• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

[Benchmark3D] Microstutter Case Study: Skyrim

:D

Can you just sum up what those links are saying? ;)

It seem that playing with the CPU does indeed help and even on APU

But on NV it makes it worse which is shown in the first link, i think its down to NV having better multi threaded drivers and is why in some CPU + GPU reviews NV gained by having a stronger CPU while AMD didn't so much.

Also when [H] did a TriFire AND TRISLI review the TriFire won, but when it was re done on a beefier platform the TRISLI won.


http://hardocp.com/article/2011/04/28/nvidia_geforce_3way_sli_radeon_trifire_review/1
http://hardocp.com/article/2011/04/11/amd_radeon_69906970_crossfirex_trifire_review/

Two graphics cards, ten processors, ten benchmarks

If you like playing games in Full HD resolution with all fancy graphical effects enabled, you obviously need to invest in a high-end graphics card. If you're looking at buying one of the current high-end cards, such as the AMD Radeon HD 7970 GHz Edition or the Nvidia GeForce GTX 680, it's worth taking a closer look at the rest of your computer. Will your PC be able to squeeze the most out of these expensive powerhouse video cards, or do you have a bottleneck hiding somewhere? With a speedy Core i7 processor you obviously need not worry, but what if you have something like an Intel Core i3 or AMD A6? Hardware.Info examined just that scenario, so read on to find out.
http://uk.hardware.info/reviews/3714/radeon-hd-7970-and-geforce-gtx-680-tested-with-10-cpus
 
Last edited:
That would be testable probably by changing the render threading option in the nVidia drivers tho some of the backend is probably still threaded even with it disabled.
 
I've always thought that Skyrim had some kind of stutter. It always felt weird. Some have said it's an AMD problem, but I used to get some weirdness in BF3 on my 570 which went away with my 7970 so it's swings and roundabouts.
 
This thread isn't about purchase justification of a 660Ti it's a thread for Skyrim frame latency issues on AMD cards for people affected. Try to keep it that way. Do you even have Skyrim? :)

Yes, I have around 130 hrs on it before I got bored.
I played it at 1080p with GTX 470 SLi and it was a good experience.

I was the one that brought the 12.11 frame latency issues to the attention of these forums.

Wasn't it you who said that your GTXs were 'buttery smooth'?
 
It doesn't matter how many people tell me the HD7950 is a better card than the GTX660Ti.

The nVidia card 'feels' like a better card.

Weehamish, puts his points across badly, but he has been honest with his experiences.

I will be sticking with nVidia, so no surprises there!



Yes, I have around 130 hrs on it before I got bored.
I played it at 1080p with GTX 470 SLi and it was a good experience.

I was the one that brought the 12.11 frame latency issues to the attention of these forums.

Wasn't it you who said that your GTXs were 'buttery smooth'?

Edited: i found it.
http://forums.overclockers.co.uk/showthread.php?t=18466444&highlight=startername_SkodaMart on Windows 8

And your issue was before those drivers.
http://forums.overclockers.co.uk/showthread.php?t=18446828&highlight=startername_SkodaMart

No NV has not always been smoother in Skyrim.

http://techreport.com/review/23419/nvidia-geforce-gtx-660-ti-graphics-card-reviewed/7
 
Last edited:
Yes, I have around 130 hrs on it before I got bored.
I played it at 1080p with GTX 470 SLi and it was a good experience.

I was the one that brought the 12.11 frame latency issues to the attention of these forums.

Wasn't it you who said that your GTXs were 'buttery smooth'?

I did as is my 7950 although they were marginally choppier in multi GPU. What's your point?

No No No.... Rusty please... don't get him started, again! :o


I agree (now)! :D
 
It seem that playing with the CPU does indeed help and even on APU

But on NV it makes it worse which is shown in the first link, i think its down to NV having better multi threaded drivers and is why in some CPU + GPU reviews NV gained by having a stronger CPU while AMD didn't so much.

Also when [H] did a TriFire AND TRISLI review the TriFire won, but when it was re done on a beefier platform the TRISLI won.


http://hardocp.com/article/2011/04/28/nvidia_geforce_3way_sli_radeon_trifire_review/1
http://hardocp.com/article/2011/04/11/amd_radeon_69906970_crossfirex_trifire_review/

http://uk.hardware.info/reviews/3714/radeon-hd-7970-and-geforce-gtx-680-tested-with-10-cpus


What they actually say is quite interesting...

Conclusion

Several things have become very clear based on the benchmarks. To get the most out of a high-end single-GPU graphics card, you need an Intel Core i5 3570K or higher. So only if you really feel that you have to get the maximum frames per second, do you need to spend the money on a Core i7.
It's also interesting to note that AMD's low-end processors score better than Intel's cheap processors in combination with high-end graphics cards. That means you're better off with an AMD A6-5400K (about £56) or AMD A10-5800K (about £99), than an Intel Pentium G860 (about £52) or an Intel Core i3 (about £91). Particularly the Pentium and the Core i3 have a negative impact on the performance of the graphics cards.
The last conclusion we can draw is that the AMD Radeon HD 7970 GHz edition suffers less from CPU scaling than the Nvidia GeForce GTX 680. That means that the AMD card performs better with slower processors compared to the Nvidia card. We don't have a clear explanation for this, but it's possible it has something to do with the driver overhead.
So now you know. Should you upgrade your CPU before getting that high-end graphics card? Or do you instead go with an AMD card that runs better with slower processors?
If you look at the benchmarks again it clearly shows the AMD GPU performance goes up the more threads the CPU has, while the Nvidia GPU stay around the same or even drop.

So 'that' could explain it, AMD's GPU drivers might be better threaded than Nvidia, but like turning HT on on your Intel CPU this can cause micro stutter.
 
Last edited:
^^ Could be the other way around nVidia drivers could be better threaded but it results in some compromise in single thread situations in order to have better multi-thread support.
 
Has anyone gone back to old drivers to see if it fixes the problem?

Skyrim frametimes(max everything ingame), Labyrinthian Thoroughfare Dungeon.
bb3d953fd23040910aebec8f48eba4a6.jpg


Rig in the sig all@stock for the test(7970 MSI OC, CrossFireX disabled) 12.8 drivers.

My Skyrim isn't vanilla, it has quite a few high res mods etc.
 
^^ Inside area where its easy to hold 60fps and then using vsync isn't going to show much in the way of microstutter unless something is very very broken anyway.
 
If you look at the benchmarks again it clearly shows the AMD GPU performance goes up the more threads the CPU has, while the Nvidia GPU stay around the same or even drop.

So 'that' could explain it, AMD's GPU drivers might be better threaded than Nvidia, but like turning HT on on your Intel CPU this can cause micro stutter.

^^ Could be the other way around nVidia drivers could be better threaded but it results in some compromise in single thread situations in order to have better multi-thread support.

This makes me want to re-run my frame time benchmarks with HT on and off. The question is can i be bothered?
 
Skyrim frametimes(max everything ingame), Labyrinthian Thoroughfare Dungeon.
bb3d953fd23040910aebec8f48eba4a6.jpg


Rig in the sig all@stock for the test(7970 MSI OC, CrossFireX disabled) 12.8 drivers.

My Skyrim isn't vanilla, it has quite a few high res mods etc.

The time before this when Skyrim was amongst the games that was test by the same people it did not have any latency issues either :)

Its just this Driver, it can be fixed.
 
Good info Tommy. Goes to show if you are mainly playing Skyrim, 12.8 are the drivers to use.

:)

Although, it was meant to be 12.10's I installed, I never noticed that I was using 12.8's until I checked before posting.:o

^^ Inside area where its easy to hold 60fps and then using vsync isn't going to show much in the way of microstutter unless something is very very broken anyway.

12.10's with vscync off in the fresh air:

75f4b655d8006ce545a8a869f57edd01.jpg
 
Okay, already saw a difference however - I also saw stutter on the GTX 680, just not as bad haha. I'll post the CSV info on an edit, but for now here is the actual graph (I zoomed in.)

I tried to normalize the two runs as best I could, both rigs are set to ingame ultra, with 8xAA, disabled any outside optimizations (CCC to default, NV panel to default), and both at 1080p full screen
EDIT for more info:

I used the same account, so its the same character, same gear, same mount, yada. I disabled all mods, so its all WoW stuff.
Resolution was set to 1080p, full screen, 8xAA, all ingame options Ultra.
CCC/NV panel reset for the run.
I found a two landmarks easy to identify and faced the wall, then strafed between the two. CSV file for both runs was similar in size.

I noticed stutter on both rigs, but far more on mine (the HD 7970.)

GTX 680 Results:
8mqXL.png

HD 7970 Results:
MZqHL.png


I will update my CSV file with the two runs, clearly there is a difference.
EDIT #2: Posted the info in my CSV files:
https://docs.google.com/spreadsheet/ccc?key=0AhY9sOddohlTdGdsbHUyb1dfUm9RWnBlTFJKWnljQkE#gid=0


http://forums.anandtech.com/showpost.php?p=34490495&postcount=399
 
Back
Top Bottom