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

[sc-users] Efficiency in multi-channel generative piece



Hi list,

First of all, a big thank you to all of you, this is my first post here but I have been reading the list for some time and have learned very much from it.

I am in the process of building my first 'big' project in SC and am encountering issues which I am having trouble pinpointing. The whole code is a bit long to post here but I will try to explain as best as I can. I am working in an 8-channel setup:

1) a Pbind is selecting between three synths for each event, routing it to four possible different bus 'effects' groups; respectively a PitchShift, reverb [FreeVerb], a delay bus and a resonator bus with a pitch-shifted copy (Resonz into JPverb split into dry and pitch-shifted = 1), in order of execution.

2) In addition to the effects groups, the Pbind is selecting from bus 0-7, essentially choosing between a 'dry' signal out and a 'dry+wet' where the signal is routed both to the output busses and the effects busses.

3) For all but the reverb bus, I would like to be able to create an instance of the effect for each loudspeaker, and modulate the parameters for each instance independently. 

This works, but is very computationally heavy. The server averages 40-50%, with spikes in activity >100%, accompanied by obvious sample dropouts and even sudden silence (with the interpreter and server still both running the same processes and even generating synths in the node tree). 

I should mention that some of the events are of very short duration (~0.001 sec). I am not sure how to best address this issue. From what I have read and researched on the list and forums, I am not sure if I should be looking at Supernova, Pfx or JITlib, so I wanted to ask here if anyone had any idea where I might be going wrong? 

Apologies for the long post, if this is too verbose and a copy of the whole code would be better, please let me know.

Have a great day,

V