[citation][nom]Curnel_D[/nom]Chris, it's a decent article, but why in the world would you use 512mb models in everyting aside from the 250 and 260. If you would have shown the 1gb 4870, along with a 1gb 9800+, it would have showed a clearer picture of how the 250 is identical to the 9800+/9800/8800GT.Meh. And there are MASSIVE rumours saying that Nvidia is hand-picking the review models sent to reviewers, even confirmed by HardOCP. Addressing that in this article would have been great.[/citation]
Why? Because that's what we have in the lab. However, for the sake of technicality, let's take a look at NewEgg and see what stepping up to 1GB models of everything else does:
Radeon HD 4850 1GB lowest price: $161
Radeon HD 4870 1GB lowest price: $214
Now, the comparison against the 512 MB 4850 and 4870 cards was perfect because the GeForce GTS 250--the card being reviewed here--fell right in between the performance and pricing of those two boards, making the comparison a particularly good one.
The 512 MB Radeon HD 4870 was faster than the GTS 250, so it's fair to assume that the 1 GB card will be even faster than that. In fact, its price tag puts it at the same level, essentially, as BFG's overclocked GeForce GTX 260 C216.
At $161, the 1GB 4850 costs as much as a 512MB 4870. Given the differences in memory technology, I'd go with the 4870 and overclock it to the best of my ability before buying a *slower* card with more memory that's going to get hung up that much faster at high resolutions.
Personally, I think comparing the GTS 250 *directly* to a GTX 9800+ paints the picture that they're the same clear as crystal.
And as far as the rumors are concerned, why buy in to the sensationalism? I didn't overclock the card in the event that there were memory ICs onboard not representative of a board you'd buy at BestBuy. But to be honest, to cry foul about one graphics card launch where boards are "hand-picked" ignores the fact that *all* vendors hand-pick the best of the best to put on show. It's the reviewer's responsibility to take that into account when they set up their testing and neutralize any advantage that gives in the results.
Incidentally, that's what makes our System Builder Marathon series so special. Where else can you find a trio of systems with hardware all bought off the shelf at retail and compared?