Review Index:
Feedback

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

Summary and Conclusions

If you haven't read my previous article, as I have referenced several times before, you need to do so.  Not only does it have all the background on the testing process used here and some more information on what exactly each separate graph represents, it also includes some information on how Vsync can affect the issues that AMD CrossFire sees as well as some problems we found with Eyefinity.  

Performance

There are some interesting things that we need to take into consideration based on all the data we are showing you here.  First, the same problem that existed in our Frame Rating launch article shows up here with the Radeon HD 7970 cards - which makes perfect sense as they are the same results!  In three of our six tested games, Battlefield 3, Crysis 3 and Sleeping Dogs, the HD 7970s in CrossFire showed significant runt frames that tell us the animation smoothness and frame rate is not nearly as good as reported FRAPS average frame rates would like you to believe, AND that in our opinion buyers are being shorted on performance they are paying for with a second card.  Three other games, Dirt 3, Skyrim and Far Cry 3 didn't have this problem. 

View Full Size

 

Benefits of Single GPUs

The GeForce GTX 690 and the GeForce GTX Titan paint an interesting picture if you just consider them against each other.  Both cards will cost you a cool $999, but offer very different hardware configurations.  The GTX 690 is a dual-GK104 product with 4GB of total frame buffer and 2GB for each GPU; essentially slightly slower GTX 680s in SLI.  The GTX Titan is a totally different beast with a single, larger GK110 GPU (2,688 cores in a single GPU, 1,536 in each GK104) and a larger 6GB frame buffer.  When it launched many people complained that the GTX Titan was overpriced because it offered lower performance than the GTX 690 - which shows to be the case in most of our FRAPS benchmarks and even the observed FPS data. 

What you might not have seen before though is what happens when we crank up the resolution to triple monitor setups - our 5760x1080 benchmarks.  In those cases, pretty much across the board, we found that the GTX Titan was able to produce much more smooth and consistent frame times compared to the GTX 690.  This is likely a combination of both the large frame buffer of the Titan, (triple the memory) as well as not having the need to worry about balancing frames across multiple GPUs and doing any kind of frame metering.  When the GTX 690 is running games at higher settings on 5760x1080, it struggles to keep up.

But on single monitors the GTX 690 is able to bypass any inherint problems with multi-GPU configurations and produce a smooth frame rate.  It's not quite AS SMOOTH as the GTX Titan but in general I would say the additional frame rate performance of the GTX 690 out weighs the frame time smoothness advantage that the Titan has.  Take Sleeping Dogs at 2560x1440 - the average between the two cards goes from 35 FPS to nearly 45 FPS, and maximum variance only goes from 1 ms to about 3 ms. 

View Full Size

 

What this means for the upcoming Radeon HD 7990 Release

While we aren't actually testing a Radeon HD 7990 here, we are basically testing the exact same configuration with a pair of Radeon HD 7970s running in CrossFire.  This setup emulates the ASUS ARES II and the Powercolor Devil 13 pretty closely and from what we are hearing will be very close to what you'll find on the official reference HD 7990 as well.  This comparison today wasn't in response to AMD's tease of information at GDC this week, but it is well timed none the less. 

The results shouldn't surprise you, and it won't surprise AMD any more either - if released today, the HD 7990 would not perform well in our tests.  AMD has told me that they are working on an option to meter frames in the way that NVIDIA is doing it, but offering users the options to enable or disable it, but we are months away from that fix. Until then, any dual-GPU Radeon HD 7000-series cards are going to show these problems represented as runts and dropped frames.  We have many more pages of results to go over for the HD 7950/7870/7850/etc and those will be published in the coming days - but the story will look very similar as you'll find.

In all honesty, when AMD told me they were planning this card release I recommend they hold off until its driver fix is in place - myself and other reviewers are going to be hitting them hard on these issues until then, and any dual-GPU option with the Radeon name is going struggle to live up to any placed expectations. 

 

Final Thoughts

The second part of our final reveal of the Frame Rating performance methods have shown us some interesting results for the $999 and above card lineups as they stand today.  The Radeon HD 7970s in CrossFire, representing the currently available and upcoming HD 7990s don't look great in our testing as we mentioned above, and I would seriously consider your buying decision before picking up this configuration. 

View Full Size

NVIDIA's GeForce GTX 690 and GTX Titan present another debate though on the issue of single versus multi-GPU options.  In our testing the GTX 690 definitely introduces more frame time variance than the GTX Titan, but the cards are so fast as it is that running the GTX 690 at single monitor resolutions like 1920x1080 and 2560x1440 don't cause a big enough problem to be a factor, thus giving it the edge because of the higher average frame rates.  The GTX Titan is a powerful card and performs admirably in the single display testing but it really stands away from the GTX 690 on multi-monitor resolutions like 5760x1080 where the 6GB frame buffer can help a TON.  Not having to worry about moving frames between GPUs at that resolution also helps produce a smoother animation as well.  If you are buying a $1000 card and you think you might want to run NVIDIA Surround, then the GTX Titan is your better solution.

 

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:

March 30, 2013 | 01:40 AM - Posted by John Doe (not verified)

Personally, I'll be the first to say that I could care less about ANY of this shit since I own a Sparkle Calibre 680 and a Galaxy 680 White to complement it.

:)

March 30, 2013 | 01:44 AM - Posted by Anon (not verified)

Just because you can post doesn't mean you have to.

March 30, 2013 | 06:35 AM - Posted by John Doe (not verified)

The same applies to you as well.

June 2, 2013 | 11:11 PM - Posted by clearlice (not verified)

Soon after all these years I had been undertaking this exercising
pondering it was going to whip me in shape, wow!

March 31, 2013 | 12:02 PM - 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 | 03:20 PM - Posted by renz (not verified)

60 fps won't matter if it doesn't play as smooth as it should be. hence the whole point of these article

April 4, 2013 | 03:22 AM - Posted by Anonymous (not verified)

you mean at a reported 60 fps. as you can see by the charts above, reported versus actual is about a 2:1 ratio. so take that as the radeon running at 30 fps.

April 5, 2013 | 09:50 AM - Posted by Anonymous (not verified)

how stupid is this hole page comparing two 7970's and not a 7990

July 8, 2013 | 09:39 AM - Posted by Anonymous (not verified)

How stupid are you for not spelling whole right?

March 30, 2013 | 02:00 AM - Posted by rezes

you using still old drivers!!

AMD: 13.2 beta 7 ????

March 30, 2013 | 03:38 AM - Posted by Edge86 (not verified)

As you should have noticed also the NV Drivers where older (314.07 & 314.09 for TITAN). This is because he started a time ago to make these time-consuming tests. At that Moment the AMD Driver wasn't old. :)

Thanks Ryan for the interesting Frame Rating tests so far (and upcoming). Would it be possible the add 3-way & 4-way MGPU-Setups for an extra review in future?

Greetings,
Edge

March 30, 2013 | 05:37 AM - Posted by Ryan Shrout

LOL 13.2 beta 7 is SOO OOLLDD!!  :)

All of our testing was completed as of March 15th, and it was the latest driver as of that date.  And trust me, nothing that affects what we are seeing here is changed with the latest beta drivers. 

As for 3/4 card configs, we can do it, its just a time concern now.  

March 30, 2013 | 06:27 AM - Posted by John Doe (not verified)

Talk about being subjective.

Not everyone updates their drives like they change their underwear, dummy.

June 2, 2013 | 11:11 PM - Posted by Darlene (not verified)

Soon after all these years I had been carrying out
this physical exercise considering it was going to whip me in shape,
wow!

March 30, 2013 | 07:03 AM - Posted by Prodeous (not verified)

Totally agreed with you.

March 30, 2013 | 09:38 AM - Posted by rezes

Thank you for answering

March 30, 2013 | 03:46 PM - Posted by technogiant (not verified)

Would be interesting....I've heard else where that tricrossfire eliminates stuttering.....at least subjectively....would be good to shine the truth light of "frame rating" on that statement.

March 30, 2013 | 05:27 PM - Posted by arbiter

I doubt it eliminates the stuttering problem as it just increases FPS to a point you can't see it anymore. nvidia doesn't have stuttering cause its what they do on the hardware to prevent it which AMD doesn't but supportable gonna make a fix for in July which would have to be software fix.

April 2, 2013 | 01:16 AM - Posted by ThorAxe

I'd like to see that too. I used to run 4870x2 + 4870 in Tri-fire and I don't recall stutter. However, I have run 6870s in Crossfire and did notice issues in BFBC2.

I don't have problems with my GTX 570 SLI or GTX 680 SLI PCs.

It would be great to see older cards tested too.

April 1, 2013 | 10:30 AM - Posted by mfpterodactyl (not verified)

Oh, Ryan you rascal! Are you not familiar with the MO of the AMD diehards? These guys are more deluded than the Westboro baptist church.

Let me bring you up to speed.

The latest beta driver, released at 3am the night before, not available on the official AMD website, and available only through obscure links passed around by the AMD diehards, is the JESUS DRIVER THAT WILL FINALLY CHANGE EVERYTHING! It will unlock all the untold power that AMD diehards just KNEW was in their cards all along. This situation repeats EVERY SINGLE TIME a new beta driver is released, publicly or not.

To be perfectly honest you're throwing the diehards way too much of a bone by using beta drivers to begin with. You wouldn't review beta (non-production) hardware, now would you? But that's another discussion for another time.

I really am sorry Ryan for all the venom some people will spit at you now and in the future because you're the first tech journalist in quite a while to actually do some investigative journalism instead of being a press release parrot. There are a lot of people that appreciate all your hard work, please don't forget that. The real value in what you're doing is not just in evaluating current hardware and making better buying decisions, but in shaping the future of the industry. Because of your hard work, future hardware WILL run games more smoothly than they would otherwise, and for that we thank you. Thanks, Ryan.

April 3, 2013 | 12:32 PM - Posted by Anonymous (not verified)

Hope and beta driver change. Praise be the new frame rate!
The future is here and this is the driver we have been waiting for! Honor be upon Catalyst Maker, the oceans are receding and all crossfire rigs now rise with this tiding of good joy!
Banished are the runts frames, in Abu Dhabi's name.
AlluAMDahkbar!

April 1, 2013 | 09:48 AM - Posted by steen (not verified)

I agree, 3/4-way multi gpu setups might show interesting results.

People are getting too hung up depending on their preference. This is genuinely interesting stuff, that has't been applied by end-user review sites before. Both IHVs have resources that are beyond anyone else. I don't buy for a minute that any of the major IHVs doesn't analyze the render pipeline in great detail.

March 30, 2013 | 02:22 AM - Posted by pdjblum

Ryan,

Great work you have done. Has Scott at techreport had anything to say about it?

March 30, 2013 | 01:19 PM - Posted by pdjblum

Disappointing that I never get a response from the staff. I have been a loyal reader for some time now, and do not get why you guys cannot once in a while respond.

March 30, 2013 | 01:39 PM - Posted by Josh Walrath

I love you, and I care for you.  I just have no idea what Scott has communicated with Ryan about.

March 30, 2013 | 01:43 PM - Posted by pdjblum

Back at you Josh. Thanks for replying.

March 30, 2013 | 02:22 PM - Posted by Ryan Shrout

I'd rather keep conversations between Scott and I private. 

March 30, 2013 | 03:01 PM - Posted by pdjblum

I was just hoping he was impressed with how far you have taken this thing. No doubt, your work has nvidia and amd taking notice, which is quite a thing.

March 30, 2013 | 05:29 PM - Posted by arbiter

nVidia made some tools that are used in this testing which pcper is in process of making their own up to move away from using nvidia made tools

April 2, 2013 | 08:45 PM - Posted by Tom Petersen (not verified)

I watch everything Ryan does:)

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.