I'll agree with you on the drivers, and perhaps even the 'most powerful card part' but, if I am thinking of the same article, the entire point of that article was to show how much of a CPU bottleneck would affect card performance. It was most likely aimed at the people who think they can own a P4 or what have you and slap in a GTX 280 as an 'upgrade' instead of spending less for the same performance. It even says along those lines in the title if I'm not mistaken. The article set out to prove what was intended, seeming that if not even the greatest graphics cards are severely limited, then one can infer that even more powerful graphics cards would be even more limited. In fact would most likely stop at the same ceiling, thus rendering spending the additional money to be pointless.
Also, good article here.