MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

how to create those sounds (François Bayle, Rashad Becker..)
MUFF WIGGLER Forum Index -> Modular Synth General Discussion  
Author how to create those sounds (François Bayle, Rashad Becker..)
ketem13
I encounter this amazing piece by François Bayle called Tremblement de terre très doux.

I was wondering how you can synthesise those fm(?) sound types appear in minute 07:30 and throughout the all piece.
https://www.youtube.com/watch?v=eDctfUyGI1c

I also found it similar to some sounds Rashad Becker makes in is album Traditional Music of Notional Species Vol. II
(for example - https://www.youtube.com/watch?v=neemYTM5mNg)

and also in some Ben Vida work such as Damaged Particulates minute 05:55


any leads to that direction of sounds ?
Navs
I would guess it's more of what you have already been investigating: feedback/cross-modulation.

The sounds you linked reminded me of this:
https://navs.bandcamp.com/track/buchlasnore

The 'groaning/indigestion' sounds change in timbre as well as pitch. You can do this by self-modulating a sinewave to get something resembling a saw. Put the auto-modulation through a VCA and you can control the depth ... by itself, or another oscillator etc.

All of this is just mucking about really, just patch and twiddle until you find something nice, then document what you did so you can repeat it; it's usually more fun than goal-oriented patching smile
ketem13
Navs wrote:
I would guess it's more of what you have already been investigating: feedback/cross-modulation.

The sounds you linked reminded me of this:
https://navs.bandcamp.com/track/buchlasnore

The 'groaning/indigestion' sounds change in timbre as well as pitch. You can do this by self-modulating a sinewave to get something resembling a saw. Put the auto-modulation through a VCA and you can control the depth ... by itself, or another oscillator etc.

All of this is just mucking about really, just patch and twiddle until you find something nice, then document what you did so you can repeat it; it's usually more fun than goal-oriented patching smile


Hi and thanks for the suggestions. edit: I just listen to your stuff in the link you put and it is much a like! so it is a self-modulated sinewave or something else? (as much as you can recall)

I'm not quite sure what it is, or how you self-modulate sinewave? could you please expand on that ?

definitely will also just patch around smile
Dave Peck
Yes, sounds like a lot of self-modulation and cross-modulation at audio rates.

Also, some timbres seem like they may be using a dual band pass filter
- resonant LPF and resonant HPF in series, with the LPF cutoff set somewhat higher than the HPF cutoff, each filter modulated separately, to create complex vowel-ish sounds. The audio source being fed through the filter can be a saw or pulse wave that is set to a very low frequency and has some pitch mod that moves it just above and below the audio range, around 20 Hz, so sometimes you're hearing this audio osc, sometimes you're mostly hearing the resonant filter 'pings' when the osc pitch drops lower. This technique can be heard on some Robert Rich tracks, especially on the CD "Bestiary".
Navs
I need to look at the patches again for the specifics, but I mean just patching the audio output back into its FM input.

If you have a quadrature oscillator, patching the 90 degrees output back to the FM input should bend the sine into something sharper sounding. Try it with the Toppobrillo Multifilter set for self-oscillation if you don't have a quadrature VCO: try feeding the BP or HP output back into the CV input.

Even a 'normal' oscillator and exponential FM will give you some changes. Feed it back via an attenuator or polarizing attenuator to get different shapes, in the same way you would with an envelope like Maths:

https://navsmodularlab.blogspot.com/2010/04/patch-tips-8-more-feedback .html

It's peanut butter jelly time!
ketem13
Navs wrote:
I need to look at the patches again for the specifics, but I mean just patching the audio output back into its FM input.

If you have a quadrature oscillator, patching the 90 degrees output back to the FM input should bend the sine into something sharper sounding. Try it with the Toppobrillo Multifilter set for self-oscillation if you don't have a quadrature VCO: try feeding the BP or HP output back into the CV input.

Even a 'normal' oscillator and exponential FM will give you some changes. Feed it back via an attenuator or polarizing attenuator to get different shapes, in the same way you would with an envelope like Maths:

https://navsmodularlab.blogspot.com/2010/04/patch-tips-8-more-feedback .html

It's peanut butter jelly time!


Thanks for the explanations and tips! will try those out.
btw the audio example in the link you put unfortunately are not working
ketem13
Dave Peck wrote:
Yes, sounds like a lot of self-modulation and cross-modulation at audio rates.

Also, some timbres seem like they may be using a dual band pass filter
- resonant LPF and resonant HPF in series, with the LPF cutoff set somewhat higher than the HPF cutoff, each filter modulated separately, to create complex vowel-ish sounds. The audio source being fed through the filter can be a saw or pulse wave that is set to a very low frequency and has some pitch mod that moves it just above and below the audio range, around 20 Hz, so sometimes you're hearing this audio osc, sometimes you're mostly hearing the resonant filter 'pings' when the osc pitch drops lower. This technique can be heard on some Robert Rich tracks, especially on the CD "Bestiary".


thanks!
you think I can achive those kind of sound characters with Toppo multifiter and the Epoch Twin peak ?
Dave Peck
The Twin Peak should work well for these effects. The Multifilter is only one filter with only one frequency peak so you would need a second filter in series with it for a second peak. But yeah, you could still get some sorta-similar effects with just a single LPF or BPF with high resonance and an audio osc source that is dipping into and out of the low end of the audio range, going from a low tone to a series of resonant pings. But having the two filters in series (either two BPFs or one HPF and one LPF) will give you more vowel-ish sounds.
ketem13
Dave Peck wrote:
The Twin Peak should work well for these effects. The Multifilter is only one filter with only one frequency peak so you would need a second filter in series with it for a second peak. But yeah, you could still get some sorta-similar effects with just a single LPF or BPF with high resonance and an audio osc source that is dipping into and out of the low end of the audio range, going from a low tone to a series of resonant pings. But having the two filters in series (either two BPFs or one HPF and one LPF) will give you more vowel-ish sounds.


Hi
when you say filter in series, you mean to fid the bandpass put of one filter to the in put of the other filter also in bandpass?
Dave Peck
Aha, I did not word that correctly.

If you are using a HPF and a LPF, they should be in SERIES. Patch the audio to the input of the HPF, patch the output of the HPF to the input of the LPF. Then set the cutoff of the HPF low and the LPF high. This creates a variable bandwidth BPF that has a resonant bump at both the high and low cutoff frequencies.

If you are using two BPF filters, they should be in PARALLEL and their cutoffs should be set with one higher than the other (doesn't matter which one is high and which one is low). Send the audio to the inputs of both filters. patch the two filter outputs to a mixer and combine them, creating a PARALLEL filter configuration. This creates a somewhat similar response to the first type, with two resonant peaks (Like the Hordijk / Epoch TwinPeak module). But, unlike the first example, the signal in the frequency range between these peaks will be filtered out. With the HPF/LPF/SERIES setup, the audio between the peaks does NOT get filtered out. But they can both create these kind of vocal-ish timbres when you modulate the two filter cutoffs separately.
Navs
ketem13 wrote:
btw the audio example in the link you put unfortunately are not working


Yes, several clips have been 'lost' over the years as Google, Dropbox etc. changed directories or the ability to use the Reader etc. I have stopped updating the links but I just had a listen - this one attached is probably the most relevant as you can hear the sine being bent and becoming more buzzy.
pixelmechanic
I've got some of those sounds here...

[s]http://soundcloud.com/pixelmechanic/jr-1019-131113-wav[/s]

Made with (one of) my Nord MicroModular(s)



Key things here are:

1. 4 voices (but each one randomised a small amount)
2. Feedback and more feedback

Begins with two sine oscillators that are cross modulating frequency, one has a clocked random frequency, the other uses the envelope for glissandi effects, the envelope is also acting as VCA for the carrier

When each voices envelope ends I'm using a compare function to check it's finished then trigger new random values.

The VCA out is then feeding a mixer into SVF filter with high and low pass values fed back into mixer and I'm using the bandpass output for audio out.

Overall it means there's a variety of things here causing timbral shaping.

I've got other similar patches where I use the Vowel Filter instead (or as well)
ketem13
Navs wrote:
ketem13 wrote:
btw the audio example in the link you put unfortunately are not working


Yes, several clips have been 'lost' over the years as Google, Dropbox etc. changed directories or the ability to use the Reader etc. I have stopped updating the links but I just had a listen - this one attached is probably the most relevant as you can hear the sine being bent and becoming more buzzy.

Thanks sound good!! and thanks for your tip with the bandpass out into the cv of the MF it actually working pretty well ! I think I'm getting there smile so exited
ketem13
pixelmechanic wrote:
I've got some of those sounds here...

[s]http://soundcloud.com/pixelmechanic/jr-1019-131113-wav[/s]

Made with (one of) my Nord MicroModular(s)



Key things here are:

1. 4 voices (but each one randomised a small amount)
2. Feedback and more feedback

Begins with two sine oscillators that are cross modulating frequency, one has a clocked random frequency, the other uses the envelope for glissandi effects, the envelope is also acting as VCA for the carrier

When each voices envelope ends I'm using a compare function to check it's finished then trigger new random values.

The VCA out is then feeding a mixer into SVF filter with high and low pass values fed back into mixer and I'm using the bandpass output for audio out.

Overall it means there's a variety of things here causing timbral shaping.

I've got other similar patches where I use the Vowel Filter instead (or as well)


love this tune! thanks for sharing!
"Begins with two sine oscillators that are cross modulating frequency" - could you pls explain what does it mean two osc that are cross modulating frequency ? are they modulating each other fm input ?

"one has a clocked random frequency" - it means that a random frequency (lets say sampled white noise) goes to the 1v/oct input?
pixelmechanic
^ Not sure I'd describe it as a tune (or track) - It was just intented to document what the patch does - but thanks!

If you look at the image you should be able to trace the various feedback paths. Looking at it again I'm not actually crossmodulating the fm inputs of the two oscs.

The first oscillator is fm-ing the second oscillator, but in fact the second oscillator is amplitude modulating (rather than frequency modulating) the first, which at audio rates creates a bunch of other sidebands.

I'm feeding back the output of the vca (built into envelope here) to the fm input of oscillator one, AND also patched the envelope the
Fm input of oscillator one.

By four voices, that means that the micromodular is running this patch as four voices, and even though when I turn the knobs it affects each one in the same way, the random values will be different enough to create micro variations in each voice.

In particular I'm cv-ing various things in the envelope time, so each voice has micro-variation in timing giving it a granular feel.

It's these kinds of things that really draw me to working with the micromodular.

The other thing that is in use here is morph groups - where a single knob is mapped to many controls but in different amounts and ranges...

Yes, 'clocked' wasn't the right word to use, triggered is better. Each time the envelope finishes (per voice) it triggers a new random value (like sample and hold with white noise). In fact I'm doing this three times, one for vco and filter, one for the env gen attack time and one for panning.
ketem13
Much clear now! also with the Nord modular manual open aside smile

edit: I'm using modular synth with osc that doesn't have AM input.

Could you say that it will be the same has taking the out of the top osc into a dc vca input and the output of the second osc (that modulating the AM of the first one, like in the picture) to the cv input of the dc vca ?

then the output of the dc vca is actually the first osc AM by the second one ?

that will be right ?

thanks
pixelmechanic
Yes, that's right, VCO 1 out into VCA, VCO 2 out int VCA cv input will give you AM sidebands

srsly never
MUFF WIGGLER Forum Index -> Modular Synth General Discussion  
Page 1 of 1
Powered by phpBB © phpBB Group