The Logic Off Topic list

> >It was an interesting sound.  He created a simple line of two or
> >three notes at the lowest resolution (so the total length of the line
> >was the length of the entire piece).  Then, each note in the original
> >line was replaced with a copy of the line, transposed to the note and
> >sped up so it fit in place of the note.  Continue this process until the
> >resolution is high enough to sound the way you want.
> 
> This is a fun idea :-).  The process is much akin to how fractals 
> like the Koch-curve are generated.  Did he use just plain chromatic 
> notes, or did he use micro-tonalities as well?  I would think that 
> with chromatic notes you would very soon run out of sufficient 
> different pitches to be able to iterate the process any longer.

Just chromatic notes.  He didn't actually subdivide pitch changes,
they were instead transposed, does this make sense?  So if
your original motif was (pitches only) C E, then:

0th iteration:  C                                    E
1st iteration:  C                E                  E                   G#
2nd iteration: C      E       E       G#       E       G#       G#      B#
3rd iteration: C  E  E G# E  G# G# B# E  G# G# B# G# B# B# D##


So, at each iteration, each note in the original motif is replaced with
a copy of the motif at half the duration, transposed to start at the
note we are replacing.  Your suggestion sounds interesting, but that
would make it much harder to hear the self-similarity, I think.

I always wanted to carry this process down to the sample level,
but I never had time to try it.

   -smeet