We are in the process of wrapping up Surge XT VST 1.3 and along with it will drop Surge XT Rack 2.2. You can always grab the latest nightly from our GitHub with the latest, but its probably a month or so until we are done.
But one of the new modules which we would love some feedback on is the unison module, which allows you to pull any polyphonic VCO into the Surge unison and drift VCO pattern.
Since it is a potentially confusing module, I made a short video. Sorry for the background noise. Did it with my MacBook mic in my office just now. But happy for any feedback on this (or any other 2.2 feature) as we start to think about the next release. Thanks as always!
The transition isn’t smooth when adding unison voices - how cool would that be ?
You mean like move from 3 to 4 and it cross fades as opposed to jumps? Tricky but will think about it
Although note the modulation rack there so you can modulate drift detune and the filters like any other surge module. I didn’t show that in my vid
Love the features you brought to the Unison! From prior discussion I had a sense of some of the features, but this is impressive.
My Venom Poly Unison is much more compact (3 hp), but is also much more primitive
- No return with channel summation back to the original channel count.
- No ability to overflow excess channels into multiple VCOs
- No expander for additional CV, though my Poly Clone can do the job if configured properly
- No stereo, drift, or filter
Poly Unison does have extended detune range, with options for 1 semitone, 1 octave, or 5 octaves. And you can choose which direction to detune.
Your expander module makes all the sense in the world - I think I may steal that idea and create a Venom 3hp CV expander with support for 4 in/out pairs that works with both Poly Unison and Poly Clone.
I will leave all your other great ideas alone. I’ll keep Venom Poly Unison lean and mean, for simpler applications.
Tricky indeed! I struggle trying to think of a practical solution.
I suppose it could be possible if your input channels remain as they are, and the clones are tacked on at the end. (1, 2, 3, 1’, 2’, 3’, 1’‘, 2’‘, 3’‘) But if you did it like I did with all the clones clustered together, then I don’t see how that can work. (1, 1’, 1’‘, 2, 2’, 2’‘, 3, 3’, 3’')
Assuming the channel distribution is worked out, I think you would need a virtual VCA to fade in the new voice, and you would need to fade/slew the detune changes. I don’t know about the expander CVs.
Great work Paul, thanks a bunch! You really are a model citizen in the Rack universe. Unison and drift are favorite features in the Surge VCO’s and this is very welcomed, opens a lot of doors.
Using the CV expander I can address all voices over multiple modules. But I found myself in a situation where I did want to have them split in left and right channels per module if that makes sense.
So instead of outputting them sequentially, sort them so that all left channels would end up in modules 1, 3, 5, etc and all right channels would end up in 2, 4, 6, etc.
I could solve this with a bunch of merge/split modules, but an alternative ordering option in the menu would be nice.
Yeah definitely it would imply that reordering be a path dependent map not an immediate algo which is the hardest bit.
Stealing is the greatest form of flattery! Go for it!
We were just chatting in surge discord how your modules (including recurse) and this one are all sort of in a space of interesting ideas of reusing polyphony in fun ways and I’m sure there’s loads more we could do.
The underlying vcos are all assumed mono and placed across the stereo field so I’m not quite sure I follow. Could you expand a bit on what you mean by left and right? Like you want to skew cv based on resulting position in the field?
Ah yes they aren’t hard panned but spread across the stereo field panned. I overlooked that essential detail.
Still… I’ll explain my erroneous thinking:
I wanted to have the V/O signals of channels that end up left of center be bundled in one V/O-n output and V/O signals that end up right of center to come out of V/O-(n+1). The reason this came up is because I tried making my own chorus/flanger/phasor like effects by filtering/delaying/etc. each voice individually but with clear differences of left and right channels, which could easlily be done if the output of my polyphonic filter/delay/whatever could be sent to a single VCO-n input of the Unison module. Same left/right bundling then would be true for the CV’s and the XPand module.
But I can see this might not be trivial or possible since individual voices are not hard panned.
Ahh well the surge xt stereo delay should be useful for that since you can modilare the left and right delay separately! But yeah unison is not for building flangers - although I like the idea of a similar module for flanger chorus and phaser with bring your own delay line