Sign in with
Sign up | Sign in

Nvidia Kepler GK104: More Leaks, Rumors on Specifications

By - Source: 3DCenter.org | B 5 comments

Nvidia's Kepler goes big with its 1536 CUDA cores, a vast improvement over the Fermi's 512 CUDA cores.

As reported on February 7th, we got a first glimpse of the rumored specifications for Nvidia's Kepler based graphics cards. The leaked specifications were met with both "wow.. can't wait" and "wow... those are so fake why even post" from both sides of the comment fence. Now, we are starting to get more pieces of information on the upcoming Kepler series. Based on information coming out of German-based 3dcenter.org, we may have a clearer picture of the true specifications for Kepler GK104.

Outside of the switch to the 28 nm process, one of the major changes in the Kepler architecture is to allow for more CUDA cores. This is achieved by no longer having shader frequency, just GPU frequency. Each Stream Multiprocessor will contain 96 CUDA cores, unlike the 32 - 48 that Fermi had. This change in layout of the CUDA cores will have the GK104 sporting up to 1536 CUDA cores, which is a big boost from GF110 and GTX 580. The number of texture units have doubled from 64 to 128 on GK104. The GK104 will only have 32 ROPS versus 48 in GF110 but it shouldn't affect performance compared to the Fermi.

The above GK104 architectural overview comes from Bright Side of News.

Nvidia Kepler GK104:

  • 28nm production at TSMC,
  • Die size 340mm²
  • 4 Graphics Processing Clusters (GPC)
  • 4 Streaming Multiprocessors (SM) per GPC = 16 SM
  • 96 Stream Processors (SP) per SM = 1536 CUDA cores
  • 8 Texture Units (TMU) per SM = 128 TMUs
  • 32 Raster OPeration Units (ROPs)
  • Chip clock (top model): 950 MHz
  • 1250 MHz actual (5.00 GHz effective) memory, 160 GB/s memory bandwidth
  • 256-bit DDR memory interface (up to GDDR5)
  • 2048 MB (2 GB) memory amount standard
  • 2.9 TFLOP/s single-precision floating point compute power
  • 486 GFLOP/s double-precision floating point compute power
  • Elimination of Hotclocks

  

The GK104's performance is expected to exceed the GTX 580 at the $350 to $400 price range. In addition, it is expected to outperform AMD's HD 7950 at similar price point and challenge the HD 7970 for the performance crown. The GK104 looks to be the similar to the current generation GTX 560 Ti with regards to price to performance in its category. 

Please keep in mind, of course, that these specifications are from 3dcenter's supposed reliable source. We won't know for sure until Nvidia shows its hand. Stay tuned!

Discuss
Display all 5 comments.
This thread is closed for comments
  • 0 Hide
    Anonymous , 14 February 2012 18:12
    thats all good but whats the point when 99.9% of game are ports from the consoles and dont use these cards properly dont get me wrong i am no console lover infact i hate them they are holding the games
    industry back in what they can do crysis 2 enyone nvidia and ati should take heed
  • 1 Hide
    deadjon , 14 February 2012 18:50
    768, 1024 and now 1536 SPs. The rumor mill has certainly been doing some work. I think this is getting out of hand now, Nvidia are clearly happy with letting the rumors trail completely off course. Remember this image was created by the source website and is very much a mockup.

    I don't think any sources can be trusted now, they all seem to have different ideas or they change their ideas every 5 seconds. The smart folk will ignore Kepler rumors from now on and the idiots/fanboys will continue to praise prematurely and inflate the rumors. By release we might have a 3072 SP GK104!

  • 2 Hide
    kunjar , 14 February 2012 19:54
    •2.9 TFLOP/s single-precision floating point compute power

    Gamers could be utilising such processing capabilities for solving scientific problems of tomorrow, today. I think both Nvidia and Amd should work alondside researchers to create corresponding applications which take advantage of the huge untapped resource of computational power within these GPUs when, an individuals machine is idle or say when the GPU is idle to help solve complex problems. Obviously the user has a choice to participate or not, similar to the SETI@home program i believe.
  • 1 Hide
    Lewis57 , 14 February 2012 20:21
    kunjar•2.9 TFLOP/s single-precision floating point compute powerGamers could be utilising such processing capabilities for solving scientific problems of tomorrow, today. I think both Nvidia and Amd should work alondside researchers to create corresponding applications which take advantage of the huge untapped resource of computational power within these GPUs when, an individuals machine is idle or say when the GPU is idle to help solve complex problems. Obviously the user has a choice to participate or not, similar to the SETI@home program i believe.


    I don't believe many will join in unless there is some sort of payment. Ok Folding@Home hasn't been semi successful, but I don't believe many would be ok with their GPU running @ 100% load using electricity and ultimately degrading the life of the hardware with not even a reward to cover the electricity cost.

    That's just my feeling anyway, even if my PC had one of these cards, I doubt I'd contribute considering the cost of electricity and the state of the jobs market.
  • 1 Hide
    wifiwolf , 14 February 2012 21:36
    1536 SPs in 340mm²???
    from 512 SPs in 520mm² (40->28nm)

    die size seems right though
    http://semiaccurate.com/2012/02/06/how-big-is-the-keplergk104-die/