A proposal: "Polyphonic Inhomogeneity"

TL/DR: I would love to see a C/V-controllable Inhomogeneity- / Anomaly- / Irregularity- / [Whatever-it-is-called in english]-Parameter on polyphonic modules, which slightly alters the other parameters per voice.

Story: My constantly changing interest in outboard gear (Midlife-GAS) has recently shifted from “small modular system” to “polyphonic analog synth”. And with christmas around the corner I am on the brink of ordering a Sequential Take 5. As I would do normally I started to build the synth engine in VCV to better understand the modulation options and the overall sound design possibilities without owning it (sounds strange, but helped me a lot with my decision to buy a digitone). One feature of the Sequential Take 5 (and the Prophet 5/10 reissue) is the vintage knob, which is not only a slight detune between each individual voice, but also alters the filter and envelope parameters per voice.

J3PO shows a very nice example here on the prophet 10:

While building the patch I stumbled across this feature and essentially had to restart. And there is a lot to research when it comes to polyphonic CV for a module: Is the CV-Input polyphonic? Is the Position of the Parameter on the module corresponding to 0V or 10V on the CV-Input? Bipolar? After that you have to find out how to insert offsets into polyphonic CV and how to control the offsets (aka “turning the vintage knob”).

I know that rebuilding a synth is not the “purpose” of eurorack, but we do have polyphony, right? And with VCV as VST around the corner, I think a few people will build synth voices for use in a DAW with VCV. I don’t know how difficult such a feature would be to implement, in my simple understanding of math it seems to be something like “Knob Value” x (1 - (“Vintage Value” x “Max. Deviation”)) where Vintage Value is [0 - 1] and “max. Deviation” is perhaps something like [0, 0.1, -0.1, 0.03, -0.03, 0.07, -0.07, 0.13, -0.13, …] (one value for each polyphony-channel).

Does this sound interesting? Is this a feasible proposal? Has anybody realized something like that in a module already?

Thanks for reading and your opinions!

  • mo
1 Like

There’s this, that sounds a bit like what you are looking for :

2 Likes

Yes!

This one sets a slow changing random variation on each channel of the 16-poly V/OCT.

Spread on the ML module is divided with 100 and used for the “scale” on the “Walk” module. These values are shown in the “Multivoltimetro”, controls the “CV-Map”, controlling the 16 “Affix” knobs (5V input corresponds to the middle position, output = input), used to add offsets to the “poly” V/Oct input. This gives a varying detuning of 16 oscillators.

  • Speed of the variations can be adjusted with “Walk” → “Rate”
  • Amount of the variations can be adjusted with the “Cloner” → “Spread”

It would take a lot of modules to make it work for all parameters (if available for either CV or CV-MAP) .

Perhaps some new modules will appear that makes it possible with fewer modules.

vintage_tuning.vcv (14.4 KB)

edit: I just realised that there is no variation on channel 1

2 Likes

This is probably better, using the rate input on the walk to set the polyphony.

“WALK” → “SCALE” sets the amount of drift.

I added drift on the filter cutoff frequency too.

vintage_tuning_2.vcv (23.4 KB)

and a variant with reduced module count: vintage_tuning_3.vcv (19.1 KB)

I probably won’t buy any Sequential synths - but when the Behringer Pro-800 can be bought, I’m probably going to get that. I allready have their Model-D, K-2 and Neutron.

Behringer-Pro-800

1 Like

@stoermelder was making a module called “dirt” with cross-talk between polyphonic cables, which is related to what you’re talking about

If actually want to emulate cross-talk between the channels of a polyphonic cable, you could use a SubmarineFree LT-116

1 Like

Thank you all for your suggestions. But perhaps this was not clearly enough explained by me. I would love to see this in future polyphonic modules directly integrated not as a separate module. That’s why I posted this in the Development Section and not in Modules.

I think it would be so much easier to have this directly as an option on VCOs / VCAs / VCFs and Envelopes.

My current workaround for this requires a parameter which has a polyphonic CV-Input, an extra Vult KNOBS and something like the LindenbergResearch Voltage Controlled Signal Spread - for each parameter!

I would like to simply set the Release of an Envelope with the Button on the Envelope Module and the actual release slightly differs from voice to voice, depending on a “vintage-amount” (either in the context-menu or even better set via CV-Input.

Before I start to open GitHub-Issues / -FeatureRequests on all my favorite modules I thought I ask here, if this would be something interesting.

But again, thanks for your suggestions! Sequential-Patch is coming to life. Next up is the task of finding a Filter…

  • mo

midi-cv —> polyphony 16 channels

submarine LT-116 —> right click on the background of the module (not on knobs) —> presets —> identity

doing so you have everything perfectly tuned, so now is the moment to modify another knob for every column…for example, on fist column I have “tuned” another knob on 0.001, on the second column -0.016, on the third 0.003, on fourth -0.014 and so on.


doing so, the result was a little too “drifted”, so I put a “nysthi scale module”, set to 0.993, and the result is a little bit less drifted…furthermore with this “scale module” it’s easier to set a kind of “global drift” with only 1 parameter


hope it helps!

cheers :broccoli: A


You could accomplish this a bit more succinctly with Computerscare Knoly Pobs:

Set the little “scale” knob to a small value, and it will scale the polyphonic knobs accordingly. The larger “offset” knob on top would be used for the main CV value. When “scale” is set to zero, the output of all polyphonic channels will be identical to the “offset” knob.

1 Like

Try Bogaudio Polycon:

Looking forward to hearing what you come up with in Rack. Maybe you will save some money :slight_smile:

Here’s my take on it: this shouldn’t be built into Rack itself, but clever developers would build per voice offsets as options into their modules.

Example:

  1. Polyphonic Oscillator module takes in a polyphonic CV signal and has an option to apply per voice offsets or maybe even has a “Slop” type of knob on the GUI to control depth.

  2. Polyphonic Filter module does the same, only with offsets for Cutoff, Resonance, CV depth…again, can be a dropdown option or can be a Slop or Offset knob.

  3. Polyphonic Envelope module applies per voice offsets to attack and release times…

…Different developers may implement things differently. For instance, in polyphonic synths with decent tuning, it’s more often an offset than a completely random tuning thing (though the tuning of an osc may still drift). On my Prophet-10 Rev 4, you really hear the filter cutoffs and envelope offsets quite audibly. With the Prophet’s voice assignments, sometimes you’ll just get a bum note repeating every time you play it because maybe voice 6 is getting assigned to B4 and that voice is a little flat but the envelope is a little longer than the others so it really sticks out.

Anyway, I really think it should be up to the developers to figure out ways to do this (if they so chose) or for users to replicate this via modules as has been proposed here. Just my two cents!

3 Likes

another easy way to “drift” easily, different from the 1st method I have posted: this one “moves” constantly the voltages:

midi-cv —> polyphony 16 channels

computerscare Knoly Pobs —> nothing to touch/set

Bogaudio Walk —> rate 0% / offset 0.0001 / scale 0.12345 (to be set as you want, of course)

Studio Six Plus One Eva —> nothing to touch/set


of course you can mix the 2 methods I have explained…I have already done it

hope it helps!

cheers :broccoli: A

1 Like

Some of the proposed methods will work for offsets and/or drift on Oscillators and VCF’s, but are there any envelope modules that would allow for polyphonic per-voice offsets?

Here’s a “16x vintage envelope”

vintage_envelope.vcv (25.7 KB)

Very nice Jens! Thanks! Illustrates how to do it with existing modules and also why it would be so nice if developers built in similar functionality to the modules themselves! :slightly_smiling_face:

That is a nice setup, but with one misconception. The “Vintage Factor” is not modulation. If you think about the voice cards in an analog synth: There is one card, that is always 3 cents sharp and has an envelope with a 5% longer decay and release, while the filter is 4% less resonant as the next one. So the deviation is always static. And that has a very musical effect with let’s say a single note played on a six voice analog synth where every note is assigned to the next polyphonic channel / voice card. There is a pattern that is repeated every sixth gate or every third gate when you play simple intervals or every second when you play a triad.

LindenbergResearch Spreader does exactly that, the spread is static. And the deviation amount can be set with attenuated CV and you can even reseed the deviation if you want to change the pattern. I really like that and would really love to see this integrated in a few polyphonic modules to reduce the clutter and to be able to use the dial on the module itself.

This is my approach at the moment for Decay and Release on both Envelopes and Cutoff and Resonance on the Filter.

I don’t own vintage synths - but drift ? Isn’t that a temperature co-variance of RLC-circuits and transistors, vacuum tubes etc. and not just a circuit sligtly off calibrated ? And why more recent analog synths do (user initiated) auto-calibration.

From my understanding Oscillator Drift is temperature-related and occurs on every analog synth, but the vintage control on sequentials synths is based on the calibration of different voice cards and their parts degeneration over time and not temperature related. But I may be wrong here. This is from the Manual of the Take 5: “To add even more vintage character, we designed the vintage knob, which introduces parameter variations from voice to voice.”

"VINTAGE KNOB

Recreates the characteristics of vintage synthesizers by introducing micro-fluctuations in oscillators, filter, and envelopes per voice."

https://modwiggler.com/forum/viewtopic.php?t=252254#p3588839

I don’t know exactly what it does - not open source.