Review Index:
Feedback

Frame Rating: GeForce GTX Titan, GeForce GTX 690, Radeon HD 7990 (HD 7970 CrossFire)

Summary Thus Far

Because of the complexity and sheer amount of data we have gathered using our Frame Rating performance methodology, we are breaking it up into several articles that each feature different GPU comparisons.  Here is the schedule:

Welcome to the second in our intial series of articles focusing on Frame Rating, our new graphics and GPU performance technology that drastically changes how the community looks at single and multi-GPU performance.  In the article we are going to be focusing on a different set of graphics cards, the highest performing single card options on the market including the GeForce GTX 690 4GB dual-GK104 card, the GeForce GTX Titan 6GB GK110-based monster as well as the Radeon HD 7990, though in an emulated form.  The HD 7990 was only recently officially announced by AMD at this years Game Developers Conference but the specifications of that hardware are going to closely match what we have here on the testbed today - a pair of retail Radeon HD 7970s in CrossFire. 

View Full Size

Will the GTX Titan look as good in Frame Rating as it did upon its release?

If you are just joining this article series today, you have missed a lot!  If nothing else you should read our initial full release article that details everything about the Frame Rating methodology and why we are making this change to begin with.  In short, we are moving away from using FRAPS for average frame rates or even frame times and instead are using a secondary hardware capture system to record all the frames of our game play as they would be displayed to the gamer, then doing post-process analyzation on that recorded file to measure real world performance.

Because FRAPS measures frame times at a different point in the game pipeline (closer to the game engine) its results can vary dramatically from what is presented to the end user on their display.  Frame Rating solves that problem by recording video through a dual-link DVI capture card that emulates a monitor to the testing system and by simply applying a unique overlay color on each produced frame from the game, we can gather a new kind of information that tells a very unique story.

View Full Size

The capture card that makes all of this work possible.

I don't want to spend too much time on this part of the story here as I already wrote a solid 16,000 words on the topic in our first article and I think you'll really find the results fascinating.  So, please check out my first article on the topic if you have any questions before diving into these results today!

 

Test System Setup
CPU Intel Core i7-3960X Sandy Bridge-E
Motherboard ASUS P9X79 Deluxe
Memory Corsair Dominator DDR3-1600 16GB
Hard Drive OCZ Agility 4 256GB SSD
Sound Card On-board
Graphics Card NVIDIA GeForce GTX TITAN 6GB
NVIDIA GeForce GTX 690 4GB
AMD Radeon HD 7970 CrossFire 3GB
Graphics Drivers AMD: 13.2 beta 7
NVIDIA: 314.07 beta (GTX 690)
NVIDIA: 314.09 beta (GTX TITAN)
Power Supply Corsair AX1200i
Operating System Windows 8 Pro x64

 

On to the results! 

Continue reading our review of the GTX Titan, GTX 690 and HD 7990 using Frame Rating!!

April 1, 2013 | 05:09 PM - Posted by Pendulously (not verified)

To put it simply: 'NUMBER OF FRAMES WHICH EXCEED THRESHOLD' should be on the X-Axis, and the THRESHOLD (in milliseconds) will be specified by the corresponding Y-Axis value.

March 31, 2013 | 04:17 AM - Posted by gloomfrost

Fantastic article Ryan. I think this will be the reference for future benchmarks and keep all GPU manufactures honest! In the podcast you mentioned having uploaded a non you tube video without the compression artifacts of flash and you tube. Is there anywhere i can download the video for the Skyrim comparison?

Thanks

April 3, 2013 | 10:03 AM - Posted by Anonymous (not verified)

I think you meant "keep the lying amd con artist CF foobar for years, incapable of getting away with straight out fps lying, by sheer force of facts."

In the case of nVidia, they already gave all of us the benefits of smooth frames for years.

This means of course, all the screaming denials by the amd fanboys and websites have been utter and total crap, for years on end.

I said so, and was attacked relentlessly.
Delicious amd fanboy tears, years of raging failure, total VINDICATION.

March 31, 2013 | 09:01 AM - Posted by Anonymous (not verified)

http://www.youtube.com/watch?v=U8HVQXkeU8U&list=PLLpjKzWIuEfmHcyTF7Qxuh1...

^

Battlefield 4 running at 60FPS, 3K resolution on an AMD Radeon 7990.

PCPER=Fail!

March 31, 2013 | 10:37 AM - Posted by bystander (not verified)

What does that video have anything to do with this article?

April 2, 2013 | 10:58 AM - Posted by Anonymous (not verified)

According to the auther 7970x2 is bad, yet they are playing nicely!

April 3, 2013 | 03:14 AM - Posted by rrr (not verified)

I still see some very little hiccups throughout the video. Not saying it can't be acceptable for anyone, but there are people, who would like better fluidity than this.

April 3, 2013 | 03:15 AM - Posted by rrr (not verified)

I still see some very little hiccups throughout the video. Not saying it can't be acceptable for anyone, but there are people, who would like better fluidity than this.

March 31, 2013 | 03:42 PM - Posted by db87

AMD: 13.2 beta 7????

Catalyst 13.2 Beta 7 is from 2/26/2013

The latest drivers are Catalyst 13.3 Beta3, dated 3/20/2013

March 31, 2013 | 04:29 PM - Posted by bystander (not verified)

As said a few times, it takes time to do all the data collecting, analysis, graphs and write an article. They used the most up to date drivers when they started the benchmarking.

Do you think this stuff is done instantly?

June 5, 2013 | 12:14 PM - Posted by Santo (not verified)

But this, οf сourse," you'd only be partly correct. One Missed ShotIn the beginning, it's common to miss a lot of ground to cover. Remember to give your class a reward. We get into the caves. Some people own swamp buying used cars with bad credits as a form of outdoors competition for the working line. 4mpg and if you want the wheels to look sporty, classy, or clean. Signal services are usually offered on a monthly basis.

My site ... сom/groups/buy-a-caг-with-bаd-сredit-and-nο-money-down/">autocrat

March 31, 2013 | 05:00 PM - Posted by Anonymous (not verified)

Honestly running RadeonPro with DFC and vsync fixs most of these games. I have played sleeping dogs at 60fps and it was glass smooth with RP. Most of the CFX problems are without vsync on - which in my opinion is not an option. So i've never had these so called runt frames affecting my gaming by default. I'm playing crysis 3 atm with radeon pro DFC and vsync and its pure smoothness. Sure the current drivers are a problem for people who game without vsync but i don't see why anybody would game without vsync.

Its interesting they recommend people not to buy AMD cards just because they can't be bothered to run a 3rd party app. Try suggesting things to fix the issue rather would help gamers more.

April 2, 2013 | 10:59 AM - Posted by Anonymous (not verified)

That wouldn't please the sponsor (nvidia).

April 3, 2013 | 09:01 PM - Posted by Anonymous (not verified)

Well, then, we need a totally new set of benchmarks for the past year plus, where 60FPS with vsync on is the bench, and either it does at the settings or it doesn't.

That means scrapping every single claim 7970 cf overclocked or whatever has higher fps, just throw ALL OF IT OUT FOREVER.

Thanks for "the solution", and you're welcome for the "implications of your solution" mr amd fan boy.

March 31, 2013 | 06:09 PM - Posted by Anonymous (not verified)

All the AMD fanboys are pissed. Oh no.

April 3, 2013 | 05:09 AM - Posted by Panta (not verified)

which put you in the same category.. a fan boy :|

April 3, 2013 | 10:07 AM - Posted by Anonymous (not verified)

WRONG.

Noticing the lying liars and their rager reactionary scunge brains borging out about the calamity of their full facepalm after YEARS of hearing them scream two amd crossfire For The Win !!!!!! does not a fanboy maketh.

April 3, 2013 | 05:09 AM - Posted by Panta (not verified)

which put you in the same category.. a fan boy :|

April 1, 2013 | 02:16 AM - Posted by CaptTomato

As I said at Techreport, when it comes to single GPU, framerates are all that matter on most occasions, so don't be put off buying a 7870/7950/7970 because of XF results.
I have a single 7950 and it's killer

April 3, 2013 | 10:14 AM - Posted by Anonymous (not verified)

That's fine but all those years of squacking future proof and buy one amd card now because in the near future you can go with CF and have a huge boost when you need it....

BLAH BLAH BLAH BLAH BLAH is what those endless YEARS of amd justifications now mean.

I tell you one thing, I noticed a year ago how some of these websites started shying away from dual card recommendations, which of course means they all had the info and inkling on this problem with amd crossfire, and no doubt nVidia was hammering their little eardrums and deep down inside they knew - not to mention when one of the did talk, they posted how CF failed in no less than 50% of the popular games they were using in their testing suites... it became IMPOSSIBLE to recommend CF with a clear conscience, but the fanboy in them made certain they included SLI in the "do not recommend" category as well.

This is what I saw develop, and there is no doubt why it went that way. When the crap underdog card is failing miserably but the cover is still not blown, make ceratin to cut down the competition, too, that way, you never really fully recommended the failing dirty amd dog double fps lying CF setups...

It's so sad. It's so facepalm. It has made every single site that does reviews look like clueless fools, and in fact, they WERE for YEARS.
Credibility ?
Not so much, so very little. So very, very little.

April 1, 2013 | 02:26 AM - Posted by CaptTomato

Damn, I tried registering for the forums and I couldn't do it.

April 1, 2013 | 06:43 AM - Posted by svnowviwvn

So where is the "3/31: Radeon HD 7950 vs GeForce GTX 660 Ti (Single and Dual GPU)" review?

It is 4/1 and I do not see it up on the site.

April 1, 2013 | 08:53 AM - Posted by Josh Walrath

Sometimes life does not always cooperate with review release schedules.  I should know, I'm the champ when it comes to that!

April 1, 2013 | 11:12 AM - Posted by Anonymous (not verified)

Nvidia looks good under frame rating with either single GPU or multi-GPU. How does Nvidia do with Physx enabled?

Just wondering...

April 1, 2013 | 01:21 PM - Posted by GarTheConquer

Ryan, thank you so much for bringing this issue to light!
I have an Xfire-7970 setup and because of your research, AMD will have no choice but to remedy this problem (eventually).
Thank you, your hard work is greatly appreciated.

April 1, 2013 | 03:00 PM - Posted by svnowviwvn

Ryan thanks for the very detailed reviews.

I for one would also like to see "Frame Rating" reviews done for all single GPUs, both discrete and integrated. That includes AMD APUs, Intel's HD series, Nvidia's Optimus, AMD's Endora. Laptop & Desktop both Single GPU and/or CF/SLI.

Since we can now see that Fraps numbers can be suspect and that drivers/implementation can result in runts/missing frames I want to see if any of the above mentioned systems result in lower actual FPS than the Fraps number shows.

April 1, 2013 | 04:23 PM - Posted by Pete (not verified)

Amazing work Ryan. Much appreciated. PCper is now my #1 site for hardare reviews. Can you possibly show Titan SLI frame times in the future. Thanks bud.

April 1, 2013 | 07:13 PM - Posted by seravia

Where is the next part?

3/31: Radeon HD 7950 vs GeForce GTX 660 Ti (Single and Dual GPU)

April 2, 2013 | 04:08 AM - Posted by rezes

At the Skyrim game. Fraps FPS and observed FPS equal for AMD radeon cars! however of frame time variance. How can we explaining that situation? Skyrim used DX9.0! is answer it?

It seems to AMDs drivers problem at the only on the DX10-DX11

In the meantime, I am using gtx690 in my system.I was just wondering cause of the problem.

April 2, 2013 | 11:27 AM - Posted by ZoranICS

Great article guys! I hope AMD is watching and are doing all they can to fix this issue that has been known for years.

I am amazed how many ignorant people actually post in this discussion. It seems that even hard facts aren't good enough for some dumb people. :|

I prefer AMD cards (Though I run a GTX670 at the moment) and hope they fix this ASAP.

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Lines and paragraphs break automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd> <blockquote><p><br>
  • Web page addresses and e-mail addresses turn into links automatically.

More information about formatting options

By submitting this form, you accept the Mollom privacy policy.