[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [sc-users] will there ever be real multiprocessor support?



On Mar 5, 2009, at 7:28 AM, nescivi wrote:

On Thursday 05 March 2009 09:14:58 Josh Parmenter wrote:
On Mar 5, 2009, at 4:46 AM, nescivi wrote:
Since synths run fairly independent of each other

Really?The all share a connection between input and output busses, and
often buffers. They are very connected. It is not uncommon for one
synth to depend on the output of another, and the only way to tell
this is order of execution really.

Hi Maije,

I did read the whole thing... but the problem doesn't really change. In the end, you need to share memory between the processes (even for mixing). To say one synthdef is not dependent on another is a problem, since one synth created by a synthdef may change how another synth created by the same synthdef is used.

A value written to a control bus by one synthdef may actually change where the output of another synth may be written to, and there really isn't a way I can can think of to pre-analyze the graph and where things going to be written or read from without actually running the graph.

In other words, what if a synth changes which channel it outputs to based on something inside the synthdef?

I didn't mean to sound as though I am jumping on you. That wasn't my intention. My apologies.

Josh


please read the email as a whole, and don't just jump on one opening sentence.

The order in which they go is defined in the node tree.
And the in-/outputs do not change within one processing cycle.

One synthdef is not dependent on others, once it has its turn of being
processed. At that moment it takes the data from its input buses, and
produces output on its output buses.

So given that the dependencies are known within each process cycle, some of it can be parallellised, I imagine. Of course depending on the complexity of the
node tree, and how intertwined the different synths are.

The final mixing stage may have to happen on one CPU, depending on how the
audio driver works.

sincerely,
Marije

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: https://listarc.bham.ac.uk/marchives/sc-users/
search: https://listarc.bham.ac.uk/lists/sc-users/search/

******************************************
/* Joshua D. Parmenter
http://www.realizedsound.net/josh/

“Every composer – at all times and in all cases – gives his own interpretation of how modern society is structured: whether actively or passively, consciously or unconsciously, he makes choices in this regard. He may be conservative or he may subject himself to continual renewal; or he may strive for a revolutionary, historical or social palingenesis." - Luigi Nono
*/


_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: https://listarc.bham.ac.uk/marchives/sc-users/
search: https://listarc.bham.ac.uk/lists/sc-users/search/