Hi!

This is what the core distribution looks like when using the multiport template:

https://www.dropbox.com/s/74nzr4h6b9svjcc/Sk%C3%A4rmavbild%202015-10-02%20kl.%2014.42.57.png?dl=0

This is what it looks like when using a Tempo Ramp:

https://www.dropbox.com/s/tzo9ebgiepb2vpn/Sk%C3%A4rmavbild%202015-10-02%20kl.%2014.41.46.png?dl=0

The computer is a Mac Pro 2013 8-core version running 256 buffer and 1 buffer in the VEP instances. I'm using 14 VEP instances, some uses up to 10 midi ports each, some only 1 midi port. I'm using mostly Kontakt-libraries, but also Hollywood Strings, which is divided (after recommendation from Jay Asher) into several smaller VEP instances, which have worked without a flaw.

It has been suggested in other forums that this is due to performance issues running the multiport template, and that it's better to divide the template into several instances instead for better core distribution, in opposition with what is recommended by VSL themselves.

Before I redo the entire template, I wanted a second opinion about this, so here are two questions:

1. Does the amount of midi ports affect the CPU performance in a single VEP instance? And,

2. Is this the experience other have about this as well, that VEP over several channels works better rather than using a multiport setup.

Best,

Andreas

Hello Andreas,

I will get right to your questions: 

"1. Does the amount of midi ports affect the CPU performance in a single VEP instance?"

=> Not to my knowledge.  

2. Is this the experience other have about this as well, that VEP over several channels works better rather than using a multiport setup. 

=> We get different feedback on that question. I assume it has to do with different working styles, from the size of the template to the arrangement...

@Cyril:

Any new VSL ?

=> No. 

Are you working on that ?

=> No. We believe we have done everything we can at this point. 

Best, 
Paul