I’m still searching for self-similarity in sound. Have had 2 weeks of furious development and experimenting (in SuperCollider). I’m looking for some kind of *sonic* fractal order that resonates “truth”. Havn’t found it. Sure you can do self-similary in time, through pitch, rhythm, timbre etc. But the results dont seem to resonate true. At least not like Perlin Noise does for 2D and 3D physical forms.

Paper fractal by Polly

It seems to be a characteristic of sound that orders/parameters from graphics or visual media just dont seem to translate well. In media arts they call this ‘mapping’. Mapping data from one medium to an other. It is notoriously difficult. Its easy to get data from images and videos and turn them into sound, but it is difficult keep a perceptual bond between the two. I’ve been trying this for years actually.

It might even be ‘une fausse demarche’ … a wrong turn, a bogus endeavour.

Perlin Noise is a fractal order typically spread over 2 dimensions. And its easy to spread a fractal order over 2 dimensions in sound. Just pick the ‘time’ axis … and any other axis: volume, pitch, rythm, timbre, etc. But the 2 axes involved in Perlin noise are isomorphic… they represent the same thing: space. And this is what I have been trying to find in sound: what aspect of sound has 2 isomorphic axes?

There’s this kind of thing … which involves ways to structure pitches and scales over a 2 dimensional grid. But I think this is the wrong kind of isomorphism. Its not a description of the real world, its a method for organising something that has already been abstracted.

I have actually implemented an instrument that uses Perlin Noise to choose sounds spread over a 2D grid in space. I have 100 sounds, spread over a 100 x 100m grid. I use Perlin Noise in 3 dimensions. 2 for the grid, and the third is time. It works well. The sounds are beautifully, and there is a natural progression in time which is natural and comfortable. But its not a new sound. I’m looking for a new sound.

I feel like what I need to do is discover where there is self-similarity in sound… rather than just *apply it as I see fit*. If Mandelbrot’s hypothesis, that nature displays fractal orders everywhere, extends to sound then I suspect that it is going to be in dimensions *other* than time.

### Like this:

Like Loading...

*Related*

I agree that if nature displays fractal orders everywhere, it’s not going to be in the time dimension. In one area I happen to have thought about, cosmology, it’s amazing how LITTLE similarity there is between phenomena at different scales.

I may have written something very stupid there. When I mentioned cosmology, I was really thinking about spatial scales. There’s very little in common between the various spatial scales that cosmologists study. I have no idea why I thought that had anything to do with time.

There’s a sense in which space IS time in cosmology, but I don’t think I meant anything that smartipants.

So cosmology displays little similarity at different spatial scales? Is there any apparent order displayed at a cosmological level?

Was thinking about the time thing a bit more … sound is wave energy … it seems to make sense to say that wave energy cant exist without time … but a mountain can … kinda-thing.

hang on … nah …. this is made with Perlin noise … time domain aspect seems to work: http://www.noisemachine.com/talk1/imgs/flame200.html

Yes, lots. The most obvious kind of order at large scales is gravitational clumping, but there are others as well.

Mm, nice.