AT Exclusive Preview: The GeForce 7800 GS Hits the Scene
by Derek Wilson & Kristopher Kubicki on November 22, 2005 8:00 PM EST- Posted in
- GPUs
Splinter Cell: Chaos Theory Performance
ATI's cards perform really well under the Splinter Cell: Chaos Theory SM3.0 path. Not even the 7800 GT can keep up with the X1800 XL in this benchmark. The X850 XT was left out due to shader path we decided to use for the test. The 7800 GS might need to be run with a few quality options turned down if resolutions of 1600x1200 and up are desired.
Again, at 1280x1024, the 7800 GS is playable. Moving higher than that would require a faster card. The ATI cards really shine here and put even more distance between themselves and the NVIDIA lineup. The X1800 XL remains playable at 1600x1200 with 4xAA enabled and all the SM3.0 options turned on.
ATI's cards perform really well under the Splinter Cell: Chaos Theory SM3.0 path. Not even the 7800 GT can keep up with the X1800 XL in this benchmark. The X850 XT was left out due to shader path we decided to use for the test. The 7800 GS might need to be run with a few quality options turned down if resolutions of 1600x1200 and up are desired.
Again, at 1280x1024, the 7800 GS is playable. Moving higher than that would require a faster card. The ATI cards really shine here and put even more distance between themselves and the NVIDIA lineup. The X1800 XL remains playable at 1600x1200 with 4xAA enabled and all the SM3.0 options turned on.
Facebook
Twitter
RSS
43 Comments
View All Comments
AmberClad - Monday, November 28, 2005 - link
Sorry if this was already mentioned in the article, but is there a reason that there were no pictures of the actual card? Were there distinguishing marks/logos on the card that would have revealed to Nvidia which company had leaked the card, or was it just due to legal issues?gulizi - Saturday, November 26, 2005 - link
http://forums.anandtech.com/messageview.aspx?catid...">http://forums.anandtech.com/messageview...amp;thre...vailr - Wednesday, November 23, 2005 - link
This 7800GT http://www.newegg.com/Product/Product.asp?Item=N82...">http://www.newegg.com/Product/Product.asp?Item=N82... is ($289.00, after $50.00 MIR).
NullSubroutine - Wednesday, November 23, 2005 - link
thankfully the 12x10 4x 8x are on there. anyone know the availbility date since this was a sneak peak?yacoub - Wednesday, November 23, 2005 - link
The only thing you didn't mention in your conclusion are the prices of the ATI cards, which would really help drive the point home that the 7800 GS or GT are the way to go.LoneWolf15 - Wednesday, November 23, 2005 - link
That Anandtech has posted that cards in a test (I saw this last week too, but didn't mention it then) include the Radeon X800XL, and then didn't. I'm getting really tired of it. It seems like a simple thing, but a lot of people purchased the X800XL as a performance/value solution, and the comparison is worth it to us. I'd like to see Anandtech tes with it, but if you decide not to, REMOVE IT FROM THE LIST OF CARDS IN YOUR TESTING SO THAT NO-ONE IS LED ON. Don't you guys have ANY PROOFREADERS?(pauses, takes several deep breaths) Rant over. Carry on.
P.S. It appears that in looking closer, your tests include the X1800XL, which is not listed in the test specs. Perhaps the specs have a typo, but if so, it's rather an egregious one...maybe you could correct it?
yacoub - Wednesday, November 23, 2005 - link
I'll second the notion that the X800XL would be nice to include since that was the hottest bang-for-the-buck card just a couple months ago.DerekWilson - Wednesday, November 23, 2005 - link
Would you guys rather see an X800 GTO or an X800 XL in future reviews of this nature?Also, it was a typo. It should have read X1800XL. I'm sorry about the confusion.
As always, we very much appreciate our readers pointing out errors we need to correct. Thanks!
Matkun - Wednesday, November 23, 2005 - link
There seems to be something wrong with the x1800 parts in the Black and White 2 Benchmarks, since they perform quite a bit below the x850 xt, even though they are faster in every way.Is it really proper to include scores that seem to be bugged?
DerekWilson - Wednesday, November 23, 2005 - link
If a hardware/software combination has a problem isn't it even more important that we report it? This is an accurate measure of the performance of the different parts with equivalent settings. We checked and rechecked our benchmarks this time and there's no doubt that R4xx hardware runs B&W2 better thant R5xx hardware with the latest ATI drivers.This is a problem and we hope ATI or Lionhead will take notice and fix the issue.