It seems your issue is not at all related to what the OP was encountering, and you should probably take it to a new thread, and maybe submit a problem report with demo project to the Bakers. Have you tried rolling back to Ipswich as the OP did to verify it's not something coincidentally gone wrong with your system/configuration or that particular project?

Agreed new thread please this is quite an important thread for some of us best stay on OP's topic. Many thanks.

I try not to troll (in fact I don't post much at all), but I'm honestly a bit irritated.
 
As for relevance to the OP, as I expressed in #11, my performance meter exhibited identical behavior upon playback post JP update. But please, pardon me for imagining that my issue might be related. @mdages sorry for apparently hijacking your thread.
 
As for posting a new thread, Dok would have simply referred me back to this thread had I posted my symptoms and repro steps and/or given me his usual call-center canned answers as he has a history of doing.
 
If this thread wasn't important to me, I wouldn't bother following or posting in it. Maybe scale back a bit on the passive-aggressive, Richard scale. You are the reason I don't participate in forums as a general rule.

 
@Noel, thanks for sending the patch and I will try it out.

bronsoncox
I try not to troll (in fact I don't post much at all), but I'm honestly a bit irritated.
 
As for relevance to the OP, as I expressed in #11, my performance meter exhibited identical behavior upon playback post JP update. But please, pardon me for imagining that my issue might be related. @mdages sorry for apparently hijacking your thread.
 
As for posting a new thread, Dok would have simply referred me back to this thread had I posted my symptoms and repro steps and/or given me his usual call-center canned answers as he has a history of doing.
 
If this thread wasn't important to me, I wouldn't bother following or posting in it. Maybe scale back a bit on the passive-aggressive, Richard scale. You are the reason I don't participate in forums as a general rule.

 
@Noel, thanks for sending the patch and I will try it out.

You have every reason to be irritated. For some reason I read the posts and didn't see you posted before. Stupidly I went on the last post. It could be related. I totally apologise for this, my bad. In my defence you neglect to mention that it wasn't just me who suggested this and I did ask your politely. Sorry again.
 
(Edited for clarification).

@Noel
Thanks for your tip on using the Kontakt-8Out, this makes really a huge difference when using a lot of instances.
Never thought that unused outputs of a virtual instrument can have so much effect on performance.
 
I've replaced in my current working project all Kontakt instances (>40) to the 8Out version and the performance is much better.
 
Great.
-Markus

Great, also let me know if the test build resolved the issue with the original project. It is fine here even with 1024 synth outs now.
Despite the unintentional extra load in JP (which is now fixed), in general having fewer outs will greatly improve CPU load. Even from just the synth point of view it has to allocate more memory and handle filling all the output buffers each time even if they are unused. It will also make the number of ports in menus much shorter so less scrolling around. 
 
In JP we handle unused synth buffers much better since we detect zeros reducing CPU load for unnecessary mixing.
Another fix in JP that was not listed, is that there is no limit on the number of ports. In prior versions of SONAR there was an internal hard limit of 64 outputs. It was a dumb restriction that shouldn't have been there.
 
 

Noel Borthwick [Cakewalk]
Great, also let me know if the test build resolved the issue with the original project. It is fine here even with 1024 synth outs now.
Despite the unintentional extra load in JP (which is now fixed), in general having fewer outs will greatly improve CPU load. Even from just the synth point of view it has to allocate more memory and handle filling all the output buffers each time even if they are unused.
BTW in JP we handle unused synth buffers much better since we detect zeros reducing CPU load for unnecessary mixing.

I PM'd you Noel trying to get a build appreciate you are busy. I observed earlier that drum maps in existing templates white out and eventually crash Sonar, if I removed them very fast then I would experience the CPU load issue (no crash). So it could be an offshoot of the existing problem or some other issue. Have you tested drum maps created in a previous/current version of Sonar? Cheers...

This thread exemplifies what a fantastic forum this can be, and fantastic crew the Bakers are

bronsoncox
But please, pardon me for imagining that my issue might be related.

Personally, I don't really have a problem with the way you got involved. It certainly sounded like the issue was the same on the surface, and the distraction was minimal. I only suggested you start a new thread when it seemed to become clear that too-many-ports-with dynamic-arming-enabled wasn't the cause in your case
 
In any case, I'll assume your irritation is not primarily directed at me, and continue to try and help if I can, though I'm kind of out of suggestions until you can confirm whether a rollback to Ipswich fixes it or not. If it does, it might be worth trying Noel's "quick fix", though it seems unlikely.
 
Can you share a copy of the project somewhere?
 
Cheers,
Dave

brundlefly
Personally, I don't really have a problem with the way you got involved. It certainly sounded like the issue was the same on the surface, and the distraction was minimal. I only suggested you start a new thread when it seemed to become clear that too-many-ports-with dynamic-arming-enabled wasn't the cause in your case
 
In any case, I'll assume your irritation is not primarily directed at me, and continue to try and help if I can, though I'm kind of out of suggestions until you can confirm whether a rollback to Ipswich fixes it or not. If it does, it might be worth trying Noel's "quick fix", though it seems unlikely.
 
Can you share a copy of the project somewhere?
 
Cheers,
Dave

 
As stated earlier, I don't have a problem either. I just agreed with what you said (my bad, happy to take all the heat as I made a mistake). Thanks for clarification.