NVIDIA Tesla: GPU computing gets its own brand

Discussion in 'GPGPU Technology & Programming' started by B3D News, Jun 20, 2007.

  1. B3D News

    B3D News Beyond3D News
    Regular

    Joined:
    May 18, 2007
    Messages:
    440
    Likes Received:
    1
    Being able to witness the earnest birth of a new computing industry is a special thing. In the last few years, efforts to use programmable commodity graphics hardware for things other than graphics have gained pace, and now there''s a legit multi-million dollar industry surrounding it.

    Read the full news item
     
  2. Geo

    Geo Mostly Harmless
    Legend

    Joined:
    Apr 22, 2002
    Messages:
    9,116
    Likes Received:
    213
    Location:
    Uffda-land
    I can't believe we didn't throw Nikola Tesla a bone somewhere in that article! Shame, shame! :razz:
     
  3. Arun

    Arun Unknown.
    Moderator Legend Veteran

    Joined:
    Aug 28, 2002
    Messages:
    5,023
    Likes Received:
    299
    Location:
    UK
    In terms of financial oppotunity, I just love this answer from Andy Keane because it is such a perfect summary of what problem GPGPU can solve... :)
     
  4. Vincent

    Newcomer

    Joined:
    May 28, 2007
    Messages:
    235
    Likes Received:
    0
    Location:
    London


    It seems to me that Nvidia did solve the MUL issue which in turn give fully performance of more than 512 GFlops.
     
  5. Vincent

    Newcomer

    Joined:
    May 28, 2007
    Messages:
    235
    Likes Received:
    0
    Location:
    London

    The name in economic recession???:razz:
     
  6. INKster

    Veteran

    Joined:
    Apr 30, 2006
    Messages:
    2,110
    Likes Received:
    30
    Location:
    Io, lava pit number 12
    A typo or a sign of things to come also on the desktop side... ? ;)


    http://www.pcper.com/article.php?aid=424&type=expert
     
  7. Techno+

    Regular

    Joined:
    Sep 22, 2006
    Messages:
    284
    Likes Received:
    4
     
  8. cho

    cho
    Regular

    Joined:
    Feb 9, 2002
    Messages:
    416
    Likes Received:
    2
    "http://www.beyond3d.com/content/articles/77/3"

    "double precision on G92 will be limited to the Tesla line"

    is this your guess only?
     
    #8 cho, Jun 21, 2007
    Last edited by a moderator: Jun 21, 2007
  9. Geo

    Geo Mostly Harmless
    Legend

    Joined:
    Apr 22, 2002
    Messages:
    9,116
    Likes Received:
    213
    Location:
    Uffda-land
    Read the linked interview with Andy Keane:

    http://www.beyond3d.com/content/interviews/41/3

    It's an interesting contrast with what Eric Demers responded to a similar question in our recent interview, which is that AMD sees GPU features as complimentary to both 3D and GPGPU; tho he didn't address this specific one.
     
  10. AnarchX

    Veteran

    Joined:
    Apr 19, 2007
    Messages:
    1,559
    Likes Received:
    34
    Special chips for the Tesla line? :???:
     
  11. Tim Murray

    Tim Murray the Windom Earle of mobile SOCs
    Veteran

    Joined:
    May 25, 2003
    Messages:
    3,278
    Likes Received:
    66
    Location:
    Mountain View, CA
    I'm certainly not convinced of that, but I do imagine that it will be limited via the BIOS.
     
  12. Geo

    Geo Mostly Harmless
    Legend

    Joined:
    Apr 22, 2002
    Messages:
    9,116
    Likes Received:
    213
    Location:
    Uffda-land
    I would imagine that practitioners of the dark arts amongst us will be eager to engage in validation testing of that point. :cool:
     
  13. Arun

    Arun Unknown.
    Moderator Legend Veteran

    Joined:
    Aug 28, 2002
    Messages:
    5,023
    Likes Received:
    299
    Location:
    UK
    Depending on how it is implemented, if part of the extra FP64 cost was neatly divided as another logic block (I'm really not sure how that'd work but heh) then they could just make it redundant, so that even if it increased die size by 5%, it wouldn't affect yields but only the number of chips on the wafer.

    As I said, I completely fail to see how that kind of division could be implemented, but perhaps an EE would have a better idea of whether it is possible. Another thing to take into consideration is that only G92 will be sold as a GPGPU part. G94 and G98 will not. So this could just mean they're not including FP64 on the die of G94 and G98.
     
  14. Rufus

    Newcomer

    Joined:
    Oct 25, 2006
    Messages:
    246
    Likes Received:
    60
    It wouldn't make any sense to have a Tesla-only GPU, there's simply not enough volume. Just as Quadro boards use the exact same chips (though most likely a special binning of them), Tesla boards will also use the same chips. DP will be enabled/disabled the same way line AA and the like are enabled for Quadro (BIOS / jumpers / fuses / drivers / etc).

    As for the mid- and low-end physically not having DP support, that depends on 1 thing: how much custom work is being done on the cores. We know they're doing some level of semi-custom flow in order to hit 1.5ghz, and with rumors of 2.5ghz for G9x it's going to be more custom. Once they've done a custom layout of the G92 core that has DP support, it's unlikely they'd to touch it in any significant way for the lower end units. A hybrid DP/SP FPU only has about a 5% overhead from two SP FPUs. I'm thinking burning 5% on the ALUs (so like 2% area on the whole chip?) is cheaper than redoing a custom layout.
     
  15. Rufus

    Newcomer

    Joined:
    Oct 25, 2006
    Messages:
    246
    Likes Received:
    60
    The official press release says the same thing:
    http://www.beyond3d.com/content/pr/43

    It'll be interesting if we only have a Tesla GX2 and no Geforce GX2. That goes against the standard routine of using the Geforce volume to fund the Tesla (/Quadro) development.
     
  16. silent_guy

    Veteran Subscriber

    Joined:
    Mar 7, 2006
    Messages:
    3,754
    Likes Received:
    1,379
    I don't think 'redundant' is the right word here. 'Ignored' may be a better fit.

    In theory, you could create test vectors that don't cover the lower bits of multipliers, adders etc, and a strap that ties the outputs of the LSB's to zero. So the unnecessary logic wouldn't lower yield.

    In practice, it may not be worth the effort.

    Also, stuck-at DFT coverage is typically in the upper 99%, say 99.6%. That's high, but it still leaves millions of connections uncovered. However, one counts on the fact that, statistically, a fault will also impact surrounding locations and that this increases effective coverage.
    For an ALU, the unnecessary logic is probably placed very close to necessary logic. Masking out unused logic may indirectly also mask real problems in useful logic.
     
  17. _xxx_

    Banned

    Joined:
    Aug 3, 2004
    Messages:
    5,008
    Likes Received:
    86
    Location:
    Stuttgart, Germany
    I just hope there will be a Simulink library (autocode generation) for this, I'd be all over it.
     
  18. Jawed

    Legend

    Joined:
    Oct 2, 2004
    Messages:
    10,873
    Likes Received:
    767
    Location:
    London
    But why bother to have that overhead on lower-performance GPUs?

    As far as NVidia is concerned, single-GPU Tesla is "low-performance" entry level, evaluation kit. The real McCoy is a rack of em. A single G86 is nothing more than "rounding error" in comparison to that...

    Jawed
     
  19. Voltron

    Newcomer

    Joined:
    May 25, 2004
    Messages:
    192
    Likes Received:
    3
    Great article and interview!!

    When you guys asked about whether NVIDIA would support CUDA desktop applications built using GeForce were you envisioning anything specific?

    Obviously NVIDIA is focusing on HPC right now, but could you sense, either from those interviews or just chatting, whether or not they were excited about the potential of CUDA on the desktop?

    Offline video processing for movies was an interview topic, but was there any discussion of processing for networked video such as YouTube or Joost?
     
  20. Megadrive1988

    Veteran

    Joined:
    May 30, 2002
    Messages:
    4,633
    Likes Received:
    143
    this is something that I would like to see in next-gen consoles, next-gen arcade platforms or next-gen consumer computers (PC or non-PC) -- 2 or 4 GPUs working together providing a linear increase in performance without the overhead of SLI or CrossFire.
     
Loading...

Share This Page

  • About Us

    Beyond3D has been around for over a decade and prides itself on being the best place on the web for in-depth, technically-driven discussion and analysis of 3D graphics hardware. If you love pixels and transistors, you've come to the right place!

    Beyond3D is proudly published by GPU Tools Ltd.
Loading...