This is a strange problem... perhaps you could send some small snippet of chuck source that reproduces the issue? One idea that might be worth trying is doing an un-chuck on all of the shred's ugens from the dac before killing that shred. If I recall correctly, the virtual machine should be doing this for you for free, but maybe something is going wrong with that. Doing it explicitly is supposed to stop all audio coming out of the ugens that you unchuck. For example, for a patch like this: Blit b => LPF lowpass => dac; BlitSquare sq => PRCRev rev => dac; you would unchuck everything directly connected to the dac before exiting, like this: lowpass =< dac; rev =< dac; spencer On Oct 29, 2006, at 12:03 PM, wrote:

I sent a message earlier today but didn't see it. if it comes through twice for some reason, forgive.

My problem is this: I spork a shred in code. I do this with a reference, like, spork ~ function () @=> Shred @ reference

then when it's time to remove, i use Machine.remove( reference,id() );

the problem is, with oscillators and stks, the shreds do not stop audio cleanly. for example, when i have phasor => lpf => dac and remove the shred, the notes stop, but a high pitched tone lingers. also, when BlowBotl => rev => dac and remove the shred, the notes stop, but a medium pitched noise sound lingers forever.

the time is synchronized to beat with now % T etc. I tried chucking and unchucking the ugens at the end of every loop. i tried with machine.add and remove from file instead of spork. i tried setting the gain and freq to zero at the end of every loop, but all these methods result in lingering sound.

how do you completely kill the audio from a shred in code?

any help?

thanks in advance, brian

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