Help Understanding This Phasing Patch

Anything modular synth related that is not format specific.

Moderators: Kent, luketeaford, Joe.

Post Reply
User avatar
Snufflepuff
Wiggling with Experience
Posts: 291
Joined: Sat Dec 19, 2020 11:53 pm

Help Understanding This Phasing Patch

Post by Snufflepuff »

I tried messaging the author and another person who replied to the post, but it was a few years old. . .maybe they've disappeared.
  • Today I learnt the value (again) of the simplest of utilities, such as attenuation and VCAs.

    Having a Sine wave feeding a Sample and hold that is just a bit slower than the clock of the sample and hold. So you get repeats, but those repeats become out of phase, so you get repetitions that evolve on a scale-level. Attenuating the Sample and hold voltage outputs leads to more repetitions (less choices for the Quantizer to quantize to). Then sending instead that through a VCA, I almost had a Turing Machine thing happening where I could voltage control the chance of repetition.

    The melody was nice too... should have recorded it all.
I'd like to try the patch too, but I'm struggling to understand it. In the time it takes for a sine wave to complete one cycle, the Sample and Hold outputs a constant voltage. During the next repetition, the S+H samples a slightly different part of the sine, and again outputs a constant voltage. What then are the "repeats" or "repetitions"? A Sample + Hold can't repeat a sine unless its clock frequency is nearly infinite.

And how does attenuating the S+H output change their frequency? And then the "Quantizer" shows up all of a sudden. Does that sound like an actual quantizer or is he referring to the S+H quantizing the continuous sine wave voltage? Thanks for your help!
User avatar
KSS
Super Deluxe Wiggler
Posts: 7656
Joined: Mon Jan 25, 2016 7:28 am

Re: Help Understanding This Phasing Patch

Post by KSS »

You're probably over-thinking it. I know ! was, at first read!

This is simply feeding the clock slightly out of phase with the input.

Consider if you fed the CLK at the same freq as the input. You'd get the same voltage every sample.

By changing the clock relative to the input frequency, you get samples that vary over time as they catch different portions of the input WF. SIN in the example, but there's nothing inherently special about it in this case. Except maybe its symmetry.

Over time the clock and input will 'match' phase and you get your repeats. This is no different than the way different outputs of a divider come to match up at times and not at other times.

Apparently the patch involves taking the S/H output into an attenuator and then to a quantizer.

As the attenuator scales the S/H output, portions of the sine wave are effectively cut off. They're clipping the output. You could instead attenuate -and offset!- the input WF for similar results. As they wrote, the smaller S/H output range means the quantizer has a smaller pool of choices, so the choices which do appear come up more often -relatively to the non-attenuated version.

The patch is a variation on the ascending and descending -arpeggiating- voltages you get when you feed a ramp or saw into an S/H.

you'll learn more in five minutes playing it than I could type in 30. And yet you can enjoy patches of this type for many years.
User avatar
Snufflepuff
Wiggling with Experience
Posts: 291
Joined: Sat Dec 19, 2020 11:53 pm

Re: Help Understanding This Phasing Patch

Post by Snufflepuff »

You're a life-saver, KSS. I didn't understand that the whole point was producing a pitch sequence, and that repetitions meant within that sequence, not between the sine/sequence pairing. I drew it out, if others want a visual.

https://ibb.co/cQSh0Bb. (Image embed is not working for me).
Joe_D
Common Wiggler
Posts: 202
Joined: Sun Sep 01, 2019 11:09 pm

Re: Help Understanding This Phasing Patch

Post by Joe_D »

Isn’t this basically aliasing near twice the Nyquist frequency but at sub-audio rates? Manifest as more or less repeating longer pitch CV patterns.

Edit: referring to Snufflepuff's diagram, which shows the sampling rate at approximately twice the frequency of the sampled wave.

Snufflepuff, your diagram shows two samples per cycle of the sampled wave, not one. You can cut the sampling frequency in half to get different effects (presumably as the OP did, and as KSS's write-up suggests), which will be more of a stair-step. Either way can yield interesting results.
Last edited by Joe_D on Sat Jan 15, 2022 2:30 am, edited 2 times in total.
User avatar
BaloErets
Super Deluxe Wiggler
Posts: 1421
Joined: Sat Jul 30, 2016 7:00 pm
Location: Montreal, Quebec
Contact:

Re: Help Understanding This Phasing Patch

Post by BaloErets »

Snufflepuff wrote: Fri Jan 14, 2022 1:58 pm I tried messaging the author and another person who replied to the post, but it was a few years old. . .maybe they've disappeared.
  • Today I learnt the value (again) of the simplest of utilities, such as attenuation and VCAs.

    Having a Sine wave feeding a Sample and hold that is just a bit slower than the clock of the sample and hold. So you get repeats, but those repeats become out of phase, so you get repetitions that evolve on a scale-level. Attenuating the Sample and hold voltage outputs leads to more repetitions (less choices for the Quantizer to quantize to). Then sending instead that through a VCA, I almost had a Turing Machine thing happening where I could voltage control the chance of repetition.

    The melody was nice too... should have recorded it all.
I'd like to try the patch too, but I'm struggling to understand it. In the time it takes for a sine wave to complete one cycle, the Sample and Hold outputs a constant voltage. During the next repetition, the S+H samples a slightly different part of the sine, and again outputs a constant voltage. What then are the "repeats" or "repetitions"? A Sample + Hold can't repeat a sine unless its clock frequency is nearly infinite.

And how does attenuating the S+H output change their frequency? And then the "Quantizer" shows up all of a sudden. Does that sound like an actual quantizer or is he referring to the S+H quantizing the continuous sine wave voltage? Thanks for your help!
Hey! Yeah that was my post I wrote awhile back. Apologies, I need to update my profile as the email you sent was linked to an email I don't use all that often.

I have nothing to add as KSS explained it to perfection :tu:
Post Reply

Return to “Modular Synth General Discussion”