Visualizing Runts and Lower Frame Rates – Video Examples
I know that with the publication of this article finalizing our testing methods for Frame Rating, we are going to get lots of comments from gamers about how they have these configurations and that they don’t see the issues that we are discussing and penalizing AMD for. Whether or not they are telling the truth or there is a bit self-preservation / fan-boyism at work, we wanted to do our best and try to provide example of these phenomenon and we made a handful of videos to share with our readers.
Keep in mind that these videos are using content captures through our Frame Rating system from Radeon HD 7950s in CrossFire and GeForce GTX 660 Ti cards in SLI – but the effect on visuals is identical to what we see with the HD 7970s and GTX 680s. We used records from our 1920×1080 testing. Also, because we are doing manual run throughs of all of these games now, no two runs are exactly alike; there will be differences in the overall gameplay sequences but we have done our best to make sure they are as close as possible.
If you would prefer to download this video file, you can do so here.
This is the longest our example videos that shows 50% speed and 30% speed. Keep an eye on the smoothness of the animations on the screen including while walking straight ahead as well as while turn and rotating. Because we have Vsync disabled on these systems you will see some horizontal tear on the screen on BOTH configurations and that is fully expected. In order to show the examples as exactly as possible we had to leave that feature turned off but you should try to ignore those large “tears” and focus on the “micro” stutter on the grass on the ground, etc.
Did you see it? We debating making this video a blind A/B test but instead decided it was better to just be up front with the results.
If you would prefer to download this video file, you can do so here.
This video captured from Sleeping Dogs shows the same kind of animation smoothness advantage that NVIDIA SLI has over the Radeon HD 7950s in CrossFire. This animation difference is not due strictly to stutter but that the fact that AMD configuration is seeing every other frame a runt, thereby cutting the number of frames in each time period in half compared to the 660 Tis in SLI.
If you would prefer to download this video file, you can do so here.
Our next video is of Far Cry 3 and while I admit that neither experience is a smooth as we would like, the HD 7950s in CrossFire are showing a lot more stuttering than the 660 Ti cards.
If you would prefer to download this video file, you can do so here.
The visual animation issues are bit harder to see in our Battlefield 3 video due to some darker than normal scenes, but if you keep an eye on the sprint section down the alley way closely you will clearly see the slower animation fresh rate of the AMD CrossFire system.
If you would prefer to download this video file, you can do so here.
Our capture of Crysis 3 is another case where neither configuration is running as smooth as we should expect but it is very apparent that the AMD HD 7950s are running at a much lower observed frame rate than the 660 Ti SLI solution.
If you would prefer to download this video file, you can do so here.
DiRT 3 is a bit different – this compares single card performance and smoothness on SINGLE cards, showing you that without taking multi-GPU issues into account, both vendors can provide high quality real-world experiences. Both sides of the video look equally smooth and look to be providing a solid game play experience, which more or less backs up the results we found in our 1920×1080 data in today’s article.
Even though we are showing videos slowed down to 50% and even 30% of their full frame rate, these animation differences are very apparent to me in real time, while playing the games. Without a doubt there will be some variability in what kind of annoyance thresholds for each gamer but even if you can’t see the difference when looking at ONLY your video, you should be able to see the differences when looking at the two options side by side.
One of the advantages of our capture system is that we keep literally every video we record, though obviously in a compressed format. (Keeping 25GB, 60 second benchmark runs would hilarious.) This allows us to verify our data analysis results through the extractor and Perl scripts with the original recorded video to see if the animation stutters, variances, runts, etc are all there.
I realize we need some more videos and demonstrations of these animation issues and we are working on producing some more that run at various percent speeds including 100%. If you have any more ideas for how to use standard video edit tools to create comparison that would help our readers understand the experience we are having with our hardware, please email me or let us know in the comments below!
What would a game look like
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)?
(forward note: i have bad
(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.
Nice review. I’m interested
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?