AMD Framepacing Driver 13.8 review

Page 6 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

bystander36

Diamond Member
Apr 1, 2013
5,154
132
106
Just as an example, we have websites claiming that spikes less than 40ms to 50ms don't matter and we have people claiming to see differences as low as 3ms. Both can't be correct.

...and yes, fps does matter. :thumbsup:
40-50ms seems ridiculously high. That's like 20-25 FPS. For most people, I could see 33ms as being tolerable, but not ideal. I personally will experience nausea with that kind of frame time.

I've heard from some sites that it takes about 8ms before you can't perceive a difference.

I believe some of the differences is a result of watching a video vs. playing a game. When playing, those hickups are more disruptive.
 

BrightCandle

Diamond Member
Mar 15, 2007
4,762
0
76
40-50ms seems ridiculously high. That's like 20-25 FPS. For most people, I could see 33ms as being tolerable, but not ideal. I personally will experience nausea with that kind of frame time.

I've heard from some sites that it takes about 8ms before you can't perceive a difference.

I believe some of the differences is a result of watching a video vs. playing a game. When playing, those hickups are more disruptive.

They are different measures. One is measuring a single spike, the other is a measure of continuous variance in the frame times. Don't conflate the two they are completely different problems and have completely different impacts on gameplay.
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
They are different measures. One is measuring a single spike, the other is a measure of continuous variance in the frame times. Don't conflate the two they are completely different problems and have completely different impacts on gameplay.

This is why I think we need better defined parameters for frequency and intensity. We can have a consumer look at graphs and not have any idea what they mean. Is it visible? Is it below perceptible thresholds? Most people have no idea. You will have people coming to completely wrong conclusions as well as marketing divisions making huge issues out of nothing. This is all quite new. Many (most) sites aren't even measuring frame times yet.

I sold audio gear for ~15 years. In that time I sold everything from boom boxes to rack systems to Levinson, Cello, Apogee, Quad, etc... We had people coming in looking at Apogee ribbon speakers and thinking that if those Cello performance amps in our showroom could drive them then surely their Carver Cube would too. After all, Julian Hirsch told them that all amplifiers of similar power will sound the same, and they're both rated at 200wpc.

This is what happens when people are only given some of the information and have no reference point to apply it to.
 

adnank77

Member
Jul 7, 2013
125
0
0
I got a performance boost by upgrading from 13.6 to 13.8 in Firestrike Extreme test .. With dual 7970 CF configuration, my score increased from 5691 to 5796 ..
 

7stars

Member
Apr 18, 2013
36
0
0
in my opinion there is a huge improvement (opinion, visually, no look at graphs or benchmarks) the idea of smoothness is definitely more clear to me...and not only, these drivers also prove that game developers have to find the right way against stutter issues...e.g. if Metro LL developers had not released the 1.4-1.5+ update which fix the stuttering these drivers would have been able to do very little difference... that is good drivers don't mean a good experience with no doubts. Try to play a previous Metro LL version (e.g. 1.3) with CF enabled and AMD 13.8. Then instead try with the latest MLL update and you will know what I mean for sure...same thing for Far Cry 3, that game is really bad developed
 
Last edited:

bystander36

Diamond Member
Apr 1, 2013
5,154
132
106
They are different measures. One is measuring a single spike, the other is a measure of continuous variance in the frame times. Don't conflate the two they are completely different problems and have completely different impacts on gameplay.

Let's just say, at least with v-sync on, I notice a stutter the moment my FPS drop from 60 to 58 or lower. That means one or two 33ms frame times is quite disruptive to me. If this continues, I get nauseated, which is why I shoot for higher FPS and don't use v-sync very often (some games have it forced on). I have a 120hz monitor now, so I have more leeway.
 

Dankk

Diamond Member
Jul 7, 2008
5,558
25
91
Are there any other 13.8 users getting crippling performance in their games, only to have it fixed by restarting their computer?

Strange anomaly I've been noticing since I installed this driver. The other day I booted up Sleeping Dogs and for some reason I was getting an absolutely dreadful framerate. Like ~15 FPS. Highly unusual, considering my 7870 typically performs excellently in Sleeping Dogs... usually a solid 60 FPS with everything on Very High (sans AA). But this had never happened before. After mulling around it, I simply rebooted my PC and fired up the game again. Ran perfectly smooth after that.

Now today, I booted up Call of Duty: Black Ops II. Right after dropping into a multiplayer match, I immediately noticed that my framerate was unusually low. About ~45 FPS. Obviously better than Sleeping Dogs, but still unacceptable for a game that essentially runs on the Quake 3 engine. So... I restarted my computer, and my framerate was magically back up to where it should be.

Anyone else having these issues? It's a beta driver, so I guess I won't complain too much. But it is a bit odd. I'm using Windows 8.
 

sushiwarrior

Senior member
Mar 17, 2010
738
0
71
Are there any other 13.8 users getting crippling performance in their games, only to have it fixed by restarting their computer?

Strange anomaly I've been noticing since I installed this driver. The other day I booted up Sleeping Dogs and for some reason I was getting an absolutely dreadful framerate. Like ~15 FPS. Highly unusual, considering my 7870 typically performs excellently in Sleeping Dogs... usually a solid 60 FPS with everything on Very High (sans AA). But this had never happened before. After mulling around it, I simply rebooted my PC and fired up the game again. Ran perfectly smooth after that.

Now today, I booted up Call of Duty: Black Ops II. Right after dropping into a multiplayer match, I immediately noticed that my framerate was unusually low. About ~45 FPS. Obviously better than Sleeping Dogs, but still unacceptable for a game that essentially runs on the Quake 3 engine. So... I restarted my computer, and my framerate was magically back up to where it should be.

Anyone else having these issues? It's a beta driver, so I guess I won't complain too much. But it is a bit odd. I'm using Windows 8.

Could be a background process doing something weird. Check Task Manager next time you notice it. Could also be a power state glitch, open up afterburner/other OC'ing program and check what the clocks are at while in game.
 

Dankk

Diamond Member
Jul 7, 2008
5,558
25
91
Could be a background process doing something weird. Check Task Manager next time you notice it. Could also be a power state glitch, open up afterburner/other OC'ing program and check what the clocks are at while in game.

MSI Afterburner reports my clocks correctly; 1100/1400. Also shows my GPU under 100% load. I'm not sure what the deal is.

Thanks for the tip about possible background process hog. I'll look at my task manager next time it happens.
 

ICDP

Senior member
Nov 15, 2012
707
0
0
MSI Afterburner reports my clocks correctly; 1100/1400. Also shows my GPU under 100% load. I'm not sure what the deal is.

Thanks for the tip about possible background process hog. I'll look at my task manager next time it happens.

I had the same problem a few months ago. A program called wintime (or something similar) was running on startup and causing my GPU to go to 100% usage. It was a bitcoin malware as far as I can tell.

Run task manager and go to the process tab, right click on the offending process and select the "open file location" option. THat should bring you to the folder the file resides in. You can end the process and rename the file and try a reboot so see if it fixes the problem.

http://www.bleepingcomputer.com/forums/t/492923/problem-with-bitcoin-mining-virus/
 

Rikard

Senior member
Apr 25, 2012
428
0
0
Many years ago they determined that 30 fps was an effective minimum frame rate for games. Nowadays 60 fps has most people happy but we have a lot of people seeing the difference from 60 to 120. Its far to say any such number would be just as flawed as the 30 fps number. That isn't to say I don't value such a thing, but one number is not going to be any more use than the 30fps one was. Simply put less variance is always going to be better, there isn't a cut off where it no longer matters (as far as I can tell), its going to be continuous improvement just as fps goes up.
I do not agree with this, there really is a cut off where improvements no longer matter. Ultimately it is about the human body and its brain, and have more than 50 years of research that shows that ~60 light pulses per second is the human limit for observing it as a solid light or flickering. The reason why 120 FPS feels smoother than 60 FPS is that we do have some variation in the individual frame times, so a longer delay than 16 ms will be observed under ideal conditions. This is very common occurance for a GPU at 60 FPS, but a 120 FPS (8 ms average) needs a lot more microstuttering to exceed 16 ms. This means that you can have a smoother experience even if the variance is larger if the average frame interval is much smaller...

Light conditions, viewing angles, color, strobing light intensity etc degrade your capacity to observe a 16 ms delay, so in most cases I personally only get annoyed if there is a significant fraction of frames at ~20 ms or larger frame intervals. (Not counting regular stuttering, which occur rarely, but at larger frame time intervals.) Add some allowance for individual differences of the test subjects, but there is no way that we will benefit from higher FPS or smaller variance in frame time intervals indefinitely.
 

Black Octagon

Golden Member
Dec 10, 2012
1,410
2
81
more than 50 years of research that shows that ~60 light pulses per second is the human limit for observing it as a solid light or flickering. The reason why 120 FPS feels smoother than 60 FPS is that we do have some variation in the individual frame times, so a longer delay than 16 ms will be observed under ideal conditions.

If there's more than 50 years of research to this effect, presumably you could provide a link or 12 to back this up? Moreover, these links could establish your implication that x frames/second from a computer monitor is the same as x light pulses/second?

I ask this not to be nasty, but rather because countless internet threads over the years have argued this very point, and there seems to be a pretty huge school of thought that:
a) the human eye does not see in terms of frames per second, and
b) to the (questionable) extent that one can talk about the limits of what the eye can see in terms of frames per second, the limits would in the HUNDREDS of frames per second

To cite just one possible source in this long, old and incontinent debate: http://amo.net/NT/02-21-01FPS.html
 
Last edited:

BrightCandle

Diamond Member
Mar 15, 2007
4,762
0
76
I suspect one of the reason I notice the difference between 60 and 120 so readily (along with other stuttering) is that I turn off motion blur, depth of field and post processed anti aliasing. Basically anything that blurs the screen I get rid of it because I like my games to be sharp. Combine that with monitors where my primary intent was high quality images with low motion blue and I have both the kit and the type of image to see artefacts more easily than the average. I suspect if everyone bought a lightboost monitor and turned off the blur they would see a lot more of these problems than they do today.
 

Plimogz

Senior member
Oct 3, 2009
678
0
71
Are there any other 13.8 users getting crippling performance in their games, only to have it fixed by restarting their computer?

Nothing quite like that here. But one day after installing 13.8s the computer did 'forget' how I like my desktop monitors to be configured. (resolution, orientation, postion and priority) Which never happened with the last couple of drivers. So there's that.

It hasn't done it a second time, mind you. So it's not too bad.
 

Vesku

Diamond Member
Aug 25, 2005
3,743
28
86
I suspect one of the reason I notice the difference between 60 and 120 so readily (along with other stuttering) is that I turn off motion blur, depth of field and post processed anti aliasing. Basically anything that blurs the screen I get rid of it because I like my games to be sharp. Combine that with monitors where my primary intent was high quality images with low motion blue and I have both the kit and the type of image to see artefacts more easily than the average. I suspect if everyone bought a lightboost monitor and turned off the blur they would see a lot more of these problems than they do today.

That would explain a lot, most of the perception issues at high perceivable refresh rates are interpreting motion. That's one of the qualities tested with video codecs how well objections in motion blur appropriately. It's also one of the reasons old school fluorescent lamps with ~100Hz refresh rates were actually annoying for people with sensitive eyes, mild motion sickness.

There is a lot of research into vision but there is no go to compilation to reference. So Wikipedia actually has some of the most citations: http://en.wikipedia.org/wiki/Flicker_fusion_threshold

In the case of a few (3 in one of the reviews) frames out of 1000s being delivered at >33ms, it's very unlikely to be perceivable as long as they are well separated from each other (which they have been in the reviews I read). I'd also think that for someone with eyes very sensitive to motion the primary gaming concern would be having high minimum fps.
 
Last edited:

bystander36

Diamond Member
Apr 1, 2013
5,154
132
106
As someone very sensitive to motion sickness, a high minimum FPS is very important. At least 60 FPS is required, but I still get a little nausea even then, if I game 30-60 mins. It takes 80 FPS before I no longer experience nausea.

Of course this only applies to mouse driven games in 1st person view, so I don't think it is entirely a FPS problem, and may be a latency problem.
 

Dankk

Diamond Member
Jul 7, 2008
5,558
25
91
I had the same problem a few months ago. A program called wintime (or something similar) was running on startup and causing my GPU to go to 100% usage. It was a bitcoin malware as far as I can tell.

Run task manager and go to the process tab, right click on the offending process and select the "open file location" option. THat should bring you to the folder the file resides in. You can end the process and rename the file and try a reboot so see if it fixes the problem.

http://www.bleepingcomputer.com/forums/t/492923/problem-with-bitcoin-mining-virus/

I definitely don't have any malware running on my PC.

I'm so confused. Black Ops II was running slowly again today, so I alt-tabbed out and looked at Windows Task Manager. The most CPU-intensive application I had running was... Black Ops II. And it was only using ~25% of my CPU. I then went into the Performance Manager, and looked at the CPU tab. No one core was getting pegged. CPU performance looked pretty evenly distributed among cores 1-6 (7 and 8 were parked for whatever reason). I didn't see anything suspicious.

So I rebooted my PC and the game ran properly again. Alt-tabbed out again, opened up task/performance manager again. Everything looked identical. Black Ops II was still consuming ~25% CPU and I didn't see anything getting pegged.

Although, it just occurred to me that I forgot to look at my actual CPU speed when the problem was happening. Maybe AMD Cool 'n' Quiet wasn't properly deactivating when I ran the game? But then my CPU graphs probably would've looked a lot different. Whatever; I'll double-check it next time.

I'm still inclined to chalk this up as a driver issue. But I'll keep my eye on it some more.

Edit: Happened again. CPU speed looks fine. I'm tempted to just revert back to 13.6 at this point.
 
Last edited:

Dankk

Diamond Member
Jul 7, 2008
5,558
25
91
What about GPU utilisation?

I mentioned this a few posts up. Basically, MSI Afterburner reports the correct clocks (1100/1400), and it also reports 100% GPU usage. Despite this, I'm sometimes getting awful performance in games, even in low-end game like Call of Duty. The only solution is to reboot my computer and then suddenly everything runs fine again.

I think I've ruled out both my CPU and GPU. Something else weird is going on.
 

Black Octagon

Golden Member
Dec 10, 2012
1,410
2
81
I mentioned this a few posts up. Basically, MSI Afterburner reports the correct clocks (1100/1400), and it also reports 100% GPU usage. Despite this, I'm sometimes getting awful performance in games, even in low-end game like Call of Duty. The only solution is to reboot my computer and then suddenly everything runs fine again..

Ok, I asked because in cases where GPU usage is at 100% and CPU usage is low, this usually indicates either:
(a) a GPU bottleneck, or
(b) some kind of software bug that is hogging GPU resources
 
Feb 19, 2009
10,457
10
76
Looks like the 7990 just got a price cut to $699. Still needs some driver work though based on the AT 13.8 review.

http://www.techpowerup.com/188552/radeon-hd-7990-malta-prices-slashed-to-699-targets-gtx-780.html

Its about time, they are joking if they think they can pull many sales giving away TWO SMALL DIES for the price of $1000. Major rip off, two 7970Ghz can be had for cheap and blow it away.

$699 is what it SHOULD have been on day 1. They dont have good out of the box CF experience, they are NOT going to be able to charge anywhere near NV's top dog.
 

Black Octagon

Golden Member
Dec 10, 2012
1,410
2
81
Its about time, they are joking if they think they can pull many sales giving away TWO SMALL DIES for the price of $1000. Major rip off, two 7970Ghz can be had for cheap and blow it away.

$699 is what it SHOULD have been on day 1. They dont have good out of the box CF experience, they are NOT going to be able to charge anywhere near NV's top dog.

Except they now DO have good out of the box CF experience (at least on a single non-4k monitor in DX10/11 games), to nudge you back towards the topic of this thread...
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |