Document public GP and DMA HLE regression gate
Clarify scope of scene and screenshot checks
Record GP and EP game isolation evidence
Record game diagnostic control and its evidence limitation
Record DSP bypass and compressed audio checks
Document experimental DSP bypass and measured PCM proof
docs: publish DSP native CPU memory and instruction comparisons Record the balanced Deck fixed-work efficiency result and actual hardware counters with output coverage. Separate isolated context memory from live fallback, batch cost from frame tails, and interrupted boot-resource diagnostics from gameplay speedup. Link all samples and review repairs.
docs: document concurrent DSP shadow proof and OpenAL evaluation Publish measured route coverage, fallback and injected-fault controls, exact artifact identities, review fixes and reproduction commands. Explain the limited OpenAL gain parity and missing game/state/performance qualification; preserve DSP authority and prior failed evidence.
docs: Showcase public DSP XISO and isolated native qualification Publish the verified guest GP/DMA, exact-program capture, 100-repeat replay and native arithmetic comparison. Preserve negative controls, interpreter timeout, artifact identities and the boundary before live game HLE.
docs: Document experimental DSP capture and replay tools Explain PR #384 availability, build and capture controls, headless comparison, seeded fuzzing, failure reduction, measured evidence and remaining issue #383 gates. Add audio/development/debugging navigation without promoting branch work to released features.
wiki: Record reviewed fix-and-benefit column publication Bind the publication record to the owner-requested concrete fix/benefit bullets and unchanged measured counters.
wiki: Restore concrete fixes and benefits beside counters Add a Fix / benefit column with classification and short bullet outcomes on all34 candidate rows and23 counter rows. Name supported game fixes including PGR2 menu-atlas corruption, Ghoulies depth/rendering corruption and the tested Def Jam freeze path. Distinguish measured CPU/cadence benefits from component work reduction, historical or mixed results and proposed HOLD repairs. Preserve every prior numeric and summary cell; retain detailed evidence and regression limits.
wiki: Bind publication receipt to fitted counter layout Retain source-checked metrics and link the reviewed numeric layout with desktop-visible p99 counters.
wiki: Keep numeric counter columns visible on desktop Wrap each direct before/after counter pair onto two lines so the p99 column fits within the live GitHub wiki content width. Preserve the exact source numbers and missing-value markers.
wiki: Record compact contribution-page publication Bind the publication receipt to the simplified shortlist, verified direct counters and linked detailed evidence requested by the owner.
wiki: Simplify upstream shortlist and expose numeric counters Shorten candidate summaries and move detailed review, dependencies and adverse evidence to a linked evidence page. Add direct process CPU, device GPU, source-reported FPS, guest cadence and p99 columns for 17 retained campaign rows plus all six Auto-four cohorts. Verify raw values against their owning retained PR descriptions and preserve report hashes, campaign scope and missing values. Keep FPS separate from guest cadence, historical and mixed results visibly marked, and all31 candidate areas plus3 open watchlist entries intact. No new runtime measurements.
wiki: Record approved contribution-section publication Bind the publication record to reviewed content 4e52778 and its complete PR screening and local validation. Publish the expanded section under the owner instruction to double-check other worthwhile PRs and then publish.
wiki: Expand and verify upstream contribution candidates Screen the complete 239-PR inventory and add omitted lifecycle, bounds, texture and presentation repairs, bounded work-removal candidates, and an explicit open-work watchlist. Preserve metric scope, adverse results, and current-source limitations. Verify upstream SDL cleanup and worker-exit gaps; distinguish historical LRU code no longer retained, closed overlapping proposals, and the already upstreamed unsigned R6G5B5 conversion. Retain a compact review receipt and prioritize five focused correctness handoffs.
wiki: Refresh publication identity for contribution shortlist Record the reviewed content and upstream reference without changing runtime or measured evidence. Agent-assisted: OpenAI Codex / GPT-6.
wiki: Add dedicated upstream contribution candidates Separate outgoing fork contributions from upstream PR tracking. Rank focused correctness repairs and performance candidates with PR summaries, before/after metrics, retained regressions, overlap routes and qualification gaps. Link the shortlist from Home, Development and a dedicated sidebar section. Agent-assisted: OpenAI Codex / GPT-6.
wiki: Record reviewed content and publication identity Keep emulator source and wiki content identities separate, and expose current-status and release links in the live footer. Agent-assisted: OpenAI Codex / GPT-6.
wiki: Reconcile current implementation and repair navigation Repair the audit's 153 malformed navigation rows and two fragment targets, plus an additional unescaped PowerShell pipe in the validation guide. Add current status, stable/main availability, complete post-snapshot change records, source-derived controls and balanced evidence summaries. Reconcile graphics, audio/CPU and upstream research against b7140848 while preserving dated investigations and failed/withdrawn evidence. Add an offline publication checker with regression coverage for table, target, fragment, metadata and record-drift failures. The wiki Git repository is canonical and publication remains separate from source documentation. Agent-assisted: OpenAI Codex / GPT-6.
Document Xbox hardware oddities with evidence labels and subsystem links
Updated Home (markdown)
docs: qualify Deck PGR2 movement and CI artifact identity
docs: record Deck PGR2 runner binding and race gate
docs: prioritize fork code for upstream contributions
docs: publish September 26 development status
docs: record exact PR 228 validation Document the exact-head shader and pipeline priority implementation, public Windows/Linux preview, and controlled cold/warm Windows ABBA measurements. Record the narrow supported claim: fewer synchronous shader and pipeline blocking intervals without a demonstrated general frame-tail gain. Preserve Morrowind snapshot output as a correctness gate after the tester reported strange behavior. Add runner lessons for never-executed saved-test sources, waiting-upload head-of-line blocking, and immutable shared snapshot carriers with private per-run overlays. Signed-off-by: Mainkill1 <Mainkill1@gmail.com>
docs: record learned readiness validation rules