|
craig routt wrote:
> you run the outputs of two buckets into each others input following the
> convention laid down by en ( or perhaps it was pop electronics) a couple
> of months ago.
Are you refering to an Electronotes article here? I have all the issues up
through 1997; have there been any new issues?
> use a vco as a clock input to sr.
Unfortunately, I can't think of any computer that can implement true multiple,
variable sampling rates, such as you would have with a VCO modulating a delay
clock input. We are forced to deal with using a single sampling rate, and
working around it with various forms of interpolation. This is frustrating when
coming from an analog perspective, but it is just the cold hard facts.
> then fm the clock with a bit of feedback signal.
Now, THIS you can do with Csound. Just create your own pluck orchestra, out of a
delayr/delayw pair, and using deltapi for the main pluck tone, with a simple
tone filter in the feedback loop. You may find, however, that FMing the pluck
delay line itself yields somewhat unsatisfactory results, as the "ring" time
will be tied in with the length of the delay line, and rapidly changing the
delay time will do strange things.
I would recommend using a second delay after the pluck sound, and modulating the
length of the second delay. This will give you an FM sound that does not affect
the basic response time of the pluck sound. You can have the output of the
second delay modulating the delay time of the second delay (make sure sr=kr,
which holds true for all feedback-type orchestras).
For some work along the lines you are describing, check out Tim Stilson's paper,
"General Weirdness with the Karplus-Strong String," at
http://www-ccrma.stanford.edu/~stilti/papers/Welcome.html
Another suggestion: Try using a frequency shifter in the feedback loop, or after
the output of the pluck. The frequency shifter can be modulated at audio rates
to produce some very nice FM sounds. See the hilbert opcode documentation for
info on how to set this up.
Hope this is useful,
|