The Notebook Review forums were hosted by TechTarget, who shut down them down on January 31, 2022. This static read-only archive was pulled by NBR forum users between January 20 and January 31, 2022, in an effort to make sure that the valuable technical information that had been posted on the forums is preserved. For current discussions, many NBR forum users moved over to NotebookTalk.net after the shutdown.
Problems? See this thread at archive.org.

    GPU computing processor ?

    Discussion in 'Hardware Components and Aftermarket Upgrades' started by Gryffindor, Aug 20, 2007.

  1. Gryffindor

    Gryffindor Newbie

    Reputations:
    0
    Messages:
    7
    Likes Received:
    0
    Trophy Points:
    5
    NVIDIA® Tesla™ C870 GPU computing processor

    What exactly does this technology do? Can you hook one of these to your desktop/laptop via PCI slot and speed up your computer or what? :confused:
     
    Gryffindor, Aug 20, 2007
    #1
  2. Changturkey

    Changturkey Notebook Evangelist

    Reputations:
    35
    Messages:
    381
    Likes Received:
    0
    Trophy Points:
    30
    Isn't that for engineers and such?
     
    Changturkey, Aug 20, 2007
    #2
  3. Gryffindor

    Gryffindor Newbie

    Reputations:
    0
    Messages:
    7
    Likes Received:
    0
    Trophy Points:
    5
    No idea, was trying to find out about it.
     
    Gryffindor, Aug 20, 2007
    #3
  4. fabarati

    fabarati Frorum Obfuscator

    Reputations:
    1,904
    Messages:
    3,374
    Likes Received:
    0
    Trophy Points:
    105
    It's an upcoming GP (General Purpose) GPU that's based on G80.

    Here's wiki's definition of a GP GPU
    Basicly, Tesla GPGPU's=very powerful, push out more flops than a regular CPU.
     
    fabarati, Aug 20, 2007
    #4
  5. grateful

    grateful Notebook Evangelist

    Reputations:
    36
    Messages:
    380
    Likes Received:
    0
    Trophy Points:
    30
    reading is fun

    we saw three major advantages to NVIDIA's approach. First, by bypassing 3D APIs, there's no concern about future drivers breaking an application as has plagued them in the past; consider Folding@Home's initial release on R580 and the continued absence of G80 support as an example. Second, it makes GPU computing more accessible by allowing developers to write their applications in a potentially more familiar manner, as opposed to shoehorning their application to fit within a 3D API's paradigm. Finally, it allows developers to access portions of the chip that they wouldn't be able to use directly in a 3D API.

    the information above was not my own, I grabbed it from::

    http://www.beyond3d.com/content/articles/77
     
    grateful, Aug 20, 2007
    #5
  6. tebore

    tebore Notebook Evangelist

    Reputations:
    55
    Messages:
    521
    Likes Received:
    0
    Trophy Points:
    30
    Holy crap looks like Nvidia will be the 3rd player in the rendering farm field along side AMD and Intel. Things are getting interesting. Maybe AMD/ATI will hop on a similar bandwagon but hopefully puts a but of focus on the home/power user.
     
    tebore, Aug 20, 2007
    #6
  7. chinmonkie

    chinmonkie Notebook Evangelist

    Reputations:
    17
    Messages:
    369
    Likes Received:
    0
    Trophy Points:
    30
    so what is it like a hybrid between a physics card and a graphics card in one O.O ... *Imagines how hawt that card is * drools ;O~~~~
     
    chinmonkie, Aug 20, 2007
    #7
  8. Akilae Hunter

    Akilae Hunter Notebook Consultant

    Reputations:
    7
    Messages:
    222
    Likes Received:
    0
    Trophy Points:
    30
    You can think of a GPU as a CPU of as many cores as it has pipelines. So a g80, with 128 stream processors would basically operate as fast 128 (or 32 quads or whatever the hell... dont get on my back about minutia) single-core CPU's on the same instruction set with the same problems at the same clock frequencies. The parallelism is what makes quantum computing appealing.

    A side effect of this GPGPU tech is that it can greatly increase the performance of problems that have traditionally needed farms or supercomputers to run. It really wouldn't be useful as high-i/o or server processors, but would be awesome at things like protein folding.
     
    Akilae Hunter, Aug 21, 2007
    #8
  9. chinmonkie

    chinmonkie Notebook Evangelist

    Reputations:
    17
    Messages:
    369
    Likes Received:
    0
    Trophy Points:
    30
    the day they make quanitum CPUS is the day ill be the happiest man on earth... T.T
     
    chinmonkie, Aug 21, 2007
    #9