AMD FX 6100 - Bottleneck AMD R9 280X (XFX)?

Technew111

Honorable
Dec 11, 2013
5
0
10,510
Hi guys,

Just wanting to know if there's a bottleneck with my rig. If so what would be the best cpu to get to replace fx 6100 (AMD only please)

Amd FX 6100 @ 3.7GHZ
Hyper Evo 212+ CPU Cooler
Corsair CX600M PSU
1TB HDD
8GB DDR3 1333mhz
AMD r9 280X XFX
asus m5a78l-m/usb3 motherboard.
 

wdmfiber

Honorable
Dec 7, 2012
810
0
11,160
Without a bit more money +plus Intel, there is honestly nothing for you to get.
But your bottleneck should be minimal anyway(now if you had two 280's in crossfire, different story).

Your system is decent. If you need better fps for some games, just lower your resolution a bit.
What is your monitor? 1080p?
 

wdmfiber

Honorable
Dec 7, 2012
810
0
11,160
Ya, even a 8350 wouldn't be much good to you. AMD piledriver architecture is just weak. And old... with 2014 right around the corner you don't want a "new" CPU's dating to 2012.

Stick with what you have. Or get a new Asus Z87 board and an Intel 4670K.
 

spladam

Distinguished
Nov 23, 2011
75
3
18,640


That is incorrect, an 8350 would work absolutely fine with 280x, There seems to be a very big misunderstanding of what the bottleneck phenomenon is and how it works. It's very dependent on what you are running first of all, and it takes quite a bit of a performance gap before you can actually NOTICE anything without staring at meter numbers for an hour. Most modern titles will not come close to maxing out an 8350, no matter what GPU you've paired it with.
 

DubbleClick

Admirable


Good joke.

http://www.techspot.com/review/875-intel-core-i7-5960x-haswell-e/page8.html

or literally every other valid benchmark out there.
 

spladam

Distinguished
Nov 23, 2011
75
3
18,640


"Literally" huh? "Valid" benchmarks, as opposed to all the invalid ones. Ok, that link you referenced, did you notice how the reviewers avoided using the word "bottleneck", and how they referred to CPU "utilization". That's because that's what it is. You are confusing "better CPU = better performance" with bottle necking, which is what I was saying about there being a common misconception. If they showed the CPU-Z or HWiNFO sensor readout you'd most likely see that the FX 8350 did not hit 100% during test. Yes, better CPU means better performance, but that does not mean that the GPU was not using 100% during game-play. My R9 280 hits 100% under heavy load paired with my weak and tiny FX-6100 at 4GHz. What we refer to as Bottleneck happens when your CPU hits 100% and your GPU doesn't, under heavy load. Again, bottleneck is when your GPU fails to utilize 100% use under heavy load, limiting it's performance, and my R9 280 hits 100% during the Hitman Absolution benchmark every time. However, if I upgraded my CPU, I would have a better frame count.

Let me quot the AMD crossfire thread from THIS website:

"CPU Bottlenecking misconceptions: There is ONLY a bottleneck in your system if your CPU is at or close to 100% load, OR if it is throttling at a way to high temperature. If that is not the case with your current CPU, there is simply no bottleneck with your processor. If your current CPU is running close to 100% load, it is best that you soon invest in a new CPU to take maximum advantage of the cards in your chosen configuration."

You see, no joke.

-Edit: Let me correct myself, in a moment of self doubt, I ran the Hitman Absolution bench to double check, my 280 hits 97% while my CPU hits 79%. It appears I am indeed slightly holding that GPU back, but then again, it's a FX 6100. It's really time to upgrade.....
I average 42FPS on Ultra settings by the way.
Tomb Raider bench on Ultra: 58 FPS average, CPU: 67% peak GPU: 100%

 

mdocod

Distinguished
All it takes is one core pegged at 100% utilization to trigger a CPU bottle-necked condition. When viewed on whole, that means that an FX-6100 reporting as low as ~17% utilization can be the source of a performance bottleneck.

----------

The CPU bound performance cap (FPS) is going to be very nearly the same whether you are running an R7 250X at 720P or an R9 290 at 1440P. Performance originates with the CPU, not the GPU. The size of the GPU configuration manifests as differences in VISUAL QUALITY in actual implementation, not performance.
 

slyu9213

Honorable
Nov 30, 2012
1,054
0
11,660
You're motherboard will pretty much support every Bulldozer and Vishera CPU excluding the 220W TDP variants.

I would just OC the 6120 as high as possible but do it safely. Do you already have this setup? If you do why don't you actually play the games you are currently before you make a decision. This is something I hope every newcomer does. If you have the hardware, play the games you want at the settings you want to. If the performance stinks then come to Toms or other forums and ask people on what is the best option. That will save time and money on upgrading unnecessarily.
 

spladam

Distinguished
Nov 23, 2011
75
3
18,640


Was wondering how that stress test went with that Evo 212+ ? They are really cheap on the CM website, think I might get one.

 

DubbleClick

Admirable


First of all, there is no exact definition of the term bottlenecking. And when cpu #1 gives 40 fps and cpu #2 gives 120 fps at the same settings, I do call that a bottleneck.
 

mdocod

Distinguished
It's starting to catch on though..

A year or 2 ago, if you visited these forums, or pretty much any other computer hardware enthusiast forum, you'd read post after post implying that gaming performance originates with the GPU and the CPU doesn't matter.

It has taken literally THOUSANDS of people starting threads complaining about performance problems after following the bad advise of the community for people to finally start catching on to the reality that CPU performance is the dominant force for gaming performance. Though we have a long way to go...

I'd say that at this time, we've reached perhaps a 10% saturation of hardware enthusiasts who really "get it," about 40% who recognize the importance of the CPU but still placate to fundamentally flawed premise regarding CPU and GPU "balancing" efforts, and a remaining 50% who still think it's 2002.