Couple of questions about using Titan X (Maxwell) and GT120...

Discussion in 'Mac Pro' started by Marmotta, Nov 27, 2016.

  1. Marmotta macrumors newbie

    Joined:
    Nov 27, 2016
    #1
    I was planning on getting a Titan X flashed to use in my 2010 Mac Pro, but given how much the poor GBP/USD exchange rate has pushed up the cost of getting it done by MacVidCards, I was considering doing without the Mac EFI, as I already have a couple of GT120s I can use for the boot menu. I do have some questions though that I couldn't find specific answers for online.

    Can both cards be installed simultaneously so that the driver can be loaded for the GT120 so that I can view boot options (as I intend to use Bootcamp often), but otherwise use the Titan X?

    How would this setup affect Windows use of the cards in Bootcamp?
     
  2. CactusJoe macrumors newbie

    Joined:
    Oct 12, 2015
    #2
    About that I don't really know.

    I do have tried a GT120 and 3 GTX Titan black and it was working fine. Only the fact that Apple blocked my 3rd GPU one update later...

    The only thing I must warn you about is that to be careful with your PSU you should power your TitanX with a different PSU.

    For my configuration I used a Netstor http://www.netstor.com.tw/_03/03_02.php?MTEx
    Really easy to use.
     
  3. Synchro3 macrumors 65816

    Synchro3

    Joined:
    Jan 12, 2014
    #3
  4. Marmotta thread starter macrumors newbie

    Joined:
    Nov 27, 2016
    #4
    Ok, so on the power side, no issue installing both a GT120 and a Titan X, as the combined TDP should be around 300W and the GT120 won't require either of the GPU power ports.

    However, my main concern is how OSX will work with both cards - will they both need connecting separately to a monitor? Can I select what applications will use the Titan X or set it up so that beyond the boot menu everything will run on it?
     
  5. h9826790 macrumors 604

    h9826790

    Joined:
    Apr 3, 2014
    Location:
    Hong Kong
    #5
    My personal experience (dual HD7950), there is almost no control of that. For OpenGL stuff, basically only the GPU that connected to the corresponding monitor will work. For others stuff, unless the software has ability to choose GPU, most likely it will only use the primary GPU (the one that connected to the monitor), or the GPU at slot 1 (if more then 1 monitors connected to different cards) to do the computation.
     

Share This Page