From the FUD article:

SuperSuper clocked should work at, at least 666MHz, while we suspect that it might even reach 700MHz.
I don't get where's the big deal then. Regular -65nm- 260s were able to touch the 700Mhz mark and the 55nm gpus -at least as shown from g92b- gain an extra 7%-12% overclocking headroom, which means that these new cards should be reaching -at least- the 750MHz to worth the wait. If not are we talking about a bad shrink, a bad batch? And if so, with the $300 price mark I see everywhere (while -in fact- the 55nm process makes the cards cheaper to produce than their 65nm counterparts), I don't see why the 55nm shrink is worth the wait, or worse how nVidia -even now, that does not have the crown- still tries to rips us off. This card should have cost closer to the $200 mark (30% less), no questions asked, even the cooler uses less material (and as such HAS to be cheaper to produce).

nVidia HAS the chance for a comeback, it would be really sad if they throw it -again- away. Their arrogance will turn even people like me, who were supporting them for years to come, to ATI fanboys