View Single Post
  #1  
Old 2011-06-01, 21:41
Tokeijikaku's Avatar
Tokeijikaku Tokeijikaku is online now
 
Join Date: Jul 2010
Posts: 337
Polyphonic interface and the "Redundantator" device

There's a lot of complaining about the monophonic nature of CVs and the limitations of stand-alone filter modules when dealing with polyphonic instruments and whatnot, and it's bitten me hundreds of times as well. So I figured - a fully polyphonic interface for everything and anything.

What if we could have polyphonic audio go through a loose filter or through the filters of a Thor and have said filters with keyboard tracking and envelopes allocated appropriately to each voice? What if we could mix the signals from several sequencers into a single polyphonic CV? Well that would be two CVs but my point stands.

This would allow some mighty interesting engineering, and bring us one step closer to the modular paradigm of Reason.

---

I even thought up a first-step, clunky albeit parsimonious way of implementing it without too much hassle and reprogramming - I call it the "Redundantator". Simply put, it's like a combinator with a polyphony dial. All the items inside the Redundantator are forced monophonic, but are instead "cloned", in real time and out of sight, according to the Redundantator's set polyphony, with the exact same settings. Only one instance of each instrument is visible, while in reality, there's a multitude of exact copies of the entire configuration - one instance for each "voice". And you can probably see where this is going. All MIDI signals introduced to the Redundantator are allocated to the appropriate "voice" and the whole deal plays like a polyphonic modular instrument. Polyphonic filter chains, polyphonic keyboard-tracking EQ, polyphonic unison, polyphonic shaping and distortion, voice-specific panning - everything is made possible. The audio is eventually combined through the Redundantator's interface. A Redundantator patch would be the size of its "prototype voice" alone, and given the matching settings it could probably be optimized in a great many ways. It even starts with "Re", although the name is an obvious placeholder.

I'm more than certain that Reason's architecture, which already owes its high efficiency to its redundancy (this, by the way, I kind of believe is one of the reasons we never get external plugin support), would handle this with excellence. There are a few hassles to be considered with the Redundantator, of course - what if you wanna use a signal generated from within to modify something outside? What if you wanna interconnect Redundantators in some fashion? But most importantly, how to play them polyphonically with CV notes? Anyways, the Redundantator is just a thought, and the wishful me would rather see Reason shift completely to a polyphonic paradigm. Which, of course, would be difficult to implement without screwing over everything previously produced - and the problem of polyphony counts being mismatched - but yeah, one can hope.

---

Also, make CV signals full resolution. Try converting a CV signal to audio with a Thor or modifying something "strictly-CV-driven" with high-frequency audio signals and you'll hear what I mean. That pesky quantization.

Last edited by Tokeijikaku; 2011-06-02 at 12:47.