Author |
Message |
Rob
Joined: Mar 29, 2004 Posts: 580 Location: The Hague/Netherlands/EC
G2 patch files: 109
|
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18195 Location: Durham, NC
Audio files: 211
G2 patch files: 60
|
Posted: Wed Apr 07, 2004 7:06 pm Post subject:
|
|
|
I like this. I studied it visually trying to understand it. That's a challenge.
I found I like it better with the reverb wet/dry mix reduced. The reverb moves the sound back and obscures a lot of the high frequency events, at least to me. Tell me, does the patch set it back after a while?
Does this change variations automatically? |
|
Back to top
|
|
|
Rob
Joined: Mar 29, 2004 Posts: 580 Location: The Hague/Netherlands/EC
G2 patch files: 109
|
Posted: Thu Apr 08, 2004 1:49 am Post subject:
Chaotic patterns |
|
|
It does not change variations. In fact the PM osc at 0 Hz is used as a nonlinear waveshaper. What happens is that the osc receives a seed value at the same time it is synced, setting the osc to a certain initial phase position at the syncing moment. The feedback on the osc will quickly make the output go to an attractor, which reveils itself as a 'pseudo random' pattern of random length and random form. When this pattern is very short, like a couple of samples, it creates a steady tone for a while, when the pattern is long it sounds more like some assembly of little motors/machines.
The fun of this type of thing is to find formula's that have a lot of attractors. The time it takes before the varying signal 'arrives' at the attractor (where the signal becomes periodic) and the length of the periodic cycles at the attractor are of interest here. When both are about a couple of hundred to a couple of thousand samples long it can create nice raw material to be processed by filters, resonators, etc. E.g. in this patch the vocoder is used to give the raw sound a slightly organic, sometimes even watery texture.
There is something fishy going on with using the PM osc this way. Theoretically, when set to 0 Hz there should be a predictable attractor for each seed, but the seed stays the same in this patch and still there are different patterns all the time. I suspect the anti-aliasing seems to slightly change the value of the seed on every reset.
It all has to do with how the noise behaves that is created with feedback of a PM oscillator. The noise appears as it is a nonlinear system, and so there must be attractors. Still, PM noise in general does not generate periodic signals, as the basic frequency of the oscillator forces the osc to a different attractor on each sample. Setting the frequency to 0 Hz eliminates this effect, but then a seed must be inserted at the phase input to get the thing going and generate periodic signals.
If the osc is not reset through the sync input, at which time the seed must be inserted, the periodic signal will go on forever, the form depending on the seed. But as said, it is not as predictable as it should be, which can only be caused by either soem anti-aliasing, where a previous sample influences the seed value, or some quirk that has to do with the calculation method. Which is all fine by me as I like the unpredictability. :)
:idea: Here is another noodle that shows the basics of a chaotic pattern generator. It is pretty bare bones, though resetting and inserting new seeds might look complex at first.
Description: |
|
Download |
Filename: |
BasicChaos.pch2 |
Filesize: |
2.64 KB |
Downloaded: |
2377 Time(s) |
|
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18195 Location: Durham, NC
Audio files: 211
G2 patch files: 60
|
Posted: Thu Apr 08, 2004 6:56 am Post subject:
|
|
|
Musically, these patches are interesting because they seem to be patterns, but they are always changing. It sounds like there are fragments that repeat, but the over all melody line is not periodic. Still, I can't understand how they are made.
Is there a tutorial or something about PM and attractors. I'm not sure I get how they work together. I have read about attractors in chaos theory - that was big about ten years ago. If I remember correctly, attractors are sets of values where a chaotic function appears to be periodic, at least for a while. Is that what you mean by attractors here?
I can see that you are implimenting formulas with the G2's logic, math, and sample/hold modules. Wow... |
|
Back to top
|
|
|
ian-s
Joined: Apr 01, 2004 Posts: 2669 Location: Auckland, New Zealand
Audio files: 42
G2 patch files: 626
|
Posted: Sun Apr 11, 2004 12:08 am Post subject:
|
|
|
Thanks Rob, nothing like actually hearing what you describe. I must admit that I had passed over the chaos stuff from the NM pdf book because it didn't look that interesting. Happy to be proven wrong. I am very impressed by audio rate textures. I wonder if Clavia could do a module that would make doing math a little less cumbersome without detracting from the 'analogue' metaphor? |
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18195 Location: Durham, NC
Audio files: 211
G2 patch files: 60
|
|
Back to top
|
|
|
Rob
Joined: Mar 29, 2004 Posts: 580 Location: The Hague/Netherlands/EC
G2 patch files: 109
|
Posted: Fri Apr 23, 2004 1:36 pm Post subject:
|
|
|
[quote]I am very impressed by audio rate textures.[/quote]
The classic NM was already quite nice, but the G2 is a beast when it comes to textures, well imho.
E.g. the whole idea of the new mixer section is to allow for e.g. neat filtertricks and do all sorts of shortcuts in a patch. And when making feedback systems to create textures it is very important to have control over the very subtle. Actually I think that audio rate textures is the strongest point of the G2, but as that is not necessarily a good selling point it is not mentioned in the brochures. |
|
Back to top
|
|
|
jack
Joined: Feb 06, 2003 Posts: 23 Location: East Coast USA
|
Posted: Wed Oct 27, 2004 9:37 am Post subject:
|
|
|
ROB:
I'm new to the forum. This is a great patch {noodel}. Very organic. Nice
Jack _________________ JACK |
|
Back to top
|
|
|
|