A8 5600k & e8500

Page 4 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

SPBHM

Diamond Member
Sep 12, 2012
5,056
409
126
fc3%20proz.png


I think we can safely agree now that Intel cpus bellow $300 for gaming = no no.

[/sarcasm]

another test,
4VwfH.png


it's hard to make any conclusion, but that's not the first time I'm seeing "strange results" from GameGPU, in the past it was pretty clear that their test methodology was not great, perhaps it still the case, or maybe they tested under significantly different conditions...


again for the AMD fans, wouldn't be better to just buy the FX 6100 (since power usage is not a great concern anyway...)?
at the same clock it should be as fast as the a10 for gaming, the A10 had some improvements, but it lacks the l3 cache (and comparing the 5800K with the FX 4300 it makes a big difference for gaming http://www.anandtech.com/bench/Product/675?vs=700), but the 6100 have also 3 modules, cpu clock is low, but it shouldn't be hard to OC it to 4GHz?

only $10 more
http://www.newegg.com/Product/Produc...82E16819103962
 

inf64

Diamond Member
Mar 11, 2011
3,702
4,030
136
another test,
4VwfH.png


it's hard to make any conclusion, but that's not the first time I'm seeing "strange results" from GameGPU, in the past it was pretty clear that their test methodology was not great, perhaps it still the case, or maybe they tested under significantly different conditions...


again for the AMD fans, wouldn't be better to just buy the FX 6100 (since power usage is not a great concern anyway...)?
at the same clock it should be as fast as the a10 for gaming, the A10 had some improvements, but it lacks the l3 cache (and comparing the 5800K with the FX 4300 it makes a big difference for gaming http://www.anandtech.com/bench/Product/675?vs=700), but the 6100 have also 3 modules, cpu clock is low, but it shouldn't be hard to OC it to 4GHz?

only $10 more
http://www.newegg.com/Product/Produc...82E16819103962

In your test the resolution is much lower(12x7 Vs 19x10). What AtenRa linked was high resolution gaming ,what one with high end hardware will use(and not low resolution).
 

Mallibu

Senior member
Jun 20, 2011
243
0
0
In your test the resolution is much lower(12x7 Vs 19x10). What AtenRa linked was high resolution gaming ,what one with high end hardware will use(and not low resolution).

FX delivering < 60 fps at low resolution, will still deliver < 60 fps at high resolution. It's not magicaly going to get better.

And the worst part is, with GPU bottlenecks you can just set on or two setting sfrom "ultra" to "high" with minimal cost in quality to ensure enjoyable gaming experience (60 fps).

With the FX you are stuck with 40 fps no matter what.
 

AtenRa

Lifer
Feb 2, 2009
14,001
3,357
136

GPU limited even at 1280x1024 with a 4 year old Nehalem at 3.75GHz and GTX680.

Every FarCry 3 review have produced different results so far because they bench with different hardware(GPUs) and different IQ settings.
 

Mallibu

Senior member
Jun 20, 2011
243
0
0
GPU limited even at 1280x1024 with a 4 year old Nehalem at 3.75GHz and GTX680.

Yes indeed.
With the FX however you are not, since it delivered 10 frames less. It's a bit embarassing that a 200$ 8core cpu launched a month ago, is 10-15% behind a 4 year old cpu.
 
Last edited:

AtenRa

Lifer
Feb 2, 2009
14,001
3,357
136
Yes indeed.
With the FX however you are not, since it delivered 10 frames less. It's a bit embarassing that a 200$ 8core cpu launched a month ago, is 10-15% behind a 4 year old cpu.

An OverClocked 4 year old CPU. OC the FX and you will be GPU limited as well ;)
 

Mallibu

Senior member
Jun 20, 2011
243
0
0
An OverClocked 4 year old CPU. OC the FX and you will be GPU limited as well ;)

Well, if you look at the graph SPBHM posted, the FX 8350 = i5 760 in Far Cry 3, so even with the i7 965 at stock it would probably still be 10 frames ahead.

Anyways, it's a classic example of todays games, where IPC > Core count.
 

Torn Mind

Lifer
Nov 25, 2012
11,646
2,654
136
To me, one concern with suggesting Trinity is that his RAM clocked at 1333Mhz. IF the APU's sensitivity to RAM clockspeeds applies here--it might not since he is using a discrete card; this would be a moot point in that case--, what would be the implications of it for him?
 
Last edited:

SPBHM

Diamond Member
Sep 12, 2012
5,056
409
126
In your test the resolution is much lower(12x7 Vs 19x10). What AtenRa linked was high resolution gaming ,what one with high end hardware will use(and not low resolution).

increase the resolution and the framerate is not going to increase, it will only decrease or stay the same, it's a valid test, it shows the CPU limiting the performance.

An OverClocked 4 year old CPU. OC the FX and you will be GPU limited as well ;)


the thing is, to compensate for the GPU limitation you can lower a few quality settings (which doesn't have a huge visual impact, but a significant performance hit), while the CPU performance varies a lot less normally... so there is not much you can do if you want higher framerate with a slow CPU,

nehalem at 3.7ghz is more than fast enough for most the games, when overclocked the old i7 920 is still quite good, apart from the terrible power consumption,

4 years ago nehalem was by far the best CPU around, so "a 4 year old cpu" being the i7 at 3.7GHz hardly means that the game works fine with any CPU.

I don't think the A8 (2 modules, no l3 cache) will do amazingly well in this game,