VCO as LFO: an oscillator's low-frequency range, below what coarse tuning and the pitch routes reach #137

New issue

Open

opened 2026-10-06 00:13:39 +00:00 by bcox · 0 comments

bcox commented

2026-10-06 00:13:39 +00:00

Owner

bcox asked on 2026-10-05 for a ticket for this gap in his moregaps.md feature table. Engine claims are from reading the tree at f6aa8eba on main. Instrument claims are from that document's research, read mostly as archive.org OCR of the manuals by research agents and not re-checked; the sources are listed there. Nothing here is built, measured or listened to.

Summary

A fixed-pitch oscillator exists: key follow is an amount around a pivot key, and zero holds the oscillator at the pivot (dsp/va_oscillator.h:1088-1098). What does not exist is the range. Coarse tuning reaches 48 semitones either way (dsp/va_oscillator.h:1311, the editor at editor/patch_view.h:2882), so from middle C a fixed oscillator bottoms out near 16 Hz; a pitch-lane route adds two octaves at full scale (dsp/va_layer.h:2428), so about 1 Hz with routes spent on it. The instruments' low ranges go far below that, and an oscillator down there is a modulator with every oscillator feature: a waveform, sync, a phase that other oscillators can lock to, and a place as a matrix source (dsp/va_layer.h:499-517).

#126 is the fixed-frequency ticket for the FM pair's operators; this is the VA oscillator's range.

What the instruments do

Prophet-5: oscillator B's low-frequency mode with keyboard off, 9 of the original 40 programs. Pro-One: the same switch.
Jupiter-8: VCO-2's low-frequency mode, 9 of 128 in a soundset.
ARP 2600: each VCO's low range is 0.03 to 30 Hz, off the keyboard. Odyssey: VCO 1's low mode is 0.2 to 20 Hz.
Memorymoog: oscillator 3's Low switch drops it about five octaves, with keyboard control off.
Pro 3: a Low Freq switch on oscillator 3. Pro 2: per-oscillator key follow (the range was not read).
Korg 700S: Modulator 3 runs "without following the keyboard". CS-80 and GX-1: the ring modulator's oscillator, 1.7 to 28 Hz on the GX-1.

The work, as a sketch

A low-range switch per oscillator that shifts the increment down by a fixed number of octaves, five like the Memorymoog's or more, so coarse and fine then tune inside that range. One shift on the increment, no new arithmetic; the phase-shape machinery is unchanged, and the band-limiting is moot at those increments.
Below some rate the oscillator's edge corrections do nothing useful and the pulse's PWM lane becomes a duty cycle: nothing to change, just nothing to expect.
Its value as a modulator is through the oscillator sources the matrix already has; the LFO's own sample-and-hold and random walk stay the LFO's.

Open

Whether the shift is a switch (the instruments) or a wider coarse range (fewer fields, worse editor travel).
Interaction with lock: a locked oscillator divides a source's phase by octaves already (dsp/va_layer.h:546-553), which is a low range of a kind for a locked slave only.

bcox asked on 2026-10-05 for a ticket for this gap in his `moregaps.md` feature table. Engine claims are from reading the tree at `f6aa8eba` on `main`. Instrument claims are from that document's research, read mostly as archive.org OCR of the manuals by research agents and not re-checked; the sources are listed there. **Nothing here is built, measured or listened to.** ## Summary A fixed-pitch oscillator exists: key follow is an amount around a pivot key, and zero holds the oscillator at the pivot (`dsp/va_oscillator.h:1088-1098`). What does not exist is the range. Coarse tuning reaches 48 semitones either way (`dsp/va_oscillator.h:1311`, the editor at `editor/patch_view.h:2882`), so from middle C a fixed oscillator bottoms out near 16 Hz; a pitch-lane route adds two octaves at full scale (`dsp/va_layer.h:2428`), so about 1 Hz with routes spent on it. The instruments' low ranges go far below that, and an oscillator down there is a modulator with every oscillator feature: a waveform, sync, a phase that other oscillators can lock to, and a place as a matrix source (`dsp/va_layer.h:499-517`). #126 is the fixed-frequency ticket for the FM pair's operators; this is the VA oscillator's range. ## What the instruments do - Prophet-5: oscillator B's low-frequency mode with keyboard off, 9 of the original 40 programs. Pro-One: the same switch. - Jupiter-8: VCO-2's low-frequency mode, 9 of 128 in a soundset. - ARP 2600: each VCO's low range is 0.03 to 30 Hz, off the keyboard. Odyssey: VCO 1's low mode is 0.2 to 20 Hz. - Memorymoog: oscillator 3's Low switch drops it about five octaves, with keyboard control off. - Pro 3: a Low Freq switch on oscillator 3. Pro 2: per-oscillator key follow (the range was not read). - Korg 700S: Modulator 3 runs "without following the keyboard". CS-80 and GX-1: the ring modulator's oscillator, 1.7 to 28 Hz on the GX-1. ## The work, as a sketch - A low-range switch per oscillator that shifts the increment down by a fixed number of octaves, five like the Memorymoog's or more, so coarse and fine then tune inside that range. One shift on the increment, no new arithmetic; the phase-shape machinery is unchanged, and the band-limiting is moot at those increments. - Below some rate the oscillator's edge corrections do nothing useful and the pulse's PWM lane becomes a duty cycle: nothing to change, just nothing to expect. - Its value as a modulator is through the oscillator sources the matrix already has; the LFO's own sample-and-hold and random walk stay the LFO's. ## Open - Whether the shift is a switch (the instruments) or a wider coarse range (fewer fields, worse editor travel). - Interaction with lock: a locked oscillator divides a source's phase by octaves already (`dsp/va_layer.h:546-553`), which is a low range of a kind for a locked slave only.