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

SEM family: read Oberheim's manuals and check what was built #14

New issue
Open
opened 2026-10-12 02:40:06 +00:00 by bcox · 1 comment
bcox commented 2026-10-12 02:40:06 +00:00
Owner
Copy link

Nothing in instruments/sem/ was built from a manual. This is the largest uncertainty in the instrument.

What each part rests on today (docs/references.md has the links):

  • SEM, Two Voice, Four Voice, Eight Voice, OB-X: a 1978 magazine review, encyclopedia articles, a specification of the SEM's reissue, and recollection. No manual was found at a link that answered. Two leads are in docs/references.md.
  • OB-6 (sub oscillator, VCO 2's triangle, low frequency mode and keyboard switch, sync from VCO 2 to VCO 1, the LFO's sawtooths and its routes to the amplifier and filter mode, velocity, aftertouch): the specification as a shop's page reproduces it. Sequential publishes the manual.
  • TEO-5 (cross modulation through zero, the second LFO, the envelopes' delay): Oberheim's specification, seen as a search result's summary and shops' listings. Oberheim publishes the manual.
  • OB-X8 and the OB-8's lever box (pink noise, the unison's voice count, the bender and lever to one VCO, the lever's own LFO): recollection only. Oberheim publishes the manual.

To do: read the three published manuals, find the older ones, check each file's opening comment and CLAUDE.md's "The SEM family" against them, and correct what is wrong.

Placeholders a manual may settle, each marked where it is:

  • kLeastDamping: how sharp the filter's resonance gets, and the curve of its knob.
  • kCrossModulation and kMostCross: how deep cross modulation goes, and through zero.
  • kLowFrequencyOctaves: how far down VCO 2's low frequency mode goes.
  • Where VCO 2 sits with its keyboard switch off (kFixedKey, middle C).
  • How far velocity and aftertouch reach, and that the OB-6's aftertouch goes to the amounts of its two envelopes, where ours goes to the cutoff and the level.
  • Which envelope each VCO's modulation switch offers on the SEM.
  • Which VCO the OB-X's cross modulation moves.

What a manual cannot settle is tone, which waits on measurement: #4.

Nothing in `instruments/sem/` was built from a manual. This is the largest uncertainty in the instrument. What each part rests on today (`docs/references.md` has the links): - **SEM, Two Voice, Four Voice, Eight Voice, OB-X**: a 1978 magazine review, encyclopedia articles, a specification of the SEM's reissue, and recollection. No manual was found at a link that answered. Two leads are in `docs/references.md`. - **OB-6** (sub oscillator, VCO 2's triangle, low frequency mode and keyboard switch, sync from VCO 2 to VCO 1, the LFO's sawtooths and its routes to the amplifier and filter mode, velocity, aftertouch): the specification as a shop's page reproduces it. Sequential publishes the manual. - **TEO-5** (cross modulation through zero, the second LFO, the envelopes' delay): Oberheim's specification, seen as a search result's summary and shops' listings. Oberheim publishes the manual. - **OB-X8 and the OB-8's lever box** (pink noise, the unison's voice count, the bender and lever to one VCO, the lever's own LFO): recollection only. Oberheim publishes the manual. To do: read the three published manuals, find the older ones, check each file's opening comment and CLAUDE.md's "The SEM family" against them, and correct what is wrong. Placeholders a manual may settle, each marked where it is: - `kLeastDamping`: how sharp the filter's resonance gets, and the curve of its knob. - `kCrossModulation` and `kMostCross`: how deep cross modulation goes, and through zero. - `kLowFrequencyOctaves`: how far down VCO 2's low frequency mode goes. - Where VCO 2 sits with its keyboard switch off (`kFixedKey`, middle C). - How far velocity and aftertouch reach, and that the OB-6's aftertouch goes to the amounts of its two envelopes, where ours goes to the cutoff and the level. - Which envelope each VCO's modulation switch offers on the SEM. - Which VCO the OB-X's cross modulation moves. What a manual cannot settle is tone, which waits on measurement: #4.
bcox commented 2026-10-12 22:36:33 +00:00
Author
Owner
Copy link

The OB-X's owner's manual (second edition, second printing, June 1980) and its service manual (September 1979) turned up on archive.org while #37 was done, with the OB-Xa's and OB-8's. Links are in docs/references.md. The owner's manual was read as far as its cassette section; the service manual was skimmed. Written by Claude. Nothing in the instrument was changed for this, since each difference would change what a stored patch means; they are listed in CLAUDE.md under "The SEM family".

Borne out

  • Two VCOs into a two-pole low-pass that "cannot be put into oscillation" even at full resonance.
  • Cross modulation is the second VCO modulating the first, which is the direction built.
  • Sync is the second VCO following the first.
  • Two ADSRs with a release of their own; one LFO for all voices with sine, square and sample and hold; noise; polyphonic portamento, in unison too; a pulse width from 50% to 5% (kNarrowestPulse).
  • OSC 2 ONLY on the bender, which bend.target was built for from recollection.

Different from what is built

  • Unison plays the lowest key held ("low note rule"). mini-SEM's plays the latest. voice.low_note, added for #37, is the manual's rule and starts off.
  • The square LFO moves a pitch downward only ("a discrete downward interval"). Here it swings either side.
  • X-MOD is a switch, with no depth. vco1.xmod is a level, and its depth at 1 is still a guess.
  • The first VCO is tuned in octaves over four octaves and the second in half steps over five. Both are semitones here, three octaves either way.
  • Each VCO is a sawtooth or a pulse, by a switch; here a level each, both at once, which is the OB-Xa's and OB-8's.
  • One pulse width knob serves both VCOs. Here one each, which is the OB-8's.
  • The second VCO and the noise each have HALF and FULL into the filter, half being about 5 dB down. Levels here.
  • The filter follows the keyboard by a switch, KBD. filter.tracking is an amount.
  • The bender is a whole step or an octave; the lever adds vibrato to both VCOs and does nothing pushed the other way.
  • The LFO runs from about 0.1 Hz to 20 Hz; one depth goes to either VCO's frequency and the filter's by switches, and another to either pulse width. Here a depth for each destination.
  • The service manual sets the portamento to 1 to 2.5 seconds for an octave at the top of its knob.

Still unread: the SEM's and the Two, Four and Eight Voice's manuals, the OB-6's, the TEO-5's and the OB-X8's; and the OB-X's schematics, for the cross modulation's depth and which envelope does what.

The OB-X's owner's manual (second edition, second printing, June 1980) and its service manual (September 1979) turned up on archive.org while #37 was done, with the OB-Xa's and OB-8's. Links are in `docs/references.md`. The owner's manual was read as far as its cassette section; the service manual was skimmed. Written by Claude. Nothing in the instrument was changed for this, since each difference would change what a stored patch means; they are listed in CLAUDE.md under "The SEM family". **Borne out** - Two VCOs into a two-pole low-pass that "cannot be put into oscillation" even at full resonance. - Cross modulation is the second VCO modulating the first, which is the direction built. - Sync is the second VCO following the first. - Two ADSRs with a release of their own; one LFO for all voices with sine, square and sample and hold; noise; polyphonic portamento, in unison too; a pulse width from 50% to 5% (`kNarrowestPulse`). - OSC 2 ONLY on the bender, which `bend.target` was built for from recollection. **Different from what is built** - **Unison plays the lowest key held** ("low note rule"). mini-SEM's plays the latest. `voice.low_note`, added for #37, is the manual's rule and starts off. - **The square LFO moves a pitch downward only** ("a discrete downward interval"). Here it swings either side. - **X-MOD is a switch**, with no depth. `vco1.xmod` is a level, and its depth at 1 is still a guess. - **The first VCO is tuned in octaves** over four octaves and the second in half steps over five. Both are semitones here, three octaves either way. - **Each VCO is a sawtooth or a pulse**, by a switch; here a level each, both at once, which is the OB-Xa's and OB-8's. - **One pulse width knob serves both VCOs.** Here one each, which is the OB-8's. - **The second VCO and the noise each have HALF and FULL** into the filter, half being about 5 dB down. Levels here. - **The filter follows the keyboard by a switch**, KBD. `filter.tracking` is an amount. - **The bender** is a whole step or an octave; the lever adds vibrato to both VCOs and does nothing pushed the other way. - **The LFO** runs from about 0.1 Hz to 20 Hz; one depth goes to either VCO's frequency and the filter's by switches, and another to either pulse width. Here a depth for each destination. - The service manual sets the portamento to 1 to 2.5 seconds for an octave at the top of its knob. **Still unread**: the SEM's and the Two, Four and Eight Voice's manuals, the OB-6's, the TEO-5's and the OB-X8's; and the OB-X's schematics, for the cross modulation's depth and which envelope does what.
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#14
No description provided.