bcox/mini-800
1
0
Fork 0
Code Issues 36 Pull requests Projects Releases Packages Wiki Activity Actions

A new instrument: Korg's MS-10 and MS-20 #25

New issue
Closed
opened 2026-10-12 03:07:39 +00:00 by bcox · 4 comments
bcox commented 2026-10-12 03:07:39 +00:00
Owner
Copy link

Korg's MS synthesizers as a new instrument: the MS-10 and the MS-20, and perhaps the MS-50. Not started.

The MS-2000, which the owner asked about with a question mark, is a different thing: a digital virtual analog of 2000 that has the name and little else. Four voices of two modelled oscillators, a modelled filter in several modes, sequences of modulation, a vocoder and digital effects. On 2026-10-11 the owner said fully digital synths were out when proposing instruments. The owner decided on 2026-10-11 that it is left out, as an instrument and as a source of features for this one: it is a general virtual analog more than a virtual MS-20.

Everything below is from recollection, with nothing read. Find and read Korg's owner's and service manuals before designing anything, and put them in docs/references.md.

The instruments, as recalled

  • MS-10 (1978). One voice of 32 keys: one VCO (triangle, sawtooth, a pulse of set or modulated width, noise) at 32' to 4', a two-pole low-pass with a resonance that self-oscillates, one envelope with a hold before its attack, one modulation generator whose shape is a knob (sawtooth through triangle to reverse sawtooth, and a pulse of matching width), portamento, and a small patch panel.
  • MS-20 (1978). One voice of 37 keys: two VCOs, the first a triangle, sawtooth, pulse or noise and the second a sawtooth, square, narrow pulse or a ring modulation of the two. A high-pass and then a low-pass, each with a resonance ("peak") that self-oscillates and distorts as it does, which is the instrument's sound. Two envelopes: one of delay, attack and release, usually to the pitch, and one of hold, attack, decay, sustain and release. A modulation generator like the MS-10's. A patch panel with a sample and hold, an extra amplifier, white and pink noise, a wheel and a button, and the external signal processor: a preamplifier, a band-pass, an envelope follower and a pitch-to-voltage converter, so that outside sound can play it.
  • Its filter changed during its run: an early one built round Korg's own chip and a later one round an amplifier chip, which sound different.
  • MS-50: the same parts as a module with no keyboard and everything on jacks.

To settle with the owner first

  • One instrument, most likely: the MS-10 is an MS-20 with one VCO, no high-pass and one envelope, as a SEM is an OB-X with things at nothing.
  • The patch panel, which goes with #7 as the ARP 2600's (#22) and the PS's do. Both play with no cables in.
  • More than one voice, as the SH-101 was given.
  • Its name, and its programs' names. mini10 and mini20 say little, as mini1 would have.

What it needs that is not here yet

  • The two filters and how they overdrive. Two poles each, in a row, is arrangement, and engine/state_variable_filter.h is two poles; but that one stops short of self-oscillation and is linear, and here the oscillation and what limits it are most of the character. The slopes want the service manual. The two revisions are a switch by the owner's rule of a switch per quirk, and the rest is tone, which waits on measurement (#4).
  • An LFO whose shape is a knob. Goes with #8.
  • An envelope with a hold, and one of delay, attack and release. engine/envelope.h has a delay stage now.
  • The external signal processor: #17 for the outside sound, then an envelope follower and a pitch follower, which nothing here has.
  • A ring modulator in the voice, which the SH-5 has.

What is already here to build on

  • instruments/sh5/ and instruments/sh101/, the other one-voice instruments: engine/analog_oscillator.h, engine/noise.h with pink, the SH-5's sample and hold, engine/portamento.h, and the SH-101's way of having more voices than the hardware.
  • engine/one_pole.h, engine/state_variable_filter.h, engine/variance.h, the shared arpeggiator and effects.
  • instruments/README.md says what an instrument must provide.

Provenance

  • Rule 3: no firmware disassembly and nothing extracted from a ROM. Timings and curves come from the manuals or from measuring a real instrument's audio, or are placeholders.
  • Rule 4: no factory patches. Reading a user's own dump at run time is allowed.
  • Rule 2: emulations exist. None's code may be read for more than how a problem is structured, and nothing lifted.
  • The names are trademarks, which matters for the programs' names and anything sold.
Korg's MS synthesizers as a new instrument: the MS-10 and the MS-20, and perhaps the MS-50. Not started. **The MS-2000**, which the owner asked about with a question mark, is a different thing: a digital virtual analog of 2000 that has the name and little else. Four voices of two modelled oscillators, a modelled filter in several modes, sequences of modulation, a vocoder and digital effects. On 2026-10-11 the owner said fully digital synths were out when proposing instruments. The owner decided on 2026-10-11 that it is left out, as an instrument and as a source of features for this one: it is a general virtual analog more than a virtual MS-20. Everything below is from recollection, with nothing read. **Find and read Korg's owner's and service manuals before designing anything**, and put them in `docs/references.md`. ## The instruments, as recalled - **MS-10** (1978). One voice of 32 keys: one VCO (triangle, sawtooth, a pulse of set or modulated width, noise) at 32' to 4', a two-pole low-pass with a resonance that self-oscillates, one envelope with a hold before its attack, one modulation generator whose shape is a knob (sawtooth through triangle to reverse sawtooth, and a pulse of matching width), portamento, and a small patch panel. - **MS-20** (1978). One voice of 37 keys: two VCOs, the first a triangle, sawtooth, pulse or noise and the second a sawtooth, square, narrow pulse or a ring modulation of the two. A **high-pass and then a low-pass**, each with a resonance ("peak") that self-oscillates and distorts as it does, which is the instrument's sound. Two envelopes: one of delay, attack and release, usually to the pitch, and one of hold, attack, decay, sustain and release. A modulation generator like the MS-10's. A patch panel with a sample and hold, an extra amplifier, white and pink noise, a wheel and a button, and the **external signal processor**: a preamplifier, a band-pass, an envelope follower and a pitch-to-voltage converter, so that outside sound can play it. - **Its filter changed** during its run: an early one built round Korg's own chip and a later one round an amplifier chip, which sound different. - **MS-50**: the same parts as a module with no keyboard and everything on jacks. ## To settle with the owner first - **One instrument**, most likely: the MS-10 is an MS-20 with one VCO, no high-pass and one envelope, as a SEM is an OB-X with things at nothing. - **The patch panel**, which goes with #7 as the ARP 2600's (#22) and the PS's do. Both play with no cables in. - **More than one voice**, as the SH-101 was given. - **Its name**, and its programs' names. `mini10` and `mini20` say little, as `mini1` would have. ## What it needs that is not here yet - **The two filters and how they overdrive.** Two poles each, in a row, is arrangement, and `engine/state_variable_filter.h` is two poles; but that one stops short of self-oscillation and is linear, and here the oscillation and what limits it are most of the character. The slopes want the service manual. The two revisions are a switch by the owner's rule of a switch per quirk, and the rest is tone, which waits on measurement (#4). - **An LFO whose shape is a knob.** Goes with #8. - **An envelope with a hold**, and one of delay, attack and release. `engine/envelope.h` has a delay stage now. - **The external signal processor**: #17 for the outside sound, then an envelope follower and a pitch follower, which nothing here has. - **A ring modulator in the voice**, which the SH-5 has. ## What is already here to build on - `instruments/sh5/` and `instruments/sh101/`, the other one-voice instruments: `engine/analog_oscillator.h`, `engine/noise.h` with pink, the SH-5's sample and hold, `engine/portamento.h`, and the SH-101's way of having more voices than the hardware. - `engine/one_pole.h`, `engine/state_variable_filter.h`, `engine/variance.h`, the shared arpeggiator and effects. - `instruments/README.md` says what an instrument must provide. ## Provenance - Rule 3: no firmware disassembly and nothing extracted from a ROM. Timings and curves come from the manuals or from measuring a real instrument's audio, or are placeholders. - Rule 4: no factory patches. Reading a user's own dump at run time is allowed. - Rule 2: emulations exist. None's code may be read for more than how a problem is structured, and nothing lifted. - The names are trademarks, which matters for the programs' names and anything sold.
bcox commented 2026-10-12 08:02:27 +00:00
Author
Owner
Copy link

Sources read (Claude, 2026-10-12, working on this overnight), before any design:

Not found: a titled service manual for the later revision; what is inside the KORG35 block; any printed filter slope; minimum envelope times, the portamento's time, and a clearly printed MG range.

What the reading changed from the issue's recollection:

  • HOLD is not a wait before the attack. It keeps the gate open after the key is let go, up to 20 s.
  • The filters are not driven by the same EG 2 the amplifier is: they get the envelope less the sustain level, so the cutoff rests below, peaks above and settles at the knob's setting.
  • EG 1 reaches the pitch as -5 V to 0 V: the note starts flat and climbs to pitch.
  • The ring modulator is an exclusive-OR of VCO 1's pulse, at its width, and VCO 2's square.
  • Korg prints no slope for either filter. Both revisions' schematics read as the same network: a two-pole low-pass, and a high-pass that is the same loop fed through its second capacitor, which falls at 6 dB an octave under a two-pole peak. What differs is where the diodes are: in the earlier one across a high-gain amplifier that the sound itself passes through, in the later one across the amplifier that carries only the resonance.
  • The keyboard is Hz/V, the highest key has priority, and neither filter follows the keyboard.
  • The original has no sync and no pulse width modulation on the MS-20 (the MS-10 has a PWM jack).
Sources read (Claude, 2026-10-12, working on this overnight), before any design: - **MS-20 owner's manual**, Korg's reprint of the 1978 original: https://www.korg.com/us/support/download/manual/0/134/2441/ - **MS-20 service manual** (10 pages: specifications, both circuit diagrams, block diagram, adjustment), which is the earlier filter's, the KORG35 one: https://archive.org/details/korg_MS-20_SERVICE_MANUAL - **The later filter board's schematic, KLM-307** (one sheet, the 13600 one): http://machines.hyperreal.org/manufacturers/Korg/MS-synths/schematics/KLM-307.GIF - **MS-10 owner's manual** (https://www.korg.com/us/support/download/manual/1/384/3996/) and **service manual** (https://archive.org/details/korg_MS-10_SERVICE_MANUAL) - **MS-50 owner's manual**, its specification and its VCO and VCF pages only: https://www.korg.com/us/support/download/manual/1/385/3998/ - Korg's manuals for the MS-20 mini (2013), MS-20M Kit (2014) and MS-20 FS (2020), for what they say of the original: highest-note priority, and that there are two filters. Not found: a titled service manual for the later revision; what is inside the KORG35 block; any printed filter slope; minimum envelope times, the portamento's time, and a clearly printed MG range. What the reading changed from the issue's recollection: - **HOLD is not a wait before the attack.** It keeps the gate open after the key is let go, up to 20 s. - **The filters are not driven by the same EG 2 the amplifier is**: they get the envelope less the sustain level, so the cutoff rests below, peaks above and settles at the knob's setting. - **EG 1 reaches the pitch as -5 V to 0 V**: the note starts flat and climbs to pitch. - **The ring modulator is an exclusive-OR** of VCO 1's pulse, at its width, and VCO 2's square. - **Korg prints no slope for either filter.** Both revisions' schematics read as the same network: a two-pole low-pass, and a high-pass that is the same loop fed through its second capacitor, which falls at 6 dB an octave under a two-pole peak. What differs is where the diodes are: in the earlier one across a high-gain amplifier that the sound itself passes through, in the later one across the amplifier that carries only the resonance. - The keyboard is Hz/V, the highest key has priority, and neither filter follows the keyboard. - The original has no sync and no pulse width modulation on the MS-20 (the MS-10 has a PWM jack).
bcox commented 2026-10-12 08:34:29 +00:00
Author
Owner
Copy link

Built overnight on 2026-10-12 by Claude, with no one to ask. It is in one local commit, not pushed. It has been rendered offline and benched on the Pi and not yet played or heard by anyone. The issue is left open for you to close after playing it.

What was built

mini-MS: the MS-20 and the MS-10 as one instrument, in instruments/ms20/. Programs minimsd, minimsctl, minimsweb, minimsrender, minimsbench; patches are .mms files; the web port is 8812.

  • Two VCOs (the first a triangle, sawtooth, pulse of set width or white noise at 32' to 4'; the second a sawtooth, square, narrow pulse or the ring modulator at 16' to 2', with its pitch knob), the mixer, portamento and master tune.
  • The high-pass and then the low-pass filter, each with a cutoff of 50 Hz to 15 kHz, a peak that self-oscillates and two modulation knobs. filter.revision is the earlier (KORG35) or the later filter.
  • Both envelopes: EG 1 of delay, attack and release, to the pitch; EG 2 of hold, attack, decay, sustain and release, to the amplifier and the filters.
  • The modulation generator whose one knob leans both of its waves.
  • The patch panel, as .jack parameters, the 2600's way: 15 inputs and 16 outputs, with the sample and hold, the second amplifier, pink and white noise, the wheel and the button. --mixer.ext.jack=16 is the cord from PHONES to EXT SIGNAL IN.
  • An MS-10 is settings: mixer.vco2 at 0, hpf.on off, and fm.eg.jack at 17 (EG 2 to FREQ). Its PITCH knob, PWM IN and EXTERNAL SIGNAL LEVEL are here for it.
  • More than one voice, off by default, to twelve (voice.count), with Poly 1, Poly 2 and unison as the SH-101 has.
  • Pitch bend is the control wheel, the sustain pedal is the momentary switch, controller 1 is vibrato.

Shared, in engine/:

  • sallen_key_filter.h, the two-pole filter both instruments' filters are.
  • patch_panel.h, made from what the 2600 had: the jack numbering and the wires. The 2600 now uses it, its tests pass unchanged, and its bench on the Pi did not move (8.2%, 10.4%, 10.9%, 13.0% before; 8.2%, 10.3%, 10.9%, 13.0% after).
  • web/page/app.js now knows a jack control, which was the 2600's own page.js.

Tests: bazel test //... passes, 120 of 120, with no warnings. The Pi build of every instrument succeeds.

Bench on the Pi 3B, no daemon running beside it: one voice 4.2% of a core with no cords, 5.6% with every part at work; 5.6% and 6.9% with the effects; four voices 22.9%; eight 47.2%; twelve 72.2% (73.5% with the effects). A voice costs as the 2600's does, since every part is worked out a sample at a time, which is why twelve is the most.

Levels: one VCO at full is about full scale, and about 1.5 on low notes, since the high-pass filter at the bottom of its knob (50 Hz) still leans a sawtooth over. Two VCOs with the later filter reach about 2.9, and a filter singing at full peak about 2.

Deliberately not done

  • How the filters distort. Each has one soft limit where its revision's circuit has its diodes, at a placeholder level. This is most of what an MS-20 is known for and waits on a measurement (#4).
  • The external signal processor, which wants sound from outside (#17). EXT SIGNAL IN is there, and is useful from inside.
  • The MS-50.
  • The envelopes' curves and their shortest times, the knobs' tapers, the pulse vanishing at the end of its knob, the Kit's "FM VCO 1->2" switch (no document gives its depth), and the MS-10's modulation switch and button, which its manual describes and its panel and circuit diagram do not show.

Open questions, and the option chosen for each

  1. One instrument or two? One: the MS-10's manuals show the MS-20's circuits with fewer of them.
  2. The MS-50? Left out. It is a cabinet of separate modules with nothing wired behind its panel. It could be a fourteenth instrument, or some of its modules could join this patch panel.
  3. The high-pass filter's slope. 6 dB an octave under a two-pole peak, in both revisions, as the circuits read. Korg prints no slope at all, and what is inside the KORG35 is drawn nowhere I found. The common account, 6 in one revision and 12 in the other, is not what the two diagrams show. A measurement would settle it.
  4. One revision switch or one for each filter? One for both, as Korg's Kit has it.
  5. Where self-oscillation begins. 0.8 of the knob, as for every filter here.
  6. More than one voice? Yes, off by default, to twelve, each a whole MS-20, one modulation generator for all.
  7. The patch panel. .jack parameters, as the 2600's; the same questions are open as there (#7).
  8. The wheel. Pitch bend drives the wheel's jack and also bends the pitch two semitones with no cord, which the hardware's does not. bend.range at 0 makes it only what it is patched to. Controller 1 is half a semitone of vibrato; both could as well start at nothing.
  9. EG 1 OUT's polarity. The panel and the circuit diagram say 0 to +5 V; both specification lists say -5 V to 0. The jack is taken as the first and the wire to the pitch as the second, so a note with EG1/EXT up starts flat and rises.
  10. The modulation generator's range. 0.1 Hz to 20 Hz, from a figure the service manual prints a line out of place.
  11. MASTER TUNE. Two semitones either way, the owner's manual's words; Korg also prints half an octave and 100 cents.
  12. Every level: a VCO beside EXT SIGNAL IN, PHONES beside SIGNAL OUT, the filters' headroom, where a trigger turns on.
  13. The lowest key is MIDI's 48, playing that note at 8'.
  14. The glide is in pitch; the hardware's capacitor is on a voltage that is hertz.
  15. The name: minims, mini-MS, .mms.

All of these are under "Open decisions" in CLAUDE.md, and the plain choices are listed in full under "The MS-20 and MS-10" there.

Not found

A service manual for the later revision (only its filter board's diagram was read; the later main board's was fetched and not read); the MS-50's service manual; any dated edition; the envelopes' shortest times and the portamento's time.

Built overnight on 2026-10-12 by Claude, with no one to ask. It is in one local commit, not pushed. **It has been rendered offline and benched on the Pi and not yet played or heard by anyone.** The issue is left open for you to close after playing it. ## What was built **mini-MS**: the MS-20 and the MS-10 as one instrument, in `instruments/ms20/`. Programs `minimsd`, `minimsctl`, `minimsweb`, `minimsrender`, `minimsbench`; patches are `.mms` files; the web port is 8812. - Two VCOs (the first a triangle, sawtooth, pulse of set width or white noise at 32' to 4'; the second a sawtooth, square, narrow pulse or the ring modulator at 16' to 2', with its pitch knob), the mixer, portamento and master tune. - **The high-pass and then the low-pass filter**, each with a cutoff of 50 Hz to 15 kHz, a peak that self-oscillates and two modulation knobs. `filter.revision` is the earlier (KORG35) or the later filter. - Both envelopes: EG 1 of delay, attack and release, to the pitch; EG 2 of hold, attack, decay, sustain and release, to the amplifier and the filters. - The modulation generator whose one knob leans both of its waves. - **The patch panel, as `.jack` parameters, the 2600's way**: 15 inputs and 16 outputs, with the sample and hold, the second amplifier, pink and white noise, the wheel and the button. `--mixer.ext.jack=16` is the cord from PHONES to EXT SIGNAL IN. - **An MS-10 is settings**: `mixer.vco2` at 0, `hpf.on` off, and `fm.eg.jack` at 17 (EG 2 to FREQ). Its PITCH knob, PWM IN and EXTERNAL SIGNAL LEVEL are here for it. - More than one voice, off by default, to twelve (`voice.count`), with Poly 1, Poly 2 and unison as the SH-101 has. - Pitch bend is the control wheel, the sustain pedal is the momentary switch, controller 1 is vibrato. Shared, in `engine/`: - `sallen_key_filter.h`, the two-pole filter both instruments' filters are. - `patch_panel.h`, made from what the 2600 had: the jack numbering and the wires. The 2600 now uses it, its tests pass unchanged, and its bench on the Pi did not move (8.2%, 10.4%, 10.9%, 13.0% before; 8.2%, 10.3%, 10.9%, 13.0% after). - `web/page/app.js` now knows a `jack` control, which was the 2600's own `page.js`. Tests: `bazel test //...` passes, 120 of 120, with no warnings. The Pi build of every instrument succeeds. Bench on the Pi 3B, no daemon running beside it: one voice 4.2% of a core with no cords, 5.6% with every part at work; 5.6% and 6.9% with the effects; four voices 22.9%; eight 47.2%; twelve 72.2% (73.5% with the effects). A voice costs as the 2600's does, since every part is worked out a sample at a time, which is why twelve is the most. Levels: one VCO at full is about full scale, and about 1.5 on low notes, since the high-pass filter at the bottom of its knob (50 Hz) still leans a sawtooth over. Two VCOs with the later filter reach about 2.9, and a filter singing at full peak about 2. ## Deliberately not done - **How the filters distort.** Each has one soft limit where its revision's circuit has its diodes, at a placeholder level. This is most of what an MS-20 is known for and waits on a measurement (#4). - **The external signal processor**, which wants sound from outside (#17). EXT SIGNAL IN is there, and is useful from inside. - **The MS-50.** - The envelopes' curves and their shortest times, the knobs' tapers, the pulse vanishing at the end of its knob, the Kit's "FM VCO 1->2" switch (no document gives its depth), and the MS-10's modulation switch and button, which its manual describes and its panel and circuit diagram do not show. ## Open questions, and the option chosen for each 1. **One instrument or two?** One: the MS-10's manuals show the MS-20's circuits with fewer of them. 2. **The MS-50?** Left out. It is a cabinet of separate modules with nothing wired behind its panel. It could be a fourteenth instrument, or some of its modules could join this patch panel. 3. **The high-pass filter's slope.** 6 dB an octave under a two-pole peak, in both revisions, as the circuits read. Korg prints no slope at all, and what is inside the KORG35 is drawn nowhere I found. The common account, 6 in one revision and 12 in the other, is not what the two diagrams show. A measurement would settle it. 4. **One revision switch or one for each filter?** One for both, as Korg's Kit has it. 5. **Where self-oscillation begins.** 0.8 of the knob, as for every filter here. 6. **More than one voice?** Yes, off by default, to twelve, each a whole MS-20, one modulation generator for all. 7. **The patch panel.** `.jack` parameters, as the 2600's; the same questions are open as there (#7). 8. **The wheel.** Pitch bend drives the wheel's jack and also bends the pitch two semitones with no cord, which the hardware's does not. `bend.range` at 0 makes it only what it is patched to. Controller 1 is half a semitone of vibrato; both could as well start at nothing. 9. **EG 1 OUT's polarity.** The panel and the circuit diagram say 0 to +5 V; both specification lists say -5 V to 0. The jack is taken as the first and the wire to the pitch as the second, so a note with EG1/EXT up starts flat and rises. 10. **The modulation generator's range.** 0.1 Hz to 20 Hz, from a figure the service manual prints a line out of place. 11. **MASTER TUNE.** Two semitones either way, the owner's manual's words; Korg also prints half an octave and 100 cents. 12. **Every level**: a VCO beside EXT SIGNAL IN, PHONES beside SIGNAL OUT, the filters' headroom, where a trigger turns on. 13. **The lowest key** is MIDI's 48, playing that note at 8'. 14. **The glide** is in pitch; the hardware's capacitor is on a voltage that is hertz. 15. **The name**: `minims`, mini-MS, `.mms`. All of these are under "Open decisions" in `CLAUDE.md`, and the plain choices are listed in full under "The MS-20 and MS-10" there. ## Not found A service manual for the later revision (only its filter board's diagram was read; the later main board's was fetched and not read); the MS-50's service manual; any dated edition; the envelopes' shortest times and the portamento's time.
bcox commented 2026-10-13 00:59:01 +00:00
Author
Owner
Copy link

The MS-20's voice.low_note is gone: the shared switch has three positions, voice.key_priority, 0 the latest key, 1 the lowest, 2 the highest, and the MS-20 starts at 2, its hardware's rule, so it needs no special case. On branch family-extensions, not merged.

The MS-20's `voice.low_note` is gone: the shared switch has three positions, `voice.key_priority`, 0 the latest key, 1 the lowest, 2 the highest, and the MS-20 starts at 2, its hardware's rule, so it needs no special case. On branch `family-extensions`, not merged.
bcox commented 2026-10-13 04:27:03 +00:00
Author
Owner
Copy link

Closed: Ben played it on his laptop on the night of 2026-10-12 and found it looks and sounds fine.

Confirmed by him the same night: the MS-10 as settings of the MS-20; more than one voice, off by default, to twelve; the patch panel as .jack parameters; filter.revision as one switch for both filters; the wheel, the button and the controllers as they start; minims, mini-MS, .mms and port 8812; and its plain choices as they stand.

What is left, and where it is kept:

  • The MS-50: not an instrument; its modules go on this one's patch panel. #61.
  • The external signal processor waits on sound from outside: #17, where it is noted.
  • How the page shows a cord: #58. A patch file naming an output: #59.
  • How the filters distort, the high-pass filter's slope and EG 1 OUT's polarity want a real one measured, and are under "Open decisions" in CLAUDE.md, with #4.

Everything said here is on main, pushed, on the night of 2026-10-12. What was heard was heard on Ben's laptop, not on the Pi.

**Closed: Ben played it on his laptop on the night of 2026-10-12 and found it looks and sounds fine.** Confirmed by him the same night: the MS-10 as settings of the MS-20; more than one voice, off by default, to twelve; the patch panel as `.jack` parameters; `filter.revision` as one switch for both filters; the wheel, the button and the controllers as they start; `minims`, mini-MS, `.mms` and port 8812; and its plain choices as they stand. What is left, and where it is kept: - **The MS-50**: not an instrument; its modules go on this one's patch panel. #61. - **The external signal processor** waits on sound from outside: #17, where it is noted. - How the page shows a cord: #58. A patch file naming an output: #59. - **How the filters distort**, the high-pass filter's slope and EG 1 OUT's polarity want a real one measured, and are under "Open decisions" in CLAUDE.md, with #4. Everything said here is on `main`, pushed, on the night of 2026-10-12. What was heard was heard on Ben's laptop, not on the Pi.
bcox closed this issue 2026-10-13 04:27:03 +00:00
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
main
No results found.
Labels
Clear labels
No items
No labels
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
bcox/mini-800#25
No description provided.