NVIDIA Maxwell Speculation Thread

Discussion in 'Architecture and Products' started by Arun, Feb 9, 2011.

Tags:
  1. ToTTenTranz

    Legend Veteran Subscriber

    Joined:
    Jul 7, 2008
    Messages:
    11,076
    Likes Received:
    5,626
    $999 for the Titan X.

    For me, it seems they're anticipating that Fiji will put up quite a fight.
     
  2. Newguy

    Regular Newcomer

    Joined:
    Nov 10, 2014
    Messages:
    257
    Likes Received:
    113
  3. A1xLLcqAgt0qc2RyMz0y

    Veteran Regular

    Joined:
    Feb 6, 2010
    Messages:
    1,431
    Likes Received:
    1,108
    Haven't all the Titan's (not the dual GPU one) been released at the $999 price?
     
  4. ToTTenTranz

    Legend Veteran Subscriber

    Joined:
    Jul 7, 2008
    Messages:
    11,076
    Likes Received:
    5,626
    Yes, but this is a card with 12GB GDDR5, so many were expecting it to go towards $1500.
     
  5. ToTTenTranz

    Legend Veteran Subscriber

    Joined:
    Jul 7, 2008
    Messages:
    11,076
    Likes Received:
    5,626
    Pascal news:

    - 2x Perf/Watt of Maxwell (I think this is related to neural network learning performance though, not games)
    - Will inherit the FP16 "mixed precision" from Tegra X1's GPU.
    - Claims 4x better FP16 performance than Maxwell. Probably means 2x more FP32 ALUs together with the theoretical 2x better FP16 performance for the mixed precision capability.
    - Stacked memory for up to 1TB/s bandwidth
     
  6. iMacmatician

    Regular

    Joined:
    Jul 24, 2010
    Messages:
    787
    Likes Received:
    215
    Maybe it's cheaper than it otherwise would be due to the slow DP rate?

    Also, Pascal is apparently "10x Maxwell," when counting FP16, NVLINK, and other improvements.

    [​IMG]
     
  7. RecessionCone

    Regular Subscriber

    Joined:
    Feb 27, 2010
    Messages:
    501
    Likes Received:
    178
    This is for training neural networks, and includes 2x more GPUs.
     
  8. silent_guy

    Veteran Subscriber

    Joined:
    Mar 7, 2006
    Messages:
    3,754
    Likes Received:
    1,380
    It's a bit like explaining day to day ups and downs in the stock market: the urge to find a deeper reason in everything.
    But sometimes everything can simply be explained by "it's just marketing". There is no rule book that requires Titan to have DP. It's just a label to separate mainstream from ultra high-end. And similarly, there's no magic formula that translates the presence, or not, of DP into a sticker price.
    A bunch of people got together in a conference room and decided on a price. It's really that simple.
     
  9. fbomber

    Newcomer

    Joined:
    Jun 9, 2004
    Messages:
    156
    Likes Received:
    17
    When will the Titan X be available?
     
  10. AnarchX

    Veteran

    Joined:
    Apr 19, 2007
    Messages:
    1,559
    Likes Received:
    34
    Is double-precision some how broken in Maxwell architecture? Building a ~600mm² GPU monster with ~200GFLOPs DP?
    Could be there some internal fight about the superscalar approach used in Fermi Gen2 and Kepler, that GK210 was produced?
     
  11. RecessionCone

    Regular Subscriber

    Joined:
    Feb 27, 2010
    Messages:
    501
    Likes Received:
    178
    No, there's nothing broken. From what I heard, the Maxwell with double precision got extra features added and was renamed Pascal. Evidently Nvidia wanted the best possible gaming and deep learning performance for Maxwell, and was content with letting the traditional HPC market wait a little longer.
     
  12. xDxD

    Regular

    Joined:
    Jun 7, 2010
    Messages:
    412
    Likes Received:
    1
    in the configuration of the chip , with the low ratio sp / dp , could have played a role the "cancellation" of 20nm ?
     
  13. jlippo

    Veteran Regular

    Joined:
    Oct 7, 2004
    Messages:
    1,449
    Likes Received:
    577
    Location:
    Finland
    Moving less data around should be the big win for FP16 and a reason why it resurfaces again.
     
  14. ToTTenTranz

    Legend Veteran Subscriber

    Joined:
    Jul 7, 2008
    Messages:
    11,076
    Likes Received:
    5,626
    Fun fact: The Titan X has about the same DP throughput as a Geforce GTX 580.

    I think the GK110 is "young enough" to be kept in the market a bit longer for DP, and being stuck to 28nm means they had to cut somewhere.
    Now nVidia is trying spin that FP32 and FP16 are spectacular for neural networks, which is why they spent 80% of the keynote talking about neural networks.
     
  15. Barbarian

    Regular

    Joined:
    Jun 27, 2005
    Messages:
    289
    Likes Received:
    15
    Location:
    California, USA
    You can pack your data in FP16 even right now. There's already hardware support for encoding/decoding to FP16.
    What they're adding is hardware support for directly evaluating calculations in FP16.
    The big reason that helps performance is that they can pack double the number of variables in the same number of registers (2xFP16 in each 32bit register).
    And if you use less registers you gain more latency hiding capabilities, which translates to more throughput.
     
  16. lanek

    Veteran

    Joined:
    Mar 7, 2012
    Messages:
    2,469
    Likes Received:
    315
    Location:
    Switzerland
    1/35 DP ? ( i suppose the gpu is 1/32 as GM204 ), i know for Raytracing FP32 this card will be excellent, but even at 999$, you just buy 2 old gpu who had 4-5Tflops FP32 ( 780TI or whatever )...

    Arg, they will need to disable at least 8SM on the "gaming" cutdown version for make peoples buy a gpu with 6GB more at this price ( before Evga and other release the "980TI" with 12GB instead of 6.. ( )
     
    #3216 lanek, Mar 17, 2015
    Last edited: Mar 17, 2015
  17. RecessionCone

    Regular Subscriber

    Joined:
    Feb 27, 2010
    Messages:
    501
    Likes Received:
    178
    It's not that they're "trying to spin" this. I work in deep learning, and nobody uses FP64. We've been doing FP16 experiments instead, so far they look promising.

    Also, they're not pushing GK110 for DP. Rather, GK210, a really different chip (2x register file, 2x shared memory).
     
  18. Malo

    Malo Yak Mechanicum
    Legend Veteran Subscriber

    Joined:
    Feb 9, 2002
    Messages:
    7,683
    Likes Received:
    3,757
    Location:
    Pennsylvania
    Since when has Nvidia ever cared about brand nomenclature confusion?
     
  19. ToTTenTranz

    Legend Veteran Subscriber

    Joined:
    Jul 7, 2008
    Messages:
    11,076
    Likes Received:
    5,626
    I understand that, but GTC has always been their place to brag about their GPU compute and they used to dedicate a good part of their time to FP64 performance.
    It's good that they found an area where they could brag about FP16 performance, otherwise it'd be really awkward to talk about a card with very low FP64 performance in there.


    I'd bet that GK210 and GK110 come from the very same wafer, only difference being a bit of laser trimming here and there.
     
  20. AnarchX

    Veteran

    Joined:
    Apr 19, 2007
    Messages:
    1,559
    Likes Received:
    34
    With Maxwell NV get rid of the superscalar structure of ALUs. 1:4 or even 1:8 should be very cheap, at least you did not make some mistakes in design. 1:32 is just ridiculous.

    GK110 was around since end of 2012 so they needed 2 years to enable SM_37 with bigger caches? All data (CUDA, A1-stepping, time-frame) says it a new chip.
     
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...