[Quality][Hardware] Run strict RTX 4080 Windows/Vulkan qualification

[proggeramlug]

Parent: #128
Validation branch/PR: #147
Related hardware/release gate: #140

Objective

Run the complete nine-case Bloom renderer qualification on a quiet Windows x64 machine with an NVIDIA GeForce RTX 4080, using Vulkan and hardware ray-query GI. Return the complete evidence bundle whether the strict run passes or fails. Do not change thresholds or use report-only mode.

Exact source

• Branch: codex/rendering-wip-integrated-20260727
• Commit: 4508eaff082b849203ac60f9d8f3a327b71d95c9

git fetch origin
git switch --detach 4508eaff082b849203ac60f9d8f3a327b71d95c9
git submodule update --init --recursive
git status --short

The final command should print no tracked changes. The qualification runner rejects the wrong host OS, forces Vulkan, opts into hardware GI, records the exact adapter/backend, and applies the governed RTX 4080 Vulkan budgets.

Host prerequisites

• Windows 10 or 11 x64 and NVIDIA GeForce RTX 4080
• Current NVIDIA driver with Vulkan support
• Git, Python 3, Rust stable with the MSVC target, Node/npm, CMake
• Visual Studio 2022 Build Tools with Desktop development with C++ and a Windows SDK
• Perry compiler from https://github.com/PerryTS/perry available as perry on PATH
• Close games, builds, indexing, and other sustained workloads before capture

Confirm the tools:

python --version
cargo --version
node --version
npm --version
perry --version

Pinned Bistro asset

If examples/bistro/assets is absent:

git clone https://github.com/zeux/niagara_bistro examples/bistro/assets
git -C examples/bistro/assets checkout --detach 7c9f9f9ac0915024ccf3dddbccd8bfc643a42607
Copy-Item -Force examples/renderer-test/assets/outdoor.hdr examples/bistro/assets/outdoor.hdr

Then validate all nine assets and approved baselines:

python tools/quality/run.py check

Do not start the measurement unless this reports PASS.

Strict measurement

New-Item -ItemType Directory -Force tools/quality/out | Out-Null

python tools/quality/run.py run full `
  --machine-class nvidia-rtx4080-windows-vulkan `
  --host-idle-timeout 600 `
  --out tools/quality/out/windows-rtx4080-vulkan `
  2>&1 | Tee-Object -FilePath tools/quality/out/windows-rtx4080-vulkan-console.log
$runExit = $LASTEXITCODE
Write-Host "quality exit code: $runExit"

Exit code 3 means the machine never became sufficiently idle; retry later without modifying the budgets. Any other nonzero result is still useful and must be returned with its evidence.

Return bundle

Compress-Archive -Force `
  -Path tools/quality/out/windows-rtx4080-vulkan,tools/quality/out/windows-rtx4080-vulkan-console.log `
  -DestinationPath tools/quality/out/windows-rtx4080-vulkan.zip
Get-FileHash tools/quality/out/windows-rtx4080-vulkan.zip -Algorithm SHA256

Attach the ZIP to this issue or provide a durable download link. Also paste the exit code, ZIP SHA-256, Windows version, NVIDIA driver version, and whether any unrelated process started during capture.

Acceptance checklist

• Exact commit checked out and tracked worktree clean
• python tools/quality/run.py check passes
• Strict full command run without --report-only
• result.json records Windows, Vulkan, and NVIDIA GeForce RTX 4080
• Required ray-query cases report the hardware path or a clear strict failure
• Host preflight and all per-case postflight records are present
• Complete output ZIP and SHA-256 returned
• Result and any failures summarized on [Quality][P0] Add deterministic reference-scene visual and GPU performance qualification #128

A strict failure is not permission to update baselines, budgets, or noise bounds. Return the evidence first so the renderer can be fixed and the same command rerun.

[proggeramlug]

Automated handoff verification at the pinned SHA:

• Windows hosted build: PASS — https://github.com/Bloom-Engine/engine/actions/runs/34357267801/job/102485008262
• Windows shared tests: PASS — https://github.com/Bloom-Engine/engine/actions/runs/34357267801/job/102485007753
• Renderer Quality workflow: PASS — https://github.com/Bloom-Engine/engine/actions/runs/34357267858

The aggregate Tests workflow is red only in pre-existing non-Windows contract/mobile jobs: the package check sees Jolt submodule files omitted by hosted checkout, and Android/iOS/tvOS/visionOS lockfiles are stale under --locked. FFI parity itself reports 0 failures and 0 warnings. These do not block the Windows RTX 4080 pickup command in this ticket.

[proggeramlug]

Completed the Windows qualification on the available AMD Radeon 760M, following the request to use this machine because an RTX 4080 is unavailable.

• Two complete strict nine-case Vulkan runs: 7 pass / 2 fail, exit 1 each; no --report-only.
• Required Sponza/Bistro cases report native hardware ray-query GI. All captures have the requested pixel dimensions, telemetry, intermediate images, and accepted preflight/postflight CPU records.
• Reproducibility: PASS, exit 0, all 257 compared artifacts byte-identical.
• All five seeded negative controls detected their intended regressions.
• Remaining visual failures: Sponza SSIM 0.972333014 against 0.975; skinned/alpha SSIM 0.931997895 against 0.970 and luminance RMSE 0.047519531 against 0.040.
• The exact renderer revision used to create the approved baselines reproduces both failing images byte-for-byte on this Windows/GPU/compiler configuration. Software GI, bound material bindings, and DirectX controls do not remove the discrepancy. Reduced anisotropy worsened results and was reverted.

Windows execution fixes cover relative executable resolution, missing Vulkan HINSTANCE, RGBA/BGRA screenshot export, native CPU load sampling, and DPI-independent headless capture. The Radeon profile measures performance without borrowing RTX/Apple budgets. No approved images, thresholds, existing budgets, or reproducibility bounds were changed.

Host: Windows 11 Pro x64 10.0.26200; AMD Radeon 760M Graphics; AMD Vulkan driver 24.12.1 (LLPC), Windows driver 32.0.12033.1030. CPU load snapshots passed, but unrelated process starts and background GPU activity were not continuously monitored.

Measured source: 8309985bf9f5358c38e5dd2cbc0e328d2bccb6f8, clean in both final runs, based on requested revision 4508eaff082b849203ac60f9d8f3a327b71d95c9 plus necessary Windows execution fixes.

Measurement report | Portability investigation | Complete evidence ZIP

ZIP SHA-256: aa099676ecd801ebb7ed586ddb9f9732b8f913f13a6d926604ad193b078d82a4

The original RTX 4080 certification is still pending that hardware. These results provide a repeatable local Radeon qualification and preserve the two strict failures for review.

Draft Windows fixes PR #154. The uploaded ZIP digest was verified against GitHub asset metadata.

[proggeramlug]

The Windows image discrepancy is fixed in draft PR #158. Scene and shadow shaders now compute the approved cutout phase with integer arithmetic, preserving existing baselines and thresholds. Exact Metal/Windows cutout-input captures identified the previous phase-extent difference.

On source 662a44f, both complete Radeon 760M/Vulkan runs pass all nine image gates, and 257/257 compared artifacts are byte-identical. The two canonical Metal images and all 24 hosted checks also pass. The GPU regression passes on Vulkan, DX12, and Metal and rejects the old Vulkan shader.

The full runs still report FAIL where postflight host load invalidated timing: Bistro in run 1; Bistro, draw/light stress, and weighted transparency in run 2. Reproducibility does not override those rejections. Named RTX 4080/Apple M1 Max qualification and the remaining engine plan stay open.

Report and evidence ZIP, including raw inputs, source, commands, host snapshots, controls, and independent checksums. ZIP SHA-256: 0cfe31d256cece617e15addba3ffd6d35039f04aec3831021e65283b924d7dfb.

[proggeramlug]

The next renderer fix is in draft #159: correct SSGI depth receivers and normal reconstruction. The original 16-frame HD startup limits now pass on Radeon/Vulkan, verified Radeon/DX12, and hosted Metal; the new analytic-plane regression rejects either old calculation.

At d610d6a1ad4ab0ea9c7768c1a30acd2cce3fc2e4, all nine Radeon images pass in three complete runs. Runs 2 and 3 pass all configured checks, including host load, and reproduce all 257 compared artifacts byte-for-byte. Their metadata and timing differences also meet the unchanged reproducibility limits. Run 1 retains its background-load failures. RTX-specific budgets and named physical-hardware acceptance remain unperformed.

Report, captures, logs, source, and checksum receipts are published. ZIP SHA-256: 0ea0d9ff4769c7ee9eaed2ed977bf2205f6eda52d328120819743037a5f5645a.

Next-work audit found seven canonical Windows link/compile failures (six outdated palette uses and one unavailable Perry UI function). It also found that hosted Windows CI reports green without Cargo output because its Bash entry point is invoked through PowerShell without an explicit Bash shell. The published local Windows tests were executed directly and remain valid; the hosted invocation and example matrix need correction. The broader engine plan remains open and no PR is merged.

[proggeramlug]

Windows build/example follow-up is published in draft PRs #160 and #161. Windows CI now actually executes Bash/Cargo and requires execution summaries; the corrected hosted native build and all 20 canonical native example links pass. Exact local source 59244b9 also passes 20/20. The pinned matching Perry source build resolves the release-library and per-app profile link failures; all failed profiles remain in the evidence.

The Windows shared suite is still failing: the latest hosted run finishes with 486 passed, two DX12 GPU failures, and one ignored test. Both failures reproduce on the Radeon 760M and Windows' software adapter; Vulkan passes. The preceding access violation remains unresolved. These findings are under separate investigation, and startup/clean-install qualification remains next.

Report and evidence ZIP. ZIP: 9,940,053 bytes; SHA-256 2ebcb258a72e473723331dcd9fbaad1781b36d0ea70b0e95a3bcfd90fcf0a9b9. All 287 payload hashes and archive CRCs were verified, along with the uploaded asset digest and tag source. No unavailable RTX qualification or completed broader engine acceptance is claimed.

[proggeramlug]

Draft #162 and its published report/evidence preserve DX12 indirect draw identities through the existing bounded fallback and restore the visibility reconstruction GPU oracle without changing its pixel limits. The original two oracles pass in five Radeon/WARP backend/compiler configurations; the new production virtual-visibility check passes on Radeon Vulkan, Radeon DX12 and WARP. The complete local Vulkan shared CI component passes at ab8019b.

Hosted Windows native building and all 20 example links pass again. The Windows shared process still crashes with STATUS_ACCESS_VIOLATION (exit 139) before completing the suite; this remains unresolved. The expanded local DX12 golden suite also exposes a camera-cut equality failure that reproduces on unchanged parent 59244b9. Camera-history diagnosis and the hosted crash are being handled separately; neither failure is waived.

ZIP: 2,006,103 bytes, 201 verified payloads. SHA-256 ff50b4653062f2070508b825fc47f032e5907d5a054285e193ff33d2cb74975e. The uploaded digest and exact source tag were verified. No full platform/hardware acceptance or merge is claimed.

[proggeramlug]

Published the camera-history reset evidence and draft PR #163, stacked on #162.

The unchanged shader leaves visible residue after 40 old-history frames (about 16,900 pixels, maximum channel error 232). Guarding old color/confidence reads with the existing history predicate restores exact fresh/reset equality at both eight and 40 frames under Radeon Vulkan, DX12/FXC and DX12/DXC. All six captured RGBA pairs match byte for byte; thresholds and baselines are unchanged.

The complete local DX12 and Vulkan shared suites pass, including all 93 goldens. Hosted macOS passes the same expanded camera-cut check and 93 goldens; real Windows native builds and all 20 example links pass. Hosted Windows shared tests still crash with 0xc0000005 before goldens. A local WARP shader-optimizer crash is now under investigation; #163 does not claim to fix it. Named RTX/M1 Max full-quality jobs remain skipped.

Report and evidence ZIP — camera-history-c643b85-windows-metal-20260911-v1.zip (3,052,783 bytes), SHA-256 c03b55eafd7c8d2cf17c08c958440395d4aa65fd37f9beeef1241a53e247d4d7. Every payload hash and ZIP CRC was verified, and GitHub asset digests match. The archive includes original failures and exact source/hosted-tree receipts. No merge or unavailable-hardware result is claimed.

[proggeramlug]

Draft PR #164 and its evidence report are published. Hosted Tests run 34552411807 passes all 22 jobs, including actual Windows shared execution, native engine compilation and all 20 example links.

The Windows lane now explicitly uses DX12/FXC. Original WARP/DXIL controls reproduce an access violation during concurrent traversal shader compilation; the debugger stack and failed controls are retained. All ten hierarchy tests pass serially with DXC, while both four-thread DXC controls crash. FXC passes the concurrent controls. The underlying WARP/DXIL defect remains open.

Complete local Radeon DX12/DXC and Vulkan shared components pass all 93 goldens and remaining integration checks. Hosted Windows goldens skip the CPU adapter: their nominal 93 "ok" results do not qualify physical images. The separate physical Radeon FXC run still has four layered-material pipeline compilation failures (X3570/X3511); an explicit-LOD correction is being validated without changing assertions or baselines.

Archive warp-compiler-f96af2e-local-hosted-20260911-v1.zip (1,980,437 bytes), SHA-256 0c181559edb073296bd74367c865f457b067bdfafc0436dbda48c25b32adf407. ZIP CRC, every payload hash and both GitHub asset digests are verified. Source and hosted merge trees match. Native/browser starter startup, installed-package usability, wider engine requirements and named hardware acceptance remain open; no PR is merged.

[proggeramlug]

Draft PR #165 and its evidence report are published. The sheen lookup now samples explicit level zero, fixing four DX12/FXC material pipeline compilation failures (X3570/X3511) without changing rendering assertions or baselines.

The complete local FXC shared suite passes, including all 93 goldens; DX12/DXC and Vulkan each pass all 93 goldens too. Hosted Tests run 34555112391 passes all 22 jobs. Windows library tests pass 489 with one ignored; its goldens skip the CPU adapter and are not physical-image qualification. Hosted Metal supplies additional rendering evidence. The native engine build and all 20 example links pass.

Archive fxc-layered-5d99fdd-local-hosted-20260911-v1.zip (1,847,832 bytes), SHA-256 8fc1a6af07b975a14ecc4d3b09b35c3db1967038b590d8ebca869528a9810b7b. Every payload hash, ZIP CRC and both GitHub asset digests are verified. The committed shader matches the tested candidate; the hosted merge and PR head have identical trees.

The WARP/DXIL concurrency defect remains open. Starter rendering, installed native startup, broader engine/performance acceptance, unavailable hardware qualification and draft review remain separate work. No PR is merged.

[proggeramlug]

Draft #166 and its evidence report are published. The installed web command now runs through Node on Windows, preserves compiler failures and requires fresh build outputs. The original Bash-shim failure and initial hosted directory-copy failure are retained.

The final asynchronous-copy implementation passes all nine command regressions and a fresh installed Perry-plus-engine WASM build locally. Hosted Tests run 34556841600 passes all 22 jobs, including the actual Windows pack/install/help check on Node 22.23.2, native engine build and all 20 example links. PR head and hosted merge checkout have identical Git trees.

Archive portable-web-f30ebd0-local-hosted-20260911-v1.zip (1,805,739 bytes), SHA-256 f0d74f30f4ff5c04b9a3a4c5e64d33c8b8f5765f264a9cd7fed34328ef7a54a2. All payload hashes, ZIP CRC and both uploaded asset digests are verified.

This is installed command/build evidence. Compiled-game browser rendering, the full starter and shared lifecycle remain open; native installed startup is covered separately by draft #167. Stable npm 0.4.16 still has the old entry point. No package is published to npm and no PR is merged. Radeon work continues; RTX-specific qualification remains unperformed.

[proggeramlug]

Draft #167 and its evidence report are published. The installed Windows package now finds Jolt's published win32-x64 archives and omits redundant final-link metadata that pointed to an absent CMake directory.

A fresh actual package installs, links, runs a falling Jolt body and captures all 16,384 expected pixels on Radeon DX12 and Vulkan. Both PNGs are byte-identical. Hosted Tests run 34557843801 passes all 22 jobs, including the fresh installed native physics/image check on Windows, native engine build and all 20 example links. Hosted source trees match the PR head. The hosted startup uses Microsoft Basic Render Driver; it is not physical-GPU performance evidence.

Archive installed-native-73b1e42-local-hosted-20260911-v1.zip (2,363,657 bytes), SHA-256 2b53e68cf459bb5d8c6efb74384c0791f11795e43e97fba36ac3cf7f808b8f0a. Every payload hash, ZIP CRC and both uploaded asset digests are verified. Earlier path, harness, link and missing-capture failures remain explicitly distinguished in the report.

The acceptance is native headless startup using the qualified Perry 0.5.1220 runtime profile. Visible presentation, packaged DXC, general long paths and the complete native/browser starter remain open. Direct-2D capture is handled in draft #168. No PR is merged or npm package released; RTX-specific qualification remains unperformed.

[proggeramlug]

Draft #168 and its evidence report are published. Direct native frames now service queued output captures using the existing readback implementation. Render-target-only frames keep the request pending; scene diagnostic captures wait for a scene frame.

A fresh package at dc77ba5 renders the exact Jolt physics fixture through scene and direct-2D paths on Radeon DX12 and Vulkan. All four PNGs match all 16,384 expected pixels and are byte-identical. Both backends also pass all five render-target tests and all 93 goldens with original assertions. The initial WASM module-guard error is corrected and its failed/passing checks are retained.

Hosted Tests run 34559712338 passes all 22 jobs, including actual installed scene/direct-2D captures on Microsoft Basic Render Driver, the native build and all 20 example links. The hosted merge checkout and PR head have identical Git trees. Windows goldens skip the CPU adapter; its installed captures are software-rendering startup evidence, not physical-GPU performance results.

Archive direct-capture-dc77ba5-local-hosted-20260911-v1.zip (2,770,207 bytes), SHA-256 f0f2a03b41d1f926c900a9901831f2a96a5311cab55f22c63e60f0b7f542dd34. All payload hashes, ZIP CRC and both uploaded asset digests are verified.

Visible presentation, browser startup/capture, packaged DXC and full starter/lifecycle acceptance remain separate. Draft #169 advances shared cleanup and canonical example startup. No PR is merged or npm package released; RTX-specific qualification remains unperformed.

[proggeramlug]

Draft #169 and its evidence report are published. runGame(updateAndDraw, cleanup) now supports final cleanup on native and web. Pong uses that shared loop and edge-triggered pause input. All 51 invalid palette references in six examples are corrected and checked against the exported palette.

A fresh installed package simulates Jolt, produces all 16,384 expected pixels and records exactly one cleanup in scene/direct-2D modes on Radeon DX12 and Vulkan. A real callback-only native compatibility check passes. Compiled Pong records RPRPRPRPR with the old held-key pause check and RPPPPPPPP with the correction. All 20 native examples link; corrected Pong and the engine complete the web build. The initial palette crash and intermediate input-staging audit mistake remain in the archive.

Hosted Tests run 34561587869 passes all 22 jobs. Both installed Windows rendering modes produce the exact image and one cleanup on Microsoft Basic Render Driver. The merge checkout and PR head have identical Git trees. Windows goldens skip the CPU adapter; installed captures prove software-rendered startup, not physical-GPU performance.

Archive starter-lifecycle-b43ea2c-local-hosted-20260911-v1.zip (4,558,218 bytes), SHA-256 db70fbe196446d3648c238c5e770ceaeabcb8b28282e8012f08c3cdf447e928d. All payload hashes, ZIP CRC and uploaded asset digests are verified.

Draft #170 adds actual Perry-compiled browser acceptance; its result is still pending and its separate Windows shared-test job has an access violation despite the FXC setting. This publication reports #169's exact successful run, not a resolution of that continuing instability. Browser rendering, visible native presentation, packaged DXC, fixed updates and the complete starter remain open. No PR is merged or npm package released; RTX-specific qualification remains unperformed.

[proggeramlug]

Draft #170 now passes all 22 hosted Tests jobs at 9e6f4ab82f50db97d66385adf4c67f4e08fdae07; its report and evidence are published.

The real Perry-compiled game renders eight frames, records one cleanup and produces all 16,384 expected pixels in hosted macOS Chrome. A compiled FFI call with an explicitly injected startup error is correctly rejected. The PR head and hosted checkout have identical Git trees. Physics is omitted from this rendering fixture; starter assets/text, browser physics and the full example runtime matrix still need acceptance.

The archive retains the initial compiled-game timeout caused by unresolved imports despite a successful compiler exit, plus the original Windows shared-test access violation despite FXC. The corrected preparer verifies exact dependency resolution and actual WASM imports. The serial Windows library passes 489 tests with one existing ignored test; all 20 native examples link and both fresh installed Windows rendering modes produce the exact frame and one cleanup on Microsoft Basic Render Driver. Windows raster goldens skip the CPU adapter. The serial setting is a qualified CI mitigation, not an identified crash root cause. General Perry throw propagation remains a separately observed limitation.

Archive compiled-web-9e6f4ab-local-hosted-20260911-v1.zip (3,943,175 bytes), SHA-256 d8989d608b8fb5d790fdd1088bdfedc662b511e56a10163e58d6c495551dd507. ZIP CRC, every included payload hash and uploaded asset digests are verified. No PR is merged or npm package released; RTX-specific qualification remains unperformed.