Review Index:
Feedback

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

AMD CrossFire and Eyefinity Concerns

It will become painfully apparent as we dive through the benchmark results on the following pages, but I feel that addressing the issues that CrossFire and Eyefinity are creating up front will make the results easier to understand.  We showed you for the first time in Frame Rating Part 3, AMD CrossFire configurations have a tendency to produce a lot of runt frames, and in many cases nearly perfectly in an alternating pattern.  Not only does this mean that frame time variance will be high, but it also tells me that the value of performance gained by of adding a second GPU is completely useless in this case.  Obviously the story would become then, “In Battlefield 3, does it even make sense to use a CrossFire configuration?”  My answer based on the below graph would be no.

View Full Size

View Full Size

 

Another issue that cropped up with the AMD configurations in CrossFire mode showed its face when we tried 5760x1080 testing.  In nearly every case, an AMD CrossFire dual-GPU configuration running an Eyefinity configuration at 5760x1080 showed alternating dropped frames.  Whereas with single monitor gaming we were seeing some full game frames being turned into runts at the display, with Eyefinity those frames were missing completely (seen as missing overly colors from the expected pattern). 

This actually caused two things to happen.  First, our Perl scripts and data generation would sometimes error out completely (after the capture and extraction steps) because the code was never able to find the initial sequence of 16 colors in the correct order to validate the video capture.  We are still working on a fix for this in order to present that information in a useable format, but for now we will point out specifically when that occurred.  Secondly, the frame rates were smoother!  Without the runts getting in the way of the animation sequences the motion on the screen was actually more fluid than it would have been otherwise, but don’t take this as a vindication of what AMD’s Eyefinity is actually doing.  While the animation is smoother, you are still not seeing any benefit from the second card in your CrossFire configuration and you could view it simply as wasted investment.

Finally, we saw some interesting visual problems in our captures of Eyefinity that cause us to raise some eyebrows.  Because the overlay changes colors with every frame I was able to notice some instances where the digital scan out of the graphics cards were not matching up; frames were being split by other frames.

View Full Size

View Full Size

This second screen uses an alternate overlay for specific multi-GPU scenarios

At first I was afraid something was going on with our capture hardware, that somehow the EDID of the Datapath VisionDVI-DL card was incorrectly communicating with the AMD GPUs.  But in fact, we saw several problems and inconsistencies with AMD’s graphics performance when more than one display was attached to the system, even if we were NOT in an Eyefinity setup!  As I later learned, enabling Vsync actually does not work at all with Eyefinity and that, combined with the results I have seen (of which the screenshot above is an indicator) with our testing, lead me to believe that something is fundamentally wrong with AMD’s Eyefinity implementation.  And if it’s not “wrong”, it is definitely counter intuitive.  We’ll be asking AMD for more information in the coming weeks and hope to get more information from them as our Frame Rating process evolves. 

 

For our first set of data we are going to be looking at the Radeon HD 7970 GHz Edition and the GeForce GTX 680 in both single and dual-GPU combinations.  By looking at the high-end graphics cards first we can set a baseline for what to expect moving forward into the mid-range options and even mainstream.

April 9, 2013 | 11:18 AM - Posted by Anonymous (not verified)

Thanks Dan,

As you mentioned, there is certainly something to this stuttering. It is worthwhile to continue to optimize the test criteria so that it completely removes any Nvidia bias.

Ryan, can you comment about the possibility of raising the size of the scan lines for runts?

Thanks!

April 9, 2013 | 05:52 AM - Posted by Danieldp

Sorry, double post. XD

April 11, 2013 | 06:46 AM - Posted by Cookie (not verified)

Dan,

Just because Toms hardware is longer around does not mean that they do a better job. My vote goes to Pcper, I prefer to read something proper, no offence to Toms hardware.

April 11, 2013 | 11:53 PM - Posted by Danieldp

Hi,

Not really the point I was making, obviously both sites have sufficient expertise. The thing I was pointing at is the vastly different results...

Dan

April 11, 2013 | 08:22 PM - Posted by Dominic (not verified)

Can you do a test with an APU like the A10-5800K in crossfire with like an HD 6670 to see if this frame rate discrepancy occurs in this circumstance as well?

I would assume it would based on your results but nonetheless I'm curious on its results.

April 13, 2013 | 03:34 AM - Posted by JCCIII

Dear Mr. Stroud and friends,

Thank you for a tremendous amount of work, diligence, and integrity...

I thoroughly enjoyed the video with you and Tom Petersem. Although, I have to mention; I have been very disappointed in Nvidia since my purchase of a group of GTX 480s, believing, from day one, I had thrown away more than $1500 for three unmanageable 1500 Watt hairdryers marketed as graphics cards, which were subsequently relabeled GTX 580s, once the bugs were worked out, kind of like Microsoft's Vista to XP, kind of like scamming on people--no, definitely scamming.

I have always been an enthusiast of the Nvidia since the days of 3dfx and had likewise always enjoyed anticipating and buying Nvidia's new products, and the GTX Titan is awesome.

With memories of ATI, Matrox, and Nvidia (I still have my RIVA 128), a home has been found within my memories, and that is why I am excited about what you and the rest of PC Perspective have done and are going to do. With collaboration you-all are moving a beloved industry onward toward a better future for us and for the companies we want to succeed.

Sincerely,
Joseph C. Carbone III; 13 April 2013

April 18, 2013 | 04:57 AM - Posted by Anonymous (not verified)

All you have to do is use RadeonPro and it will fix all these issues...

April 18, 2013 | 11:29 PM - Posted by CoderX71 (not verified)

This

April 22, 2013 | 09:07 PM - Posted by Anonymous (not verified)

I dont see this huge problem i guess i am blind or only run 1 screen but i have played all titles listed and got better FPS in all of them using my GTX680's and My 7970's.

I still prefer my AMD cards for now for these reason:

1)In benchmarks my AMD cards kill my 680's in crossfire overclocked.

2)Graphics just look allot nicer on the AMD cards.

3)Biggest problem is Nvidia cards Cant Mine!!! (That's the big killer there for me)as id rather make $500.00 off my cards a week if i feel like it and be able to game as well.

I am not an Nvidia or AMD Fanboy as i have 680's in one of my builds and 7970's in a couple others plus have bought many of both cards in between.

I think maybe there is a problem for those running triple screens that hopefully AMD fixes as you have to admit they did a hell of allot on drivers recently that gave huge performance boosts.But 1920 x 1080 60 hertz i have no issues and def a big difference when i add a 2nd 7970 as i have swapped a card to other builds and put it back in do to loss of performance. The other thing i use my cards for is overclocking and Benchmarking and they def show huge performance there.I had my 680's over 1300mhz and they couldn't come close to 2 7970's at only 1225mhz.

So id have to say for working computers i will run my 7970's until there is no more money to be made and i will game on the 680's.

Pretty much not many games need more then 1 card anyway unless running multiple screens and high resolutions, Hell an APU can Max most console port games on the Market but the very few true PC games we actually have.

Yes i agree AMD fix the damn problem. But i also don't think Nvidia fan boys should be rooting for this because if they do the 7970 will be a nasty card all around that's capable of allot more then playing just games.

Just my opinion and experience on the to many cards i have owned to count in my life from both brands.

Take Care

May 5, 2013 | 12:13 AM - Posted by Evo (not verified)

If only we had a tool such as radeon pro to tweak the crossfire to make it operate properly. If only it existed...

The crysis 3 results are a bit questionable as in game vsync was causing havoc for nvidia (input lag) and amd (stuttering). Also crossfire was not engaging properly unless you alt-tabbed (this still occurs half the time). Not to mention the weird fix of opening an instance of google chrome to fix some of the problems with frame rate people were having with AMD setups.

My 7970 cf with radeon pro and other fixes works perfectly for me with Crysis 3, but there are some people still having issues. Also depends when the testing was done as when patch 1.3 was first released it caused massive problems for AMD cards that were later fixed.

May 20, 2013 | 02:17 AM - Posted by tigerclaws12894

Might have a typo or grammar error in the paragraph before the last in the Vsync topic. Anywho, have you ever considered triple buffering on AMD solutions as well as that config on 60Hz vs 120Hz as well? Input latency sure will be an issue but it'd be nice to know if it's better with 120Hz monitors.

May 20, 2013 | 07:11 PM - Posted by ServerStation668 (not verified)

This Dell T5600 has good video benchmarks: https://www.youtube.com/watch?v=3uK1J3o1fks

August 13, 2013 | 09:04 PM - Posted by Anonymous (not verified)

I have read this article four or five times and I find it intriguing. I must admit to not understanding most of it though. However, I must say being the owner of THREE 7970's, MSI Lightning BE, MSI Ghz OC Edition & Club3D RoyalAce at a cost of around £1300.00 GBP, just shy of $2000.00 USD I feel somewhat cheated. I hope AMD's forthcoming "Fixes" will redress these issues. Brilliant article and I am looking forward to all the follow ups.

September 1, 2013 | 02:35 AM - Posted by BryanFRitt (not verified)

What would a game look like if

it got a smooth 120+ fps,
was on a 60Hz display,
and in addition to the regular 'vsync' spot, it would 'vsync' at the '1/2' way spot?
aka update the display at the top and middle, updating at these same two spots every time, and only updating at these two spots.

Would the middle 'vsync' spot be annoying? helpful? noticed? informative? etc...? (This sounds like a good way to see how important fps is)

What's a good name for this?
1/2 x vsync, 2 x vsync, vsync 1/2 x, vsync 2 x, or something else?
What's the logic behind your pick(s)?

September 2, 2013 | 04:39 PM - Posted by lyo (not verified)

(forward note: i have bad english.)
1) is there a diff in observer fps between cards with more ram?
i.e. sli of 2x gtx770 2g vs sli of 2x gtx770 4g?

2) can you publish a min/max/var of partial frame per frame?
insted of runt i wanna know how many different "color" are per frame, and if they are evenly spread.

January 30, 2014 | 12:03 PM - Posted by BBMan (not verified)

Nice review. I'm interested as to how this tech is evolving.
But now I'm curious- I've read some of your test methods- but I may have missed something. I've seen mostly games that are more single player/first-person. Is that part of your methodology? I'm thinking of more intensive object rendering titles like Rome Total War II that has to render myriads of objects and stress memory more. Have you considered something like that?

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.