FILTERING WHITE NOISE WITH COMB FILTERS
I am currently working on a project that uses the findbounds object in jitter to track the position of a color. The tracking information will then be used to change the parameters of comb filter.
It is my intention that the input to the comb filter will be the noise~ object. So in essence, the movement of the color will be changing the parameters of the comb and filtering the white noise.
I have examined the comb~ help patch and noticed that the number boxes controlling each parameter of the filter do not have a maximum or minimum value. Surely for the task I have described, each parameter (delay time, gain coefficient, feedforward coefficient, feedback coefficient) should be configured in a way that the range of possible values is limited to a particular range. And what would this range be for each parameter in order to get the most usable results from white noise?
Any advice anyone can give would be greatly appreciated...
"MSP Tutorial 31 - Comb Filter" gives some explanation and offers some example parameter settings (and allows you to try various settings on white noise).
For more on the theory of how the delay time, feedforward, and feedback parameters affect the spectral effect of the filter, the book Computer Music by Dodge and Jerse explains quite plainly.
http://www.amazon.com/Computer-Music-Synthesis-Composition-Performance/dp/0028646827
For a more technical explanation of the theory, of course there is Wikipedia
http://en.wikipedia.org/wiki/Comb_filter
and Julius Smith's site on digital filters
https://ccrma.stanford.edu/~jos/pasp/Comb_Filters.html
Try using the [scale] object to scale values coming from the [findbound] object going to [comb~].