So I realized that there is a segmentation error when I use zfilter2.....but don't know what that means.  I'm guessing it is important.

I've attached a noise_zfilter2.csd file which generates noise and passes it to the zfilter2 opcode.  The filter coefficients (generated in Matlab) correspond to two conjugate reciprocal pole-zero pairs at +/-pi/4. This should be an all-pass filter.  My interpretation of setting kdamp and kfreq to 0.01 is that the poles and zeros will be slightly offset, but otherwise fixed in location on the z-plane.

In anycase, the segmentation error is present and the output is zero so long as the zfilter2 opcode is uncommented in the code.  Try commenting it out and switching the output to the noise (a1).  It should work then.  If you leave it uncommented and change the output to the noise (a1), the output is still zero.

Thanks!

-Nik

----- Original Message -----
From: jpff [mailto:jpff@codemist.co.uk]
To: csound@lists.bath.ac.uk
Cc: csound@lists.bath.ac.uk
Sent: Sun, 16 Nov 2008 18:40:31 +0000
Subject: [Csnd] Re: zfilter2 problems

zfilter2 has previously (1998) been identified as not always stable,
but since then some recoding were thought to improve things.  I cannot
see any reason why it should behave as you report.  Could you send be
a complete csd file?
==John ffitch


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"