Page 1 of 2 12 LastLast
Results 1 to 10 of 14

Thread: David Airlie Announces "Reverse Optimus" Multi-GPU

  1. #1
    Join Date
    Jan 2007
    Posts
    13,429

    Default David Airlie Announces "Reverse Optimus" Multi-GPU

    Phoronix: David Airlie Announces "Reverse Optimus" Multi-GPU

    David Airlie has announced work on "Reverse Optimus", which is a new approach he's taking for poking laptops with multiple GPUs, namely the NVIDIA Optimus laptops with a discrete NVIDIA GPU and integrated Intel graphics...

    http://www.phoronix.com/vr.php?view=MTI2ODc

  2. #2
    Join Date
    Dec 2012
    Posts
    404

    Default

    This isn't even remotely funny anymore... . Once I heard the first messages of two videocards in a laptop, I thought it was an april's fools joke. Turned out to be bitter truth.

    Are there any dual video card laptop configuration that work without a hitch? I don't mind if it takes up an effort to setup up though...

  3. #3
    Join Date
    May 2007
    Location
    Third Rock from the Sun
    Posts
    6,532

    Default

    Quote Originally Posted by Rexilion View Post
    This isn't even remotely funny anymore... . Once I heard the first messages of two videocards in a laptop, I thought it was an april's fools joke. Turned out to be bitter truth.
    There are even laptops that run Crossfire/SLi setups as well.

  4. #4
    Join Date
    Nov 2008
    Location
    Old Europe
    Posts
    885

    Default

    Quote Originally Posted by Rexilion View Post
    This isn't even remotely funny anymore... . Once I heard the first messages of two videocards in a laptop, I thought it was an april's fools joke. Turned out to be bitter truth.
    Why? It makes perfectly sense for an IGP + dedicated GPU combination.

  5. #5
    Join Date
    Sep 2008
    Location
    Vilnius, Lithuania
    Posts
    2,399

    Default

    "Reverse optimus"? So it's "sumitpo"? Or, since anagrams are popular nowadays, how about "imp to us" or "imps out"!

  6. #6
    Join Date
    Nov 2008
    Location
    Old Europe
    Posts
    885

    Default

    Quote Originally Posted by GreatEmerald View Post
    "Reverse optimus"? So it's "sumitpo"? Or, since anagrams are popular nowadays, how about "imp to us" or "imps out"!
    Or "positum". It's latin. But what does it exactly mean?

  7. #7
    Join Date
    Oct 2008
    Posts
    143

    Default

    Quote Originally Posted by Rexilion View Post
    This isn't even remotely funny anymore... . Once I heard the first messages of two videocards in a laptop, I thought it was an april's fools joke. Turned out to be bitter truth.

    Are there any dual video card laptop configuration that work without a hitch? I don't mind if it takes up an effort to setup up though...
    Afaik optimus with bumblebee works faily smooth. Just set an env and run your application on the dedicated gpu.

  8. #8
    Join Date
    Feb 2008
    Location
    Linuxland
    Posts
    4,734

    Default

    Quote Originally Posted by entropy View Post
    Why? It makes perfectly sense for an IGP + dedicated GPU combination.
    Only if the big GPU can't go low enough in idle/2d modes. This is constantly improving, so in a generation or two only the most high-end mobile GPUs should require an integrated one too to keep power in check.

  9. #9
    Join Date
    Dec 2012
    Posts
    404

    Default

    Quote Originally Posted by entropy View Post
    Why? It makes perfectly sense for an IGP + dedicated GPU combination.
    Erm, not in my world. I agree with what curaga is mentioning: it's a scaling issue.

    But that is outright insane. 10 years ago, laptops carried 1 card. Right now, it's 2. In 10 years time, it will be 1 card again.

    Huh?

  10. #10
    Join Date
    Nov 2008
    Location
    Old Europe
    Posts
    885

    Default

    Quote Originally Posted by curaga View Post
    Only if the big GPU can't go low enough in idle/2d modes. This is constantly improving, so in a generation or two only the most high-end mobile GPUs should require an integrated one too to keep power in check.
    I couldn't find any recent numbers. It would be really interesting to see a comparison of the
    power consumption of typical mobile IGPs and dedicated mobile GPUs running in 2d mode.
    Even more interesting is the trend over the last say five years (for low- to high-end components).

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •