Skip to content

M2 Studio console: unified CH1–10 FX Rack + real BUS/FX BUS native routing workflow - #235

Merged
masarray merged 5 commits into
mainfrom
feat/m2-studio-console-unified-fx-and-bus-workflow-20261031
Oct 31, 2026
Merged

masarray merged 5 commits into
mainfrom
feat/m2-studio-console-unified-fx-and-bus-workflow-20261031

Conversation

@masarray

Copy link
Copy Markdown
Owner

M2: unified native console FX presentation + real subgroup/FX-bus workflow (Studio One routing semantics)

Base before editing: current main at branch creation a7fb61511ffe27fdef1f9923695daf6899646dff (merged #231, #233). Review fresh main before merge; #232 device PR independently owns hardware I/O and must not be overwritten.

Product intent: official Fender Studio Pro Effects Signal Routing: one insert/send model for Audio, Bus, FX channels, group output vs parallel FX send, no feedback loops. We implement workflow semantics rather than copying branding/art.

Implemented in this PR

  • web/bridge.js: one fxLaneSnapshot/ fxSlotMarkup/ renderRack presenter, same eight-insert workspace and same six-row compact FX strip for ALL existing CH1–10. No two CH1/routed HTML skins. Shows only real JUCE chain or accepted native routed VST3, correct bypass state, disabled when unpatched. Keeps old CH1 vendor/Orbit identity and routed leases separate as a temporary command adapter; this PR intentionally does not claim completed native engine unification.
  • Operator + BUS: from selected connected channel, use available native unpatched CH as stereo BUS receiver. Stage exclusive source output to subgroup (disable its MAIN, POST unity send to BUS, BUS → MAIN). + FX BUS: add parallel POST .25 send and preserve source direct MAIN. Both are REAL existing native graph edges, not fake UI; output remains unmodified until guarded MUTE → APPLY → accepted epoch/revision ACK. Existing native VST3 rack is usable on the BUS after actual routing.
  • Safety: reject stale epoch, dirty routing/FX draft, active/secondary output, no spare channel, unsupported source, send limit, route feedback, rolled back exact previous values on failed preflight. No audio-thread changes or another callback.
  • Updated regression contracts that incorrectly required two different CH1/CHN renderers; added shared presenter and routing wizard guards, plus focused handoff docs/M2_UNIFIED_CONSOLE_BUS_WORKFLOW_20261031.md and AGENTS precedence.

Explicitly NOT claimed

Live ADD CHANNEL/CH11+; separate arbitrary named BUS counts; complete all-channel VST3 instance ownership, dynamic ChannelId graph cutover, native MASTER FX, PDC, fully identical FX command semantics, physical #15 crackle closure, certified 0ms device latency or copied Studio One UI. Existing 10CH+4AUX native fixed graph is the only live source; this first integrated GUI + actual native BUS route milestone is a safe bridge to #192/#95.

QA / CI and hardware qualification

Copy link
Copy Markdown
Owner Author

Cross-thread reconciliation (2026-10-31): keep M2 isolated until it passes physical/software gates. Latest canonical main now includes merged DEVICE2+SOURCE2 integration #234 at 7db4ab70a4c971487e0b921df5a4e8cc805f0183, so re-check/rebase before merging M2. Your PR's latest tested head cc4a2d648cc02749f635f1a1b30b56dc954d3797 has UI run 41173956993 FAILED in tests/ui_smoke.py:1464: assert await native.locator('[data-strip="0"] .rack [data-plugin-picker="1"]').count()==6. Earlier head's Node tests had stale renderer assumptions; those were corrected, but the real Chromium smoke still indicates CH1 compact FX rack/picker visibility/count drift. Fix the DOM contract or correct the test only after checking the new single-renderer intended 6-row behavior. Verify CH1 legacy VST3 ADD/BYPASS and CHN routed FX are functional; preserve 10CH/4 AUX and no second engine. Do NOT merge with failing UI; re-run CI on one stable head, then Windows CTest and hardware #15 remain separate.

@masarray
masarray force-pushed the feat/m2-studio-console-unified-fx-and-bus-workflow-20261031 branch from 78e3103 to 4fa4e27 Compare October 31, 2026 05:54
…cs, unused-strip admission and no optimistic native APPLY

Copy link
Copy Markdown
Owner Author

Cross-thread integration gate 2026-10-31: concurrent #236 M1 Operator Live Console now touches web/monitor.js and tests/mixer_io_smoke.py, overlapping this M2 PR. Our current exact head 7c81fe132bcf52f53255a90f91590e14db1058ed passed full Chromium UI #41174540452, Windows #41174540488 running. DO NOT merge #235 and #236 independently against stale main. After the FIRST merge, the other MUST reconcile both overlapping files against latest main and pass exact-head native Windows + Chromium. #235 only stages real subgroup and FX BUS in unused current 10CH native lanes, unifies presentational FX Rack, leaves actual CH1 legacy vendor lease intact. No full ChannelId parity or physical #15 FAT claimed.

Copy link
Copy Markdown
Owner Author

Exact-head validation checkpoint: 7c81fe132bcf52f53255a90f91590e14db1058ed was rebased on current main c4bd931f9be671cd6a100c235c499a5d86036879, preserving the newest hardware/device/AGENTS changes. Full Chromium UI #41174540452 SUCCESS, including native FX editor, CH1 ADD/picker, 11-strip geometry, G1/G2 source/bus routing and added subgroup-vs-parallel-FX-bus draft test (no auto-APPLY). Windows #41174540488 currently IN PROGRESS at native CMake configure/build gate; no claim of native green or installer yet. #236 M1 touches the same web/monitor.js and tests/mixer_io_smoke.py; merge-order reconciliation is required if #236 lands first. Preserve original CH1 vendor/Orbit/Show while awaiting true ChannelId/native FX lease cutover #192/#95. Draft until Windows full MSVC/CTest/ZIP gate passes. Physical #15 remains open.

Copy link
Copy Markdown
Owner Author

Cross-thread integration gate (2026-10-31): M1 live SOURCE / continuously ACK-coalesced AUX controls are being implemented and exact-head CI-tested in #236 (source 259331438d2d). This M2 PR #235 also touches web/monitor.js and tests/mixer_io_smoke.py. Please do not merge either branch by replacing those shared files. Preferred order: merge #236 after exact-head native/GUI CI, then replay the unique M2 workflow changes onto new main and rerun CI, retaining routingSourceSwitch same-format 64-sample native source crossfade, one route authority, live sends backpressure, named sources, DEVICE2 and #15 safety. M2 cannot mark GROUP/DCA or VST parity complete from GUI alone; report exact native evidence. Main currently moved c4bd931 docs only, no audio conflict yet.

Copy link
Copy Markdown
Owner Author

SOURCE/AUX M1 #236 has now merged to main@0835806b89ecef650faa989a2545c52a0b2885aa after exact-head Windows+Chromium success. PR #235 branched before M1 and shares web/monitor.js and tests/mixer_io_smoke.py. MUST reconcile M2 branch against new main before merging. Do not overwrite native routingSourceSwitch, ACK/crossfade, live send/backpressure, named device-verified SOURCE or DEVICE2.

Copy link
Copy Markdown
Owner Author

M2B hardware FAT operator script — real ASIO/WASAPI, NOT certified yet (2026-10-31)

Source under test: 1097fc8eb4d78f842726230ff26e4ed4302295bf. Chromium workflow #41195235235 SUCCESS; Windows MSVC/CTest/package #41195235215 is RUNNING at this comment, so no Windows pass/release claim. This PR combines merged M1 named hardware SOURCE + continuous AUX POST controls with M2B real BUS, routed VST3 racks, and fixed 10CH+4AUX native graph.

FAT on physical Windows 10/11 x64 with WebView2 Runtime and installed sound-interface driver:

  1. Download only exact-head main Windows x64 CI artifact after merge and successful package verification. Unpack whole archive. Confirm BUILD_PROVENANCE.json:sourceCommit matches run head and exeSha256 matches EXE, not an old main build. Run with headphones at low volume.
  2. AUDIO DEVICE: choose real ASIO/WASAPI input/output, verify pins/rate/buffer. Keep OUTPUT MUTED. First 0FX + zero route; use SIGNAL CHECK OUTPUT_ONLY_TONE, then physical Mic 1→CH1 dry→MAIN and Mic 2→CH2. Verify only the chosen physical input is audible, no clip.
  3. FX Rack CH1: create + BUS. Native BUS should be allocated on a spare existing lane (typically CH10), CH1 direct MAIN disabled, POST unity send to BUS→MAIN; it must remain DRAFT/silent until MUTE → APPLY → accepted native ACK. Unmute manually and verify single dry path, not doubled sound. The GUI must reveal BUS identity.
  4. From CH2's FX Rack select SEND TO EXISTING BUS → OUTPUT → BUS (the same BUS). Again MUTE/APPLY/ACK and manually unmute. Verify both mics audibly sum, BUS fader controls both. Confirm a second BUS wasn't consumed and CH1/CH2 insert names/slots remain stable.
  5. Parallel FX test: prepare a separate FX BUS if a spare channel exists; CH2 FX SEND → BUS should retain dry MAIN plus .25 linear POST send. Add a licensed, manually installed VST3 (factory approval lock is still EMPTY) to the FX BUS, confirm editor/bypass/reorder, no phantom LIVE lamp before ACK.
  6. Rejection probes while output muted/armed: duplicate edge CH1→BUS, BUS→same BUS, BUS cycle, no free lane, changing device epoch, prior dirty draft, output armed. A failure must NOT alter accepted sound or device; no implicit unmute. Confirm Show save/recall compatibility and that CH1 vendor Orbit still works.
  7. Reliability: baseline 0FX first at 44.1/48 kHz and 64/128/256 buffer; observe time to live sound, peak meters, CPU, dropouts, audible clicks during guarded structural change and live gain/bypass, 30-min idle/playback. Only lower buffer (32) if your driver supports. Save audio/device/buffer, video/screenshots, exact BUILD_PROVENANCE, failures and whether crackle P0 — Crackling with 0 VST3: diagnose WASAPI clock/input/output buffering before engine expansion #15 occurs.

Not yet claimed: true unbounded CH11+, all-channel VST instance-owner migration, MASTER VST/PDC, zero physical roundtrip latency, full Studio One pixel parity, professionally certified hardware, or #15 crackle resolved. These require subsequent engine and physical work.

@masarray
masarray marked this pull request as ready for review October 31, 2026 11:11
@masarray
masarray merged commit 54e4229 into main Oct 31, 2026
2 checks passed
@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Oct 31, 2026 •

Copy link
Copy Markdown

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review Status Commit Review trigger
📝 Code Review ✅ Completed 2026-10-31T11:12:40.777311Z 1097fc8 Draft marked ready
ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

Copy link
Copy Markdown
Owner Author

RELEASE/PHYSICAL FAT HANDOFF — M2B Windows artifact now AVAILABLE (31 Oct 2026)

Merge commit 54e4229bd3e0d1038aad39b5839116bedaf18515 on main has both exact-head main CI PASS:

  • Chromium UI run #41195859419: SUCCESS.
  • Windows Native VST3 Host run #41195859441: SUCCESS, including CMake MSVC compile, native CTests, Windows GUI packaging and independent ZIP/provenance checks.
  • Download actual compiled candidate: Windows-x64 FAT artifact #13041385252, 4,405,014 compressed bytes at upload. Extract full Actions artifact, confirm BUILD_PROVENANCE.json sourceCommit and SHA256 EXE match; run with WebView2 Runtime and real ASIO/WASAPI driver.

The workflow's legacy artifact NAME contains Verified-Hardware-FAT; this means only the automated packaging verification passed. It does NOT mean a human ran a real external hardware FAT. Follow the preceding physical script and #15 0FX crackle gate. FactoryFX approval lock is EMPTY: this ZIP does not silently contain unapproved sibling plugins. Formal Pro Studio parity still needs stable dynamic ChannelId, unified vendor instance owner on CH1, PDC/native MASTER FX, and actual low-latency measured testing.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant