Review Index:

Frame Rating: GTX 660 vs HD 7870, plus HD 7790, HD 7850, GTX 650 Ti BOOST

Author: Ryan Shrout
Manufacturer: PC Perspective

Battlefield 3 - GTX 660 vs HD 7870

Battlefield 3 (DirectX 11)


Battlefield 3™ leaps ahead of its time with the power of Frostbite 2, DICE's new cutting-edge game engine. This state-of-the-art technology is the foundation on which Battlefield 3 is built, delivering enhanced visual quality, a grand sense of scale, massive destruction, dynamic audio and character animation utilizing ANT technology as seen in the latest EA SPORTS™ games.

Frostbite 2 now enables deferred shading, dynamic global illumination and new streaming architecture. Sounds like tech talk? Play the game and experience the difference!

Our Settings for Battlefield 3

Here is our testing run through the game, for your reference.

View Full Size

View Full Size

Right away with the Radeon HD 7870 we see a difference between the FRAPS measured frame rates and the observed frame rates of our Frame Rating capture based analysis system.  In many areas we found the performance of the CrossFire configuration to be no faster than the single GPU, when we see large amount of alternating full frames and runt frames. 

Worth noting of course though is that the single GPU results are showing the HD 7870 as the superior card compared 1 on 1 with the GTX 660.

View Full Size

Our frame time diagram shows where those drops in observed frame rates happen and WHY we measure them that way.  The GeForce GTX 660s in SLI though result in a pretty steady frame time sequence with very little variance.

View Full Size

This graphs shows the percentile distribution of frame rates, effectively your minimum frame rate level at any given percentile.  You can see that the "average" frame rate of the SLI configuration jumps from 59 FPS to 110 FPS or so, while the CrossFire only increases from 61 FPS to 80 FPS. 

View Full Size

Our data shows a high level of frame variance on the CrossFire result as you would expect based on our frame time plot above. 


View Full Size

View Full Size

The same pattern is seen at 2560x1440 as we witnessed above at 1920x1080 - the CrossFire result sees significant runt frame counts and because of that the "observed" frame rate drops noticeably.  Also, the GTX 660 is running consistently slower than the HD 7870 in a single GPU scenario.

View Full Size

The frame time graph looks even worse at 25x14 than it did above with more frequent episodes of alternating runt and long frames displayed to the gamer.  The SLI results do show a bit more variance as well though nothing close the ballpark of the CrossFire numbers.

View Full Size

Comparing the orange (CF) an the black lines of the HD 7870 there is very little difference between the observed performance based on FPS percentile data.  In fact, towards the end we see the CrossFire results dropping under that of a single card.  The GTX 660s running in SLI though are much better with the average FPS scaling from 36 FPS to 68 FPS, an increase of 88%.

View Full Size

Clearly the frame time variance and potential stutter seen in the Radeon HD 7870s in CrossFire are a problem, shown here in another light.  The SLI configuration does see more variance than the single GPUs but it doesn't really get bad until we hit the 95th percentile.


View Full Size

View Full Size

At 5760x1080 running in either AMD Eyefinity or NVIDIA Surround, the HD 7870s in CrossFire exhibit more problems, this time in the form of dropped frames.  Once we take those out of the calculations we see performance at or below the results from the single HD 7870.  NVIDIA's GTX 660s in SLI though scale consistently as you would expect.

For single card results though, the GTX 660 is definitely falling behind the HD 7870.

View Full Size

That blob or red shows you the dropped frames that are missing from the colored overlay that we expected to see.  One of the Radeon HD 7870 GPUs is basically doing work that is never seen by the user...

View Full Size

The CrossFire results look really bad in our frame time plot but even the GTX 660s in SLI are showing some higher than expected frame time variance.  Compare that to the slower, yet more consistent frame times on the single HD 7870 and GTX 660 and you can see why multi-GPU simply produces more complicated issues for developers and GPU vendors. 

View Full Size

Our frame rate graph definitely indicates that SLI is scaling with an average frame rate that increases from 22 FPS to 37 FPS but that SLI does tail off at the end.  In the worst cases here (99th percentile) we only see scaling jump from 16 FPS to 20 FPS.  CrossFire never is able to separate from the single HD 7870 results.

View Full Size

Because we remove the dropped frames all together from the CrossFire data, it appears that the frame variance levels of CrossFire HD 7870s is LOWER than the GTX 660s in SLI - and that is true in the purest sense.  Frames that never show up on the screen can't "vary" from the running average of previous 20 frames!  

April 8, 2013 | 12:00 AM - Posted by Anonymous (not verified)

I just saw a video from valve discussing a tool called Telemetry from RAD games, it seems like it could be useful for analysis of game performance.

April 8, 2013 | 10:41 AM - Posted by Pendulously (not verified)


You have already noted elsewhere that Percentile (1-99) is insufficient, for tracking large SPIKES in Frame Times, spikes which may occur only once every few seconds (i.e., once every few hundred frames, thus well beyond the 99th percentile threshold).

So you need a different dependent variable (x-axis), with 'milliseconds' remaining the independent variable (y-axis).

In a 10 minute gampeplay section (36000 Frames at 60 FPS), 36 Frames is equivalent to the 99.9th percentile (36000 divided by 1000).

If you put '36' on the x-axis (or perhaps multiples of 10), you could easily show people the effects of LARGE SPIKES IN FRAME TIMES. In many cases/circumstances, more illuminating than Frame Variance Graphs which show nothing beyond the 99th percentile.

April 8, 2013 | 11:52 AM - Posted by Stennan (not verified)

Are there any redeeming features to AMDs Crossfire solution?

April 8, 2013 | 02:06 PM - Posted by Anonymous (not verified)

Yes, they drop every other frame in crossfire and count it anyway and sell it to their fans as a superior frame rate than SLI.

AMS CF Frame numbers sold 80!(actual screen rate 40+stutter ! )

Wow it sure beats 76 from SLI !

Yes, a very "redeeming" quality.

April 10, 2013 | 09:12 AM - Posted by Stennan (not verified)

Thank you for not answering my question :)

I just wanted to know if there might be an upside with regards to perhaps input lag. Kind of hard to measure though.

April 11, 2013 | 04:17 PM - Posted by Anonymous (not verified)

Ok, here's a few thoughts. With the AMD runts and drops, and the fans now trying to claim "less input lag", my question is of course, "What are you aiming at?" !

With every other frame dropped or runted, you're getting ~50% of the game screen opponent positions from the frames displayed (first person shooter as an example of course since most games are that ) - so your opponents position on screen is "jumping" from one frame over to two frames ahead, changing position in a lurch - with that type of inaccuracy of on screen opponents, how is "less lag" going to help when one cannot see half the positions of the opposition, and aiming at those in between spots just is not directly occurring? you have no visual data for those positions, half the positions of your opponent !

You can see how that may present some real issues for the gamer, as far as firing accuracy.
Yes, one could develop a sort of compensation with practice, just like a similar amount of consistent input lag is gotten used to as the mind and muscles acclimate and compensate accordingly.

So, it appears to me missing frames and runts are anything but helpful, and the lag thenno lag, or the lag not running in sync with what appears on screen could be a real problem.

With equalized frame time outputs to screen, the "smoothness" factor certainly helps aiming at opponents and learning to time things correctly.

April 10, 2013 | 09:13 AM - Posted by Stennan (not verified)

Thank you for not answering my question :)

I just wanted to know if there might be an upside with regards to perhaps input lag. Kind of hard to measure though.

April 9, 2013 | 06:45 AM - Posted by fadeout (not verified)

Are we seeing that crossfire is inherently borked? Or are we seeing that the current drivers are not leveraging the potential of the hardware?

I'm thinking that drivers can be updated.

A note on the fanboy thingy, surely we need both companies to succeed to push one another to create ever faster and more powerful cards. No?

April 9, 2013 | 10:44 AM - Posted by bystander (not verified)

The problem has persisted at least two series of cards, but based on AMD's response, it likely has always existed. AMD had said they never considered frame metering. Instead, they had been focusing on delivering frames as fast as possible.

That to me would say the drivers were not borked per say, they just weren't trying to do anything about it. They weren't even aware of it.

Nvidia has their metering done with special hardware, but AMD believes they can do the same with software, which they probably can, though it may not be as ideal.

April 9, 2013 | 10:23 PM - Posted by Anonymous (not verified)

Not sure why the 660 is being compared to a 7870. MSRP is over $100 difference. 7850 makes more sense to me but there must be a good reason the 7870 was chosen.

April 9, 2013 | 10:57 PM - Posted by bystander (not verified)

The GTX 660 is the cheaper of the two cards. You must have the GTX 660ti in mind.

April 13, 2013 | 03:32 PM - Posted by Anonymous (not verified)

The reason is a big fat smooch on the behind of amd and their fanboys.
Keep the raging natives a bit less restless.
You see, only 2 pages of comments, they didn't come in screaming bias while their radeon I don't believe any of this and "if it's true" it doesn't matter heads exploded.

They have all been made absolute idiots, as the years of their stupidity is apparent.

The years of them screeching no one can see over 30 fps anyway is exposed. It doesn't matter what they knew or did not know, they made every excuse and lie and fanboy promo they could over the course of years and now it is absolutely clear they were seeing only half the frames gobbled with screen wretching runts, but still claimed it was beauty incarnate.

They are fools, and we all know it.

April 11, 2013 | 03:15 PM - Posted by Arek (not verified)

It would be nice to see how a pair of gtx 580's fares, as I have these in my setup. Pretty please! *_*

April 13, 2013 | 07:21 PM - Posted by Johnny Rook (not verified)

I can't stress enough how amazing this is.

I have one or two questions though:

1. Is the ATI/AMD Dual-GPU cards like the HD7990 have the same "problems" has dual cards in SLi?

I have an HD5970 and I do recognize there are problems as far as stutter is concerned but, I notice, I have a clear visual "feeling" that something is not right in a very few games. Mostly those "heavy" titles like "Metro2033" and the new "Crysis 3".

2. Is the problem something related to the hardware itself, is it drivers or is it a mix of both?

April 13, 2013 | 07:31 PM - Posted by Johnny Rook (not verified)

Sorry, I meant "problems has dual cards in Crossfire?"

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.