Hi everyone,
I got this info about the anolog audio output being cheap PWM with filtering from this thread : http://www.raspberrypi.org/phpBB3/viewtopic.php?f=28&t=8684 I'm assuming these guys know what they're talking about (I think the first reply metioning this is by a member of the Raspberry Pi team.
Cheers,
Pierre
2012/8/14 Michael Zacherl sdiy-mz01@blauwurf.info
On 14.8.2012, at 01:47 , Miller Puckette wrote:
Thanks for the info... some quick ideas:
good to know but that didn't help.
I can't tell what it is but apparently it's a timing or order of execution issue at startup in batch-mode: Delaying the ; pd dsp 1 message by 200ms helps avoiding the "device busy" error. Same patch loaded in the gui, no problem. :-\
According to 'top' the cpu load drops from 75% to about 70% with -nogui for a 4k-window. Just about enough to avoid glitching. But any system activity brings the sound to a halt for a short moment.
So that was just an ssh-session and Pd running in batch-mode.
At some point I'll go through this and see what can be done: http://puredata.info/community/pdwiki/Optimize/
BTW, is it possible to quit Pd via the patch? Just in case Pd has such high priority that the system becomes
inaccessible.
Yep: the message is "; pd quit"
ah, of course!
thanks, Michael.
-- feed your perception: http://blauwurf.at http://soundcloud.com/noiseconformist
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list