From 4908eb2e31e0dc76020bb1520625e10e8def24c9 Mon Sep 17 00:00:00 2001 From: sindoring Date: Sun, 12 Jul 2026 02:16:09 +0400 Subject: [PATCH] test(M00): require WMO-ready pose captures --- .../claims/M00-QAR-WMO-READINESS-001.md | 78 +++++++++++++++++++ docs/CAMERA_POSE_SWEEP.md | 8 +- docs/modules/world-renderer.md | 3 +- tools/sweep_render_checkpoint_camera_pose.ps1 | 4 +- 4 files changed, 88 insertions(+), 5 deletions(-) create mode 100644 coordination/claims/M00-QAR-WMO-READINESS-001.md diff --git a/coordination/claims/M00-QAR-WMO-READINESS-001.md b/coordination/claims/M00-QAR-WMO-READINESS-001.md new file mode 100644 index 0000000..479237f --- /dev/null +++ b/coordination/claims/M00-QAR-WMO-READINESS-001.md @@ -0,0 +1,78 @@ +# M00-QAR-WMO-READINESS-001 — Goldshire WMO readiness + + + +## Ownership + +- Target: M00 +- Program: QAR +- Owner/Agent ID: sindo-main-codex +- Branch: `work/sindo-main-codex/m00-wmo-readiness` +- Lease expires UTC: 2026-07-14 +- Integrator: milestone integrator + +## Outcome + +Prevent pose calibration against an under-loaded WMO scene and replace invalid Goldshire evidence with a baseline-ready comparison. + +## Non-goals + +- Changing runtime streaming budgets. +- Claiming all WMO queues must globally drain before a checkpoint. +- Changing manifest camera pose without valid paired evidence. + +## Paths + +- Exclusive: pose-sweep readiness default and Goldshire timing evidence +- Shared/hotspots: renderer baseline/module fidelity findings +- Generated/ignored: timing and sweep images under `user://` + +## Contracts and data + +- Public API/events: pose sweep default wait becomes 8 seconds, matching the M00 manifest +- Schema/format version: unchanged +- Migration/compatibility: explicit `-WaitSeconds` remains supported +- Consumers: M00 paired-fidelity workflow + +## Dependencies + +- Requires: Goldshire pose evidence and private build 12340 reference +- Blocks: valid Goldshire camera-pose ranking +- External state: private reference and generated captures remain outside Git + +## Verification + +- Commands: 2/8/15 second timing audit, ready-scene 3x3 pose sweep, baseline dry-run, repository gates +- Fixtures: private Goldshire Inn screenshot +- Fidelity evidence: visual WMO presence plus streaming snapshot and paired metrics +- Performance budget: offline diagnostic + +## Documentation deliverables + +- Inline public API docs: PowerShell default +- Module specification: corrected readiness/fidelity finding +- Data-flow diagram: readiness precondition +- Sequence/state/dependency diagrams: unchanged +- Source map/status updates: camera-pose operational guide + +## Simplicity and naming + +- Important names introduced: none +- Simplest considered solution: reuse the manifest's measured 8-second stabilization default +- Rejected complexity/abstractions: global queue barrier that would include unrelated distant work +- Unavoidable complexity and justification: WMO publication is asynchronous +- Measured optimization evidence: 2/8/15 second snapshots + +## Status + +- State: ready-for-review +- Done: timing audit invalidated the 2-second evidence; sweep default now matches the 8-second manifest; ready-scene 3x3 grid and human inspection completed +- Next: jointly refine negative pitch and FOV around the ready-scene Goldshire candidate +- Blocked by: + +## Handoff + +- Commit: this work-package commit +- Results: WMO instances 9/113/328 at 2/8/15 seconds; inn absent at 2 and visible at 8/15; ready grid improved mean error from 0.101402 at (0,0) to 0.087952 at (10,-10) +- Remaining risks: global WMO queues still contain unrelated work; best pitch is at the tested boundary and building scale indicates unresolved FOV/framing +- Documentation updated: `docs/CAMERA_POSE_SWEEP.md`, `docs/modules/world-renderer.md` diff --git a/docs/CAMERA_POSE_SWEEP.md b/docs/CAMERA_POSE_SWEEP.md index fe253ca..1346fda 100644 --- a/docs/CAMERA_POSE_SWEEP.md +++ b/docs/CAMERA_POSE_SWEEP.md @@ -15,10 +15,10 @@ This offline M00 diagnostic recovers reproducible checkpoint framing when an ori -YawOffsets -15,-10,-5,0,5,10,15 ` -PitchOffsets -10,-5,0,5,10 ` -ViewportWidth 1280 -ViewportHeight 960 ` - -WaitSeconds 2 + -WaitSeconds 8 ``` -Each candidate receives its own output directory and comparison report. The viewport must exactly match the reference image dimensions; the runner fails on `size_mismatch` instead of inventing a score. `ranking.json` sorts candidates by mean perceptual error and then changed-pixel ratio. Use `-PlanOnly` to validate the Cartesian grid and output paths without rendering. Run a coarse grid first, then a finer grid around the best candidate. +Each candidate receives its own output directory and comparison report. The viewport must exactly match the reference image dimensions; the runner fails on `size_mismatch` instead of inventing a score. The default 8-second wait matches the M00 manifest and is a readiness precondition: lowering it requires evidence that the checkpoint's WMO/M2 content has already appeared. `ranking.json` sorts candidates by mean perceptual error and then changed-pixel ratio. Use `-PlanOnly` to validate the Cartesian grid and output paths without rendering. Run a coarse grid first, then a finer grid around the best candidate. ```mermaid flowchart LR @@ -43,6 +43,8 @@ The headless capture dry-run must report requested yaw and pitch offsets. Compar The private `2560x1440` build 12340 `goldshire_inn_large_wmo` reference was evaluated at FOV 62 degrees. A coarse 5x5 grid over yaw/pitch `[-20, -10, 0, 10, 20]` reduced mean error from `0.099632` at `(0, 0)` to `0.077575` at `(-10, -20)`. Extending pitch produced `0.070901` at `(0, -40)` and `0.063574` at `(0, -60)` with changed-pixel ratio `0.549556`. -This is not a camera-pose solution. Human inspection shows the original-client reference centered on the Goldshire Inn, while the zero-offset OpenWC candidate contains forest/roads and no inn; negative pitch increasingly fills the frame with grass. The monotonically improving full-frame metric rewards similar green color coverage rather than landmark alignment. Therefore the manifest pose is unchanged and this checkpoint remains spatial/placement-composition evidence. A future registration metric must use approved landmark or region masks before it can select camera pose automatically. +These values are invalid camera-pose evidence because the sweep used a 2-second wait. A subsequent readiness audit showed 9 WMO instances and no inn after 2 seconds, 113 instances with the inn visible after 8 seconds, and 328 instances after 15 seconds. The apparent monotonic improvement rewarded grass while the landmark was still absent; it does not demonstrate a spatial/placement defect. The manifest pose remains unchanged pending a ready-scene rerun. + +The ready-scene 3x3 rerun at FOV 62 degrees ranked `(yaw=10, pitch=-10)` first with mean error `0.087952` and changed-pixel ratio `0.665527`, compared with `0.101402`/`0.702489` at zero offsets. Human inspection confirms that the inn is present and the direction is plausible. Pitch remains on the tested boundary and the building scale differs from the reference, so these offsets are coarse evidence only; a joint pitch/FOV refinement is required before changing manifest camera values. During this run the sweep orchestration was hardened in two ways: expected comparator exit code `1` is now collected through an explicit child process, and viewport dimensions are explicit. A reference/candidate `size_mismatch` now fails before ranking instead of producing an empty metric. diff --git a/docs/modules/world-renderer.md b/docs/modules/world-renderer.md index 2634ffc..ed3182a 100644 --- a/docs/modules/world-renderer.md +++ b/docs/modules/world-renderer.md @@ -230,7 +230,8 @@ Exact exported settings and cache versions remain documented in [`../../RENDER.m - Camera-occluder probe не нашёл camera containment в пяти точках; paired mismatch локализован прежде всего в manual look direction/target/FOV calibration, с явным ограничением по RID-only geometry. - Empirical FOV sweep выявил, что checkpoint camera должна явно вызывать `make_current()`; после исправления projection ranking остаётся inconclusive из-за неизвестного manual yaw/pitch/framing reference. - Camera-pose sweep can now rank bounded yaw/pitch grids without changing manifest defaults; perceptual ranking remains diagnostic and requires human framing approval. -- Goldshire Inn pose evidence has no landmark-aligned optimum: full-frame error decreases as negative pitch replaces the missing inn composition with grass. The checkpoint remains a spatial/placement gap, not a validated camera offset. +- The first Goldshire pose grid was invalidated by WMO readiness: the inn is absent after 2 seconds but visible after the manifest-standard 8-second wait. Pose comparisons must stabilize asynchronous checkpoint content first. +- A ready-scene Goldshire 3x3 grid improved mean error from `0.101402` at zero offsets to `0.087952` at yaw `10`/pitch `-10`; pitch/FOV refinement remains required before manifest calibration. - D3D12 descriptor issues remain; the capture-path anonymous `RefCounted` shutdown leak is regression-covered by a clean verbose dry-run, while other RID/resource diagnostics still require independent evidence. - M2/WMO/material/particle/ribbon/portal parity incomplete. - Public API is mostly exported configuration rather than stable contracts. diff --git a/tools/sweep_render_checkpoint_camera_pose.ps1 b/tools/sweep_render_checkpoint_camera_pose.ps1 index 8ee2fd0..3a6bbf6 100644 --- a/tools/sweep_render_checkpoint_camera_pose.ps1 +++ b/tools/sweep_render_checkpoint_camera_pose.ps1 @@ -11,7 +11,7 @@ param( [double]$CameraFov = 62.0, [int]$ViewportWidth = 1280, [int]$ViewportHeight = 900, - [double]$WaitSeconds = 2.0, + [double]$WaitSeconds = 8.0, [double]$MeasureSeconds = 0.1, [switch]$PlanOnly ) @@ -121,6 +121,8 @@ try { checkpoint = $Checkpoint reference_directory = [IO.Path]::GetFullPath($ReferenceDirectory) viewport = @($ViewportWidth, $ViewportHeight) + wait_seconds = $WaitSeconds + measure_seconds = $MeasureSeconds candidates = $rankedCandidates } New-Item -ItemType Directory -Force -Path $absoluteOutput | Out-Null