Originally Posted by Rollo
I have no quarrel with you either grey_1, and I can appreciate your frustration waiting fo this launch. (and everyone else's)
A lot of people are comparing this launch to the FX5800 and the HD2900 because it's six months behind the competing launch due to respins, and that it's suspected the cards will run hotter and at higher voltage.
Those things all MAY be true (don't know on heat and voltage) but that is where the similarity ends.
The FX/HD were both much weaker in AA performance than their competitor. I've watched the GTX480 run the FarCry2 demo twice as fast as a GTX285 with AA on. I've seen a PR slide showing how the 4X to 8X MSAA penalty has been cut by a huge percentage. NVIDIA is also launching a new mode with more samples that performs VERY well. So that is different.
The FX/HD launched at performance significantly lower that their competitors, from what I've been told and what I've seen, I don't believe this to be the case with the GTX480. I think it will launch as the market leader. That's a BIG difference.
The FX/HD lacked significant features their competitors had, like usable DX9 performance for the FX and AF angle invariant AF for the HD. The GTX480 has everything the HD 5870 has, and much more in terms of checkbox features.
NVIDIA is launching a better part than the HD5870 on 3/26, they're just launching it half a year plus after the part it competes with. If 3/26 is too long to wait for a better part, you should buy a 5870 now. It's not like they're "bad" cards, they just won't be the "best" cards in a month.
Last- my guess is that NVIDIA withholds clocks and performance to keep ATi in the dark and make them either guess with their refresh, or wait and lose some sales. You would too if you controlled NVIDIA- your responsibility would be to the stockholders.
Actually not quite all in fermi is improved over the HD5870,as the best case scenario for Fermi if it managed to hit 750 Mhz clock speeds,the floating point performance is 750 gigaflop second in dual precision math and tops out at 1.5 terraflops for single precision math.
HD 5870 cards are slower at dual precision(about 560 gigaflops),but way faster at single precision math,clocking in at 2.7 terraflops,and it's single precision math that used primarily in gaming scenarios,while dual precision is used for Cuda applications,specifically scientific work that requires the absolute maximum precision in the final result.....Keep in mind though that these are the theoretical maximums for each GPU.
Then there's the texturing abilities,with the HD 5870 using 80 texture units,while fermi seems to be using only 64 texture units,but as we don't know the individual capabilities on each unit in fermi,it's hard to make comparisons,but going on the rumored amount available in fermi and the clocks speed difference(200 Mhz faster for Cypress),ATI's HD5870 has more texturing power than Femi.
Lastly,even Anantech was pissed off in a recent article that clock speeds are still completely unknown,so it's hard to predict performance,and keeping clocks in the dark does very little to intimidate ATI,as they'll ship the refresh parts at the fastest clocks they can while still having good yeilds anyhow,as the point is to make money,not winning over Fermi with refresh parts so highly clocked that they'll have crappy yeilds as a consequence,and may the best win in the end.
Oh and as Xion mentioned,ATI's high end is the HD5970 cards,so double the above stats(1 terraflop dual precision,5 terraflops single precision,160 texture units),and Fermi is also down on memory bandwith too,so they have to be pretty carefull on the pricing of fermi cards here,as if it's close to the HD 5970,no one but the most hardcore Nvidia fan would buy one.