Wednesday, March 21st 2012
GK110 Specifications Approximated
Even as launch of the GK104-based GeForce GTX 680 nears, it's clear that it is emerging that it is not the fastest graphics processor in the GeForce Kepler family, if you sift through the specifications of the GK110 (yes, 110, not 100). Apparently, since GK104 meets or even exceeds the performance expectations of NVIDIA, the large-monolithic chip planned for this series, is likely codenamed GK110, and it's possible that it could get a GeForce GTX 700 series label.
3DCenter.org approximated the die size of the GK110 to be around 550 mm², 87% larger than that of the GK104. Since the chip is based on the 28 nm fab process, this also translates to a large increment in transistor count, up to 6 billion. The shader compute power is up by just around 30%, because the CUDA core count isn't a large increment (2000~2500 cores). The SMX (streaming multiprocessor 10) design could also face some changes. NVIDIA could prioritize beefing up other components than the CUDA cores, which could result in things such as a 512-bit wide GDDR5 memory interface. The maximum power consumption is estimated to be around 250~300 Watts. Its launch cannot be expected before August, 2012.
Source:
3DCenter.org
3DCenter.org approximated the die size of the GK110 to be around 550 mm², 87% larger than that of the GK104. Since the chip is based on the 28 nm fab process, this also translates to a large increment in transistor count, up to 6 billion. The shader compute power is up by just around 30%, because the CUDA core count isn't a large increment (2000~2500 cores). The SMX (streaming multiprocessor 10) design could also face some changes. NVIDIA could prioritize beefing up other components than the CUDA cores, which could result in things such as a 512-bit wide GDDR5 memory interface. The maximum power consumption is estimated to be around 250~300 Watts. Its launch cannot be expected before August, 2012.
34 Comments on GK110 Specifications Approximated
Yeah SLI 2 of those MOFOs on one PCB :rockout: :nutkick:
According to articles/news which I have ever read about "Keplar related" information , I believe the chip of GK100 (not GK104 ) maybe come as "GTX685" which is really/fully "28nm Keplar technology". Therefore I decided to skip (pass) "buying GK104(GTX680) and wait for "the availability date of GK110" ( around of "Aug 2012" time frame) in order to get "fully 1 Teraflops on one GPU in double _precision". Do you think so that this plan is reasonable ??:pimp:
Now seriously (not), didn't you get the memo? Your card really, trully was sold at a $1000 loss per card. It's true man, that's how you do bussiness nowadays. :eek:
Also the best price for a chip like GK104 is in the sub $400 region, maybe even below $300 where it would sell like hotcakes and where it would get much higher profits (not gross margin, profits). But 28nm supply being as it is now they will probably sell as many as they can make so it would be pointless for them to price it where demand would surpass their supply. Once supply is better prices will drop.
One thing I can speculate on is making the room for all this extra while not really gain'n to many more shaders (2048 theoretically) would bring back a hot clock of sorts, were the shaders can "boost" themselves far past the base clock or boost clock (think 1308-1509 for GTX 680).