News Headlines
- Fri, May 24
- Time and Eternity Preview: All the Single Ladies
- Joe Danger 1 and 2 set to crash onto Steam later this year, Big Picture and Workshop support included
- Sony explains why Gran Turismo 6 is staying on PS3, cites PS3 potential and install base
- Phoenix Wright: Ace Attorney Trilogy HD is comign to iOS on May 30, prepare your Apple devices
- PlayStation 4 could reach Europe within 2013, according to UK newspaper ad
New Articles
Related Articles
For the first time, we decided to forgo XP, and solely rely on Vista for running our benchmarks.
As for drivers, all the Nvidia cards were tested with the 167.26 Beta drivers, except for the SLI setup, which utilized the newer 169.01 drivers. Both ATI/AMD cards used the Catalyst 7.10 drivers.
World In Conflict: We used the built-in benchmark of the demo version of this game. We ran the benchmark in DX9 rendering mode, with a high level of quality.
Unreal Tournament 3: Was benched using Unreal Tournament 3 benchmarking tool made by King Bill & MatrixMC. This was the beta demo version of Unreal Tournament 3. We tested using the DeathMatch, ShangriLa (map) fly-through, with 3 bots, running for 48 seconds.
Hellgate: London: Stricltly speaking, this is not a benchmark. We included some performance numbers from Hellgate: London to show you what kind of relative performance levels you may get in the game. The FPS performance was gathered by using a 16 second FRAPS at the begining of the single-player demo. Due to the inherent randomness of Hellgate: London, each time we did the FRAPS run, stuff was a little bit different: monsters moved in different paths, and the intial alley had slightly different dimensions. This randomness effectively nullifies any hope of benchmarking in this single-player demo version of the game. As for settings, the following options were set: Triple Buffering ON, Engine Version DX9, Shader Quality VERY HIGH, Shadow Detail HIGH, Dynamic Lights ON, Model Detail HIGH, Model Viewing Distance VERY HIGH, Texture Detail VERY HIGH, Weather ON, and Trilinear Filtering ON.
We ran each FRAPS run twice and averaged the results to counter-balance the randomness, but again: take these results with a grain of salt.
Call of Juarez: We used the stand-alone Call of Juarez DX10 benchmarking program for these results. For our AA testing, we used a setting of 2x.
PT Boats: Knights of the Sea: This is the first time we used this demanding, stand-alone DX10 benchmark. We set the Image Quality level to 'High', and tested AA at a setting of 4x.
Bioshock:: For this benchmark, all of the Detail settings were set to 'High'. All of the graphic option switches were set to 'On', with the exception of the following three settings: Vsync, Windowed mode, and Force Global Lighting. We used FRAPS to measure framerate performance. The FRAPS run was 138 seconds, triggered from pulling the switch in the sub at game's beginning. The sub's dive involves many big models moving around, which should strain the GPU's and be a good measure of the game's engine.
Article Index |
|
next
1 2
Great review once again!
Your chart does NOT support this statement.
Huh?
"8800 GT OC SLI Idle = 185, Under Load = 275"
"8800 GTS Idle = 190, Under Load = 276"
I'm no math whiz, but, yeah.
Bottom line = I am buying a pair of these for my next upgrade. Wow!
I hear revamped GTS/GTX cards will release in November.
~Ibrahim~
Problem is the "standard of error" you speak of also supports the possibility that the SLI GT numbers are lower than posted and/or the Single GTS numbers are higher than posted. Right?
But this weak argument doesn't change the fact that you stated "Your chart does NOT support this statement".
I was pointing out that the chart DOES support the statement even if you don't.
My guess is that you were looking at the BFG 8800 GT OC numbers and doubled them rather than looking at the 8800 GT OC SLI numbers. You then went back after reading what I posted, realized your mistake and are now attempting to make yourself look better.
But just to end this off-topic discussion, I will be the bigger man here and say you were right and I was wrong.
275 > 276
You win.
Thanks for all the comments. It was quite the rush to write that review after only coming back from the California launch on Saturday -- I'm glad there aren't too many typos in it
Anyways, about the 640 mb 8800 GTS thing: just wanted to mention that it looks like the 8800 gts 640 is getting a stream processor boost:
http://www.neoseeker.com/news/story/7277/
This should make the 640 mb 8800 gts a reasonable card to get if you have a high quality display, capable of high rez'es.
Not sure what is happening to 8800 gts 320mb. Rumors I've read look like it'll be refreshed mid-November -- perhaps around the launch of the RV670.
But it might be phased out.
End of November will be a good time to buy a video card though, it looks like.
thanks
It's a tough call -- theres a couple of things: the 8800gts 640mb with the added stream processors will very probably be better for you in your situation. With revised 8800gts should outclass the 8800 gt in those resolutions because of the 320 bit memory interface tied to the extra memory.
But two things: further revisions of the 8800 gts might be unveiled soon. Also, the price has not been released (so not sure how it'll compare to the 8800 gt).
One other rumor worth mentioning is that it looks like a 1 GB 8800 GT might be out before Xmas.I'd guess the revised 8800 gts would offer better performance than the 1gb 8800 gt however.
And, yet one more thing: the RV670 will be out before not too long. Even as good as the 8800 GT is, I'd recommend waiting for this card, just to see how it compares -- you never know.. it is a 55nm card, so it should be interesting.
In a snap judgement, if you wanted to buy ASAP, I'd say you'd probably want the revised GTS: it'll presumably be offered for a reasonable price in comparison with the 8800 gt, and should offer better performance at your intended resolution.
Will that be PCIe 2? exactly what will the differences be?
Besides the additional stream processors, it looks like the new g92 / 8800 gts will have a reduced 256 bit memory interface (contrary to my last post). So it looks like the new GTS will come in 512mb and 1 gig varieties.
The 1 gig 8800 gts should out-perform a current 8800 gtx by a decent margin, and should also be available for less.
It'll be pci-e 2.00, but that really isn't much of a big deal. It'll more be a factor a year+ plus from now. PCI-e 1.00 offers enough bandwidth for the next while with 16x pci-e lanes on your mobo.
By the way guys, also from I heard today from someone else, the RV670 (upcoming hd 3800 ati release) looks like it'll offer less performance than a 8800 GT, but will retail for about 50$ less.
For your situation axforts, you'll probably be most interested in the new 8800 gts 1 gb version.
NVIDIA will probably try to take some of the fire out of the launch of the RV670 by launching it on the same day. (Which is, according to SOME RUMORS, about 3 weeks away).
Maybe it'll end up being a 8900 GT?
I can't see them sticking with the 8800 GTS name.
next
1 2