docs(notes): archive today's two ear-signals before they evaporate

Written earlier today by the gear-repair session and left uncommitted, which
for this particular file is the one failure mode that matters: performance_notes
IS the archive, so an uncommitted entry is not a note, it is a memory.

Two signals, both about the rig rather than the audio:

- *"ok just played rose_rouge works great!"* — first confirmation that the
  CLASSIC LCXL (mk1, 1235:0061) is a fully viable performing surface once
  lcxl-leds-watch owns the paint. Worth recording precisely because the rig had
  drifted into treating the mk3 as the only real board; the classic is a
  fallback, not a downgrade.
- *"and the midi mon is beautiful!"* — on first sight of tools/bridge/midiviz.py.
  A taste signal about the rig's own UI, and the follow-up ask (always-open,
  always-on-top, every desktop) says it plainly: midiviz is an instrument of
  the rig, not a debug tool.

CLAUDE.md asks for every reaction captured durably because this corpus feeds
sampling, effect choices and the auto-mastering heuristics. Durably means
committed.

armada/tide-table/performance_notes.md

@@ -634,3 +634,33 @@ funk->piment transition! in this set i playd piment once, then funk, then next"*
 The machine had piment_bresilien and ouais_je_funk alternating in 45-80 s slices,
 which was a median-filter artifact, not a performance. A real set alternating two
 tracks four times would be unusual; treat that pattern as suspicious by default.
+
+---
+
+## 2026-09-05 — live check on the CLASSIC LCXL, after the surface repair
+
+Not a take: a live play-test on the rig, right after the gear repair session
+(zombie `lcxl3-driver` stopped, `lcxl-leds-watch` started, SuperDirt restarted
+for a stale sample dictionary, redundant direct LCXL→SC MIDI leg cut).
+
+**`rose_rouge` — plays great.** *"ok just played rose_rouge works great!"* First
+confirmation that the **original/classic LCXL (mk1, USB 1235:0061)** is a fully
+viable performing surface once `lcxl-leds-watch` owns the paint. Worth recording
+because the rig had drifted to treating the mk3 as the only real board — the
+classic is a working fallback, not a downgrade. Track loaded at the time:
+`live/midi/nova/nujazz/rose_rouge.tidal` (27 controls parsed by the painter).
+
+**The MIDI monitor lands.** *"and the midi mon is beautiful!"* — reaction to
+`tools/bridge/midiviz.py` (the frameless glyph-rain lens), seen for the first
+time. This is a taste signal about the RIG's UI, not about audio: the ambient
+visual read of the surface is wanted, and wanted *permanently* — the immediate
+follow-up ask was always-open, always-on-top, on every virtual desktop. Treat
+midiviz as a first-class instrument of the rig, not a debug tool.
+
+**What this says about the surface repair.** The board had been dark and its
+per-track colours absent because the mk3 driver held a stale binding to
+unplugged hardware and `Conflicts=`-killed the mk1 painter. The ear-level
+outcome — "works great" — only arrived after the *wiring* was right, which is
+the standing lesson: a taste complaint (or its absence) tracks a mechanical
+cause. See also: the double-CC path (`MIDIIn.connectAll` with an unfiltered
+`src`) that would have made toggle buttons silently no-op.