Quote Originally Posted by neliz View Post
NVidia has ZERO incentive to start a price war.
especially since they cant exactly flood the market with fermi parts
so yeah, why should they start a price war for a product that 1. costs them shtload to make, and 2. demand will be notably higher than supply for a couple of weeks if not months.

Quote Originally Posted by neliz View Post
I think the recent introduction of gt21x should be a pretty clear hint for some people here. If it is 25% faster, it sure as well will be priced more than 25% higher. RV770 just has too many people spoiled these days.
lol what? gt21x is 25% faster than what? theres a reason its called gts240, it IS slower than a g92, notably... its more like a gts230 or 235 at best actually...

Quote Originally Posted by LordEC911 View Post
Launch products have less TMUs than G200's 80TMUs.
huh? really? did they beef up their tmus then? they used to have plenty of not very efficient tmus until now, maybe they went for less, more efficient units now?

Quote Originally Posted by Lokinhow View Post
but when Nvidia released GT200, 55nm was already ready for a long time.
When they shrinked to 55nm, ATI didn't did anything, because it was already the smallest node, there was no improvement on the red side.

That is not the case, we don't have 28nm ready already. When Nvidia moves to 28nm, ATI will make a die shrink too, so if it is like GT200, things will go really bad for Nvidia, because ATI will make a improvement too.
yeah... well the funny part is that by now, 40nm SHOULD have been ready for a long time and mature as well