from what i've recently learned, you should probably only modify
blocksize in subwindows or abstractions. this is because the dac~ only works with 64 sample chunks, and thus you need to always convert back to 64 sample blocks.
i'm not sure if i got your question right, but one good advantage to
having a larger parent blocksize is that you can easily do the overlap-and-add method. this is (i think) because if you're throwing arrays around within the patch, if you tabsend an array of (for example) 128 that's in an abstraction with blocksize 128... to another array of size 256 within an abstraction with blocksize 256, then the last half of that array will be zeros. this is good for when you do fast fourier transforms and the overlap-and-add method.
i'm not positive about whether arrays are initialized to have all
zeros or not, but i'm pretty sure that this is the idea. i'll get to IIR filtering soon enough so make sure you figure it out so you can help me! so far i've only really looked at FIR algorithms.
scott
On Sat, 7 Sep 2002, Andreas Sumerauer wrote:
I am playing around with IIR filters and I would like to know if it makes any sense if the blocksize in the parent window is bigger than that of the subwindow. I assume that the Blockwise calculation anyway can't be done properly because the subpatch always breaks the mold. Is there still a reason to keep the bigger blocksize? (for waveguide synthesizers, where the filter is used in a feedback path)
Thanks
Andreas
PD-list mailing list PD-list@iem.kug.ac.at http://iem.kug.ac.at/cgi-bin/mailman/listinfo/pd-list
"640K ought to be enough for anybody." -- Bill Gates, 1981