Re: [chuck-users] Setting LPF Q or Freq to very high or low values causes bad sounds...

Thanks. For the moment, I've limited the values to what seems to be marginally sane, that is, >0 and <12000 for the frequency, and >.05 for the Q (chuck automatically caps this at 1000) - Johnathan On May 31, 2011, at 1:59 PM, Kassen wrote:

On 31 May 2011 17:57, Johnathan Bell wrote: I'm not sure if I'm just doing something wrong or what, but (for instance) if I set up a SawOsc => LPF => DAC and set the LPF's Q to something very low (< 0.05) with the Osc set to a frequency of 440Hz, there is some kind of blow-your-ears-out high gain garbage that comes out of the speakers... Does anyone know if (1) this is normal and (2) what are normal (sane) values for this? I get the same kind of behavior if I set the cutoff frequency to something > 12000ish.

There seem to be issues with the LPF, HPF, etc filters generally. They don't seem stable under all conditions that we'd like them to be stable under. I don't think this is your fault, Johnathan.

I'd like to suggest the relevant DEV's have a look at these; http://www.musicdsp.org/archive.php?classid=3#25 That's the ones that FLuxa ( http://www.pawfal.org/fluxus/docs/0.17/en/fluxa.html ) uses. I found those quite stable so far and to behave quite well under continuous modulation too.

Fluxa implementation is here; http://git.savannah.gnu.org/cgit/fluxus.git/tree/fluxa/src/Modules.cpp starting at line 568, that one includes a extra line avoiding floatingpoint de-normalisation.

Since Fluxa is GPL I assume that these too will be GPL compatible but I couldn't quickly find actual confirmation of that.

Yours, Kas. _______________________________________________ chuck-users mailing list chuck-users@lists.cs.princeton.edu https://lists.cs.princeton.edu/mailman/listinfo/chuck-users