Skip to content

M1 Operator Live Console: streaming AUX controls + device-verified named inputs - #236

Merged
masarray merged 15 commits into
mainfrom
feat/m1-live-console-workflow-20261031
Oct 31, 2026
Merged

masarray merged 15 commits into
mainfrom
feat/m1-live-console-workflow-20261031

Conversation

@masarray

@masarray masarray commented Oct 31, 2026 •

Copy link
Copy Markdown
Owner

M1 Integrated Live Console — one native realtime graph

Source: descendant of merged DEVICE2/SOURCE2 #234. No second audio callback, device manager, DSP graph, phantom input or separate route controller.

Real behavior

  1. Continuous SENDS-on-Faders: POST AUX gain dispatch while dragging (>0.25 dB, >=25ms), ACK-coalesced to at most one in-flight plus latest desired target per edge. MAIN fader remains unchanged; native DSP does <=64 sample ramp.
  2. Continuous AUX MASTER: drag input routed through same native atomic ACK channel, not structural APPLY.
  3. Named SOURCE profiles: searchable mono/stereo aliases (including non-adjacent L/R) scoped to actual JUCE backend + capture name + verified input pin/name mask; failed remap never creates phantom ports.
  4. Same-format input switch of EXISTING patched channel (MONO→MONO, STEREO→STEREO): GUI emits routingSourceSwitch; same JUCE graph validates current device epoch/revision and enabled pin IDs, then crossfades physical input PCM BEFORE the existing VST3 and send graph in 64 samples. Positive native arMixSourceSwitchAck is the source of UI LIVE state.
  5. Structural fallback remains: initial unpatched source, disconnect, channel format change, hardware I/O, FX topology, PRE/PAN routing must use guarded MUTE/APPLY and native ACK. No auto-unmute.

Verification

Truthful boundaries

Fixed 10 input +4 AUX, native per-channel VST3, MAIN/MONITOR/one extra destination. Unfinished full-product features (audio groups, DCA, Matrix, dynamic +channel, MASTER FX/PDC, complete CH1 legacy migration, Factory VST3) remain separate work. This PR is a significant M1 integration, NOT a claim of full Waves LV1 parity or passed physical FAT.

References: #170, #179, #15. Older PR #232 superseded by merged #234.

@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 2026-10-31T05:51:06.397304Z b195d5f PR opened
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.

@chatgpt-codex-connector chatgpt-codex-connector Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: b195d5f958

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".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Comment thread web/mixer-io.js
Comment on lines +483 to +485
if(cancel&&d.sent)sofStage(d.ch,'gain',+sofGain(d.startDb).toFixed(7));
else if(!cancel&&Math.abs(d.db-d.startDb)>0.005)
sofStage(d.ch,'gain',+sofGain(d.db).toFixed(7));

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 Badge Flush the final send target when a drag returns to its origin

If an operator moves a send fader, returns to its starting position within the 25 ms throttle window, and releases, sofMove skips sending the return movement and this condition skips the final write because d.db equals d.startDb. The previously transmitted intermediate gain therefore remains live despite the operator returning the fader to its original level. Reproducing with a drag from −20 dB to −10 dB and back leaves the native target at −10 dB. Once a drag has transmitted a target, flush its final value on release even when it matches the starting value; apply the same correction to the standalone copy.

Useful? React with 👍 / 👎.

Copy link
Copy Markdown
Owner Author

Cross-thread reconciliation / merge order: canonical main includes #234 DEVICE2+SOURCE2 at 7db4ab70 and docs #237 at c4bd931f. This PR owns continuous live AUX send/master control (bounded ACK coalescing) plus persistent device/pin-signature-scoped SOURCE aliases. #235 independently edits web/monitor.js for BUS/FX BUS, so do not blindly merge both controller rewrites. Recommended order: (1) certify #236 exact-head Chromium + native Windows/CTest and native ACK backpressure/alias remap negative cases, (2) merge #236 first if green, (3) rebase #235 atop updated main, reconcile only nonconflicting BUS + unified FX presenter in the same sole routing authority, and rerun exact-head gates. Keep structural MUTE/APPLY, one JUCE manager, CH1..10+4AUX, no fabricated CH11, #15 FAT open. UI run 41174095156 was green on an earlier head; need the latest head gate.

Copy link
Copy Markdown
Owner Author

Exact-head UI blocker discovered during reconciliation: at head 717d65fb0ed1e95886891b80a13db2d4e520f408, Chromium run 41174166478 FAILED tests/mixer_io_smoke.py:693: assert await page.locator('[data-source-choice="STEREO:4:7"]').count()==0. The user-defined SOURCE alias apparently remains selectable after the mock pin layout changes. Please audit deviceGeneration/active-pin-signature invalidation and DOM choice regeneration; a stale stereo alias must be hidden/disabled and never route missing hardware pins, not simply pass by weakening the test. Native Node contracts passed, but Playwright proof is still red. Preserve prior SOURCE2 and AUDIO I/O matrix; no second route controller. Rebase/validate over latest main@c4bd931f and avoid colliding with #235's web/monitor.js.

Copy link
Copy Markdown
Owner Author

Concurrent PR safety — #235 M2 FX/BUS integration: #235 modifies web/monitor.js (actual +BUS/+FX BUS native routing stage, epoch-fenced no-ghost group and parallel FX semantics) and tests/mixer_io_smoke.py in addition to one uniform CH1–10 FX presenter web/bridge.js. Your M1 live controls/aliases edit the SAME monitor and smoke files. Do not merge both stale-base PRs. Choose one merge order; then rebase/reconcile OTHER onto latest main and rerun exact-head Chromium and Windows; retain live POST AUX native control, source aliases, MUTE/APPLY/ACK, and new bus staging. #235 Chromium exact-head #41174540452 PASS; Windows running; hardware #15 remains unqualified.

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