At 6:18 +0000 21/11/01, Tim Conrardy wrote:
>While both Fractal Music Composer (FMC) and Fractal Music ST (FMST) are both
>Fractal Music programs, there is a big differance in the output.

Oh yes!

>My Question is: Why is there a differance? Isn't Fractal Music, FRACTAL
>MUSIC?

You might as well ask, isn't Music MUSIC?

>I also notice that FMC uses Mandalbrot and Julia sets. FMST does not
>mention these. Does FMST go about Fractal Music in a different manner?

Completely. The world of fractals is very broad and covers an awful 
lot more than the usual Mandelbrot/Julia sets.

>I know: some toughie questions best left to mathematicians! Some
>clarification would be great!

OK, but I'll have to be as brief as possible because I have to go 
out! First of all, I rejected the Mandelbrot/Julia model from the 
start because it takes many iterations, and therefore significant 
time, to process each pixel (or note). I wanted a process that would 
produce a note (or rest) on each iteration, in order to get instant 
music in - as near as possible - real time. I also wanted something 
three-dimensional, to give me real fractal elements to control pitch, 
time and velocity (loudness), and the Mandelbrot/Julia model struck 
me as inherently 2-D. I'm sure there are ways of coaxing three 
elements out of that model - and I don't remember whether FMC does 
this - but I couldn't see a way to do it without some sort of 
compromise or kludge.

So it doesn't even use complex numbers. Instead it uses a feedback 
loop based on a trigonometrical equation with three variables. I'm 
not at home at the moment, so I don't have all my documents with me, 
so I can't tell you what it is. I used to look upon it as a closely 
guarded secret, but I guess I can let it out of the bag now, so if 
someone wants to post a reminder in a couple of days (or Tim, maybe 
you could drop me a line off-list), I'll make the big revelation!

I can tell you this much: I saw an article in Scientific American by 
Clifford A Pickover, where he produced what he called 'fractal 
popcorn' with trig functions. I took this and extended it into three 
dimensions. He knows about it, and seemed to be happy with the idea.

BTW, I think FMC's output might be a bit closer to what FM produces 
if you didn't begin by setting up a load of musical parameters that, 
to my mind, force it into a bit of a strait-jacket and to some extent 
de-fractalise it. I do realise, though, that it produces more 
acceptable music for a lot of people.

OK - gotta go.

-- 
  Cheers ... Chris