Fellow ChucKists, With the return of "double chucking" (making more then one connection between two UGens or a UGen and the dac) something seems to have gone wrong; SinOsc s=> dac; second=> now; s=> dac; second=> now; If this is run in a VM set to loop or a MiniAudicle with it's version of ChucK replaced with the latest greatest only a single connection will be cleaned up as the shred exits, meaning "s" will stay connected to the dac and as "s" is never garbage collected this means a beep. This also means there is no way at all to get rid of this beep (short of stopping the whole VM) as there is no way at all to reach the namespace "s" is/was in. I'm adding this to the WiKi as a bug. Yours, Kas.
Hi Kassen!
With the return of "double chucking" (making more then one connection between two UGens or a UGen and the dac) something seems to have gone wrong;
SinOsc s=> dac; second=> now; s=> dac; second=> now;
If this is run in a VM set to loop or a MiniAudicle with it's version of ChucK replaced with the latest greatest only a single connection will be cleaned up as the shred exits, meaning "s" will stay connected to the dac and as "s" is never garbage collected this means a beep. This also means there is no way at all to get rid of this beep (short of stopping the whole VM) as there is no way at all to reach the namespace "s" is/was in.
Thanks for finding this! This is indeed due to a bug in the UGen disconnect code, as we discovered just now. It's fixed and in CVS, and will be in the next release. Thank you very much! Yours, Ge!
Ge; Hi Kassen!
Hey Ge! Thanks for finding this! This is indeed due to a bug in the UGen disconnect
code, as we discovered just now. It's fixed and in CVS, and will be in the next release.
I have a vague feeling we dealt with this some time back? Before double-ChucKing was defined as having no effect? I might be wrong, this would be a fairly long time ago, I might be experiencing deja-vu.
Thank you very much!
You're welcome. I ran some tests on this anyway after a discussion on the forum about ChucK as compared to CSound ( http://electro-music.com/forum/viewtopic.php?t=28866 ) which went in various ways of dealing with rather large numbers of sine generators. Yours, Kas.
Just curious, will this fix the problem using adc? I've found that often
times even after removing all shreds, the adc will stay connected and
outputting.
Thanks
On Mon, Sep 15, 2008 at 6:19 PM, Kassen
Ge;
Hi Kassen!
Hey Ge!
Thanks for finding this! This is indeed due to a bug in the UGen
disconnect code, as we discovered just now. It's fixed and in CVS, and will be in the next release.
I have a vague feeling we dealt with this some time back? Before double-ChucKing was defined as having no effect? I might be wrong, this would be a fairly long time ago, I might be experiencing deja-vu.
Thank you very much!
You're welcome. I ran some tests on this anyway after a discussion on the forum about ChucK as compared to CSound ( http://electro-music.com/forum/viewtopic.php?t=28866 ) which went in various ways of dealing with rather large numbers of sine generators.
Yours, Kas.
_______________________________________________ chuck-users mailing list chuck-users@lists.cs.princeton.edu https://lists.cs.princeton.edu/mailman/listinfo/chuck-users
-- David M When the people are being beaten with a stick, they are not much happier if it is called The People's Stick - Mikhail Bakunin
David M; Just curious, will this fix the problem using adc? I've found that often
times even after removing all shreds, the adc will stay connected and outputting. Thanks
It should, I'd say? At least the adc and dac will both be reachable from any name-space so I'd imagine that running adc =< dac; (a few times if need be) you should at least be able to get rid of that connection without restarting the VM? Was that a documented issue? I can't remember hearing about it. Yours, Kas.
Greetings! Actually, after pondering this issue for a while, I believe it had been in the code for a while, but was hard to reach. I think there is a good chance this also fixes the dac disconnect problem. Yay for optimism (and nay for optimizations)! Yours, Ge! On Tue, 16 Sep 2008, Kassen wrote:
David M;
Just curious, will this fix the problem using adc? I've found that often
times even after removing all shreds, the adc will stay connected and outputting. Thanks
It should, I'd say?
At least the adc and dac will both be reachable from any name-space so I'd imagine that running
adc =< dac;
(a few times if need be) you should at least be able to get rid of that connection without restarting the VM?
Was that a documented issue? I can't remember hearing about it.
Yours, Kas.
participants (3)
-
David M -
Ge Wang -
Kassen