Hi Jean,

there was actually a discussion few days ago here, about the [pd~] and optimization topic.

However, for audiovisual processes, apparently it's still more reliable to use two different instances of Pd.
One proceses the sound and the other the video.
The video Pd instance might gain some more cpu resources when launched with the flag -noaudio.
Then you want to exchange data between patches on a local network on your machine using [netsend] and [netreceive].

AFAIK, the main feature of [pd~] is to keep synced processes happening in different threads.
Thus, when one of the processes slows down, the others slow down too.
Which, can be good for audio projects, but audio and video patches won't work really well due to this feature.
 
cheers,
Marco


 
Is there a strategy to optimize GEM video processing and real time
audio on a single Mac using pd~  :

Audio in subprocess or in main process ?
Any startup flags ? chmod +s stuff ? shm settings ?
Run two instances of PD with messages between them ?
Subprocess fifo size ?


--
Marco Donnarumma
Independent New Media and Sonic Arts Practitioner, Performer, Teacher
ACE, Sound Design MSc by Research (ongoing)
The University of Edinburgh, UK
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Portfolio: http://marcodonnarumma.com
Research: http://res.marcodonnarumma.com | http://www.thesaddj.com | http://www.flxer.net
Director: http://www.liveperformersmeeting.net