hi marius,

> I also encourage you to go on with nova. I did not test it much (and 
> also don't know about the internals). but for me the reason, why you 
> should go on is, that pd is limited to a single processor architecture 
> and nova probably is not (true???). miller said on the pd conf, that if 
> the multi processor architecture will succeed, he would have to write a 
> total new program. I don't know if that is really true, but it was like 
> an alarm bell.

well, nova is designed as a multi-threaded system, but the computation
of the audio is still running in a single thread ... i have been
thinking about ways to split the nested dsp chains to multiple cores,
and some ideas in my mind, how one could implement it, but no actual
code ... actually, splitting the dsp chain to several cores was my
proposal for last year's google summer of code, but neither clam nor the
iem were interested :/

the only benefit you have at the moment is that everything except the
audio like dsp graph sorting or object instantiation can be computed on
multiple cpus ...

pd is designed completely synchronous, for me it was easier to write a
new program than try to make pd thread-safe and asynchronous ... but if
even miller says, that he would have to write a new program, i won't
contradict :P

cheers, tim

--
[EMAIL PROTECTED]    ICQ: 96771783
http://tim.klingt.org

A paranoid is a man who knows a little of what's going on.
  William S. Burroughs

Attachment: signature.asc
Description: This is a digitally signed message part

_______________________________________________
nova-dev mailing list
[email protected]
http://klingt.org/cgi-bin/mailman/listinfo/nova-dev
http://tim.klingt.org/nova

Reply via email to