I’ve forgotten where I read it but my understanding of fractals is that they display self-similarity at different scales. I’ve just had a quick flick through Mandelbrot’s book “The Fractal Geometry of Nature” but I couldn’t find a quick one line description.
Perlin Noise is that. Random values generated at different scales, then all added up.
When generating landscapes, it is quite clear where the different scales are. Mountain ranges, then mountains, then boulders, then rocks, then textures on the rocks, etc. But with music, or sound, where are the scales?
I think this is going to be where any discoveries are to be made. Which aspects of sound inherently contain detail at different scales?
The most obvious one is scale in time. That clearly applies to compositions… a form can be executed over 10 minutes, over 1 minute, over 10 seconds, over 1 second, over 100msecs. Xenakis was concerned with repeating the same form from the macro time structure to the micro time structure. And I believe that fractal geometries have been used in algorithmic composition successfully.
What aspects of ‘spatial audio’ inherently contain detail at different scales?