There is active discussion of this on the music-dsp list at the moment.

The 'f' in '1/f' refers to frequency, and the whole signifies that the intensity
of the noise is inversely proportional frequency. The immediate practical aspect
is that there is equal power in each octave band (referred to as 'pink' noise by
sound engineers). To achieve this given a white noise source requires a filter
with a 3dB rolloff. The difficulty is that the simplest single-stage filter will
roll off at 6dB. One suggestion offered is to generate noise in each octave,
using randh, and simply mix the results together. It is of course a moot point
what the base pitch is for these octaves - for compositional applications, it is
often required to have 1/f sequences at very low rates - eg below 1Hz.

Another suggestion from the music-dsp list today is that a quite effective
empirical method is to mix white noise with the same noise filtered by a single
6dB stage. In short, the whole topic is fair game!

Try the web site: http://linkage.rockefeller.edu/wli/1fnoise


Richard Dobson

pete moss wrote:
> 
> i recently wrote a small program to generate one over f noise.  i know
> that the values produced should have a spectrumclose to one over f.  any
> ideas how to do this?  what does one over f mean?  what does f mean?
> thanks
> 
> pete