Author |
Message |
Tusker
Joined: Feb 03, 2005 Posts: 110 Location: Texas
|
Posted: Thu Oct 05, 2006 3:29 am Post subject:
What are some methods of injecting randomness? |
|
|
What are some methods of injecting randomness?
Could we have a discussion on this?
The human ear values randomness in sound. A number of physical instruments have inadvertent complexities and randomness, merely because of the physical attributes of the insturment. For example a violinist (not having a fret board) will always miss the "true pitch" by some microscopic amount, sliding into the pitch the really want to play using the physical feedback loop (ear to mind to finger).
There are several acoustic elements in which randomness could be desirable. Things like pitch, or timbre or volume could be examples. Since the Nords are so open ended, I'd be more curious to hear about methods of creating, shaping and triggering random information, than in what we use it for (I use a NM1 but I thought this was a general topic for everyone).
Create - I use noise as a starting point typically. Are there are other effective sources of randomness within the Nords? I've thought of doing more sophisticated things like running a sequencer, sample and holding the output then running the result through a lag to smooth it . ... However I don't see the value in that over the simple noise generator. Opinion?
Shape - I would like to be able to shape the randomness more than I can today. Clearly control signals can be limited in range, lagged to reduce variability. But I'd like to get closer to being able to specify a probability density function. Has anyone used parametric eq (of white noise) or similar techniques to shape the random cloud?
Trigger - We have key on, sequencer (rhythmic) triggers, and purely machine derived things like sample hold or noise (continuous randomness). I am curious for your examples. For example to try to get pluck string nuance, I sometimes use a logic test on the output of a envelope to develop a control signal that I might use to change the tone (usually pulse width). When you pluck a guitar string that is already excited, you get a different timbre, and this help to model that a bit. I realize this is not true randomness, maybe complexity is a better word for it. Do you use any similar tricks?
I realize I haven't formed a true question. But I am free associating here, and if it triggers a thought for you, I would be glad to hear it. Are there some other considerations of randomness that I haven't framed in this "question"?
Thanks in advance,
Jerry |
|
Back to top
|
|
|
dasz
Joined: Oct 16, 2004 Posts: 1644 Location: victoria, canada
Audio files: 29
G2 patch files: 56
|
Posted: Thu Oct 05, 2006 9:22 am Post subject:
|
|
|
With the G2, I like to capture randmon voltages in a note sequencer (by recording them as notes). I can use several note sequencers and different sources - like LFO S/H or Rnd modules.
Wrt to triggering, I still use an event sequencer, but each step has a probability of being triggered (using a rndTrigg module).
/Dasz |
|
Back to top
|
|
|
cebec
Joined: Apr 19, 2004 Posts: 1098 Location: Virginia
Audio files: 3
G2 patch files: 31
|
Posted: Thu Oct 05, 2006 9:46 am Post subject:
|
|
|
One good way to capture and inject randomness in the G2 and the NM is to use a bunch of Level Amps to boost the noise from an Input module (i.e., the A/D converters). I've used this method and injected the noise into all sorts of places. I think it can have an aesthetically pleasing effect. |
|
Back to top
|
|
|
sheridan
Joined: Jun 05, 2005 Posts: 473 Location: London, England
Audio files: 27
G2 patch files: 60
|
Posted: Thu Oct 05, 2006 10:59 am Post subject:
|
|
|
In my sound design patches, I have had to use lots of random modules. Often, I would chain them, or have something else modulating the 'gain' of the random signal... that way, I could determine when and how random the modulation signal was. Usually, only tiny amounts of random variances are required for natural effects and I found that using a high speed random LFO with a sample and hold module gave useful results. _________________ Sheridan
Hear music and Nord Modular G2 patches and find out music production tips at
http://www.nitetimeproductions.co.uk |
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18197 Location: Durham, NC
Audio files: 212
G2 patch files: 60
|
|
Back to top
|
|
|
cebec
Joined: Apr 19, 2004 Posts: 1098 Location: Virginia
Audio files: 3
G2 patch files: 31
|
Posted: Thu Oct 05, 2006 12:01 pm Post subject:
|
|
|
Yeah, the recent Mandelbrot patches are loads of fun. |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Thu Oct 05, 2006 2:34 pm Post subject:
|
|
|
This one seems to focus more on structure, while here the focus seems to be more on bringing variation/liveliness into the sound (until Howard posted the mandelbrot link of course ).
Both are worthwhile I think.
Another interesting thing is to have LFO's or audio oscillators to modulate each other, this gives rise to a certain amount of unpredictability as well. _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
dasz
Joined: Oct 16, 2004 Posts: 1644 Location: victoria, canada
Audio files: 29
G2 patch files: 56
|
|
Back to top
|
|
|
Tusker
Joined: Feb 03, 2005 Posts: 110 Location: Texas
|
Posted: Fri Oct 06, 2006 1:11 am Post subject:
|
|
|
Thank you guys. You've given me a lot to chew on. Those are some nice ideas for creating and controlling random signals.
To respond to Blue Hell's interesting observation ...
Yup the question of "what the mind likes" is curious isn't it? It's seems to me that formal western music didn't choose a particularly organic "form" of composition around which to develop originally. More of a dramatic "form" actually. (Situation => Complication => Resolution). The whole idea of chaos with tendencies appears to be more reflective of nature and organic processes. I admit, I'm still playing with the dramatic form, hence my interest in subtle randomness. I am somewhat interested in sociological metaphors (elements coming together without apparent commonality yet finding purpose together, the element that does not fit which transcends and transforms the entire system, etc.) as a basis for form, also.
I'll take some time to study these patches. The Mandelbrots look delicious. I'll try to make the G2 demo work on my system. Thanks for filling my mind.
Jerry |
|
Back to top
|
|
|
dasz
Joined: Oct 16, 2004 Posts: 1644 Location: victoria, canada
Audio files: 29
G2 patch files: 56
|
Posted: Tue Oct 10, 2006 9:18 pm Post subject:
|
|
|
http://electro-music.com/forum/post-92668.html#92668
Another quasi random patch. All this talk of randomization has gotten me interested in the topic recently...
thanks for the inspiration ... Both of the patches on this topic were done as a result of our discussions ...
and my regards to mr. noodle - blue hell - Jan Punter...
/Dasz |
|
Back to top
|
|
|
davep
Joined: Jul 05, 2004 Posts: 467 Location: Oakland, CA
Audio files: 10
G2 patch files: 73
|
Posted: Thu Oct 19, 2006 4:19 pm Post subject:
|
|
|
A simple trick I often use for generating quasi-random control signals is to mix a couple of LFOs running at different rates, and mixed at different levels. Like LFO X set to .010 Hz at level 127 and LFO Y set to .13 Hz at level 75. It isn't a true random signal, but it's complex enough that you don't easily detect a pattern (assuming one rate is not a simple multiple of the other) and it allows you to create various 'behaviors' by tweaking the rates, levels, and waveshapes.
I first stumbled upon this on the NM1 when I needed a few more slow random LFOs and I was running out of DSP, and discovered that two slave triangle LFOs and a control mixer used less DSP than one random LFO. I also feel that it works better for many applications because the signal has a more constant rate of 'motion', whereas the random LFO tends to hover around zero and then make sudden swings to the extremes.
Another trick is LFO cross-modulation, using LFO X to FM LFO Y and using LFO Y to FM LFO X. By varying the base frequencies, waveshapes and the FM amounts you can get lots of different behaviors. You can then tap just one of the LFOs to use as your random signal, or mix them together and use the sum. Or add a third LFO so that X modulates Y, Y modulates Z, and Z modulates X. And if you do this in the audio range you get lots of different chaos effects.
Another trick I found was using a low frequency triangle wave with soft sync, so that it produces a rising and falling signal with a constant slope (rate), but changes direction at random intervals. Very useful when you want a slow sweeping effect with smooth constant motion, but still with an element of unpredictability. Thread & example patch is here:
http://electro-music.com/forum/topic-8949.html
DP _________________ Dave Peck |
|
Back to top
|
|
|
Tusker
Joined: Feb 03, 2005 Posts: 110 Location: Texas
|
Posted: Fri Oct 20, 2006 9:50 am Post subject:
|
|
|
Elegantly simple yet powerful trick with the LFO FM. Thank you. |
|
Back to top
|
|
|
davep
Joined: Jul 05, 2004 Posts: 467 Location: Oakland, CA
Audio files: 10
G2 patch files: 73
|
|
Back to top
|
|
|
dasz
Joined: Oct 16, 2004 Posts: 1644 Location: victoria, canada
Audio files: 29
G2 patch files: 56
|
Posted: Fri Oct 20, 2006 1:37 pm Post subject:
|
|
|
Dave,
That's exactly what I was thinking. I just did a patch prior to seeing your patch. They are quite simmilar. I like to also use sequencers (with rndtrigger) to trigger multiple envelopes at random depths ....
Nice meeting you in person at NAMM!
/Dasz |
|
Back to top
|
|
|
|