MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler Blog & NewsMW News   Muff Wiggler StoreMW Store 
 SearchSearch   RegisterSign up   Log inLog in 
CURRENTLY BROWSING IN GUEST MODE
PLEASE LOGIN TO ACCESS FULL FUNCTIONALITY

Granular euro - godsend, or largely pointless?
MUFF WIGGLER Forum Index -> Eurorack Modules Goto page Previous  1, 2, 3  Next [all]
Author Granular euro - godsend, or largely pointless?

Veqtor

I'm quite fond of the idea of having an actual modular. I could never replace it with a computer. Perhaps it is that I work with computers all the time and want to get away from them when I create. There's something about a computer screen that kind of makes it not-leisure anymore for me. I went from reaktor to REAL modulars so... I'm never going back. w00t


Monobass

yeah me too.. although If I could easily encapsulate a PD or Max patch into a standalone module that would be awesome...


Drumdrumdrumdrum

Andrino and Max?

I'm still a bit naive about all these open source programs. It all seems a bit too geeky for me. I'm researching the whole idea but it seems like a bottomless pit of research. How far does one go before they can use it for compositions?

Granular effects just take me to outer space. I have to have it in my music, there is no doubt.


Soy Sos

The Fragment Generator is not really granular in the way people are discussing, as in chopping up bits of digital audio. I just got a Tyme Sefari a couple of days ago and am having a lot of fun with it. I know it's not granular either, but it is letting me crudely shuffle and scan through small 8 bit audio samples. The Max 4 Live instrument Granulator is the shit, I doubt we'll see anything that huge in a module any time soon. I'd love something between those 2 extremes at some point for sure!

DGTom (the designer) describes the Fragment Generator:

I could call it a VC Electro Boogie Tom Fill module. . . but I don't think that will help much

its not digital but it can sound very digital.

it isn't a filter, tho it _contains filters_

it's more indebted to the Polyrhythmophone than the 258, but, has more in comman with the TR 808 than either of those.

it is, indeed, a PITA to explain without getting bogged down in technicals, but I'll do my best.

Its a 'voice module' that has VC over Frequency, Amplitude & Timbre built around 3 high Q (ringing) filters which we've called "fragments"

How often, how loudly, at which frequency & at which relative position each fragment 'rings' makes up the sound at any given moment in time.

It can sound like a 500bpm electro breakbeat, like the secret attic offspring of granulab & a CZ-101, make throaty, resonant filter sweepy sounds & insane subsonic, almost DC 'is this shit even on' speaker wobbling waveforms.

its my attempt at Marvel Comics "WHAT IF. . . Granular Synthesis Was Analogue!!?!?!" & is best used while wearing a cape with your undies on the outside of your pants for that very reason!


Monobass

Drumdrumdrumdrum wrote:
Andrino and Max?

I'm still a bit naive about all these open source programs. It all seems a bit too geeky for me. I'm researching the whole idea but it seems like a bottomless pit of research. How far does one go before they can use it for compositions?


Max, it was about 1-2 hours for me, Arduino maybe more like a day.

How far did you have to go with a modular synth before you used it for a composition?


tomerbe

if i have time i'll put together some sort of granular patch this weekend. though i'll be "cheating" as i'll be using my es3 & PD to provide the grain envelopes and CVs with random deviation. i only have enough modules to do 2 or 3 grains at once (3vcas, 3 vcos), but enough to be somewhat interesting....


amalthea

Drumdrumdrumdrum wrote:
phase ghost wrote:
I'd say largely pointless. However, cv into the computer (via Silent Way) to control parameters of a granular synth plugin would be cool.


Well, this is getting OT Euro, but what good Granulators are being used? Really smooth ones that are Hi Hi Fi? I only know of the Granular EFX in Omnisphere that is really good. All the freeware ones like Granulator, just don't sound as Hi Fi to me on the demos.


I used to use the Martin Brinkmann's reaktor ensemble Grainstates all the time, and it's quite good. Another good is White Grains, although it's harder to control. I've also just started using Camel Audio's Alchemy synth for granular stuff, and it is amazing. Really high quality, and gives you much more control than the granular function in Omnisphere (as well as the all-important ability to use your own audio material).

-Andreas


Drumdrumdrumdrum

Why have I not heard about Alchemy before now? That is one slick VST! Im particularly interested in using this for resynthesis. They seem like a fun bunch of geek programers too.


thaneco

Reaktor is an amazing program with endless possibilities. Using both modular and reaktor is the best way for me (like for example passing the audio out from my modular to reaktor's granular effects). But I have to agree that watching a computer screen is a little tiresome lolspew


Drumdrumdrumdrum

Not to dis Reactor, but it always sounds like Reactor....if you know what I mean? Im starting to feel the same about Omnisphere too. Can pick it from a mile away on a movie soundtrack or nature doco.

I agree that a balance between VSTs and modular is the key. But balancing can be very tricky.

If some of these companies like Livid could get some real hands on interfacing between modular and VST then................

MY ASS IS BLEEDING


Dr. Sketch-n-Etch

So, I'm a bit of a noob with this granular stuff. Is it basically just switching little snippets of sound in and out really quickly? For example, say that you've got two sources of audio (say, Bach's "Mass in B Minor" on Channel 1 and the Sex Pistols' "God Save the Queen" on Channel 2) going into a crossfader (like, say, the Intellijel uFade -- TM 2011, All Rights Reserved) which is being controlled by a pulse wave at 1 to 100 Hz. Now, let's say you modulate the pulse width and frequency of the pulse wave. Would this essentially be granular synthesis?


tomerbe

you are somewhat there.

granular synthesisis a pretty open idea, but usually implies randomization of parameters for each grain, possibility of randomization of grain size and start time. and a variability of grain density, so that you might have 0 to say 32 grains (no real top limit) playing at the same time.

check barry truax's riverrun for an early example



here's a simple granular csound sampling piece i did a while back

[s]http://soundcloud.com/tomerbe/i-can-hear-my-heart-singing[/s]


Mark Landman

Dr. Sketch-n-Etch wrote:
So, I'm a bit of a noob with this granular stuff.

Here's the place to start, http://en.wikipedia.org/wiki/Granular_synthesis

Typically it involves sample granulation, but some fantastic soundscapes can come from using simple single cycle waveforms too. This by nature calls for digital implementation, an analog version w/ sufficient grains seems unlikely.


Johnisfaster

How could it possibly be pointless? I put modular samples into software granular apps all the time and it sounds bitchin. To be able to do it on the modular with CV control would be amazing.

A simple sine sample can get really crazy, it doesnt have to be a complex audio source to get really complex results.


amalthea

Drumdrumdrumdrum wrote:
Why have I not heard about Alchemy before now? That is one slick VST! Im particularly interested in using this for resynthesis. They seem like a fun bunch of geek programers too.


Then there's also of course Steinberg's upcoming Padshop, which looks like it has a ton of control and functionality, but unfortunately (as far as I can tell) no way to use your own audio samples (yet?).


http://createdigitalmusic.com/2012/02/steinberg-padshop-coming-soon-gr anular-synthesis-for-the-rest-of-us-handy-intro-video-explains/


mrcharles

I've been exploring granular synthesis software for almost a year now... in addition to the Reaktor ensembles, here are a couple of applications that I recommend.

In particular, Density ( http://www.densitygs.com/ ). This comes as a standalone application or in Max4Live versions. It is a very powerful application that can morph between different "snapshots" of grain parameters, has built in reverb, compression, and a multi-band filter equalizer.

Audiomulch ( http://www.audiomulch.com/ ) has a number of granular synthesis tools in its library of contraptions.

I'm new to the modular world (trying to figure out just what my first configuration will be...), but I wonder just how analog a modular granular synth could be... it would have to handle an awful lot of data.

When you combine granular audio and video... well, it is out of this world... see:




tomerbe

here's my attempt to do granular synth on my modular - this is fairly minimal, just two vcos through two vcas. think i'll try to expand on this....

grain envelopes and pitches are generated by PD and sent through the ES-3

https://vimeo.com/37874001


ianross

Dr. Sketch-n-Etch wrote:
So, I'm a bit of a noob with this granular stuff. Is it basically just switching little snippets of sound in and out really quickly? For example, say that you've got two sources of audio (say, Bach's "Mass in B Minor" on Channel 1 and the Sex Pistols' "God Save the Queen" on Channel 2) going into a crossfader (like, say, the Intellijel uFade -- TM 2011, All Rights Reserved) which is being controlled by a pulse wave at 1 to 100 Hz. Now, let's say you modulate the pulse width and frequency of the pulse wave. Would this essentially be granular synthesis?


If that's what granular synthesis is then your new mixer is basically a granular synthesizer. : )


Dr. Sketch-n-Etch

ianross wrote:
If that's what granular synthesis is then your new mixer is basically a granular synthesizer. : )

Yes, I guess it is! What I'm curious about is if two (or more) recognizable audio sources are sequenced rapidly, are they still recognizable? Does it sound like two audio sources playing at the same time, or something else entirely? I guess we'll be able to do this kind of experiment with the Mutamix.


radiodread87

could you also do this with the Ufade? as in have two sound sources play into it and switch between them at audio rates?


earlykooka

ianross wrote:
Dr. Sketch-n-Etch wrote:
So, I'm a bit of a noob with this granular stuff. Is it basically just switching little snippets of sound in and out really quickly? For example, say that you've got two sources of audio (say, Bach's "Mass in B Minor" on Channel 1 and the Sex Pistols' "God Save the Queen" on Channel 2) going into a crossfader (like, say, the Intellijel uFade -- TM 2011, All Rights Reserved) which is being controlled by a pulse wave at 1 to 100 Hz. Now, let's say you modulate the pulse width and frequency of the pulse wave. Would this essentially be granular synthesis?


If that's what granular synthesis is then your new mixer is basically a granular synthesizer. : )


it isn't what Granular Synthesis is at all. Not even close.
To simulate Granular Synthesis in this way you would not necessarily need two soundsources. One is sufficient ( though, depending on the GS program, you can choose to run a number of granular "streams" that are granulating different audiofiles at once : different apps or plugs let you run one, eight or as many as your computer can handle. For the moment lets stick to just one. To simulate GS with a mixer and a record, first you would need to be able to scratch the record with a precision and a speed several orders of magnitude greater than any human, firstly to access the fragments that you want, then to tune them to precise intervals, or randomly, or changing according to waveforms that you are also changing in realtime, then you would need to do this on a vast number of records at once, then be able to change how many records there are. in real time, inhumanly quickly and inhumanly precisely, then you would need to be able to modulate your tens or hundreds of mixer channels with AR times in the milliseconds and AR shapes of incredible precision. Then you would need to do all of this within 'windows' ( from which the grains are drawn randomly, semi-randomly, centre-weighted or whatever) that are changing position and size constantly according to other functions, except when they aren't. Good luck mate ;-). I would advise you,to check out Density GS which is an absolutely fantastic GS app and also exists as an M4L plug, as previously mentioned. It is quite simply phenomenal.


rowman

radiodread87 wrote:
could you also do this with the Ufade? as in have two sound sources play into it and switch between them at audio rates?


Just trying this with the xfade of the Korgasmatron and it works at a moderate audio rate at least.

Am I right in thinking that the 'modular granular' goal is as simple/difficult as having your vst GS take automation from an input CV?

for example:
'Silent Way CV To OSC translates CV signals (such as those produced by Silent Way CV Input, for example) into OSC messages. These can in turn be used to control all sorts of software and hardware devices.'


edit: woops missed this
phase ghost wrote:
However, cv into the computer (via Silent Way) to control parameters of a granular synth plugin would be cool.


also incorporating something like this https://www.muffwiggler.com/forum/viewtopic.php?t=42417
could make your favourite granular softsynth a physical part of your modular, as long as it can recieve OSC code.
But don't patch actual audio or CV into it, you have to use Expert Sleepers /ADDAC for that so I guess it is a little clumsy.


Dr. Sketch-n-Etch

radiodread87 wrote:
could you also do this with the Ufade? as in have two sound sources play into it and switch between them at audio rates?

Yes. The uFade is perfect for audio-rate switching like this, because the V2164 chip is very fast. If the input signal is attenuated to a lower level (say +/-1V) and then amplified after the crossfader, then it would be even faster.


Dr. Sketch-n-Etch

earlykooka wrote:
it isn't what Granular Synthesis is at all. Not even close. <...snip...>

OK. Based on what you're saying, GS is only possible in the digital domain. That's cool. I'm not really interested in it, personally, but that's just me.


Dr. Sketch-n-Etch

rowman wrote:
Just trying this with the xfade of the Korgasmatron and it works at a moderate audio rate at least.

Warning... Technical language to follow.

According to the datasheet, the slew rate of 2164 is 700 uA/us. Typically, the output is converted through a 30k resistor. Hence, a 10V peak-to-peak signal will require 333 uA of current, and this will take 333/700 ~ 0.5 us = 500 ns. The half-period of a 10 kHz signal is 50,000 ns. Hence, the rise time of a perfect square wave fed at 10 kHz would be 1% of the duty cycle. This means that the 2164 should easily handle audio rate switching of signals at synth voltage levels with little distortion.

Of course, there are other chips in the signal chain which are slower, such as the opamps. The TL07X which converts current from the 2164 to voltage has a slew rate of 13 V/us (typically) with minimums as low as 8V/us, which means that it requires about 1000 ns for a 10-V shift, and this would require 2% of the duty cycle for transitions of a perfect square wave at 10 kHz. This is still pretty fast, as general-purpose opamps go.

This analysis also ignores response time (the time it takes for the chip to realize that it's supposed to do something) which can be about 1 us in both cases. This, again, is a pretty small number in terms of audio-rate modulations.

MUFF WIGGLER Forum Index -> Eurorack Modules Goto page Previous  1, 2, 3  Next [all]
Page 2 of 3
Powered by phpBB © phpBB Group