• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

G92 Q407

I personally can't see why Nvidia would raise the prices from the 8800 GTX launch with the G92. They didn't have any competition then with the 8800 GTX either then.

I don't expect the flops number to translate into even a two fold increase of performance, as 1.9 would be enough for me.

If its a 65nm chip...they cost quit a bit more than 90nm.
 
Yeah its quite a bit more complicated than that. You get more chips per wafer, which lowers the cost, however the tech needed to do it is more expensive. Chip for chip, the smaller the chip is, the more expensive it is to make, but like i said, since you can get a higher yield, this offsets the higher cost, but again, the technology isnt free. Also nVidia doesnt make their own cores anyway, and GPUs are much harder to shrink than CPUs. ATI learned that the hard way.
 
hah oh god.

Im going to have a late summer build around August but from the sounds of this.. it makes me unsure. Maybe just a cheap PCI-e card to tide me over before I go with the nVIDIA G92 line.. unless ATi has something up their sleeves at the end of the year
 
Perfect timing for my new rig. If performance estimates are true, it won't be cheap, but it will be worth it.
 
Yeah its quite a bit more complicated than that. You get more chips per wafer, which lowers the cost, however the tech needed to do it is more expensive. Chip for chip, the smaller the chip is, the more expensive it is to make, but like i said, since you can get a higher yield, this offsets the higher cost, but again, the technology isnt free. Also nVidia doesnt make their own cores anyway, and GPUs are much harder to shrink than CPUs. ATI learned that the hard way.

Ummm..... no!

Chip for chip the smaller the process the cheaper it is. The only exception is right at the beginning of the production cycle on a new process. During this time you often have yield problems, which sometimes (not always) offset the cost savings. However these are normally addressed within a couple of months and you start saving good chunks of change.

The fabs that Nvidia would use (and ATi use) have had 65nm processes for some time.
 
But can it run Morrowind...? [Serious question, actually]

Funny thats a question I would like answered as well. My current rig stutters for like 1/2 a sec everytime I enter a new cell in morrowind making it REALLY annoying. Post on the elder scrolls forums say its because of nvidia drivers but I'm not sure but if it is true I hope they fix it with the R92 :|.

Now hopefully this next card will handel HD playback like the 8600/8500's because it sucks that my 8800 gtx does not -.-. In anycase I must place a curse on nvidia because I dislike you and you keep kicking ati's @$$ (although ati is mostly to blame I believe) and if this does come out at the end of the year then I will again have to go nvidia :mad: .
 
Usually the nvidia cards that are hyped don't turn out as well as expected.

On another note I finally broke down a built a new system, 8800 GTS 640 included.
 
Erasmus there is other differences. The smaller the process, the more expensive the wafer (of the same size). 65nm wafers cost more than 90nm wafers, and still will in a year's time (relative to production scale of course). Smaller processes enable larger ASICs (by transistor count), but as potential defects are also a function of transistor count, as well as process complexity, defect density at 65nm will be greater than for 90nm, even when adjusted by defect/transistor ratio.
 
Ummm..... no!

Chip for chip the smaller the process the cheaper it is. The only exception is right at the beginning of the production cycle on a new process. During this time you often have yield problems, which sometimes (not always) offset the cost savings. However these are normally addressed within a couple of months and you start saving good chunks of change.

The fabs that Nvidia would use (and ATi use) have had 65nm processes for some time.


Ummm..... no!

Ummm, Tell that to ATI!

Ummm, the fabs that nVidia and ATI use? Ummm, they are fabless! Ummm, they get there chips from TSMC. Ummm, nVidia can already get 65nm chips. They are just ummm about 3x as much money!

Also, ManicOne, thank you for the further clarification.
 
Back
Top