Le 07/04/2016 06:58, Jonathan Wilkes via Pd-list a écrit :
Running 2 instance of pd communicating with network socket is very different than using pd/pd~
Right, but like you said if you have video meeting its deadlines in one process and audio meeting its deadlines in another process, the pd/pd~ approach should meet the user's needs.
Furthermore pd/pd~ offers a better user experience (i.e., run a single patch and let Pd spawn the 2nd automatically). Given that I'd think most people would be using it for simple divisions of audio/video work, but messages to the list suggest otherwise.
really? i should send more mail then. I'm a big fan of pd~.
So I'm wondering if the increased cpu usage of pd~ is
significant enough to be driving users to the worse UX in order to get the efficiency.
communication between pd and pd~ is the only drawback of pd~ usage in my opinion. but it's a sample accurate communication. something that can't be done otherwise. if you need this precision, you don't have any choice.
My use of pd~ is not for splitting audio and video, but mostly to split heavy physical modelling and audio-video rendering. physical modelling use and generate lot's of data that can't go throw pd~ inlet and outlet, that where the shermem external become useful.
cheers c
-Jonathan
Pd-list@lists.iem.at mailing list UNSUBSCRIBE and account-management -> https://lists.puredata.info/listinfo/pd-list