[PD] glitches when moving between sub patches

Roman Haefeli reduzierer at yahoo.de
Thu Oct 30 01:46:18 CET 2008

On Wed, 2008-10-29 at 19:29 +0000, forwind wrote:
> Hi all,
> At the moment I am attempting to refactor a patch. Part of this involves
> moving alot of the exposed functionality down into sub patches. I have
> noticed however after doing the patch work I get glitches when I open
> and close the sub patches. The machine should have no problems being a
> 64bit 8 gig . The distribution is 64studio 2. Audio is handled by jack.
> The limits.conf has :
>         @audio - rtprio 99
>         @audio - memlock 250000
>         @audio - nice -10
> the xwrapper.conf has:
>         allowed_users=console
>         nice_value=19  
> And fstab has:
>         shmfs           /dev/shm        shm     defaults        0
>         0
>         none            /tmp/jack       tmpfs   defaults        0
>         0
>         none            /mnt/ramfs      tmpfs   defaults        0
>         0
> The graphics card is a nvidia GEForce 8600 GT running under the nvidia
> driver version 169.07.
> I suspect this is what is causing the glitches but don't know how I can
> stop it from 'getting in the way'. A last gasp effort could involve
> raising the nice value even more which past 19 might be a bit extreme ?
> Any ideas. There didn't seem to be any settings to optimise the graphics
> card through the nvidia control panel.  

sorry, if this is obvious, but are you running jack and pd in realtime
mode? jackd -R / pd -rt (probably the latter isn't necessary when
running over jack)

if i am not totally mistaken, the pd graphics shouldn't have an effect
on the audio part, since they are running in two differennt threads, pd
and pd-gui. so, a higher priority of the pd thread over the pd-gui
thread should already help (which is the case, if you run pd/jack with
realtime priority).

unfortunately, there are cases, where there are interferences between
both threads; for instance when moving around an array with 1'000'000
elements. i get clicks when i open a subpatch containing a big array,
but not when opening a subpatch containing only message boxes and

if there are gui related drop-outs such as when displaying an array, is
that because this amount of data causes a jam on the socket? is the
socket the bottleneck here?


Telefonate ohne weitere Kosten vom PC zum PC: http://messenger.yahoo.de

More information about the Pd-list mailing list