Repository navigation
Polyphony and CV
These conventions are the same across all five modules, so they are written once here rather than repeated on every page.
Every module follows the polyphonic channel count of its audio input, up to sixteen channels, and sets its outputs to match.
Per-voice state is genuinely per-channel. Each polyphonic channel has its own delay line, its own allpass state, its own LFO phase and its own envelope. A single shared delay line across sixteen channels would sound plausible on a first listen and be wrong.
Left and right are separate paths, not a mono path duplicated.
The right input is normalled to the left. Patch a mono source into IN L only and both sides receive it, without needing a mult. Patch both and they stay independent.
In the modulation effects the two sides are deliberately offset in the LFO cycle, which is where the stereo movement comes from. In Env Follower the two sides are followed separately, so a stereo source produces two different envelopes.
Every continuous control has a CV input and its own attenuverter.
The knob sets the base value. The attenuverter scales the incoming CV. The two are summed and the result is clamped to whatever range makes sense for that parameter.
Ten volts through a fully open attenuverter moves the result by the full range of the knob. An attenuverter at zero, which is the default, leaves the knob in sole control, so an unpatched or unused CV input never changes the sound.
Attenuverters are bipolar. Turn one below centre and the CV is inverted.
Bypassing a module passes its audio input through to its output rather than muting it, so you can A and B an effect against the dry signal without repatching.
Env Follower is the exception. Its outputs are measurements rather than treated audio, so there is nothing to pass through and bypassing it leaves the outputs at zero. The reasoning is on its own page.
Modules
Reference