[H] looks at another FX5700U versus R9600XT duel

This time its by the same manufacturer Gigabyte.

Same results as last time, and same strange lighting issues in NFS:U. I still think the lighting by the FX5700 is better on the street, but the R9600 does a better job closeup. BJ, don't think so.

You decide which you like more;

<A HREF="http://www.hardocp.com/article.html?art=NTc4" target="_new">http://www.hardocp.com/article.html?art=NTc4</A>


- You need a licence to buy a gun, but they'll sell anyone a stamp <i>(or internet account)</i> ! - <font color=green>RED </font color=green> <font color=red> GREEN</font color=red> GA to SK :evil:
 
As far as reviews go, Brent's talents really shined with this one. Great review.

<A HREF="http://rmitz.org/AYB3.swf" target="_new">All your base are belong to us.</A>
<A HREF="http://service.futuremark.com/compare?2k3=1819095" target="_new"><b>3DMark03</b></A>
 
That is a well done review. Unusual, but I like that he picked settings based on the best that those cards could do without dropping min. fps too low. Some people may claim it was ATI biased, but I think it made sense.

Also, compare this to Toms VGA Charts 3, where they added up all the framerates from each benchmark (minus aquamark), and came up with a summary score. This summary score was then devided by price to get the fbucks value. From day one of that review I felt fbucks was a joke and actually joined this forum because of it. I mean look at these two cards. Th FX5700U wipes out the the R9600XT in the summary score and also beats it out in the fbucks. But the whole performance lead the FX5700U comes from its 70fps lead in Quake 3 arena. With framerates so high and Quake 3 fps being equal to all other fps, that whole fbucks thing was a joke and totally made the NVida cards look much better than they really are. Redo it minus Quake 3, and the whole story changes. I thought of these two cards as equal, but with better IQ and AA/AF, the R9600XT I'd say was better. But this review you linked to here, shows the R9600XT is really superior to the FX5700U. Huge difference from the FBucks.

ABIT IS7, P4 2.6C, 512MB Corsair TwinX PC3200LL, Radeon 9800 Pro, Santa Cruz, TruePower 430watt
 
They should do FBucks on a relative scale. Make a 9600Pro always 100% and scale everything accordingly, averaging the scaled percentages to get a final score, but still not ignoring the performance of high framerate games. Then again, that takes more work than just adding numbers...but a simple spreadsheet could have it done in about the same time if done properly.

FBucks in theory is a good idea, but it should also have more parameters...most people I know aren't going to buy a 9600XT over a 9800XT only because it's better per dollar. They really should divide them into classes (but then that gets fuzy with the 5900SE and the now über-low price of the 9800Pro). Ahhh, just scrap the whole thing I guess, people will get the best card they can afford mostly.

Maxtor disgraces the six letters that make Matrox.
 
Fbucks with Q3 is retarded. Q3 as a benchmark is retarded.

<font color=red>_________________________________________</font color=red>
<b><A HREF="http://kevan.org/brain.cgi?dhlucke" target="_new">Wiggidy Wack!!!!</A></b>
 
Fbucks with Q3 is retarded
I totally agree. I do not mind seeing Quake 3 benchmarked and seeing that a FX5700U scores much higher than a R9600XT in that one test, but to use its raw framerates to virtually cancel out all other games by treating 1 fps in Quake 3 = to 1 fps in any other low fps game, makes no sense but to make NVidia cards look like they compete. Shoot, my radeon 9800 Pro only got 200 fps in Quake 3. Oh I better return it and buy one of the FX cards that scores 227. Oh wait, but i can play games with 4X/8X with superior image quality, while the fx cards can't. I guess I'll keep the R9800 pro.

So I agree that including Q3 in their FBucks calculations was retarded and basically made FBucks a Big Joke.

ABIT IS7, P4 2.6C, 512MB Corsair TwinX PC3200LL, Radeon 9800 Pro, Santa Cruz, TruePower 430watt