refactor(core): bind generator fields once per resolved strate (L40, migration step 2); research R7

SurfaceWorld, FloatingIslands and legacy evaluators sit behind one immutable densityAt adapter held in
a variant (no virtual call per sample); BindResolvedStratumFields builds a sorted strate table at world
setup, so DensityAt, native SW/FI sampling and ClassifyBoxDetailed no longer ask the strate manager per
sample. No arithmetic change: owner hashes identical to L37/L39 at W1/W12. Cost vs L39: FI -0.66%,
SW -1.31% (noise). verify fast/medium/full PASS (full re-run alone: 407 gates).
Research R7: disk cache design for UE5/Windows/multiplayer and a zlib probe (about 1/3 size, 0.1 ms/tile).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_016dhxM43dnWvUu2B5Ph1iG1

Standalone/Research/WorldAssemblerM3/Linux40/REPORT.md

@@ -0,0 +1,74 @@
+# Linux40 — immutable field binding
+
+## Change
+
+`ShapeWorldInternal.h` now has concrete immutable SurfaceWorld, FloatingIslands and legacy adapters
+with a common `densityAt(x,y,z,out)` call shape. `FGeneratorFieldAdapterV1` stores them in a variant,
+so hot sampling uses concrete dispatch without a virtual call. FI retains its fixed-point sampler and
+captures the legacy boundary parameters while binding.
+
+`BindResolvedStratumFields` creates a sorted table of resolved strate bounds, plan references and
+evaluator objects during world construction/initialization. Native SurfaceWorld and FI sampling,
+boundary composition lookup, legacy SurfaceWorld bridging and `ClassifyBoxDetailed` select from that
+table. The manager layout width is captured once for the legacy-compatible chunk projection.
+
+## Lookup boundary and identity
+
+The direct manager selection lookups in native SurfaceWorld/FI evaluation and box classification are
+removed. `DensityAt` now selects native fields through the bound table. Legacy
+`UVoxelGenerator::GetDensityAt` still performs its own manager routing when reached through the
+legacy bridge or an out-of-plan fallback. Manager-derived layout width/FI compatibility parameters
+are captured at bind time; manager access also remains in build/planning helpers. Generator-specific
+boundary and passage policy still lives in `FStandaloneShapeWorld`.
+
+No density arithmetic, fixed-point conversion order, evaluator version or Unreal adapter changed.
+The canonical owner FI and SurfaceWorld hashes are identical to L37/L39 for W1 and W12. The owner
+layout IDs also match L39: FI `b7a7d9ecc48d8f26`, SurfaceWorld `8a52e84c26ed92d2`. Fast logs and hash
+summaries: `Saved/Linux/l40/fast-trial1.log` and `Saved/Linux/l40/fast-summary.json`.
+
+## Cost
+
+Five interleaved pairs against the L39 Clang binary, using the canonical owner recipes:
+
+| Recipe | L39 median | L40 median | Median change | Paired changes (L40 vs L39) |
+|---|---:|---:|---:|---|
+| FloatingIslands | 17.556 s | 17.440 s | −0.66% | +2.26%, −3.99%, −0.66%, −0.87%, −0.47% |
+| SurfaceWorld | 1.829 s | 1.805 s | −1.31% | −2.64%, −0.22%, +0.55%, −0.44%, −1.48% |
+
+Both deltas are within observed pair noise; no per-voxel regression was measured. Raw output is in
+`Saved/Linux/l40/cost-l40-vs-l39-run.log` and `cost-l40-vs-l39-summary.json`.
+
+## Verification
+
+| Tier | Result | Wall time | Evidence |
+|---|---|---:|---|
+| Fast | PASS | 162.0 s | Canonical owner hashes match L37 for both recipes at W1/W12; all controls pass. |
+| Medium | PASS | 468.6 s | Transition, route/proof, LOD0–8 ring, owner shape and capture matrix pass. |
+| Full | TIMED OUT | 1740 s (29 min) | GCC build and supplement passed; L25 and crossing24 determinism passed for both compilers, and Clang upper-SW style0 W1/W8/W12 passed. The run timed out during the GCC upper-SW style0 W1 cell; no full summary or cost-baseline/pair results were produced. |
+
+Clang and GCC candidate builds succeeded with source stamp
+`f35bc59c6172370e98dc83ec8ee5ba51dfab83a3faf075c26189fc83172f4ac8`. The full run used the requested
+CPU/memory limits and a 1740-second timeout. At the stop, host load was 20.5. The partial log is
+`Saved/Linux/l40/full-final.log`; a `full-summary.json` was not written. Other tier and build logs are
+under `Saved/Linux/l40/`.
+
+## Failed attempts
+
+- The first Clang build passed `unique_ptr` objects to adapter constructors expecting raw observer
+  pointers; corrected with `.get()`, then the build passed.
+- Two early cost-harness launches failed before starting measurements: one shell quoting error and
+  two Python `-c` quoting errors. The later logged five-pair run completed successfully.
+- The single full-tier run reached its 1740-second timeout while the GCC upper-SW fixture style0 W1
+  determinism cell was still running. The cell has no completion record; the run returned timeout
+  status 124. No full-tier L36 baseline build or full-tier cost pairs ran.
+
+No Unreal compile was run; no Unreal adapter or `VoxelWorld.cpp` source change was required.
+
+## Orchestrator review (Claude)
+
+Accepted. The full tier had timed out here under load (two Lunas plus host load of about 20). I re-ran it alone on
+the final source (stamp `f35bc59c…`): **PASS, 407 gates, 2857 s.** Its built-in cost comparison is against
+**L36** and shows FI +3.03% (all five pairs positive) and SW +0.38%. That FI delta is L37's mouth geometry; this
+round's own L39→L40 pairs are −0.66% FI and −1.31% SW. Two follow-ups for the tool:
+- the full tier's cost baseline is hard-coded to L36; it should follow the previous commit;
+- the full tier now needs about 48 minutes including builds, so its timeout must be above 29 minutes.

Standalone/Research/WorldAssemblerM3/Research7/REPORT.md

@@ -0,0 +1,45 @@
+# Research 7 — terrain disk cache: Windows and multiplayer
+
+Inspection date: 2026-10-23. Read-only review of `Standalone/Source/StreamingWorld.h`, `Standalone/Source/StandaloneViewer.cpp`, Core version declarations, and the verification README.
+
+## What exists and Windows blockers
+
+`FStandaloneMeshCache` is a whole-world archive, not an on-demand disk cache. Its filename is derived from seed, clip radius, and tunnel mode; the header also checks source JSON FNV-1a-64, generation settings, cache/algorithm version, tunnel mode, and voxel size. A tile lookup uses `(LOD, X, Y, Z)`. The `VFMESH1\0` v1 archive has an 80-byte raw C++ header, 40-byte raw tile records, float32 position/normal pairs, and uint32 indices. It is uncompressed and host-layout/endian dependent; the endian marker detects mismatch but does not convert fields. Save/load also omit generated `SurfaceVertexCount`, `SkirtTriangleCount`, and `TriangleSkirtFaceMasks`, so a cache hit cannot preserve skirt culling.
+
+Load reads and validates the entire archive synchronously in `ConfigureProfile` before workers start. Streaming workers then copy hits from an immutable in-memory map. Bake save writes synchronously to the final path with truncation. There is no disk eviction or file lock; leaving tiles evicts resident meshes only. The cache has no `mmap`, `fsync`, rename, or zlib use today. Existing checks reject mismatched metadata, invalid LOD/counts, truncated data, duplicate keys, non-finite positions/normals, and out-of-range indices, but there is no payload checksum, total allocation budget, or trailing-data check. Stored density hashes are not checked against the mesh.
+
+Windows concerns are the narrow `.string()` path conversion and a fixed `MAX_PATH` buffer in `DefaultDataPath`; the cache is written next to source data, which may be read-only in a packaged install. Windows long-path support needs application opt-in, and the cache path should stay in TCHAR/UE paths. Direct-to-final truncation can leave a partial archive and destroy the previous one; `ofstream::flush` is not a durable or atomic replacement. Concurrent bakes can collide. The standalone viewer itself has unguarded Win32/D3D11 headers and should not be pulled into the UE plugin; keep any platform adapter gated and use UE file APIs.
+
+## UE5 design
+
+- Store a versioned, endian-neutral per-tile payload under `FPaths::ProjectSavedDir()/VoxelForge/TileCache/<project-id>/<user-scope>/`. Use `ProjectPersistentDownloadDir()` when the packaged install’s Saved directory is not writable; append a project identity either way. [FPaths API](https://dev.epicgames.com/documentation/en-us/unreal-engine/API/Runtime/Core/FPaths)
+- Use `FPlatformFileManager::Get().GetPlatformFile().OpenAsyncRead()` and `IAsyncReadFileHandle::ReadRequest()` for tile/block reads. Parse and decompress on a worker; never wait for file I/O or decompression on the game thread. UE documents async open as non-blocking and read requests as caller-owned; keep buffers/requests alive through completion and cancellation. [FPlatformFileManager](https://dev.epicgames.com/documentation/unreal-engine/API/Runtime/Core/FPlatformFileManager), [IPlatformFile](https://dev.epicgames.com/documentation/unreal-engine/API/Runtime/Core/IPlatformFile), [ReadRequest](https://dev.epicgames.com/documentation/unreal-engine/API/Runtime/Core/Async/IAsyncReadFileHandle/ReadRequest?application_version=5.5)
+- Use UE `FCompression` with registered format names and `IsFormatValid`; its compression/decompression routines are thread-safe. This avoids a separately linked Windows zlib. Keep codec ID/version in each tile record and check availability in the packaged client/server. Oodle Data is supplied as a UE plugin and documents Kraken, Mermaid, and Selkie as distinct size/decode-speed choices; do not assume a packaging setting alone enables a runtime cache codec. [FCompression](https://dev.epicgames.com/documentation/en-us/unreal-engine/API/Runtime/Core/FCompression), [Oodle Data (UE 5.7)](https://dev.epicgames.com/documentation/en-us/unreal-engine/oodle-data?application_version=5.7)
+- Treat the generated terrain as persisted feature data, not as a speed claim. Record schema, world key, tile key, codec, compressed/uncompressed sizes, and payload digest. Include all mesh groups, masks, and collision inputs required to rebuild identical tiles. Write a unique temp beside the target, flush and close it, then replace/rename on the same volume. Do not allow a cross-volume copy fallback. Use `ReplaceFileW`/same-volume `MoveFileExW` through a Windows adapter; `FlushFileBuffers` is the file-data durability step. [ReplaceFileW](https://learn.microsoft.com/en-us/windows/win32/api/winbase/nf-winbase-replacefilew), [MoveFileExW](https://learn.microsoft.com/en-us/windows/win32/api/winbase/nf-winbase-movefileexw), [FlushFileBuffers](https://learn.microsoft.com/en-us/windows/win32/api/fileapi/nf-fileapi-flushfilebuffers), [Windows long paths](https://learn.microsoft.com/en-us/windows/win32/fileio/maximum-file-path-limitation)
+
+Cache misses, corrupt entries, and unavailable codecs fall back to cold generation on workers. Keep eviction as a background size-bounded LRU; cache eviction only causes regeneration. Continue judging the no-cache cold path separately.
+
+## Key, multiplayer, invalidation, validation
+
+Key a base tile by authoritative season/world-plan identity (include Core recipe and plan content hashes), recipe/schema version, evaluator version, selected mesher plus an explicit mesher implementation revision, LOD and absolute tile coordinate. Add the development source stamp and cache-format version. Core currently declares WorldAssembly schema 2/evaluator 208; FloatingIsland recipe schema 1/default geometry version 4; mesher choices are MarchingCubesV1, HybridV1, HybridV2, with far-horizon version 4 affecting LOD4–8. Read actual resolved recipe/plan versions and hashes from the active snapshot. The verification README defines a SHA-256 source stamp over a sorted source manifest; it is dynamic and can optionally include the Unreal adapter. The current cache uses only the source-data FNV hash and does not carry that stamp.
+
+Persist immutable base terrain only. Keep player edits in the replicated diff layer and overlay them at runtime. Do not write edited composite tiles into the base cache; if a separate composite cache is ever added, key it by the affected tile’s edit revision/content hash. A client cache is visual convenience only: the server uses its own matching cache or regenerates and builds its own collision; never accept client cache bytes as collision or gameplay authority.
+
+On load, bound both per-record and total tile/byte allocations before allocating; require exact file/block lengths, known codec/schema/key, finite and bounded geometry, valid normals, triangle/index/mask counts, and index ranges. Verify a per-payload digest and regenerate on failure. A digest detects accidental corruption, not a malicious rewrite that also changes the digest; only importable/shared caches need a server signature or MAC. Never trust a cache supplied by a client.
+
+## Compression probe and next build round
+
+A small serial Python probe used `VFMSH001` exports `linux27-fi-anchor-pillar.vfmesh` and `linux27-sw-column-wide.vfmesh`, all non-empty tiles LOD0–8, CPython zlib level 1. Payload excludes scene/tile headers. These are Linux in-memory decode numbers, not UE/MSVC or disk-I/O results; exports lack recipe/source stamps and density samples.
+
+| Export | Non-empty tiles | Mean raw → zlib bytes/tile | Payload ratio | Decode ms/tile p50 / p95 |
+|---|---:|---:|---:|---:|
+| FI | 1,380 | 75.6 KB → 24.2 KB | 32.0% | 0.109 / 0.523 |
+| SW | 1,416 | 60.0 KB → 21.5 KB | 35.8% | 0.114 / 0.459 |
+
+The serial probe peak RSS was 15.4 MiB. Per-LOD details are in `Saved/Linux/research/R7_python_zlib_probe.json`; script: `Saved/Linux/research/R7_python_zlib_probe.py`.
+
+Later compare none, zlib, Oodle Kraken/Mermaid/Selkie, and LZ4; quantized vertices (uint16 positions, compact normals) with delta indices; and density/sign-bit packing with RLE. Use the same owner FI and SW tile sets at LOD0–8, keyed to the exact active recipe/plan. Record bytes/tile (median/p95/total), single-core decode ms/tile (median/p95), peak working memory, and cold disk-read-plus-decode latency separately. Validate geometry error, seams, and server collision against uncompressed output. Sign-bit inputs are absent from these mesh exports, so that candidate needs a later density export and must include remesh time in its comparison.
+
+## Owner question
+
+Will cache files remain local to each user/install, or be copied/shared between machines?

Standalone/Research/WorldAssemblerM3/Research7/scripts/R7_python_zlib_probe.json

@@ -0,0 +1,255 @@
+{
+  "codec": "CPython zlib level 1",
+  "decode_repeats_per_tile": 3,
+  "files": {
+    "FI": {
+      "path": "Saved/Linux/linux27-fi-anchor-pillar.vfmesh",
+      "file_bytes": 104373453,
+      "scene_version": 1,
+      "voxel_size_microns": 250000,
+      "tile_records": 2482,
+      "nonempty_tile_records": 1380,
+      "raw_payload_bytes": 104284041,
+      "zlib_payload_bytes": 33400206,
+      "weighted_ratio_pct": 32.02810869210563,
+      "raw_bytes_per_nonempty_tile_median": 46104.0,
+      "decode_ms_per_tile_median": 0.109374,
+      "decode_ms_per_tile_p95": 0.523225,
+      "lod": {
+        "0": {
+          "nonempty_tiles": 136,
+          "raw_bytes_total": 11914151,
+          "packed_bytes_total": 3218452,
+          "weighted_ratio_pct": 27.01369153370643,
+          "raw_bytes_median": 95849.5,
+          "packed_bytes_median": 33744.0,
+          "decode_ms_median": 0.27110500000000004,
+          "decode_ms_p95": 0.316059,
+          "encode_ms_median": 0.610732,
+          "encode_ms_p95": 0.837242
+        },
+        "1": {
+          "nonempty_tiles": 131,
+          "raw_bytes_total": 16499945,
+          "packed_bytes_total": 5941348,
+          "weighted_ratio_pct": 36.008289724602115,
+          "raw_bytes_median": 92598,
+          "packed_bytes_median": 42032,
+          "decode_ms_median": 0.317582,
+          "decode_ms_p95": 0.818547,
+          "encode_ms_median": 0.837591,
+          "encode_ms_p95": 2.246851
+        },
+        "2": {
+          "nonempty_tiles": 172,
+          "raw_bytes_total": 6666423,
+          "packed_bytes_total": 2031078,
+          "weighted_ratio_pct": 30.467283579214822,
+          "raw_bytes_median": 45768.0,
+          "packed_bytes_median": 11943.0,
+          "decode_ms_median": 0.102419,
+          "decode_ms_p95": 0.124572,
+          "encode_ms_median": 0.2224085,
+          "encode_ms_p95": 0.256617
+        },
+        "3": {
+          "nonempty_tiles": 125,
+          "raw_bytes_total": 5555529,
+          "packed_bytes_total": 1649468,
+          "weighted_ratio_pct": 29.6905659209051,
+          "raw_bytes_median": 45768,
+          "packed_bytes_median": 11960,
+          "decode_ms_median": 0.104118,
+          "decode_ms_p95": 0.168216,
+          "encode_ms_median": 0.2313,
+          "encode_ms_p95": 0.349597
+        },
+        "4": {
+          "nonempty_tiles": 174,
+          "raw_bytes_total": 9645483,
+          "packed_bytes_total": 3115498,
+          "weighted_ratio_pct": 32.30007247952228,
+          "raw_bytes_median": 45768.0,
+          "packed_bytes_median": 13419.0,
+          "decode_ms_median": 0.1148265,
+          "decode_ms_p95": 0.259002,
+          "encode_ms_median": 0.255053,
+          "encode_ms_p95": 0.641356
+        },
+        "5": {
+          "nonempty_tiles": 144,
+          "raw_bytes_total": 4232303,
+          "packed_bytes_total": 1383894,
+          "weighted_ratio_pct": 32.69836776809222,
+          "raw_bytes_median": 24938.0,
+          "packed_bytes_median": 9315.0,
+          "decode_ms_median": 0.06914100000000001,
+          "decode_ms_p95": 0.178538,
+          "encode_ms_median": 0.1752125,
+          "encode_ms_p95": 0.460837
+        },
+        "6": {
+          "nonempty_tiles": 96,
+          "raw_bytes_total": 5255938,
+          "packed_bytes_total": 1770019,
+          "weighted_ratio_pct": 33.67655782849798,
+          "raw_bytes_median": 38395.0,
+          "packed_bytes_median": 10093.5,
+          "decode_ms_median": 0.088979,
+          "decode_ms_p95": 0.317363,
+          "encode_ms_median": 0.1975715,
+          "encode_ms_p95": 0.823547
+        },
+        "7": {
+          "nonempty_tiles": 96,
+          "raw_bytes_total": 11085771,
+          "packed_bytes_total": 3715174,
+          "weighted_ratio_pct": 33.51299607397627,
+          "raw_bytes_median": 105285.5,
+          "packed_bytes_median": 30236.0,
+          "decode_ms_median": 0.271746,
+          "decode_ms_p95": 0.588424,
+          "encode_ms_median": 0.63578,
+          "encode_ms_p95": 1.607345
+        },
+        "8": {
+          "nonempty_tiles": 306,
+          "raw_bytes_total": 33428498,
+          "packed_bytes_total": 10575275,
+          "weighted_ratio_pct": 31.635507524149006,
+          "raw_bytes_median": 100540.0,
+          "packed_bytes_median": 29556.0,
+          "decode_ms_median": 0.258287,
+          "decode_ms_p95": 0.529198,
+          "encode_ms_median": 0.644282,
+          "encode_ms_p95": 1.330977
+        }
+      }
+    },
+    "SW": {
+      "path": "Saved/Linux/linux27-sw-column-wide.vfmesh",
+      "file_bytes": 85051310,
+      "scene_version": 1,
+      "voxel_size_microns": 250000,
+      "tile_records": 2130,
+      "nonempty_tile_records": 1416,
+      "raw_payload_bytes": 84974570,
+      "zlib_payload_bytes": 30390177,
+      "weighted_ratio_pct": 35.76384911391726,
+      "raw_bytes_per_nonempty_tile_median": 38620.5,
+      "decode_ms_per_tile_median": 0.11431150000000001,
+      "decode_ms_per_tile_p95": 0.459004,
+      "lod": {
+        "0": {
+          "nonempty_tiles": 285,
+          "raw_bytes_total": 22447321,
+          "packed_bytes_total": 9051898,
+          "weighted_ratio_pct": 40.32507041708897,
+          "raw_bytes_median": 79066,
+          "packed_bytes_median": 32524,
+          "decode_ms_median": 0.244825,
+          "decode_ms_p95": 0.389183,
+          "encode_ms_median": 0.65028,
+          "encode_ms_p95": 1.155402
+        },
+        "1": {
+          "nonempty_tiles": 149,
+          "raw_bytes_total": 14395111,
+          "packed_bytes_total": 5813866,
+          "weighted_ratio_pct": 40.38778165725849,
+          "raw_bytes_median": 108155,
+          "packed_bytes_median": 46017,
+          "decode_ms_median": 0.348557,
+          "decode_ms_p95": 0.551613,
+          "encode_ms_median": 0.930073,
+          "encode_ms_p95": 1.662883
+        },
+        "2": {
+          "nonempty_tiles": 217,
+          "raw_bytes_total": 6043511,
+          "packed_bytes_total": 1736514,
+          "weighted_ratio_pct": 28.733529234909973,
+          "raw_bytes_median": 27184,
+          "packed_bytes_median": 6351,
+          "decode_ms_median": 0.058163,
+          "decode_ms_p95": 0.14959,
+          "encode_ms_median": 0.14067,
+          "encode_ms_p95": 0.37157
+        },
+        "3": {
+          "nonempty_tiles": 190,
+          "raw_bytes_total": 6058525,
+          "packed_bytes_total": 1902018,
+          "weighted_ratio_pct": 31.394076941169676,
+          "raw_bytes_median": 35120.0,
+          "packed_bytes_median": 7724.5,
+          "decode_ms_median": 0.0678105,
+          "decode_ms_p95": 0.15743,
+          "encode_ms_median": 0.17532,
+          "encode_ms_p95": 0.392782
+        },
+        "4": {
+          "nonempty_tiles": 173,
+          "raw_bytes_total": 8837946,
+          "packed_bytes_total": 2824214,
+          "weighted_ratio_pct": 31.95554713730996,
+          "raw_bytes_median": 35120,
+          "packed_bytes_median": 13578,
+          "decode_ms_median": 0.106095,
+          "decode_ms_p95": 0.267514,
+          "encode_ms_median": 0.252394,
+          "encode_ms_p95": 0.663049
+        },
+        "5": {
+          "nonempty_tiles": 144,
+          "raw_bytes_total": 6093522,
+          "packed_bytes_total": 1982234,
+          "weighted_ratio_pct": 32.530185334524106,
+          "raw_bytes_median": 35013.0,
+          "packed_bytes_median": 10745.0,
+          "decode_ms_median": 0.08670549999999999,
+          "decode_ms_p95": 0.186559,
+          "encode_ms_median": 0.230201,
+          "encode_ms_p95": 0.474221
+        },
+        "6": {
+          "nonempty_tiles": 96,
+          "raw_bytes_total": 6363816,
+          "packed_bytes_total": 2057514,
+          "weighted_ratio_pct": 32.33145018649188,
+          "raw_bytes_median": 61690.0,
+          "packed_bytes_median": 18087.0,
+          "decode_ms_median": 0.1538235,
+          "decode_ms_p95": 0.308712,
+          "encode_ms_median": 0.394647,
+          "encode_ms_p95": 0.820557
+        },
+        "7": {
+          "nonempty_tiles": 96,
+          "raw_bytes_total": 9716690,
+          "packed_bytes_total": 3399953,
+          "weighted_ratio_pct": 34.99085593962553,
+          "raw_bytes_median": 98110.0,
+          "packed_bytes_median": 30712.0,
+          "decode_ms_median": 0.2564655,
+          "decode_ms_p95": 0.53338,
+          "encode_ms_median": 0.6491285,
+          "encode_ms_p95": 1.50398
+        },
+        "8": {
+          "nonempty_tiles": 66,
+          "raw_bytes_total": 5018128,
+          "packed_bytes_total": 1621966,
+          "weighted_ratio_pct": 32.322132875048226,
+          "raw_bytes_median": 74425.0,
+          "packed_bytes_median": 23710.5,
+          "decode_ms_median": 0.187504,
+          "decode_ms_p95": 0.337573,
+          "encode_ms_median": 0.47401950000000004,
+          "encode_ms_p95": 0.885023
+        }
+      }
+    }
+  },
+  "process_peak_rss_kib": 15772
+}

Standalone/Research/WorldAssemblerM3/Research7/scripts/R7_python_zlib_probe.py

@@ -0,0 +1,123 @@
+#!/usr/bin/env python3
+"""Cheap per-tile zlib probe over existing VFMSH001 scenes; read-only inputs."""
+import collections
+import json
+import os
+import resource
+import statistics
+import struct
+import sys
+import time
+import zlib
+from pathlib import Path
+
+FILES = {
+    "FI": Path("Saved/Linux/linux27-fi-anchor-pillar.vfmesh"),
+    "SW": Path("Saved/Linux/linux27-sw-column-wide.vfmesh"),
+}
+REPEATS = 3
+
+def percentile(values, fraction):
+    ordered = sorted(values)
+    if not ordered:
+        return None
+    return ordered[min(len(ordered) - 1, int((len(ordered) - 1) * fraction))]
+
+def scan(path):
+    rows = collections.defaultdict(list)
+    with path.open("rb") as stream:
+        prefix = stream.read(60)
+        if len(prefix) != 60 or prefix[:8] != b"VFMSH001":
+            raise RuntimeError(f"{path}: not a VFMSH001 scene")
+        version, voxel_um, tile_count = struct.unpack_from("<3I", prefix, 8)
+        if version != 1 or tile_count > 1_000_000:
+            raise RuntimeError(f"{path}: unsupported header")
+        for _ in range(tile_count):
+            header = stream.read(36)
+            if len(header) != 36:
+                raise EOFError(f"{path}: truncated record")
+            key = struct.unpack_from("<4i", header)
+            _, _, vertex_count, index_count, mask_count = struct.unpack_from("<5I", header, 16)
+            if key[3] < 0 or key[3] > 8 or index_count % 3:
+                raise RuntimeError(f"{path}: malformed tile record")
+            raw_size = vertex_count * 24 + index_count * 4 + mask_count
+            payload = stream.read(raw_size)
+            if len(payload) != raw_size:
+                raise EOFError(f"{path}: truncated tile payload")
+            if raw_size == 0:
+                continue
+            encode_start = time.perf_counter_ns()
+            packed = zlib.compress(payload, level=1)
+            encode_ms = (time.perf_counter_ns() - encode_start) / 1_000_000.0
+            decode_ms = []
+            for _ in range(REPEATS):
+                start = time.perf_counter_ns()
+                decoded = zlib.decompress(packed)
+                decode_ms.append((time.perf_counter_ns() - start) / 1_000_000.0)
+                if decoded != payload:
+                    raise RuntimeError(f"{path}: round-trip mismatch")
+            rows[key[3]].append({
+                "raw": raw_size, "packed": len(packed),
+                "encode_ms": encode_ms, "decode_ms": statistics.median(decode_ms),
+                "vertices": vertex_count, "indices": index_count,
+            })
+        if stream.read(1):
+            raise RuntimeError(f"{path}: trailing data")
+    return version, voxel_um, tile_count, rows
+
+def main():
+    if hasattr(os, "sched_getaffinity"):
+        os.sched_setaffinity(0, {min(os.sched_getaffinity(0))})
+    result = {"codec": "CPython zlib level 1", "decode_repeats_per_tile": REPEATS, "files": {}}
+    for name, path in FILES.items():
+        version, voxel_um, tile_count, by_lod = scan(path)
+        lod_result = {}
+        for lod, rows in sorted(by_lod.items()):
+            raw = [r["raw"] for r in rows]
+            packed = [r["packed"] for r in rows]
+            dec = [r["decode_ms"] for r in rows]
+            enc = [r["encode_ms"] for r in rows]
+            lod_result[str(lod)] = {
+                "nonempty_tiles": len(rows),
+                "raw_bytes_total": sum(raw),
+                "packed_bytes_total": sum(packed),
+                "weighted_ratio_pct": 100.0 * sum(packed) / sum(raw),
+                "raw_bytes_median": statistics.median(raw),
+                "packed_bytes_median": statistics.median(packed),
+                "decode_ms_median": statistics.median(dec),
+                "decode_ms_p95": percentile(dec, 0.95),
+                "encode_ms_median": statistics.median(enc),
+                "encode_ms_p95": percentile(enc, 0.95),
+            }
+        total_raw = sum(x["raw_bytes_total"] for x in lod_result.values())
+        total_packed = sum(x["packed_bytes_total"] for x in lod_result.values())
+        all_decode = [r["decode_ms"] for rows in by_lod.values() for r in rows]
+        all_raw = [r["raw"] for rows in by_lod.values() for r in rows]
+        result["files"][name] = {
+            "path": str(path), "file_bytes": path.stat().st_size,
+            "scene_version": version, "voxel_size_microns": voxel_um,
+            "tile_records": tile_count, "nonempty_tile_records": sum(len(v) for v in by_lod.values()),
+            "raw_payload_bytes": total_raw, "zlib_payload_bytes": total_packed,
+            "weighted_ratio_pct": 100.0 * total_packed / total_raw,
+            "raw_bytes_per_nonempty_tile_median": statistics.median(all_raw),
+            "decode_ms_per_tile_median": statistics.median(all_decode),
+            "decode_ms_per_tile_p95": percentile(all_decode, 0.95),
+            "lod": lod_result,
+        }
+    result["process_peak_rss_kib"] = resource.getrusage(resource.RUSAGE_SELF).ru_maxrss
+    out = Path("Saved/Linux/research/R7_python_zlib_probe.json")
+    out.write_text(json.dumps(result, indent=2) + "\n", encoding="utf-8")
+    for name, info in result["files"].items():
+        print(f"{name}: tiles={info['nonempty_tile_records']}/{info['tile_records']} "
+              f"raw={info['raw_payload_bytes']} packed={info['zlib_payload_bytes']} "
+              f"ratio={info['weighted_ratio_pct']:.1f}% "
+              f"decode_p50/p95={info['decode_ms_per_tile_median']:.3f}/"
+              f"{info['decode_ms_per_tile_p95']:.3f}ms")
+        print("  LOD raw B/tile median | zlib B/tile median | decode p50/p95 ms")
+        for lod, row in info["lod"].items():
+            print(f"  {lod}: {row['raw_bytes_median']:.0f} | {row['packed_bytes_median']:.0f} | "
+                  f"{row['decode_ms_median']:.3f}/{row['decode_ms_p95']:.3f}")
+    print(f"peak_rss={result['process_peak_rss_kib']} KiB; JSON={out}")
+
+if __name__ == "__main__":
+    main()

Standalone/Research/WorldAssemblerM3/Research7/shared-notes.md

@@ -0,0 +1,2 @@
+# Shared notes (researcher R7 <-> implementer L40), newest at bottom
+