Review Index:
Feedback

Frame Rating Dissected: Full Details on Capture-based Graphics Performance Testing

Crysis 3 – HD 7970 versus GTX 680

View Full Size

Crysis 3 is the currently the biggest GPU hog and both the GTX 680 and HD 7970 handle it equally well at 1920x1080.  Even using our FRAPS metrics though, GTX 680s in SLI are scaling better than the HD 7970s in CrossFire.

View Full Size

The frame time plot from our Frame Rating system shows another instance though of CrossFire’s inability to keep consistent animation on the screen.  The single card configurations are pretty consistent with each other but both also exhibit some tiny bumps in frame times on a repeating pattern, obviously a particular of the CryEngine.  SLI does have some increases in frame time variance across the board with a few minor “hitches” as well.  CrossFire though appears to be alternating between 2ms frame times and up to 50ms frame times resulting in…

View Full Size

Not only a lower observed frame rate but a frame rate that is LOWER than the single card!  I can tell you from first-hand experience that this definitely was the case in play through as well; it felt slower than the single card experience.

View Full Size

SLI looks fantastic in this graph and is able to take the matching performance of the GTX 680 and HD 7970 up from 30 FPS average for the entire run to 57 FPS.

View Full Size

Our custom ISU rating tells me that the GTX 680 SLI configuration looks GREAT and only differs from the single card configurations at 95% and above percentile.  The HD 7970 in CrossFire though shows huge amounts of variance from the outset and in fact does exhibit a lot of stutter in game as well.

 

View Full Size

At the higher resolution the single card HD 7970 as a slight edge over the GTX 680 this time around and this time the scaling of CrossFire appears to be faster than SLI.

View Full Size

A quick glance at our results from the observed frame rate clearly shows that isn't the case though as only in short bursts does the CrossFire experience actually match that of the dual GTX 680s in SLI.

View Full Size

Our frame time plot indicates where the alternating frame times in Crysis 3 occur with CrossFire and how it relates to the performance of SLI.  Other than the single large hitch seen at the 12 second mark or so in SLI, the GTX 680s handle Crysis 3 much better.

View Full Size

The GTX 680s are able to scale from about 19 FPS on average to 36 FPS - a solid 89% scaling factor.  The HD 7970 GHz Edition cards are not so impressive, only going from 21 FPS to 25 FPS but that quickly falls down to just even performance with a single card.

View Full Size

Ouch, another very poor result here for HD 7970s in CrossFire with Crysis 3 at 2560x1440 with as much as 25 ms of frame variance (nearly two full refresh cycels). 

 

View Full Size

This is one of the few Eyefinity runs for the HD 7970 CrossFire configuration that was able to run until completion and generate the necessary graphs for us.  So we’ll finally get to see some interesting results.  Even at first glance, we can tell that something here isn't quite right.  According to FRAPS, the HD 7970 is pushing out more than 200 FPS to the screen on Crysis 3 at 5760x1080, which is obviously inaccurate. 

View Full Size

Ouch, there are definitely some problems here, not the least of which is the graphs poor setting of range maximums (will fix soon!)  Notice the spots on the plot of the orange line (HD 7970 CF) where there is no data – that indicates a dropped frame and a lot of frame time variance. 

View Full Size

Removing those and any runts we find that the observed FPS is actually right in line with that of the single HD 7970 graphics card.  Also, without the CrossFire misreported results out of the window, the update scale helps us see the scaling that the GTX 680s in SLI. 

View Full Size

Here again is another one of our RUN files to show you the affects of dropped frames on Eyefinity testing.  FRAPS based frame rates sky rocket up though the observed frame rate is much lower, in line with a single HD 7970 GHz Edition card.  There are some runts involved in this but the biggest factor is obviously the dropped frames (missing colors in our pattern).

View Full Size

Again, for comparison, here is the RUN graph for the GTX 680s running in SLI at 5760x1080.  Notice that the frame rate is consistent with no drops or runts.

View Full Size

Interesting results here with the CrossFire setup taking a lower position across the board in our percentile minimum frame rate graphs. Also, we see the pair of GTX 680s in SLI start out much faster than anything else tested but at the 94th percentile or so fall below that of the HD 7970.

View Full Size

Keeping mind that we are looking at pretty low frame rates across the board, the HD 7970 has the best overall result here in our ISU graph with the least amount of frame variance over the course of our 60 second run.  Obviously CrossFire has a big issue once again and see significant variance starting at the 50th percentile and it only gets worse from there.

March 30, 2013 | 05:47 PM - Posted by bystander (not verified)

I have a hard time trying to grasp exactly how erratic input would affect the results. I have a feeling, based on my constitution (I get simulator sickness with poor latency), that the best case is which ever has the lowest worst case interval times.

March 30, 2013 | 10:55 PM - Posted by bystander (not verified)

..But then you have occasional latency increases. Of course those increases are to remove redundant frames, and once increased, they probably don't need much adjustments most the time.

This whole topic always gets me going back and forth, but my instincts is overall, even if latency is considered, even spacing matters more as it adds more useful points of input, assuming it adds only marginal/occasional increases of latency.

March 28, 2013 | 05:20 PM - Posted by Bob Jones (not verified)

Can you address the visual quality differences between the two cards? Specifically on Sleeping Dogs, the 660 Ti seems to be missing some lighting sources outside - most noticeable is the cafe/shop lights before you go down the stairs, and then the store across the street right at the end of the vieo.

March 29, 2013 | 07:46 AM - Posted by Ryan Shrout

Time of day in the game when those videos were recorded?  They should be basically identical, I'll check.

March 29, 2013 | 12:10 AM - Posted by I don't have a name. (not verified)

Fascinating article. I think it'll take a few reads to fully comprehend everything that is being said. Thank you indeed, I found it fascinating. Certainly, as a 7970 owner, I'll be holding off on a potential second 7970 purchase for the time being.

March 29, 2013 | 02:09 AM - Posted by rezes (not verified)

The last GeForce 314.22 Beta Drivers and last Radeon drivers 13.3 beta 3. Please using these drivers on yours test.

AMD driver may be more stable on this drivers!

March 29, 2013 | 07:46 AM - Posted by Ryan Shrout

Nothing has changed on the issues in question.

March 29, 2013 | 02:15 AM - Posted by technogiant (not verified)

Thanks for the great article and all the work you've done guys.

I run 2x gtx 460's is sli and while I dislike screen tearing I've noticed that options such as vsync, active vsyn and frame rate limiters actually make the experience less smooth as appears to have been highlighted in this article.

I've considered getting a 120Hz monitor just so I can run without any of those options at a decent frame rate but use sufficiently high settings so as not to go above 120Hz and so incur screen tearing.

Thinking further I'd like Nvidia to develop a variation of their gpu boost technology that would actually down clock the gpu to prevent frame rates from exceeding the monitors refresh rate.....think this would give the benefits of no screen tearing without the negatives of vsync and the like.

Thanks again for the article guys.

March 29, 2013 | 02:37 AM - Posted by technogiant (not verified)

Actually using gpu boost dynamically to both under and overclock the gpu to achieve a target frame rate could be a very nice way of producing a smoothed experience without any of the negatives of other methods as its occurring directly at the gpu instead of in the game engine to display time line.

March 29, 2013 | 05:01 AM - Posted by Pick23 (not verified)

So is this article saying that even with the new testing methodologies:
1.Single card 7970ghz is still slightly better than the 680
2.Crossfire absolutely sucks
?

March 29, 2013 | 05:18 AM - Posted by John Doe (not verified)

7970 Ghz is slightly better than a 680 ONLY at stock.

When you're comparing 7970 Ghz to 680, things ENTIRELY depend on clock speeds since the 7970 Ghz is nothing more than a pre-OC'ed 7970.

But yes, CF does indeed sorta suck. Still.

March 29, 2013 | 04:26 PM - Posted by Anonymous (not verified)

Sweet 7970 still the best card under.. well under $1000 lol

March 30, 2013 | 06:21 PM - Posted by steen (not verified)

What's stock for a 680? ;) 7970 GE is slightly slower than Titan...

CF sucks just like SLI. What's your poison, input lag or frame metering? Do poeple understand what "runts" are? CF is actually rendering the frames, you just don't benefit as they're too close together. One frame renders the top 1/4 of the screen when the next frame starts. Your top ~200 lines are the fastest on your screen. ;)

(Sorry for the previous multi-posts. Don't know what happened there.)

April 7, 2013 | 10:15 AM - Posted by John Doe (not verified)

I have ran far more CF and SLi setups than you did FOR YEARS and understand far more abouts these things than you and your little silly mind does.

March 29, 2013 | 08:39 AM - Posted by fausto412 (not verified)

interesting piece, good job pcper.com

now I wonder if when AMD does a global fix my 6990 performance will be dramatically improved on bf3?

and what is the effect of correcting this to latency and actual framerate? will we see FPS go down at the expense of frametimes?

It is Obvious Nvidia was on top of this for some time...I just don't see a proper fix in 120 days.

March 29, 2013 | 09:24 AM - Posted by Chad Wilson (not verified)

Just out of scientific curiosity, did you do a run without the splitter in the mix to confirm that the splitter is not adding any latency to your tests?

March 29, 2013 | 09:33 AM - Posted by Anonymous (not verified)

Do people actually use vsync without some form of triple buffering?

I don't see how. Jumping between 30 and 60fps or whatever is not an enjoyable, nor smooth experience.

So, if you can enable vsync AND allow the game to sweep through a normal range of available framerates, does this negate the increased frame times of constantly switching back and forth between high fps and low fps?

March 29, 2013 | 09:56 PM - Posted by bystander (not verified)

V-sync, even with triple buffering, still jumps back and forth between 16ms and 33ms, but it does it between frames. A 120hz monitor helps here, as you can have 25ms frames too, so it is less of a variance.

March 29, 2013 | 09:36 AM - Posted by Anonymous (not verified)

Furthermore, is playing a game without vsync enabled REALLY an option?

Are you sure gamers all over the world disable it to be rid of the latency issues? I'm not so sure.

I'll happily take a little latency in a heated round of counter-strike than end up dead, or missing my shot because 50% of the screen shifted 8 feet to the right. (screen tearing).

Pretty much all games are unplayable without the use of vsync and I'm not convinced it's a personal preference, either, if you enjoy your experience while you're tearing frames - I'd just call you a mean name that insinuates you're not telling the truth.

March 29, 2013 | 11:00 AM - Posted by Marty (not verified)

If you are a competitive player, VSync is not an option, you are lagging an extra frame behind.

March 29, 2013 | 09:45 AM - Posted by rezes

Where is the new tests? and when?

March 29, 2013 | 11:20 AM - Posted by Anonymous (not verified)

So how much did nVidia pay you?

While I can see the potential in this kind of testing, and some of the issues you have mentioned are valid, you have drawn quite a bold and one sided conclusion using the competitor's software. I'll save my judgements for when this becomes open source.

March 29, 2013 | 05:12 PM - Posted by Fergdog (not verified)

It's not purely Nvidia made software, if you read the article or paid attention to this site you'd know Ryan co-developed this benchmark with Nvidia and he's been working on it for a long time.

April 2, 2013 | 08:10 PM - Posted by Anonymous (not verified)

nVidia has to do almost everything, the amd fans need to get used to it.

AMD's years long broken bottom line and years of massive layoffs and closings mean they claim they "weren't even aware of this issue !? !! "

- like every other total screw up AMD gets caught on with their dragon drawers on the floor next to their spider platform, buck naked in epic fail and presumably clueless...

Maybe we need some depositions and subpoenas of internal emails to see just how much they covered up their epic fail here.

March 29, 2013 | 05:00 PM - Posted by Fergdog (not verified)

Quick question, for adaptive vsync, you put it on half refresh rate didnt you?

March 29, 2013 | 05:02 PM - Posted by Fergdog (not verified)

Half refresh adaptive only really makes sense on 120hz monitors, not sure why you used that setting on a 60hz monitor for benchmarking.

March 29, 2013 | 08:06 PM - Posted by Anonymous (not verified)

Hoping you post the titan and and info today as promised!!!!!!!!!

March 29, 2013 | 08:07 PM - Posted by Anonymous (not verified)

Titan and amd

March 29, 2013 | 09:56 PM - Posted by MaxBV (not verified)

Waiting on that GTX 690 and Titan article still, hope you guys haven't forgotten.

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.