1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Logic 9 Questions about CPU optimization from a Logic newbie

Discussion in 'Logic 9' started by Stradibaldi, Jul 6, 2014.

  1. Stradibaldi

    Stradibaldi New Member

    Hi everyone,

    I'm a classically trained musician and taught myself Logic. Recently I got a big project which requires huge templates. I've been tearing my hair out trying to figure out how to optimize things.

    So I have some newbie questions and I would be so grateful for any advice/assistance.

    --------------------------------------------------------------------------------------------------------------------------------

    BACKGROUND

    I have a MacBookPro with 4 GB memory, 2.4 GHz processor and 2 Intel i5 cores (seen as 4 cores by Logic).

    I am streaming all my samples (PLAY and Kontakt 5) off an external drive via Firewire 800, into Logic Pro 9.1.8 in 64bit (with PLAY going over the 32 bit bridge).

    The way I work is to load an instance of PLAY or Kontakt for each instrument on its own track (e.g. Cellos Staccato, Cellos Spiccato, Cellos Pizz = 3 tracks with 3 instances of Kontakt LASS).

    The main problem I've been having is the dreaded "The audio engine was not able to process all required data in time." thanks to gigantic CPU spikes before and often during playback leading to note dropouts or freezing playback. This happens especially with PLAY to the point that I dread loading and composing with the darn thing; but it also happens when I have 25+ instances of Kontakt.

    --------------------------------------------------------------------------------------------------------------------------------

    MY QUESTIONS

    1. IS IT BETTER to host my sample library files on the same Macbook as Logic, **OR** stream them from an external disk via Firewire800?

    2. Which is better for CPU usage: multitimbral instances of Kontakt and PLAY (16 instruments loaded in one instance, with independent automation and plugins on auxes) **OR** one instance per track (16 instances of Play/Kontakt)?

    3. Which is better for CPU usage: routing all my instruments to stereo out with a reverb bus send on each track; **OR** routing each section of the orchestra to a submix with a reverb bus send from each submix? (or is there no difference)

    4. Supposing I have money to throw around, do I need to install more RAM, or get a new Macbook with better processor speed, or more cores or what?

    5. any other stupid mistakes I'm making or best practices I should be following?

    THANK YOU in advance.
     
  2.  
  3. Eli

    Eli Senior member

    Better to stream them from a separate high performance drive.

    One instance per track.

    It shouldn't make any significant difference.

    Any of those choices will improve performance. But in your case, with big orchestral libraries, and only 4 GB or ram; the ram is probably the most important upgrade at this point.

    No stupid mistakes, just a workflow suggestion. Instead of separate tracks for each articulation of an instrument, consider learning about key switching in PLAY. That way you could have all cellos loaded up in one single channel of a regular non multi timbral instrument, and then just switch articulations there to trigger the different sounds. If you do embrace this approach, the ram upgrade will be particularly useful.

    Alos, upgrade PLAY to the 64 bit version. This will allow you to take proper advantage of a potential future ram upgrade.
     
  4. Jay Asher

    Jay Asher Senior member

    Even if you solve the CPU problem, 4 GB of RAM is not going to take you very far with orchestral simulation.
     
  5. robitaille59

    robitaille59 New Member

    I second that you should double your RAM but I think you should look into freezing the tracks that you've allready tinkered with in the mix. This should free up some memory.
     

Share This Page