Interesting review... $ vs. perf. normalized

Here is a Video Card Review with an interesting twist. Instead of just posting charts with FPS and 3Dmark scores, they took the results from those tests and normalized them to the prices of the cards, so you can see which cards get you the most fps and marks per buck spent, pretty good idea if you ask me:
REVIEW

the results.... the 9600 wins, and the 9700 is second.

read for your self
________
Hemp Marijuana

Heh, thats pretty sweet. I think more comparisons should include something like this.

An interesting read but really for people who want to do a little research fast. Unfortunately prices fluctuate so rapidly that it's hard to arrive at a clear answer except in a snapshot as they have done. I agree with their conclusions, but weren't they rather obvious? The very best costs extra. The only thing of note (which is still obvious if you read more than one of the hundreds of other reviews available) is that nVidia cards are unjustafiably expensive. I assume this is largely due to manufacturing costs inherent in the FX design. Reducing prices would also give them the label "bargain brand" which nobody dealing in the high end really wants, an expectation of "premium" prices can pay off in the long run.

The article would have been more useful if:
A) the prices were closer
B) the competition more willing to compete
C) the winner less consistent -- when the winners have lower price and better performance, only an doper needs more than one chart.

Still, if you were going to read only one article, this would be the best. I'd like to see more of these, but on less clear-cut subjects.

ugmore: I definately agree that this was a fairly obvious conclusion.....but:
It is interesting to note that the high end cards sometimes provide more bang-for-the-buck than the lower end ones. These guys have done this before, and in almost every case the lowest end cards get the most performance per unit of bling. I think that is saying something when your R9700 is a better "value" than a R9100, how often do you see that?

I think that it would have been very interesting to do only the gamegauge tests and include the DX8 competeors, as there aren't any popular (are there any at all?) DX9 games. So you can see how much value in real games the new gen cards have compared to buying a ti4200.

I also think that this type of analysis would be a nice addition to normal complete reviews, so you can see detailed performance results and then have the overall be calculated in sheer perf. and value.

I also don't think that this review should be discounted because Nvidia hasn't stepped up to the plate. It illustrates an interesting concept in revewing and that is as important as the practical nature of the results, as soon (maybe) the assertions may not be as clear cut.
________
Mercedes-benz cl-class history