MMM
Results 1 to 25 of 82

Thread: AMD talks up its first Fusion chip

Hybrid View

  1. #1
    Xtreme Member
    Join Date
    Aug 2004
    Posts
    210
    Quote Originally Posted by ajaidev View Post
    Did you all notice the teraFLOP written there, if this is indeed written for the APU it means that they went from a gigaFLOP est. to a teraFLOP est.
    In my opinion, they are just referring to their visual computing breakthroughs by breaking the 1 TFLOP barrier.

    They clearly wrote 1TFLOP GPU, not 1 TFLOP APU ;-)

    Furthermore, think about the memory limitations ... ~800 Shader cores and then dual channel DDR3, which is shared with the CPU cores.

    That would be not feasible in my opinion.

  2. #2
    Xtreme Addict
    Join Date
    Jun 2007
    Location
    Thessaloniki, Greece
    Posts
    1,307
    Quote Originally Posted by Opteron146 View Post
    In my opinion, they are just referring to their visual computing breakthroughs by breaking the 1 TFLOP barrier.

    They clearly wrote 1TFLOP GPU, not 1 TFLOP APU ;-)

    Furthermore, think about the memory limitations ... ~800 Shader cores and then dual channel DDR3, which is shared with the CPU cores.

    That would be not feasible in my opinion.
    How the GPU access memory is still an unknown really. But the GPU is made on a 32nm SOI process, so higher clockspeeds than in earlier designs might be achievable.
    Seems we made our greatest error when we named it at the start
    for though we called it "Human Nature" - it was cancer of the heart
    CPU: AMD X3 720BE@ 3,4Ghz
    Cooler: Xigmatek S1283(Terrible mounting system for AM2/3)
    Motherboard: Gigabyte 790FXT-UD5P(F4) RAM: 2x 2GB OCZ DDR3 1600Mhz Gold 8-8-8-24
    GPU:HD5850 1GB
    PSU: Seasonic M12D 750W Case: Coolermaster HAF932(aka Dusty )

  3. #3
    Xtreme Member
    Join Date
    Jul 2004
    Location
    Berlin
    Posts
    275
    Quote Originally Posted by BrowncoatGR View Post
    How the GPU access memory is still an unknown really. But the GPU is made on a 32nm SOI process, so higher clockspeeds than in earlier designs might be achievable.
    According to Sam Naffziger, the GPU accesses memory through the XBar like the CPU cores. Shared memory model, same type of access. Maybe they added some prefetching logic, which works better for GPU typical access patterns.
    Now on Twitter: @Dresdenboy!
    Blog: http://citavia.blog.de/

Bookmarks

Bookmarks

Posting Permissions

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