NVIDIA Kepler Refresh GPU Family Detailed

[F0XFOUND]

Oh, perhaps it was originally, but GK100 was certainly not "held back" so they could "put out a midrange card as high-end for mad profits!!!!" as some people like to proclaim.

This is always what I thought. If nVidia could truly release a card twice as fast as what AMD has, using the same foundry, then they would, since that would ensure far more sales and profit than selling something that "saves on costs" instead.

In fact, had nVidia done this, to a degree, would amount to price fixing, and of course, is illegal.

Of course, now that both cards are here, and we can see the physical size of each chip, we can easily tell that this is certainly NOT the case, at all, so whatever, it's all just marketing drivel.

In fact, it wouldn't really be any different than AMD talking about Steamroller. "Man, we got this chip coming..."

If GK104 was truly the high end chip for the GTX 680 then why Nvidia claimed it was 3 times as powerful as the GTX 580 and could run the Samaritan demo on just one 680. Can the retail GTX 680 be 3x as fast as the GTX 580 and/or run the Samaritan demo all by itself? I remember the unreleased GTX 680 was touted as it could.

 

[HumanSmoke]

If GK104 was truly the high end chip for the GTX 680 then why Nvidia claimed it was 3 times as powerful as the GTX 580 and could run the Samaritan demo on just one 680. Can the retail GTX 680 be 3x as fast as the GTX 580 and/or run the Samaritan demo all by itself? I remember the unreleased GTX 680 was touted as it could.

IIRC the original Samaritan demo ran at 4xMSAA w/ triple GTX 580's -although I'm pretty sure it was SLI with the third card handling PhysX- in either event, it was the full screen AA that took the toll. The GTX 680 demo ran with FXAA ...FXAA wasn't implemented at the time of the original demo (May 2011), and wasn't included in the Nvidia driver until April of this year.

Nvidia never claimed that the GTX 680 was three times as powerful as the GTX 580- that's just an assumption that some people jumped to....such as this article. The suggestion is the authors based on the fact that a single 680 achieved (superficially) what previously ran with three 580's. You'll note that the author still notes the difference in AA setting.

 

[EarlZ]

Would be nice if it hits at least 25% performance gap from the 680.