You say there are bad reviews?

[swtethan]

http://www.ocworkbench.com/2007/asus/EAH2900XT/b4.htm

rofl notice anything in the benches?

 

[swtethan]

i ALMOST was convinces gtx got its ass kicked

 

[Zstream]

Originally posted by: swtethan
i ALMOST was convinces gtx got its ass kicked

That site is using the newest drivers so yes it is possible. Please read before posting next time.

 

[swtethan]

Originally posted by: Zstream

Originally posted by: swtethan
i ALMOST was convinces gtx got its ass kicked

That site is using the newest drivers so yes it is possible. Please read before posting next time.

um..... look at the AA+AF settings.... sorry you missed at what i was looking at

 

[aka1nas]

Are the Q modes super-sampled?

 

[MadBoris]

What is wierd is it's not apples to apples testing. The AA and AF comparisons bertween cards are hapazard. Considering the fps impact of AA and AF is considerable, you can just have one at 8 and one 4 or 16 and compare them. Wierd approach.

Call me a naysayer, but the benchmark on Fear running with exactly zero performance penalty with 4xAA and 8xAF (63fps) is a bit hard to believe. Image comparisons please...

 

[swtethan]

Originally posted by: aka1nas
Are the Q modes super-sampled?

yes, i think thats what it is

 

[MadBoris]

Originally posted by: swtethan

Originally posted by: aka1nas
Are the Q modes super-sampled?

yes, i think thats what it is

The Q modes are the highest quality modes, also the biggest performance hit, why not regular AA. Furthermore Nvidia is always having higher AF or AA settings. I'm calling flag on the play here.

 

[BFG10K]

Are the Q modes super-sampled?

Nope.

 

[Dainas]

The only thing that seems like B.S. to me is the card temps. Even running out in the cool breeze it should do far higher than 60c under load

 

[MadBoris]

Originally posted by: aka1nas
Are the Q modes super-sampled?

16x is 4xRGMS + 12xCS
16xQ is 8xSSMS (sparse sampled) + 8xCS

 

[Zenoth]

The more I read about HD 2900XT, and the more I can't wait for the 65nm version, in hope that it should make it at least less power-hungry and surely cooler at full load, because I find 80nm to be so-so for a next-generation card. In fact I expected it to be 65nm from the start but alas ...

 

[Raider1284]

biggest joke of a review i have ever seen, comparing one card with half the AA of another is laughable. Saying a card doesn't lose a single fps when enabling AA and AF couldn't fool anyone but the naive. And using different AA filtering when comparing cards made me lose all respect for this reviewer.

last but not least that asus 8800gtx in the review is a WATER COOLED edition...unbelievable!!!

 

[tuteja1986]

You should all wait for a Well know editor to do a non biased review... i am waiting for firingsquad's Brandon to post his gup review.

 

[Zenoth]

I hope XbitLabs will make a nice review, I liked their R520/580 reviews a lot.

 

[MadBoris]

First very favorable numbers here:
But as we already know certain games don't do so well with the 2900 currently, these 4 all apparently do. Maybe some bias in game selection especially in mentioning comparing the price of a 7800 ultra and 2900xt.

http://www.tbreak.com/reviews/article.php?cat=grfx&id=511&pagenumber=1

No word on drivers/system used that i could see.

EDIT:

I call shenanigans on this too.
They are using GTS and GTX scores from the original launch nvidia drivers from their 7 month old benchmarks
Compare old 8800 release scores: http://www.tbreak.com/reviews/article.php?cat=grfx&id=474&pagenumber=5
Compare nvidia scores on 2900xt review: http://www.tbreak.com/reviews/article.php?cat=grfx&id=511&pagenumber=2

More bad review methodology, no wonder drivers weren't mentioned.

 

[coldpower27]

I just looked at the 8x vs 8x results. Had to ignore the 4x ones since their weren't any comparable on the G80 side. I also was disappointed by the lack of a 640MB 8800.

 

[MadBoris]

double post, see edit above.