Tue Jan 14 17:29:40 CET 2014

Maybe the point is structural?

Been thinking a lot about music and sound generation these days.  I
can't really get past the old ideas that sunk in already.
I.e. patterns I like but are not terribly interesting.

Is there a way to use the processing power available today in a way
that is not trivial to see without explicitly looking for it?  I.e. is
there a way to make "expenisive" algorithms that create a lot of
character, detail, without being simulations of real systems?