Hmm... Okay. Then the problem should be in the hardware. I'm doing this kind of room impulse response estimation with nlms method and I get nice impulse response but if there is some cracks in the sound (from cpu load and dropped frames I assume) the impulse response shifts with random delay which shouldn't happen. I am using RME Fireface 800 audio interface in windows xp environment. Does anyone knwo if that might cause some problems??
-Mika Quoting marius schebella <[EMAIL PROTECTED]>: > welcome to the pd-list! > I am not sure, if many people still use the audiobuf flag, but I think > it works... (another method would be to set the buffer in the audio > settings.) > and yes, it should stay stable during the whole time. > what where the problems you had? because if the audio buffer is too low > and Pd cannot handle the amount of data, the sound is shut down. > usually a large buffer allows you to more complicated calculation (but > increases latency). > marius. > > > Mika Ristimäki wrote: >> Hi, all >> >> My first post.. how exciting =) >> >> Well anyway...just one quick question. Does the audio buffer size that >> can be adjusted with -audiobuf flag stay constant all the time? Or is >> it something that PD adjusts dynamically in case of dropped frames or >> something? I been having some delay related problems and it seems that >> sometimes it changes quite randomly. The problem might also be with my >> hardware so now I'm just trying to find the source of the problem. >> Thanks... >> >> -Mika >> >> >> >> _______________________________________________ >> [email protected] mailing list >> UNSUBSCRIBE and account-management -> >> http://lists.puredata.info/listinfo/pd-list >> _______________________________________________ [email protected] mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
