Generating Aparillo style Polyphonic CV signals

I've been playing around with Aparillo and got inspired to start tinkering, can anything like the deep control of unison voices be achieved in Drambo? You can add an extra unison module and direct the CV to a parameter other than pitch. This is interesting but the spread dial is quite limited, reasonably enough. The polyphonic CV signal can pass through a graphic shaper and this adds some scope but still doesn't allow much freedom to manipulate CV signals on each voice independently.


I've been trying to work out if it's possible to build a polyphonic CV signal from scratch. For instance taking 8 LFOs and sending the signal independently to voices 1 through 8. There's the voice selector module for demultiplexing the signal (if that's the right word) but I can't figure out how to zip it up again. Voice selector doesn't seem to have a counterpart and I've been messing around with Poly to Mono but it's not the droid I'm looking for, after ages experimenting I haven't got very far.

I'll admit I don't know exactly how polyphony works here. Does each voice have something like a separate port? It's the one part of Drambo I can't really get my head around.

Comments

  • edited December 2020

    The midi to cv module is kind of like a 'voice port'. So by using multiple midi to cv modules you could basically get extra voices to process separately. You could put 2 midi to cv modules at the front, then place 2 VCO's after that, each one routed to a different midi to cv. You'd be able to process each cv stream separately with things like the pitch, ratio or overtone modules, or even delay the pitch cv. Of course, this would still be polyphonic, note note by note, but you'd still be able to do extra layers of cv processing.

    I don't know if its what you're after, but a midi to cv module placed after a midi filter can also be used. I have a project with 8 Flexi samplers - each flexi containing 15 dynamic sample layers triggered by velocity. So with 8 midi filters + 8 midi to cv modules I now have 8 voices of polyphony per each note, not just globally. This reduces note stealing significantly. You could set up 127 note filters and 127 midi to cv modules if you want to process each stream totally individually though haha. Might break your iPad though once you start adding oscillators

  • @aleyas Thanks, I think you could even 'hack' the MIDI note to pitch CV further with a note generator. Maybe if the voice allocation is consistent (why wouldn't it be), then this might work without layering signal chains. I'm going to experiment with this.

    I'm trying to think of any virtual modular where you can process signals from individual voices in the way that Aparillo does and I can't. I don't even think NI Reaktor can do this but it's been a while since I could competently find my way around that. I don't know if a 'send to voice' module would be viable or probably just too niche.

    I like the idea of trying to open up unison to more variation than pitch spread (economy of modules) but layering does seem a lot more straightforward.

  • The beauty of the unison module is it's simplicity. It's possible to make custom spreads with (n) value generators to (n) MIDI note generators to MIDI mixer to MIDI CV with (n) voices. Using an LFO as a gate for the MIDI note generators gives a healthy refresh rate. Can use both velocity and pitch CV for any purpose so 2 for the price of 1.

    However, scrap the idea of trying to re insert a value into a specified polyphonic channel, custom distributions for the Unison spread parameter would be so elegant and simple from a user perspective.

    Sorry if this all sounds arcane. It makes sense after playing with Aparillo.

Sign In or Register to comment.