GK110 Specifications Approximated

[theeldest]

Sounds like they're following Intel's lead.

the top chip is no longer for enthusiasts but for professional use. (All of Intel's 8-core chips go to server procs) I'm sure they'll release this baby for consumers but most chips will go to workstation cards where $2k - $5k isn't out of the question. 512 bus makes more sense in that case.

I think AMD will still go mainstream-ish with their 8000 series. We'll get a performance increase but they'll be under 400 mm^2.

I also think that nVidia had the GK100 in the development cycle but didn't want another GF100. Work longer on the development cycle so that the initial release comes off without a hitch.

If they had done a release of GK100 it'd be worse than the GTX480 was at release. But I'm sure the GK110 (GTX780?) will be problem free.

 

[OneCool]

Want a GTX790.

Yeah SLI 2 of those MOFOs on one PCB

 

[function69]

This card would be a perfect replacement for my 580, can't wait!

 

[Faster4Run]

Most faveriot GPU board which I would like to wait for

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 ??

 

[MxPhenom 216]

yeah sorry amdftw, but your wrong in all respects. I have a GTX470 and its 526mm^2 die size and costed me $310 brand new from the egg. $310 != $1500

 

[Benetanegia]

yeah sorry amdftw, but your wrong in all respects. I have a GTX470 and its 526mm^2 die size and costed me $310 brand new from the egg. $310 != $1500

By his numbers, he is pretending that only 5 working chips would come out of a wafer so yeah... right... lol.

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.

 

[Chappy]

Then the GK104 (GTX 680) as we know should be priced in a mid-range level. If not, then this GK110 will be priced like $880+ for a single GPU.

 

[beck24]

This will be incredible! Competition is great!

 

[Benetanegia]

Then the GK104 (GTX 680) as we know should be priced in a mid-range level. If not, then this GK110 will be priced like $880+ for a single GPU.

No, I think Nvidia will engage in a price war in order to try to gain some market share and make room for GK110 when it launches in 5-6 months. Now that the $499 price tag is pretty much confirmed by some Newegg entries (now gone I think), I really think that's the case. They could have priced it at $550 without putting pressure on AMD, but they chose $500, which means that AMD will have to lower prices and more than likely AMD will try to put some pricing pressure on their own so that Nvidia has to adjust later on too.

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.

 

[EpicShweetness]

Just compare the GF104 to GF110 (GTX 560 ti vs GTX 580) there was a 25% increase in the shader count and a 50% increase to the L2 cache and the memory bus. The SM's were doubled as well and with it the ROP's jumped. GF110 has been speculated to carry a 512 bit bus, which would break this "mold" a little, but all totaled up its a true monster!
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).