I guess Rollo is smarter than Nvidia's Nick Stam?
Text
The 6800 series chips were designed to provide excellent performance up to 1600x1200, and although they support higher 3D resolutions, they lack certain hardware features to deliver strong performance at 2048x1536. The 7800GTX was designed to deliver much higher raw performance at 2048x1536, and better performance scaling from 1600x1200 to 2048x1536.
There are many features and performance optimizations (like zcull) that on NV4x are optimized for 2 Mpixels, and on G70 are optimized for 3 Mpixels. Like ATI, we also can operate on part of the screen at resolutions beyond our design point (the level at which the chip was originally designed to perform optimally). For example, an NV40 running @ 2048x1536 might be around 60% efficient with some of its optimizations like zcull.
But lets check out the benchmarks, shall we?
Doom 3
At 20x15, the 6800 U is 23% slower than the x850xtpe? At Nvidias flagship title?
Farcry
Scales a bit better here, the x850 is the one that has problems, even still its faster than the 6800 U.
Halflife 2
Check it out! 6800 U SLI gets beat by a single x850 xtpe, which is utterly trounced by the 7800 GTX.
Riddick
Loses again to the x850 XTPE at 20x15, SLI again is barely over a much cheaper single 7800 GTX.
SCCT
Again in SCCT, now the x850xtpe is 50% faster than the 6800 U.
3dmark05
the 6800 U and SLI actually do well for once....in 3dmark05.
Maybe your eyes are old and you can't see jaggies, but I can even at high resolutions.