The GTX 590 vs. the HD 6990 – only One is “the World’s Fastest video card”
S.T.A.L.K.E.R., Call of Pripyat is the third game in the S.T.A.L.K.E.R. series. All of these games have non-linear storylines which feature role-playing game elements. In both games, the player assumes the identity of a S.T.A.L.K.E.R.; an illegal artifact scavenger in “The Zone” which encompasses about 30 square kilometers. It is the location of an alternate reality story surrounding the Chernobyl Power Plant after another (fictitious) explosion. S.T.A.L.K.E.R., Call of Pripyat features “a living breathing world” with highly developed NPC creature AI.
Call of Pripyat is compatible with DirectX 8, 9, 10 and 10.1. It uses the X-ray 1.6 Engine with dX 11, one outstanding feature being the inclusion of real-time GPU tesselation– a Shader model 3.0 & 4.0 graphics engine featuring HDR, parallax and normal mapping, soft shadows, motion blur, weather effects and day-to-night cycles. As with other engines using deferred shading, the original DX9c X-ray Engine does not support anti-aliasing with dynamic lighting enabled, although the DX10 and DX 11 versions do.
We are using the stand-alone “official” benchmark by Clear Sky’s creators. We picked the most stressful test out of the four, “Sun shafts”. It brings the heaviest penalty due to its extreme use of shaders to create DX10/DX10.1 and DX11 effects. We ran this benchmark fully maxed out in DX11.0 with “ultra” settings plus 4xAA, including applying edge-detect MSAA which chokes performance even further.
Here we present our maxed out DX11 settings for S.T.A.L.K.E.R., Call of Pripyat DX11 benchmark with 4xAA at 2560×1600:
The HD 6990 wins the top spot by a couple of frames per second. Now we back off the AA and resolution to our usual settings.
Now we move on to 1680×1050 with 2xAA:
This game was impossible to play last year at the highest settings with a single card and much if any AA. Now we see our cards breeze through 1680×1050 and 1920×1200 with the GTX 590 leading the HD 6990. We saw that if we upped the resolution to 2560×1600 and also increased the AA to the maximum 4xAA the benchmark supports, the HD 6990 pulls ahead of the GTX 590 by nearly 2 frames per second. We are seeing a trend where the GTX 590’s lead gets cut down as the resolution and AA go way up. Part of it may be attributed to the larger framebuffer of the HD 6990, but it is just likely architectural differences.
This review was so absolutely great. Miles above the Anand’s 10 games. I really really think you go above and beyond. One of the few places you can get a full review anymore. Such a great collection of data! Most sites do a handful of games and there is no way anyone could get a true picture. Most sites have diminished their launch articles to a point of incompleteness. Where one has to go somewhere else to get a full picture.
Apoppin, i know its a lot of work for you, but i commend you for your dedication to all the data. Yours again is the most useful especially on a card that is so close in performance to the competitors. I cannot thank you enough for a fantastic article!
Very nice in-depth review, bookmarking the site now! Thank you for your hard work!
I love ABT reviews. They always seem to be pretty fair when it comes to in-depth testing. Keep up the good work guys. Really. I mean that. I plan to work for one of the many review sites when i finish my CE degree, and ABT is high on the list.
I’m impressed that Nvidia beat ATI in cooling with their dual GPU. It’s usually the opposite.
It is nice to see the cost for both cards is about the same which gives more choices for the consumers to pick from.
Nice Job!
Also a small note:
When I click “View All” it doesn’t work, could you guys get it fixed?
Our web master is aware of this issue and is working on it.
Thank-you all for your comments!
The HD 6990 is honored under warranty by most vendors with the OC switch. So I’m quite puzzled why you OC’d the 590 and kept the 6990 the same.
If you want to see the HD 6990 tested with the BIOS position No. 2, please check out our HD 6990 launch article:
http://alienbabeltech.com/main/introducing-the-worlds-fastest-graphics-card-amds-flagship-hd-6990
The same settings were used and now you have both settings completely covered.
You will also note in the Performance Summary of this GTX 580 launch article that we pitted our overclocked HD 6990 (960/1390MHz) against our overclocked GTX 580 (690/1825Mhz)
I believe we covered all of the bases.
You pointed out that bios switch 2 was tested in another review. But that does not answer why it was not tested in this review side by side with the 590(I hope that was not your answer as to why). Just saying it’s a bit weird you throw in a bunch of old games one being from 2005, but not test something most 6990 owners would be using.
BIOS position No. 2 was tested in the review immediately preceding this one and anyone can easily check the performance difference:
http://alienbabeltech.com/main/introducing-the-worlds-fastest-graphics-card-amds-flagship-hd-6990
The launch article about the GTX 590 tested the stock GTX 590 against the stock HD 6990. Then we tested the overclocked GTX 590 and the overclocked HD 6990 – both overclocked as far as they would go – the 6990 being overvolted in the BIOS No. 2 position and the GTX 590 at stock voltage.
Including a third set of Radeon slightly-overclocked numbers would have cluttered the charts and taken precious extra time that was spent in benching. Especially because these figures are easily found in the previous article that were tested at the same settings and with the same drivers.
We don’t “throw in” a bunch of old games at random. We have been following many of these games for nearly three years – when some were only a couple of years old and many top cards struggled with them – and our regular readers appreciate it. We also include a good mix of new games and are always adding more. Next added will be Bulletstorm and Shogun 2 and of course Crysis 2 when it fully debugged and running on the DX11 pathway.
magnificent review
This wasn’t a terrible review but it was definitely biased towards nVidia.
In the Metro 2033 section you actually claim that the GTX 590 “Beats” the Radeon 6990 with 39.48 FPS vs. 39.2 FPS. You even used an exclaimation point!
Comparisons within 3-6 FPS are inconclusive, let alone 0.3 FPS! Calling that any kind of defenative win is just absurd and proves you’re milking every possible edge the 590 might have.
Don’t get me wrong; I’m using an nVidia card right now (Good ol’ GTX 460 SC) but I appreciate that both companies make stellar cards and it’s only fair to try and be unbiased as a reviewer since GPU consumers tend to be so polarized.
Ultimately nVidia tends to have a slight performance edge and AMD tends to have a price edge. The only things that really make each company’s cards unique are special features like CUDA and Eyefinity.
I dont think the 590 or equivelent is such a big step from the card I have (the 480) for me really to be that excited.
The step from 8800gtx to gtx480 was a really huge step tho (3x the improvement) like this is more like .5x the improvement, so I dont really give a shit.
Sorry for mentioning all those nvidia cards, but its what I purchase, I work with CUDA for my gpgpu programming, so maybe even im a little biased, but opencl is what you use for amd cards, and its just as good.
Way too many comparisons of other GPUs, it’s 590 vs 6990, remember? I was all confused with the color schemes on the graphs.
I know quite a little about gfxcards so I didnt find answers why nVidia is better in some games and AMD in other games? Directx vs opengl? I play only Waw, BO, ET and Brink so I still dont know do I buy next 580, 590, 6970 or 6990. I have now 570 SLI but somehow my computer doesnt work well when SLI is on so single card is the solution. But which one?
I would highly recommend that you ask you question in our forum. You will get good information there and far more detail than from comments here.
SLI should work well if you have a SLI MB. And there are many reasons why Nvidia is better in some games and AMD in others.
http://alienbabeltech.com/abt/
Would it be possible to get a copy/screenshots of the current bios setting using during the tests?
Wicked not biased review. Not.
what kind of a joke is this? , we are compareing over all proformance , why the HELL is phyx disabled ? , it is fair because its an official feature of nvidia , if amd does not have it , its not nvidias fault , we want an answer , it looked like u guys wanted nvidia to look slow or someting ?
There is no other way to compare performance directly.
How many PhysX games are we comparing? Two out of nearly thirty games. Just ignore those results and look at the rest of the review. We test about 3 times more games than any other tech site.
@najeeb what kind of joke are you on about? physx makes Nvidia cards slower in performance if turned on, if physx was on it’d be unfair to Nvidia right? In your case you should say “It actually seems like they’re trying to make Nvidia cards look faster than the AMD counterparts by disabling certain features.”
Wow that was extremely bias. Xbitlabs.com comparison was perfect.
I just watched an youtube video comparing both cards and Radeon uses a lot less power from psu and gave a better or equal fps result than Nvidia. Others reviews compare both cards to be equivalent. I think this review is very outsiders and biased.
Definitely appreciate you discussing this article. Great!!
I certainly enjoyed the way you explore your experience and knowledge of the subject! Keep up on it. Thanks for sharing the info!
Can anyone able to recommend comprehensive Rice B2B Database? Thank you 😀