Vienna Symphonic Library Forum
Forum Statistics

181,953 users have contributed to 42,195 threads and 254,637 posts.

In the past 24 hours, we have 4 new thread(s), 13 new post(s) and 51 new user(s).

  1. Home
  2. Software
  3. Vienna Instruments Pro
  4. Multiprocessing

  • Multiprocessing

    M
    M

    Hi community,

    I tryied for the first time to change the multiprocessing threads in the Vienna Ensemble Server preferences, but taking a look at Logic´s CPU it turned out it didn´t have any effect on it. I changed it from 1 to 8 threads but still not impressed.

    Do I have to restart the project?

    Or doesn´t it have any effect on the CPU?

    Did I missunderstand the meaning of it?

    I have a MacPro. 3,1, 2x2.8 GHz Quad-Core Intel Xeon, OSX 10.6.8, Logic 9.1.7, VE 4.1.11515

    Any recommendations how many threads I can use? And maybe what effect it will bring me?

    MANY THANKS!!!
  • M
    M

    No comment??? [:(]
  • C
    C

    The servers and Logic are different processes. I don't know why Logic's use would be affected by distribution of cores among VEP servers. I would think Logic would be using more CPU with instruments all plugged into its process than using the server. But what VEP server is using is what it is.

    speaking generally there should be a gain in efficiency from hosting the plugins outside the process of Logic per se. but again this threads per is just redistribution of the process VEP.

    additionally to manage one instance's [heavier use] vs another has proven to be useful here. 
  • C
    C

    for instance the last project involved two instances of VEP server. One was a number of instruments, let's say half were VSL, with a couple of sends to Hybrid Reverb and there were EQ and compression on many of these instruments.

    the other involved a lot more of FX, amp sims with FX nested in them, and more automation parameters, and a lot more action in this automation. and some of instance 1, in addition to audio in the project was sent to audio input in this instance. evidently it was heavier use. So I devoted twice as many cores to it, which did evidence a need for less latency versus if I hadn't managed it like that.