This page is read from MIDI Instrument Definitions, a Python library used in the author's other projects, in which every number is traced to a document its maker published. Its source is on GitHub.

Moog Music Subsequent 37 MIDI implementation

The MIDI implementation of the Moog Music Subsequent 37: what it answers to, and the 116 controls it offers, on firmware 1.2.0.

At a glance

Firmware

This page describes firmware 1.2.0.

MIDI channel

Any one of 1 to 16.

Voices

1 or 2, as set on the instrument, and it can play paraphonically.

Note range

Not recorded.

Velocity

Not recorded.

Aftertouch

Sent. Whether it answers to any is not recorded.

Pitch bend

How far it bends is not recorded, and the range can be changed.

Program change

Received and sent, with 256 presets.

MIDI clock

Sent and received.

Start and stop

Sent.

Control change

75 controls, in the tables below, 28 of them with a fine half on a second controller number.

NRPN

Supported. 98 of the controls below have one.

System exclusive

Supported.

Controls

In the order the definition gives them, by group. Under each control is the name a program calls it by, once pymidiinstrumentdefs.load("moog/subsequent_37") has loaded them.

Wheels

Keyboard

Output

Glide

Arpeggiator

Mod 1

Mod 2

Mixer

Oscillators

Filter

Filter envelope

Amplifier envelope

Without a group

Where these facts come from

Source documents

• Subsequent 37 Manual, its user manual, read on 1 October 2026 (the file read).
• MOOG SUB 37 - FIRMWARE UPDATE 1.2.0, 1.2.0, its release notes, read on 1 October 2026 (the file read).
• Subsequent 37 | Moog Music, its product page, read on 1 October 2026 (the file read).

How they were read

User manual (63pp), pp. 9, 12, 14-15, 21-23, 25, 27-32, 34-44, 47-48, 50-59, 61, and the change log printed inside the firmware v1.2.0 package, which is where Moog keeps release notes for every instrument in this corpus.

TWO TABLES CARRY ALL OF IT, and they overlap without saying so. The CC chart, pp. 53-55, has 87 rows; the NRPN chart, pp. 56-59, has 123, numbered consecutively from 402 to 524. Every row of both is accounted for below: 75 of the 87 rows are controls here, bank select's two among them, 10 are rows the maker itself marks RESERVED and 2 are channel mode messages. 98 of the 123 NRPN rows are named and 25 are the maker's own (RESERVED).

57 PARAMETERS ARE IN BOTH CHARTS UNDER DIFFERENT NAMES, which no page points out: "FILTER EG ATTACK TIME" (p. 53) in the CC chart is "F EG ATTACK" (p. 58) in the NRPN chart, "FILTER MULTIDRIVE" (p. 53) is "FILTER DRIVE" (p. 58), "GLIDE" is "GLIDE ON". The pairings are not guessed from the spelling; they are taken from the maker's own list of parameter names on p. 51, printed there as the choices for the programmable modulation destination, which expands every abbreviation the charts use. That page is what establishes that CC 20 MOD 1 PGM DEST AMT and NRPN 444 MOD 1 PGM AMT are one parameter, since it gives both as "Mod Bus 1 Programmable Destination Amount" (p. 51), and that CC 73 ARP ON/OFF is NRPN 410, which it calls "Arpeggiator Run/Stop" (p. 51).

41 CONTROLS CAN BE REACHED BY NRPN ALONE, and they are the better half of this instrument: the arpeggiator's rate, range, pattern, gate length and clock division, each LFO's sync, clock division, clock source and keyboard tracking, each modulation bus's mod wheel, velocity, pressure and controller 4 amounts, the five mixer mute switches, and the filter envelope's sync, loop, latch and clock division. The maker says why: "Since the Subsequent 37 has over 150 parameters saved with each sound, it is not possible to assign a standard MIDI CC to every parameter. For complete MIDI output from every panel knob and button, you will need to set the KNB NRPN/CC parameter to NRPN mode" (p. 56).

THE AMPLIFIER ENVELOPE IS THE OTHER WAY ROUND, and it is the one gap worth naming: it has ten controller numbers below and no NRPN at all. The NRPN chart simply stops, four rows into reserved space at 524, having given the filter envelope twelve rows and the amplifier envelope none, while p. 51 lists twelve amplifier envelope parameters by name. So "over 150 parameters" (p. 56) is not what the published chart reaches.

THE NRPN CHART'S VALUE RANGE COLUMN IS A COUNT, not a range, and the ranges below are one less. The reading is the maker's own five times over: the column's 21 is the 21 clock divisions printed on p. 52, its 4 is CC 109's four filter slopes, its 3 is CC 85's three glide types, its 6 is CC 71's six modulation sources, and its 7 is CC 91's seven destinations. Where a parameter has both a controller number and an NRPN, nrpn_range is given only when it differs from what the controller takes.

SIX ROWS OF THAT COLUMN CONTRADICT THE MANUAL ITSELF, and they are marked in place rather than corrected. Five are consecutive: NRPN 486 to 490 print 16384, 3, 2, 2 and 21 for OSC 2 DUO MODE, OSC 2 FREQUENCY, OSC 2 BEAT, OSC 1 LEVEL and OSC 1 ON, where p. 51 gives those as an on/off control, two frequency offsets, a level and an on/off control. The rows on either side of that block are right, which is what makes it a block rather than a pattern. The sixth is NRPN 518, F EG CLK DIV, which prints 2 where p. 41 sends the same parameter to the 21 clock divisions of p. 52 in words, and where the chart's three other clock divisions all print 21. Each of the six carries its NRPN and no nrpn_range, because a count that cannot be right is not a range.

ONE NAMED ROW HAS NO VALUE RANGE PRINTED AT ALL: NRPN 500, FILTER RESONANCE. That is the page rather than the reading, and it was established twice over, because a missing cell and an unread cell look the same in text. In this manual the row's numbers are drawn as a separate piece of text whose position says which column it begins in, and on that row they begin in the NRPN column, not the value column, as they do on the 25 reserved rows. In the Sub 37 manual, whose chart is the same chart, the row is drawn as one piece and reads the same way. The other 24 named rows of this instrument's filter all carry a count.

31 ROWS ARE 14-BIT PAIRS AND 28 OF THEM ARE CONTROLS, the other three being reserved rows that carry a value range anyway. They are printed as "0-127 [MSB], 0-16383[MSB,LSB]" (p. 53) with the fine half 32 above the coarse one, which is the MMA's own pairing, so lsb carries it. Whether the instrument sends both halves is a setting and not a fact about the model: CC FORMAT, in MIDI MENU 3, offers 7-BIT and 14-BIT and "use 7-bit (default)" (p. 37). So an instrument out of the box sends the coarse half alone, as the Grandmother does.

THE ONE VALUE TABLE THE MANUAL PRINTS FOR A CONTROLLER BELONGS TO A CONTROLLER THE CHART NAMES DIFFERENTLY, and that is why its 21 divisions are quoted beside LFO 1 Rate rather than carried as its values. The table on p. 52 is headed "MIDI CC VALUES FOR THE LFO 1 CLOCK DIVIDER (CC #3), LFO 2 CLOCK DIVIDER (CC #8)" (p. 52), and the chart two pages later gives CC 3 as LFO 1 RATE and CC 8 as LFO 2 RATE, each 0-16383. Both are true: "When the SYNC button is illuminated, the LFO rate is synchronized to the Subsequent 37's internal clock, or external MIDI clock. In this mode, the LFO RATE knob selects between clock divisions" (p. 23). One controller, two meanings, and which one applies depends on another parameter. The format has no field for that, so the table is printed in full where a reader will meet it. The same is true of the arpeggiator's RATE knob (p. 15).

AND THAT TABLE CONTRADICTS ITSELF ONCE: it gives the whole note as "25-40" and the dotted half note directly below it as "31-36". The other 19 bands run consecutively from 0 to 127 with no gap and no overlap, which leaves 25-30 as the only reading that fits, and 25-30 is not what the page prints. It is quoted as printed.

THE MANUAL ALSO DISAGREES WITH ITSELF ABOUT DATA ENTRY. Page 44 says a tuning scale is selected by "RPN [Registered Parameter Number*] MSB 0, LSB 3" (p. 44) and then "send Data Entry (CC #39) value 0 -31" (p. 44), while the chart gives CC 39 as the fine half of CC 7 MASTER VOLUME and gives data entry as "<RESERVED - DATA ENTRY> 6 [MSB], 38 [LSB]" (p. 53). 38 is the number that agrees with both the chart and the MMA; the page prints 39. Neither is recorded as a control: the four numbers p. 44 uses to reach the tuning scales are CC 101 and CC 100 for the RPN, CC 39 or 38 for the value, and CC 96 and CC 97 to step through them, and the chart marks all five of those RESERVED.

BANK SELECT IS CARRIED, CC 0 and CC 32, as it is wherever a maker numbers it; what this instrument puts in them is with the two controls and under program_change below, because the chart's own value column for CC 32 is the only place the manual says how its 16 banks are addressed.

EXCLUDED, and not missed: CC 122 LOCAL CONTROL ON/OFF and CC 123 ALL NOTES OFF are channel mode messages, which belong to the MIDI specification and which the validator refuses as controls.

NOT RECORDED, and marked in place:

• the note range, since the manual never says which MIDI note numbers the instrument answers to, only that its oscillators are calibrated over "22Hz-6.8kHz. Guaranteed note range at 8' of Note 18 to 116" (p. 61) and that the calibration routine's standard range is "the full range of notes the Subsequent 37's oscillators can produce (MIDI notes 15 through 116)" (p. 48), which are two different figures for the bottom of the same thing;
• what the 89 programmable modulation destinations are numbered as, since p. 51 names them and numbers none;
• the MIDI mode, which the manual never gives;
• and a default for any control, since neither chart has a default column and the manual gives defaults only for the menu settings.

WHAT THE SUB 37 HAS TO DO WITH THIS, in full, because the two are nearly the same document and the difference is one row. Moog's own notice, on this instrument's product page, says "Aug. 28, 2017, Asheville, NC - In the spirit of continuous improvement, the award-winning Sub 37 Tribute Edition has been discontinued from production to make way for a new standard: THE SUBSEQUENT 37" (product_page), and every change it then claims is sonic rather than MIDI: "double the mixer headroom of the Sub 37" (product_page), a filter whose "gain-staging ... has been reshaped" (product_page), and "a re-tuned Multidrive circuit" (product_page). "Each enhancement maintains the magic and character of its predecessor" (product_page). Their manuals' MIDI pages were compared character for character with the model name set aside: pp. 35, 36, 37 and 52, which is the whole of the MIDI menu and the MIDI operations page, are identical, and so is the specifications page. Both NRPN charts are identical in all 123 rows, names and counts. The CC charts agree on all 86 rows the older one has, five of them differing only in how the PDF spaces the letters, and this manual adds one row the Sub 37's does not print anywhere in 62 pages: CC 123, ALL NOTES OFF, which is a channel mode message and so is not a control in either. Since the two ship the same firmware image, that is a difference between the documents rather than between the instruments, and which it is cannot be settled from documents alone.