Skip to content

P1A: deterministic native 10-strip source routing & MASTER DSP foundation (no live activation) - #108

Closed
masarray wants to merge 7 commits into
mainfrom
feature/p1-static-multichannel-routing-core-20261028
Closed

masarray wants to merge 7 commits into
mainfrom
feature/p1-static-multichannel-routing-core-20261028

Conversation

@masarray

Copy link
Copy Markdown
Owner

Milestone

Implements a real, device-independent routing primitive, not cosmetic FX mockups: explicit 10 input strips plus MASTER, physical mono/stereo source indices, channel gain/balance, solo/mute, and MASTER mono/gain/pan. All topology decisions are validated/prepared off RT. No per-sample heap/locks/trig, and no fabricated channels.

Concrete files

  • engine/core/MixerRoutingPlan.h — compile-on-control-thread route validation; extract real physical channels into preallocated stereo scratch; separate post-FX accumulate into master; master output controls.
  • tests/mixer_routing_plan_test.cpp — deterministic 10-source audio vectors, stereo isolation, mono panning, mute/solo, invalid route rejection and rollback.
  • Existing CMake Windows CTest target ARMixMixerRoutingTests included in existing workflow (no new CI workflow).
  • docs/P1A_REAL_MIXER_ROUTING_CORE_20261028.md — explicit end-to-end integration/RT/hardware gates.

Deliberate safety boundary

Not wired to the JUCE live callback or FX GUI yet. The main host still has one native VST3 chain. No unverified audio graph activation or new device calls; do not suggest CH2–10 FX are functioning. Existing #15 zero-plugin physical crackle is a blocking hardware qualification before activating P2.6. No unrelated PR #102/#106 UI/Orbit code has been copied or merged.

Acceptance

New CTest and existing Windows build must pass; no stale/dirty branch overwrite. No claim of a hardware clean audio or full multi-channel FX release from this milestone. Refs #95 and #15.

Copy link
Copy Markdown
Owner Author

Integration note: tests exercise real source buffers and master arithmetic, but this code is deliberately NOT connected to JUCE audio callbacks or VST3 processing. A CTest success cannot claim that CH2–10 or MASTER already process audible FX. Maintain 0-VST hardware gate #15 before activating an RT multichannel graph. Plan swap must honor LiveGraphHandoff retirement.

@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Oct 27, 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 2026-10-27T23:21:25.969496Z 465f5f7 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

Cross-thread reconciliation · 2026-10-28 · main c0d7c794a27fcdc7adbe77c6166a2e97b848b82c

Since this P1A branch was created from 20222c6, main added MERGED PR #107 (sample-accurate FX ON/OFF de-click, per-physical-slot envelope) while preserving PR #106 meter ballistics and PR #105 dearVR Orbit/POLAR. This PR is intentionally a pure MixerRoutingPlan foundation, not live CH2–10 DSP, so it can remain draft until qualification. Before integration, reconcile CMakeLists.txt and existing Windows workflow without dropping ARMixFxBypassEnvelopeTests / P0FXBypassSampleAccurateDeClick; do not overwrite src/LiveAudioEngine.h or GUI. Require exact-head Windows Release+CTest after rebase on current main. Hardware issue #15 (no-FX WASAPI crackle) remains open; successful synthetic routing tests do NOT qualify live multichannel sound. Integration sequence: #108 → #110 → #111; no stacked-child merge ahead of parent. Avoid starting a second CI workflow.

@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: 465f5f78fd

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 on lines +14 to +15
assert(Plan::compile(cfg,2,plan,err));
assert(plan.activeStrips==0);

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

P1 Badge Keep routing checks executable in Release builds

The inspected .github/workflows/p1-windows.yml builds this target with --config Release and runs CTest with -C Release, where NDEBUG disables every assert in this test. This removes not only all result checks but also the Plan::compile and extract calls embedded inside assertions, so the new routing acceptance test passes without exercising the configured routes. Preprocessing with -DNDEBUG confirms those calls disappear. Use always-on checks that return a nonzero exit status on failure so the required Windows native test gate actually validates this DSP foundation.

AGENTS.md reference: AGENTS.md:L26-L26

Useful? React with 👍 / 👎.

Copy link
Copy Markdown
Owner Author

Canonical next-thread handoff saved in docs-only PR #113: https://github.com/masarray/arsonmix/blob/docs/handoff-arsonmix-routing-fx-20261028/docs/HANDOFF_ARSONMIX_FX_ROUTING_20261028.md. Windows/CTest and GUI CI for #108 are PASS; however main advanced through #107 and this PR is now merge-conflicted. Preserve baseline; don't auto-merge/force rebase. New thread should port focused P1A core + tests onto current main and verify the new build; do not claim CH2–10 audio are active.

Copy link
Copy Markdown
Owner Author

P1A source and CTest are ALREADY integrated into main by P1D merge d30cf35dc74991286547765ec9927e356273785b. Older #108 is dirty against concurrent #107/#101 and should not be merged separately. Latest main also includes P1B, P1C and executable offline P1D graph; source/plan boundary remains OFFLINE (not native CH2–10). Subsequent source extraction/routing correctness fix is tracked by PR #118. Close this historical stacked PR without deleting its branch, retaining lineage.

@masarray masarray closed this Oct 27, 2026
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