Sign in with
Sign up | Sign in
Your question
Solved

FX-6200 bottlenecking Gigabyte 7950?

Last response: in CPUs
Share
March 27, 2013 2:55:17 PM

So I have built myself a gaming PC with mostly AMD components. When I play battlefield 3 however, my gpu usage is often low when my framerate is. (For example 80% core usage when I'm getting 35 fps) My main question is, would getting a new CPU help? The one I had in mind is the fx-8350, as in some ways it beat intel processors at the same price. Thanks for all of your help community! The lowest my fps has ever gotten in battlefield is 30, also as always it was paired with low gpu usage.
a b 4 Gaming
a c 210 à CPUs
March 27, 2013 3:01:11 PM

Yes, go to an 8 core 8320 if you like to OC, if you don't get the 8350...a 7950 can get bottlenecked by a mid-range CPU...(or if you just want an 8350 get it anyway, it'll OC to 5 GHz+ if you really know what you're doing and shill out big money for water cooling etc.)
March 27, 2013 3:18:54 PM

billybob42 said:
So I have built myself a gaming PC with mostly AMD components. When I play battlefield 3 however, my gpu usage is often low when my framerate is. (For example 80% core usage when I'm getting 35 fps) My main question is, would getting a new CPU help? The one I had in mind is the fx-8350, as in some ways it beat intel processors at the same price. Thanks for all of your help community! The lowest my fps has ever gotten in battlefield is 30, also as always it was paired with low gpu usage.


8350rocks said:
Yes, go to an 8 core 8320 if you like to OC, if you don't get the 8350...a 7950 can get bottlenecked by a mid-range CPU...(or if you just want an 8350 get it anyway, it'll OC to 5 GHz+ if you really know what you're doing and shill out big money for water cooling etc.)
Time to slowly earn money over the years :/ 

Related resources
a b à CPUs
March 27, 2013 4:03:39 PM

Is this issue specific to Battlefield 3? What about other games/benchmarks (e.g. Unigene Heaven or Valley)? And just to be clear, your CPU usage during Battlefield 3 isn't even at 100% - i.e. you are not maxing it out?
a b 4 Gaming
a c 108 à CPUs
March 27, 2013 5:19:39 PM

Points:
1) If the GPU is at 80% and he's not frame rate capped (limited by VSYNC) then it's likely a CPU bottlenecking issue.

2) Games will vary significantly in how much the CPU can affect bottlenecking. In some scenarios a new CPU will provide little benefit, in other scenarios it could be a 60% boost.

3) A newer CPU may not affect average FPS much but can affect the MINIMUM frame rate significantly (again, it varies a lot by game/setup).

4) CPU USAGE:
CPU's often show less than 100% usage even when they are the main bottleneck. The reasons are several, including the fact that a game often can't use all cores.

A better way to estimate CPU bottlenecking is to benchmark with your CPU set to two different frequencies (say 500MHz) apart. If the higher frequency shows improvements it's likely a better CPU would as well though how much is difficult to say.

*You can GOOGLE to see how much various games are affected by CPU scaling. As mentioned it varies a lot. BF3 can be demanding especially during Multiplayer on large maps.
March 27, 2013 5:34:53 PM

nitrium said:
Is this issue specific to Battlefield 3? What about other games/benchmarks (e.g. Unigene Heaven or Valley)? And just to be clear, your CPU usage during Battlefield 3 isn't even at 100% - i.e. you are not maxing it out?
In this screenshot everything was on low settings, but the fps stayed the same. Also this is very similar to the gpu usage with everything on ultra, just add about 10-20% usage. Also, this seems to be only with multiplayer games (I think the cpu has a hard time with a lot of players, dayz being a good example) My valley benchmarks look like this



March 27, 2013 5:42:22 PM

photonboy said:
Points:
1) If the GPU is at 80% and he's not frame rate capped (limited by VSYNC) then it's likely a CPU bottlenecking issue.

2) Games will vary significantly in how much the CPU can affect bottlenecking. In some scenarios a new CPU will provide little benefit, in other scenarios it could be a 60% boost.

3) A newer CPU may not affect average FPS much but can affect the MINIMUM frame rate significantly (again, it varies a lot by game/setup).

4) CPU USAGE:
CPU's often show less than 100% usage even when they are the main bottleneck. The reasons are several, including the fact that a game often can't use all cores.

A better way to estimate CPU bottlenecking is to benchmark with your CPU set to two different frequencies (say 500MHz) apart. If the higher frequency shows improvements it's likely a better CPU would as well though how much is difficult to say.

*You can GOOGLE to see how much various games are affected by CPU scaling. As mentioned it varies a lot. BF3 can be demanding especially during Multiplayer on large maps.
Thanks for the detailed answer! I will look into it. So, I ran a benchmark with a .6 ghz difference (Stock clocks at 3.8 and overclocked to 4.4) And the MinMaxAvg stock is: 37, 45, 42.28, the MinMaxAvg for the OC is: 41, 47, 44.053
Overall, I think a new cpu is worth looking into, as this is not a huge overclock and I've heard bad things about the L cache of my current.

Best solution

a b 4 Gaming
a c 108 à CPUs
March 27, 2013 8:38:12 PM
Share

Billybob42,

You overclocked your CPU by about 16% and achieved about a 5% improvement. It's difficult to say, but I'm not certain a new CPU would produce a huge difference in this scenario.

Having said that, I've seen CPU benchmarks online that show a huge difference between CPU's in multiplayer with large maps, and of course there are other games that are very CPU dependant. Skyrim would benefit for one.

Tweaking:
No matter what your setup, unless you can achieve 60FPS on max settings, I advise you to tweak a little like this:

1. Run game with FRAPS and tweak until you achieve 60FPS at least 90% of the time.

2. Then turn on VSYNC.

*I also used RadeonPro in the past (have an NVidia card now) and used these features:
1) Force VSYNC (for games with no native support like Witcher #1)

2) Force AA (for games with no native anti-aliasing like Mass Effect #1. In the case of ME1 I recommend SuperSampling)

3) Force 50FPS (instead of 60FPS).
*Sometimes getting to 50FPS is fairly easy but 60FPS isn't. Many monitors support 50FPS at certain resolutions like 1920x1080 or 1280x720. I had several games running at 50FPS VSYNC'd.

March 27, 2013 9:08:05 PM

photonboy said:
Billybob42,

You overclocked your CPU by about 16% and achieved about a 5% improvement. It's difficult to say, but I'm not certain a new CPU would produce a huge difference in this scenario.

Having said that, I've seen CPU benchmarks online that show a huge difference between CPU's in multiplayer with large maps, and of course there are other games that are very CPU dependant. Skyrim would benefit for one.

Tweaking:
No matter what your setup, unless you can achieve 60FPS on max settings, I advise you to tweak a little like this:

1. Run game with FRAPS and tweak until you achieve 60FPS at least 90% of the time.

2. Then turn on VSYNC.

*I also used RadeonPro in the past (have an NVidia card now) and used these features:
1) Force VSYNC (for games with no native support like Witcher #1)

2) Force AA (for games with no native anti-aliasing like Mass Effect #1. In the case of ME1 I recommend SuperSampling)

3) Force 50FPS (instead of 60FPS).
*Sometimes getting to 50FPS is fairly easy but 60FPS isn't. Many monitors support 50FPS at certain resolutions like 1920x1080 or 1280x720. I had several games running at 50FPS VSYNC'd.


The thing is, I hear the fx-6200 is more for multi-tasking, while the 8350, 8320, and 6300 are gaming oriented. I'm not really sure how this works, but I've compared my performance with some benchmarks online, as well as some youtube videos, and the difference was phenomenal. He got 45 fps recording in one seen, and I got 35 not recording in the same. I think I'll go with the 8350, but I'll research some more.
March 27, 2013 9:23:33 PM

It's decided. I'm going to go with a 8350, as I hear it outperforms the x100s of amd by a lot. Thanks for all of your time and consideration people! :D 
a b 4 Gaming
a c 108 à CPUs
March 27, 2013 11:08:22 PM

This:

http://en.inpai.com.cn/doc/enshowcont.asp?id=7986&pagei...

BF3 in the tests done by this review show no advantage using better than an FX-6200. This is obviously due to the bottleneck being at the graphics card level (note all the other CPU's at the same cap). However, this begs the question as to whether you'll see an improvement by upgrading?

The test is from 2011 and used a GTX590 but I don't have time to investigate further. I suggest finding some benchmarks yourself to confirm if you'll see an improvement by upgrading.
a b à CPUs
March 28, 2013 12:18:29 AM

photonboy said:
However, this begs the question as to whether you'll see an improvement by upgrading?
The test is from 2011 and used a GTX590 but I don't have time to investigate further. I suggest finding some benchmarks yourself to confirm if you'll see an improvement by upgrading.

Yeah, count me in also as highly sceptical too. I doubt you're going to get anything resembling "value for money" as far as frames/sec go on a CPU upgrade.
March 28, 2013 10:03:53 AM

photonboy said:
This:

http://en.inpai.com.cn/doc/enshowcont.asp?id=7986&pagei...

BF3 in the tests done by this review show no advantage using better than an FX-6200. This is obviously due to the bottleneck being at the graphics card level (note all the other CPU's at the same cap). However, this begs the question as to whether you'll see an improvement by upgrading? Most of these tests are done in low pop servers, and in the singleplayer. Both of which my cpu and gpu do fine with. The bottleneck is mostly apparent in full servers, which I'm going to be playing in. That's one of the things that irritates me about cpu benchmarks in bf3....

The test is from 2011 and used a GTX590 but I don't have time to investigate further. I suggest finding some benchmarks yourself to confirm if you'll see an improvement by upgrading.


!