Review Index:
Feedback

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

How Games Work

 

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:

 

Introduction

The process of testing games and graphics has been evolving even longer than I have been a part of the industry: 14+ years at this point. That transformation in benchmarking has been accelerating for the last 12 months. Typical benchmarks test some hardware against some software and look at the average frame rate which can be achieved. While access to frame time has been around for nearly the full life of FRAPS, it took an article from Scott Wasson at the Tech Report to really get the ball moving and investigate how each frame contributes to the actual user experience. I immediately began research into testing actual performance perceived by the user, including the "microstutter" reported by many in PC gaming, and pondered how we might be able to test for this criteria even more accurately.

The result of that research is being fully unveiled today in what we are calling Frame Rating – a completely new way of measuring and validating gaming performance.

The release of this story for me is like the final stop on a journey that has lasted nearly a complete calendar year.  I began to release bits and pieces of this methodology starting on January 3rd with a video and short article that described our capture hardware and the benefits that directly capturing the output from a graphics card would bring to GPU evaluation.  After returning from CES later in January, I posted another short video and article that showcased some of the captured video and stepping through a recorded file frame by frame to show readers how capture could help us detect and measure stutter and frame time variance. 

View Full Size

Finally, during the launch of the NVIDIA GeForce GTX Titan graphics card, I released the first results from our Frame Rating system and discussed how certain card combinations, in this case CrossFire against SLI, could drastically differ in perceived frame rates and performance while giving very similar average frame rates.  This article got a lot more attention than the previous entries and that was expected – this method doesn’t attempt to dismiss other testing options but it is going to be pretty disruptive.  I think the remainder of this article will prove that. 

Today we are finally giving you all the details on Frame Rating; how we do it, what we learned and how you should interpret the results that we are providing.  I warn you up front though that this is not an easy discussion and while I am doing my best to explain things completely, there are going to be more questions going forward and I want to see them all!  There is still much to do regarding graphics performance testing, even after Frame Rating becomes more common. We feel that the continued dialogue with readers, game developers and hardware designers is necessary to get it right.

Below is our full video that features the Frame Rating process, some example results and some discussion on what it all means going forward.  I encourage everyone to watch it but you will definitely need the written portion here to fully understand this transition in testing methods.  Subscribe to your YouTube channel if you haven't already!

Continue reading our analysis of the new Frame Rating performance testing methodology!!

How Games Work

Before we dive into why I feel that our new Frame Rating testing method is the best for the gamer, there is some necessary background information that you must understand.  While we all play games on a near daily basis, most of us don’t fully grasp the complexity and detail that goes into producing an image that makes its way from code a developer writes to the pixels on your monitor.  The below diagram attempts to simplify the entire process from the game engine to the display.

View Full Size

In the image above we have defined a few important variables based on time that will help us explain graphics anomalies.  The first is t_game and it refers the internal time that the game engine is using to keep track of its internal simulations and interactions.  This is where processes like the physics simulations, user interface, artificial intelligence and more are handled.  Different game engines keep time in different ways, but they usually fall into two categories: fixed or variable time steps.  In a fixed time method the game engine cycles the environment in its internal simulation on a regular, fixed iteration.  This is more predictable but it also forces other systems to sync up to it (drivers, GPU rendering) which may cause some issues.  The variable time step method allows a lot more flexibility, but it can be more complicated for developers to maintain a fluid feeling simulation because of simply not knowing when the next simulation update will take place. 

Following that is t_present, the point at which the game engine and graphics card communicate to say that they are ready to pass information for the next frame to be rendered and displayed.  What is important about this time location is that this is where FRAPS gets its time stamps and data and also where the overlay that we use for our Frame Rating method is inserted.  What you should notice right away though is that there is quite a bit more work that occurs AFTER the t_present command is sent and before the user actually sees any result.  This in particular is where capture method’s advantages stem from.  

After DirectX calls the game engine to recieve its time, the graphics driver gets its hands on it for the first time.  The driver maps the DX calls to its own specific hardware and then starts the rendering process on a GPU.  Once the frame is rendered, we will define another variable, t_render, which is reported when the image is ready to be sent to a display.  Finally, t_display will be defined as the time in which data from the frame is on the display, whether that be a complete frame (Vsync enabled for example) or a partial frame. 

You can likely already see where the differences between FRAPS data measurement and Frame Rating measurement start to develop.  Using capture hardware and analysis tools that we’ll detail later, we are recording the output from the graphics card directly as if it were a monitor.  We are essentially measuring frames at the t_display level rather than the t_present point, giving us data that has been filtered through the entire game engine, DirectX, driver, rendering and display process. 

It is also useful to discuss stutter and how it relates to these time definitions.  Despite some readers opinions, stutter in game play is related to the smoothness of animation and not hard wired to low frame rate.  If you have a steady frame rate of 25 FPS you can still have an enjoyable experience (as evident by the 24 FPS movies we all watch at the theater).  Instead, we should view stutter as a variance level between t_game and t_display; if the total display time runs at 50ms (20 FPS) you won’t have stuttering in animation (in most cases) but if total display time shifts between 20ms and 50ms you definitely will.

In our second video on Frame Rating, we looked at a capture from Dishonored in a frame by frame analysis and saw a rather large stutter.  I think a better term for that is a hitch, a large single frame rate issue that isn’t indicative of a microstutter smoothness problem.  Using the above variables, a hitch would be a single large instance of t_game – t_display.  As you’ll see in our results analysis pages (including games like Skyrim) this happens fairly often, even in some games that are smooth otherwise. 

Our 2nd video on Frame Rating from a couple months ago...

There is also a completely different discussion on the advantages and differences of capturing data from FRAPS versus capturing data with our Frame Rating methodology.  I believe that the advantages of hardware capture outweigh the concerns currently, but in reality the data that FRAPS generates isn’t that important, it just happens to be the closest data point to another metric we would love to know more about: game time. 

Game time is the internal game clock that the software engine uses to keep track of the physical world.  This clock could be based on a fixed time span for each tick or it could be variable or timed off of another source (like the OS or GPU driver).  An Intel GPU engineer, Andrew Lauritzen, recently made a great post over on the Beyond3D.com forums about game time, back pressure on the game pipeline and much more.  Here is a short portion of that post, but I would encourage everyone to read the entirety completely:

1) Smooth motion is achieved by having a consistent throughput of frames all the way from the game to the display.

2) Games measure the throughput of the pipeline via timing the back-pressure on the submission queue. The number they use to update their simulations is effectively what FRAPS measures as well.

3) A spike anywhere in the pipeline will cause the game to adjust the simulation time, which is pretty much guaranteed to produce jittery output. This is true even if frame delivery to the display (i.e. rendering pipeline output) remains buffered and consistent. i.e. it is never okay to see spikey output in frame latency graphs.

4) The converse is actually not true: seeing smooth FRAPS numbers does not guarantee you will see smooth display, as the pipeline could be producing output to the display at jittery intervals even if the input is consistent. This is far less likely though since GPUs typically do relatively simple, predictable work.

Clearly the best case for evaluating overall gaming performance will be to have access to the internal game and measure it in comparison the output from Frame Rating, the actual frames on your display.  Differences there could be analyzed to find exact bottlenecks in the pipeline from game code to display.  The problem is no game engine developers allow access to the information currently and the number of different engines in use today makes it difficult for even the likes of NVIDIA and AMD to gather data reliably.  There is opportunity for change here if an API were to exist (in DirectX for example) that would give all game engines reliable time iterations that we would then have access to.

 

NVIDIA's Involvement

You may notice that there is a lot of “my” and “our” in this story while also seeing similar results from other websites being released today.  While we have done more than a year’s worth of the testing and development on our own tools to help expedite a lot of this time consuming testing, some of the code base and applications were developed with NVIDIA and thus were distributed to other editors recently.

NVIDIA was responsible for developing the color overlay that sits between the game and DirectX (in the same location of the pipeline as FRAPS essentially) as well as the software extractor that reads the captured video file to generate raw information about the lengths of those bars in an XLS file.  Obviously, NVIDIA has a lot to gain from this particular testing methodology: its SLI technology looks much better than AMD’s CrossFire when viewed in this light, highlighting the advantages that SLI’s hardware frame metering bring to the table. 

The next question from our readers should then be: are there questions about the programs used for this purpose?  After having access to the source code and applications for more than 12 months I can only say that I have parsed through it all innumerable times and I have found nothing that NVIDIA has done that is disingenuous.  Even better, we are going to be sharing all of our code from the Perl-based parsing scripts (that generate the data in the graphs you’ll see later from the source XLS file) as well as a couple of examples of the output XLS files. 

Not only do we NEED to have these tools vetted by other editors, but we also depend on the community to keep us on our toes as well.  When we originally talked with NVIDIA about this project the mindset from the beginning was merely to get the ball rolling and let the open source community and enthusiast gamers look at every aspect of the performance measurement.  That is still the goal – with only one minor exception: NVIDIA doesn’t want the source code of the overlay to leak out simply because of some potential patent/liability concerns.  Instead, we are hoping to have ANOTHER application built to act as the overlay; it may be something that Beepa and the FRAPS team can help us with.

March 27, 2013 | 04:34 PM - Posted by Ryan Shrout

I'll see if we can take a look at that.

March 29, 2013 | 12:31 PM - Posted by jgstew

The whole time I was reading this article, I was more and more curious how Virtu's technology would effect things. I'm curious about more than just their Virtual V-sync, but their other options as well for both single and multiple GPUs. Virtu has not had the scaling that SLI & Crossfire have had, but perhaps their technology would show well in other areas with this analysis.

I do feel that Frame Rating & the input to display latency are much more interesting metrics.

Great work on the article.

March 27, 2013 | 02:32 PM - Posted by serpinati of the wussu (not verified)

I've read somewhere that this will not be the norm for pcper to do this type of testing with all video card reviews (too labor intensive). Is this true?

If it is true, will pcper at least record the frame data and simply give it a quick look to make sure video cards aren't doing something absolutely crazy (for example, if you were reviewing those 7970's crossfire you might not plan to actually to analyze the frame times, but you would at least look over the recorded frames anyways and catch the crazy runt frames and mention it in your reviews.)

March 27, 2013 | 04:34 PM - Posted by Ryan Shrout

No, my plan is to take this route going forward.

March 27, 2013 | 02:57 PM - Posted by Anonymous (not verified)

Greatly appreciate the work behind this! And opening up the tools / scripts to everyone, pushing it to other hardware magazines. Huge kudos! Oh, and you should slap Anandtech over the head for not even mentioning your involvment into the new metering method in their introductory article, that only mentions nVidia...

Is there any way to determine latency between t_present and t_display? If not, it should be - maybe some kind of timestamp could be worked into the overlay? Because that would be interesting not only in the context of VSync, but also regarding nVidias frame metering, which must take some time analyzing the frametimes. Supposedly, AMD has some smoothing algorithm coming up for Crossfire as well, so there it would also be valuable information.

Regarding Adaptive / Smooth VSync: They clearly come off too good in this article. They're not the focus of the article of course, still a little more thought should have been put behind this, considering the amount of time taken to create this awe inspiring effort of an article and the tools behind it. Adaptive VSync turns VSync off when tearing is most annoyingly visible, i.e. at framerates below monitor refresh rate - true triple buffering (instead of the queue nowadays called triple buffering) would be the good solution here. Smooth Vsync does not seem to do anything particularly positive judging from what little measurement is available in the article - only time (further tests that is) will tell what it actually does...

March 27, 2013 | 04:36 PM - Posted by Ryan Shrout

We have more research into the different NVIDIA Vsync options coming up, stay tuned.

As for the timestamp different to check for the gaps between t_present and t_display...we are on that same page as well.  :)

March 27, 2013 | 02:57 PM - Posted by Mawrtin (not verified)

Have you tried CF using radeonpro? Apparently it offers some kinda of Dynamic V-sync similar to nVidias adaptive if I'm not mistaken.

March 27, 2013 | 05:03 PM - Posted by Marty (not verified)

No, it does not. It's a frame limiter, which eliminates some of the stutter, but causes more lag (increases latency).

March 27, 2013 | 03:16 PM - Posted by Mangix

Hmmm. I wonder if there is a difference between Double and Triple Buffered VSync. Newer Valve games support both. Would appreciate testing there.

Also, are there driver settings that help/harm frame rating? Nvidia's Maximum Pre-rendered frames setting sounds like something that can have an effect.

March 27, 2013 | 04:25 PM - Posted by xbeaTX (not verified)

AMD has complained on Anandtech for this type of test using fraps ... now anyone know the real situation and it's shocking
this article sets the new stantard of excellence ... congratulations and thanks for the enormous work done... keep it up! :)

March 27, 2013 | 04:37 PM - Posted by Ryan Shrout

Thank you!  Help spread the word!

March 27, 2013 | 10:55 PM - Posted by TinkerToyTech

Posted to my facebook page

March 27, 2013 | 04:40 PM - Posted by Marty (not verified)

Ryan, I've a suspicion that you wanted to select Adaptive VSync on NVidia, but have selected Adaptive (half refresh rate) instead in the control panel. Would you please check it out.

March 28, 2013 | 02:20 AM - Posted by showb1z (not verified)

²

Other than that great article. Would also be interested in results with frame limiters.

March 27, 2013 | 05:23 PM - Posted by Dan (not verified)

This site is so awesome that it is one of th eonly ones that I disable Adblock in Chrome for. You deserve the ad $'s.

Rock on, Ryan and crew!

March 29, 2013 | 07:37 AM - Posted by Ryan Shrout

Thank you we appreciate it!

March 27, 2013 | 05:45 PM - Posted by Mike G (not verified)

Thanks for the all of the time taken to accurately test and then explain your frame rating methods to us. I wonder if AMD would be willing to have a representative come on and speak on how they will be addressing this issue. I for one will be holding off purchasing an additional 7970 at this time.

March 27, 2013 | 06:43 PM - Posted by bystander (not verified)

They have, with AnandTech, though the person they spoke to was the single GPU driver guy, they did mention they have a plan to offer fixes in July.
http://www.anandtech.com/tag/gpus

March 27, 2013 | 06:51 PM - Posted by Soulwager (not verified)

Is your capture hardware capable of capturing 1080p @ 120hz? The data rate should be less than 1440p@60hz.

Also, I would like to see some starcraft 2 results. It's frequently CPU limited, and I'm wondering how that impacts the final experience when compared to a gpu limited situation. I'd recommend the "unit preloader" map as a benchmark run, once to pre-load all the assets and again for the capture.

March 29, 2013 | 07:38 AM - Posted by Ryan Shrout

Actually, I think 1080p@120 is a higher pixel clock than 25x14@60; we tried to do basic 120 Hz testing right before publication without luck but we'll be trying again soon.

As for SC2, we'll take a look.

March 29, 2013 | 12:00 PM - Posted by Soulwager (not verified)

You're right, the pixel clock is higher. I guess I was only thinking about total number of pixels that need to be recorded, but the smaller resolution is more heavily impacted by overscan.

March 27, 2013 | 09:28 PM - Posted by SPBHM

great stuff.

I would be interested in seeing some results with some framerate limit, not from vsync, but another limit, something like FPS max 45 for Crysis 3 (higher than a single card, around the average for the CF), you can easily do that with dxtory

March 27, 2013 | 09:49 PM - Posted by Trey Long (not verified)

Its nice to see the truth come out. Save the excuses and fix it AMD.

March 28, 2013 | 03:41 AM - Posted by Carol Smith (not verified)

Why did you completely ignore AMD's RadeonPro freeware ?
Which offers and has offered Dynamic V-Sync that's superior to Nvidia's Adaptive V-Sync for years now.
Tom's Hardware seems to be the only site that actually used this utility to make a fair comparison between SLI and Crossfire.

RadeonPro's Dynamic V-Sync was shown to be clearly superior to Nvidia's Adapative V-Sync Implementation .

Here is the Tom's link
http://www.tomshardware.com/reviews/radeon-hd-7990-devil13-7970-x2,3329-...

I hope that you learn from your fellow journalist's to include this fantastic solution in your upcoming article.
Thanks for your hard work and good luck.

March 28, 2013 | 06:27 AM - Posted by Marty (not verified)

A frame limiter improves stuttering, but increases lag. You loose some Fps too. So you have to decide between the devil and lucifer in the case of Crossfire.

March 29, 2013 | 07:01 AM - Posted by Carol Smith (not verified)

Exact same thing with Nvidia SLI and Adaptive V-Sync, if you don't want stuttering you have to sacrifice framrates.

March 29, 2013 | 07:39 AM - Posted by Ryan Shrout

I'm going to take a look, but this is NOT "AMD's" software.

March 28, 2013 | 03:50 AM - Posted by Andre3000 (not verified)

Thanks for the eye opener! Which AMD drivers have been used for this review? I have read the article.. but i might have missed it?

March 29, 2013 | 07:39 AM - Posted by Ryan Shrout

For AMD we used 13.2 beta 7 and for NVIDIA we used 314.07.

March 29, 2013 | 08:27 AM - Posted by Steve (not verified)

Ryan: Is it possible to test AMD's CF with older drivers to see if this problem has been around for a long time or if it is a more recent problem with AMD's continuous driver upgrades to improve speeds?

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.