X2900XT Benchmarks

Radeon HD 2900 XT 512MB 512-bit GDDR3 @ 750MHz / 1.65 GHz

V.S.

GeForce 8800 GTS 640MB 320-bit GDDR3 @ 500MHz / 1.60 GHz

F.E.A.R. 1280x1024

Radeon HD 2900 XT 84.0 FPS
GeForce 8800 GTS 640MB 83.3 FPS

weak

why not compare to the GTX
Even X1950XTX can beat 8800GTS 640MB in the extreme high res
 
The 2900XT is competing against the 8800 GTS, which is why they're being compared as they're priced the same. The XTX model (or whatever it'll be called) will face off against the GTX and, judging from this, will probably outperform it, too.

Looking good. :) Nvidia finally has some competition.
 
As I said before and will say again, this card is the counterpart to 8800 GTS, not GTX.. pricewise, and unfortunately performancewise.
 
I think the point he's making is that on paper the XT should be much faster. Could be immature drivers - but given how much AMD has boasted about having good drivers I don't know.
 
The 2900XT is competing against the 8800 GTS,
Looking good. :) Nvidia finally has some competition.

Competing against GTS is no big deal. GTS is already 40% behind GTX in tests above 1600x1200. GTX is the card to beat.
 
yeah, i have to agree that is a pathetic score in fear, but in some benches you see almost double the performance of the 8800gts 640mb so there is still hope with better drivers,
 
ahhh wow, some of you guys are pretty anal about the way you compare things, LOOK AT THE REST OF THE BENCHMARKS TOO

Call of Duty 2
73.5 FPS
56.7 FPS
Company of Heroes
92.1 FPS
90.1 FPS
F.E.A.R
84.0 FPS
83.3 FPS
Half Life 2: Episode 1
112.0 FPS
57.4 FPS *
Oblivion
47.9 FPS
39.5 FPS
3DMark06
11447
9836
 
Half Life 2: Episode 1
112.0 FPS
57.4 FPS

Thats very nice to see. Source is what I've been playing mostly. I like.
 
I still think it comes down to price pure and simple. If the X2900XT is priced $50-100 more than a GTS, then you're essentially getting what you pay for.
 
X1950XTX beats the HD 2900 XT:D




Half Life 2: Episode 1

8800 GTS
147 FPS

HD 2900 XT
112.0 FPS
http://www.dailytech.com/Article.aspx?newsid=7043

X1950XTX
117 FPS


LOL
hl2-1280.gif
 

F.E.A.R. 2560x1600 4xAA
http://images.anandtech.com/graphs/nvidia%20geforce%208800%20launch_11070691155/13534.png[/mg]

Black & White 2560x1600 4xAA
[img]http://images.anandtech.com/graphs/nvidia%20geforce%208800%20launch_11070691155/13531.png[/mg]

Serious Sam 2048x1536
[img]http://www.pconline.com.cn/diy/graphics/reviews/0611/pic/xl599_thumb.jpg[/mg]

Call of duty2 2560x1600
[img]http://hardware.mydrivers.com/pages/images/20061108230745_64148.gif[/ig]

NFS Carbon 1600x1200 4xAA 16xAF
[img]http://www.pconline.com.cn/diy/graphics/reviews/0611/pic/xl607_thumb.jpg[/mg]

Splinter Cell 2048x1536
[img]http://www.pconline.com.cn/diy/graphics/reviews/0611/pic/xl601_thumb.jpg[/mg]

Half life2 1600x1200 4xAA 16xAF
[img]http://www.pconline.com.cn/diy/graphics/reviews/0611/pic/xl604_thumb.jpg[/ig]

3DMark05 2560x1600
[img]http://hardware.mydrivers.com/pages/images/20061108230254_33022.gif[/mg]

Call of duty2 2560x1600
[img]http://hardware.mydrivers.com/pages/images/20061108230745_64148.gif[/mg]

Call of duty2 2560x1600 4AA 16AF
[img]http://hardware.mydrivers.com/pages/images/20061108230749_41151.gif[/mg]

Prey 2560x1600 4AA 16AF
[img]http://hardware.mydrivers.com/pages/images/20061108230928_29768.gif[/mg]

Pacific Fighters 1280x1024 4AA 16AF
[img]http://www.firingsquad.com/hardware/nvidia_geforce_8800_gtx_gts_overclocking/images/pf1280.gif[/mg]

Call of duty2 1280x1024 4AA 16AF
[img]http://www.firingsquad.com/hardware/nvidia_geforce_8800_gtx_gts_overclocking/images/cod21280.gif[/mg]

F.E.A.R. 2560x1600 4AA 16AF
[img]http://www.firingsquad.com/hardware/nvidia_geforce_8800_gtx_gts_overclocking/images/fear2560.gif[/ig]

Oblivion HDR 2560x1600 4AA 16AF
[img]http://www.firingsquad.com/hardware/nvidia_geforce_8800_gtx_gts_overclocking/images/fol1920.gif[/mg]

Dark Messiah HDR 2560x1600 4AA 16AF
[img]http://www.firingsquad.com/hardware/nvidia_geforce_8800_gtx_gts_overclocking/images/dm2560.gif[/ig]

[COLOR="Teal"]MooseEdit: Host these images yourself. Hotlinking is a no-no...[/COLOR]
 
Wow. Where the hell did dailytech.com get their benchmark from?

nvm, im retarded: "* Our benchmarks for Half Life 2: Episode 1 showed an abnormal framerate for the NVIDIA 8800 GTS card that scaled with lower resolutions -- we are still investigating what occurred there. "

yeah that bench just kills all the hopes , blah
need more benches
 
I dont take these numbers with any seriousness, its not a real review.

Radeon HD 2900 XT 512MB 512-bit GDDR3 @ 750MHz / 1.65 GHz

V.S.

GeForce 8800 GTS 640MB 320-bit GDDR3 @ 500MHz / 1.60 GHz

F.E.A.R. 1280x1024

Radeon HD 2900 XT 84.0 FPS
GeForce 8800 GTS 640MB 83.3 FPS

weak

why not compare to the GTX
Even X1950XTX can beat 8800GTS 640MB in the extreme high res

Pretty funny how you choose just that ONE benchmark. Why not CoD2, where the XT beats the GTS by almost 20fps? Funny that.
 
Dark Messiah HDR 2560x1600 4AA 16AF
http://www.firingsquad.com/hardware/nvidia_geforce_8800_gtx_gts_overclocking/images/dm2560.gif[/ig][/QUOTE]

Could you please stop with the canned benchmarks already ?

This is getting hilarious. We already laughed the last time you posted this particular one. The X1950 XTX is almost twice as fast as an OC'ed 8800 GTX ?
Please give us a break :)
 
Wow, another set of benchmarks which still dont tell us much. Especially DX10 performance which is where this card is suppose to shine.
 
I dont take these numbers with any seriousness, its not a real review.



Pretty funny how you choose just that ONE benchmark. Why not CoD2, where the XT beats the GTS by almost 20fps? Funny that.

Probably because it's a demanding game and permissive to what we'll see in the future..

It's also quite shocking, for a card delayed for so long and --as has been said-- unflattering of it's paper specs.

I guess I'll have to wait for the XTX benches to make my descision on my Dx10 card; not like I wasn't planning on the top end from either side anyway.

Although, I was really hoping get in on a Folding@Home team, too. Is there anyone who can direct me on how Nvidia's latest does with the Folding@home stuff? That info seems rather obscure at best, and I know the Ati' cards have always done loads better.
 
These numbers are expected for a product that's a good half a year newer. Though...still a bit dismal.
 
Nvidia finally has some competition.


they have always had competition... look at the benchmarks for the last few releases of cards and it hasnt really been very close...ATI is always a step ahead aside from just releasing the cards...
 
2560x1600 meaning absolutely nothing. What should be tested is
1280x1024
1680x1050
1600x1200
1920x1200

These are the most common resolutions used throughout the community. 30: LCD's are not the most "common" used when gaming, bottom line. Sure there are those who have them but 2560x1600 is not the common resolution used in gaming. There is a big difference between real world results and benchmark results. Using the above resolutions people have an idea what to expect and not have to guess what 2560x1600 translate at 1680x1050.

Also, there are no DX10 games therefore, it's mute to even worry about it at this point.
 
ahhh wow, some of you guys are pretty anal about the way you compare things, LOOK AT THE REST OF THE BENCHMARKS TOO

Call of Duty 2
73.5 FPS
56.7 FPS
Company of Heroes
92.1 FPS
90.1 FPS
F.E.A.R
84.0 FPS
83.3 FPS
Half Life 2: Episode 1
112.0 FPS
57.4 FPS *
Oblivion
47.9 FPS
39.5 FPS
3DMark06
11447
9836

57.4 FPS *

What's the * mean?
 
57.4 FPS *

What's the * mean?

They just used it, as a sort of a mark for a note further down, explaining that that was a weird result to get and they are investigating the possible problem.
 
If I were you guys I would take this benchmark with a big bag of salt.
Test was only at 1280x1024, no other resolutions. Only said to use maximum quality, no word on AA or AF.
And most important, since when did we trust Daily Tech benchmars?

EDIT: And we still have no idea how any of these cards will perform in a DX10 game.
 
While these 'benchmarks' aren't really worth going on about, what I am impressed with is finally photos of the 9.5'' card in the wild. There is more information on the Inq about R600 here, but they're just as reliable as DailyTech. That is to say that they aren't very reliable at all.

However, if R600 puts up the performance, with all the features like HDMI, enhanced video acceleration, enhanced AA/AF, and a price point which (according to all the rumors we've been seeing) is significantly lower than that of the 8800GTX, I think this will be a winner.
 
Great...Now ATI has a card out that matches Nvidia's 8800 release 5 months ago. That XTX or whatever ATI calls it better stand way in front of the 8800GTX. I see Nvidia just waiting to trump the ATI Release with thier 8850 or 8900 in May.

Either way, DX10 benchmarks are the real test.
 
The guys at Rage3D say these benchmarks results are likely the result of early drivers. You would have thought that during the six months delay they would have worked on this, but apparently not. Will be interesting to see if the final drivers give the card a huge performance to get it to the rumored 50-70% lead over the GTX.

BTW - It was just posted this morning that the XTX has now been pushed back to Q03 of 07.
 
Test was only at 1280x1024

And that's all that was needed to be said in this thread. Clearly the CPU is the bottleneck at this resolution (can't really compare high end video cards at this resolution).
 
The guys at Rage3D say these benchmarks results are likely the result of early drivers. You would have thought that during the six months delay they would have worked on this, but apparently not. Will be interesting to see if the final drivers give the card a huge performance to get it to the rumored 50-70% lead over the GTX.

BTW - It was just posted this morning that the XTX has now been pushed back to Q03 of 07.

Rumored 50-70% lead over the GTX ? I never heard that one. Maybe you are refering to the theoretical memory bandwidth of both cards ?
And no driver update will yield that sort of performance increase. Unless of course, drivers are in such poor shape, that they are actually underperforming the card.
 
All I can say is, if the rumored power consumption of that card is true, I'm sticking to getting an 8800GTX. The added power consumption isn't worth the meager frames per second increase, especially when I am now one of those few who have 24" displays (and up) that can do resolutions higher than 1280x1024. For me, a bench at 1920x1200 resolution and up is what I need to see.
 
These benchmarks are worthless. People don't buy $400 video card to play games at 12x10 resolution and they damn sure don't spend that money to look at canned benchmarks that do not take into consideration defined levels of AA and AF.

The real comparison will be felt at 1600x1200 and beyond with heavy AF and AA. The R600's scaling here will be of the utmost importance to your gameplay in real situations. Benchmarks at 12x10 resolution are laughable for this level of hardware and if you tell me that you are buying one to play games at 12x10, well then you are laughable too and need to take a moment and evaluate your hardware purchasing decisions....because it is time for a new monitor, not a new video card.
 
The drivers are still a very early release and they still have some serious issues to work out. The last version I saw the control panel wasnt even working, and you had to use AF/AA levels that were in game only.

They are still quite buggy, hopefully within the next 3 weeks they can get them worked out. And it is meant to compete with the GTX, the drivers just arent ready yet.

And I wouldnt put full stock into the XTX launching with the XT like the dailytech article states. I have heard varied reports, and no not from "faud".
 
The drivers are still a very early release and they still have some serious issues to work out. The last version I saw the control panel wasnt even working, and you had to use AF/AA levels that were in game only.

They are still quite buggy, hopefully within the next 3 weeks they can get them worked out. And it is meant to compete with the GTX, the drivers just arent ready yet.

And I wouldnt put full stock into the XTX launching with the XT like the dailytech article states. I have heard the opposite, and no not from faud, from another reliable source. I should know more later today.

if they couldnt polish out the drivers for the past 6 months what makes you think the extra 3 weeks will help? and even if the drivers are good, the crappy CCC will drag everything down as always
 
Back
Top