~5-10% boxes, 10-20% capsules, rest ellipsoids, or a larger ShapeR.
Also closed REVIEW_FINDINGS EVoxelPassageType/Style as WONTFIX, enacting a
2026-08-16 judgement that recommended closure and was left open.
Build clean (Editor + Shipping), 25 succeeded / 0 failed, run by me.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJiTcTSDNkKBtnNdiBtdaU
COMPOSER-NOTES §6.1, titled THE BLOCKER, gated everything after it. A
source writing only Sdf did not touch density, so it answered the box
query ON BEHALF OF the converter that followed it -- fine when a human
wrote and verified the pair, unsafe the moment the composer assembles a
combination nobody checked. And the failure is the silent one: a tile
wrongly proved uniform means no geometry, no collision, no error, and a
player falls through the floor.
FVoxelBoxHypotheses now carries FVoxelBoxSdfInterval. SDF writers publish
their own [Min, Max]; replacement writers replace it, accumulating writers
MinWith it; converters and detail ops consume the prior interval through a
state-aware EffectOverBox. Each op answers only for itself and the fold
composes them. Unknown, invalid, non-finite or unimplemented bounds force
Mixed -- when in doubt the answer is Mixed, because a missed skip costs
milliseconds and a false uniform costs a player.
Bounds are SUPREMA of what Eval can produce, never the param named "max":
FBM 1.5 (not 1.0), max(Min,Max) radius envelopes, warp sqrt(2), SmoothMin
slack. [FLT_MAX, FLT_MAX] is the exact no-surface identity.
⭐ THE PROOF IS THE NOVEL STACK. Built a combination no shipping builder
produces:
ConstantVoid -> LatticeCorridorSource -> SdfRoughness -> SdfFill
brute-forced 332,750 voxels across 256 boxes: 250 proved AllAir,
0 VIOLATIONS. If isolation is real an unverified stack must still be
sound, and it is. That test -- not the refactor -- is the deliverable.
Every archetype re-proved by brute force, 0 violations everywhere:
Maze 112/256 · CrystalChamber 38/60 · Underwater 33/40 · FlatPlain 32/60
VerticalShafts 28/60 · FloatingIslands 15/60 · TunnelNetwork prod 9/40
TunnelNetwork dense 0/40 (correctly -- it really is all mixed)
COST, reported not hidden: isolation gives up some skipping. TunnelNetwork
production 12->9, VerticalShafts 32->28. Answering for yourself alone is
strictly weaker than answering for a verified pair. The classifier stays
materially useful and the trade buys safe composition, which is the point.
Eval is untouched; equivalence tests bit-identical; no per-voxel box query.
SurfaceWorld keeps its separate exact height/lattice classifier.
Also: ⛔ WITHDREW the "corrective ops must be fused" claim from §3.2.
FFloorBiasMod does NOT undo FCaveRoughnessMod -- it adds an independent
NormZ^2 x FloorBias term to the ACCUMULATED density and never reads
TotalRough. Not a correction, no fuse, no "corrects" dependency. The claim
came from reading a name and a comment; one grep would have shown the ten
Density-writing ops in between before the paragraph was written.
Build clean, 22 succeeded / 0 failed, verified in VoxelM.log.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJiTcTSDNkKBtnNdiBtdaU
The composer must assemble stacks nobody wrote by hand. Today the legal
order of operators lives in a human's head and in the hand-written
builders. COMPOSER-NOTES §3.2: an op's legal position is fixed by which
channel it READS and which it WRITES, so dependencies form a DAG and any
topological sort of that DAG is a legal stack.
EVoxelOpChannel names the two fields FVoxelOpSample actually carries
(Density, Sdf) -- read from the struct, not assumed. All 29 concrete
density ops declare reads, writes and additive/transformative, filled in
by reading each Eval rather than trusting its name. The declarations are
PURE VIRTUAL so a new op cannot silently omit them; the composer's safety
depends on every op answering.
A declaration nobody checks is a comment, so ValidateChannelOrder exists
and VoxelForge.OpStack.ChannelDAG runs all eight shipping stacks through
it. All pass: TunnelNetwork/Underwater 20 ops, Maze 8, VerticalShafts 9,
FloatingIslands 8, FlatPlain/CrystalChamber 6, SurfaceWorld 6.
Metadata is snapshotted in FVoxelOpStack::Add, so no declaration virtual
is ever called per voxel -- ARCHITECTURE §8.10 respected.
⚠️ THE FUSE WAS DELIBERATELY NOT APPLIED, and the reason is the finding I
asked for instead. §3.2 assumed FFloorBiasMod sits next to
FCaveRoughnessMod. It does not: TEN ops separate them (Terrace, LayerLine,
Ribbing, Overhang, Cliff, Scallop, Arch, RoomColumn, Dome, Pinch), and all
ten read and write Density. Moving FloorBias would reorder the live
density accumulation, and IEEE-754 addition is not bitwise commutative --
so the note's claim that fusing "leaves the field bit-identical" is false
as written.
That leaves a real open question for the composer, recorded not resolved:
channel rules alone say any Density->Density order is legal, but they
cannot express "this op CORRECTS that one". Either the fuse proceeds and
accepts a changed field (legitimate -- peer determinism never required
bit-identity across builds), or corrective intent needs its own
declaration. Settling it needs FFloorBiasMod's actual math read, which is
the next question I put to Codex.
Build clean, 22 succeeded / 0 failed, verified in VoxelM.log.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJiTcTSDNkKBtnNdiBtdaU
Closes the VerticalShafts bug. Two remaining failures were investigated
and neither was a broken world.
PART A -- my measurement window was binding, not the world.
The window was a square centred on (0,0) with a radius, while mouths sit
up to ~200 voxels out in arbitrary directions, so a box holding both was
mostly empty space paid for at full price. Seed 6 at step 1 needed
16,515,072 cells and blew the cap; fitted to the two mouths it needs
6,553,344 and measures fine.
seed 6 step 2 NOT_CONNECTED -> CONNECTED at doubled margin
seed 6 step 1 NOT_CONNECTED -> CONNECTED at doubled margin
seed 14 step 2 CONNECTED
seed 14 step 1 CONNECTED
A fitted box can cut off a route that detours outside it, so every
negative is re-run at doubled margin. 5 of 16 were binding negatives and
all 5 flipped. TRUSTED NEGATIVES: 0. There is no evidence of a single
unescapable strate in 16 seeds.
PART B -- the drainage tree has no orphans, and a third cause existed.
0 / 10,372 unreachable shaft paths (0% orphan rate)
16 strict local minima, all caught by the direct-spine fallback
neighbour fallback fired 0 times; emergency fallback 0
91.8% of non-largest components are smaller than one shaft volume --
they are roughness bubbles, not geometry
So a mouth could land in a noise bubble that is open but unreachable --
worse than useless, exactly the trap the law forbids.
VF_SuggestLandingPoint now returns the selected shaft AXIS, which is on
the drainage tree by construction.
⚠️ Codex found a third cause neither of us listed: tree capsules could
cross ledge bands, so a guaranteed connector could be blocked by a ledge.
Connector Z is now drawn from a deterministic ledge-free interval,
mirrored in the op stack.
⚠️ And an 8th error in my spec: "connect to any neighbour shaft" as a
fallback can create CYCLES -- a loop of shafts that drains nowhere and is
disconnected from the spine. The fallback must preserve strictly
decreasing origin-distance. Implemented that way.
per-archetype law 6/6 · box-verdict violations 0 · stack equivalence
bit-identical over 20,000 samples · sweep 45.3 -> 20.4 s
cache 7.970 us rebuild, 0.073 us hot call (225-cell rolls on rebuild only)
Build clean, 21 succeeded / 0 failed, verified by me in VoxelM.log.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJiTcTSDNkKBtnNdiBtdaU
Fr0zkaandClaude Opus 5.5