Archive Home arrow Reviews: arrow Processors arrow AMD A10-6800K APU Richland Processor








AMD A10-6800K APU Richland Processor E-mail
Reviews - Featured Reviews: Processors
Written by Hank Tolman   
Wednesday, 05 June 2013
Table of Contents: Page Index
AMD A10-6800K APU Richland Processor
Next Generation A-Series APUs
AMD FM2 Chipsets - A85X, A75, A55
APU Wireless and Dual Graphics Capabilities
Features and Specifications
Processor Testing Methodology
AIDA64 Extreme Edition Tests
PCMark 7 Tests
CINEBENCH 11.5
SiSoft Sandra 2013
DX11 Performance
AMD A10-6800K Overclocking
Richland Final Thoughts
AMD A10-6800K Conclusion

AMD A10-6800K Overclocking

For the unlocked K-series APUs, AMD recommends using AOD, AMD Overdrive. AOD is a utility that can be downloaded directly from AMD as a stand alone program. AOD can also be accessed and used from within the Catalyst suite, also known as the Vision Control Center.

AMD_A10_6800K_AOD_1.png

I actually ran into some trouble overclocking the AMD A10-6800K. First off, as you can see above, the APU was running at 4.3GHz from the get go. Obviously it was in turbo mode, but without anything running other than Overdrive, I'm not sure why it was pushing 4.3GHz.

AMD_A10_6800K_AOD_3.png

To manually overclock the AMD A10-6800K, you first have to turn off Turbo Core Control. You can do this by clicking the button with the same name and unchecking the box that says Enabled at the top of the screen. After that, you have full range to overclock the CPU Core Multiplier. I initially pushed the multiplier up to 46X, but when I ran the integrated stress test within AOD, it failed very quickly. I figured the CPU must have needed more juice, so I bumped the voltage up to 1.3875v. This gave me enough juice to run the overclock at 4.6GHz stably.

AMD_A10_6800K_CPUZ_OC.png

I tried a bunch of other settings, increasing the voltage slightly, decreasing the multiplier and increasing the reference clock, you name it. Nothing withstood stressing. I have to admit that I am a little disappointed that I was only able to overclock the A10-6800K to 200MHz higher than the stock Turbo speed. Of course, it was 500MHz higher than the stock clock speed, so that wasn't too bad.

In Cinebench, this translated to a score of 3.85 over a stock score of 3.41. That's about a 13% increase in performance for an extremely easy overclock.

I was also able to stably overclock the GPU to 1GHz from 844MHz, an increase of about 18.5%. This translated into increased FPS in 3DMark 11 on Test 1 from 9.2 to 11.9, up over 29%. In GT2 I went from 9.94 to 12.72, up 28%. GT3 went from 11.88 to 15.21, up 28%, and GT4 went from 7.13 to 8.89, up nearly 25%.

Overall, the increase in the GPU was much more dramatic than the increase in the CPU, and you might actually get some playable FPS on the low settings of higher end games if you can overclock to 1000MHz or better.

AMD_A10_6800K_AOD_2.png

The AMD Overdrive utility also comes with the ability to automatically overclock the APU. I tried using the presets, but failed everytime. I'd recommend slowly working your way up on your own.



 

Comments 

 
# about the RAMjebusricecakes 2013-06-04 23:12
>"Generally speaking, though, increased RAM speed doesn't amount to more than a couple of FPS difference."

this was proven false with APUs which can gain as much as a 20-25% boost from 1866 RAM over 1600 and the same with 1600 over 1333 which means which means if used bishock infinite would most likely have come in at a low but very much playable 25 FPS.

i only felt the need to point this out because anyone who is considering these new APU should be aware that RAM speed makes a much bigger difference than it does with any other setup.
Report Comment
 
 
# About the RAMHank 2013-06-05 06:47
Well, according to AMD's own testing, increasing RAM speed from 1600MHz to 1866MHz or from 1866MHz to 2133MHz results can result in FPS increases of 9 to 10%. That's a far cry from 20-25% and translates to a couple of FPS.

2133Mhz 1866Mhz 1600Mhz

Devil May Cry

1080p, Ultra, all settings max 41.351 38.051 35.691

Far Cry 3 Blood Dragon
1080p, DX11, med, low post fx, low shadows 30 26 24.932

Bioshock Infinite
1080p, low preset 31.2 27.9 25.56

Tomb Raider
1080p, low preset, texture quality normal 42.9 38.7 36

Sometimes the FPS gains are up to 4 or 5 FPS. So I should have said a few FPS difference rather than a couple. Either way, I didn't have any 2133MHz RAM on hand to test with.
Report Comment
 
 
# Memory speed mattersFasterMemoryPlease 2013-06-05 06:54
The biggest advantage of Richland over Trinity is support for DDR3-2133. It sounds like AMD should have included more information when sending these chips out, because benching Richland on DDR3-1600 is a waste of time.
Report Comment
 
 
# Hat tip from Hot Hardware reviewFasterMemoryPlease 2013-06-05 07:09
When paired with DDR3-1866 memory, the A10-6700 performs very much like the A10-5800K. The A10-6700 puts up somewhat better numbers due to its slightly higher-clocked CPU and GPU cores (3.7GHz / 4.3GHz / 844MHz vs. 3.8GHz / 4.2GHz / 800MHz), but the differences weren?t dramatic. The A10-6800K fares a bit better. Since AMD has qualified the A10-6800K for use with DDR3-2133MHz memory, and the APU is clocked at 4.1GHz (base) / 4.4GHz (Turbo) / 844MHz (GPU), its performance is better than any previous APU across the board.

#hothardware.com/Reviews/AMD-A106800K-and-A106700-Richland-APUs-Tested/?page=11
Report Comment
 
 
# RE: Hat tip from Hot Hardware reviewHank 2013-06-05 11:56
I'm a little confused. Of course the A10-6800K has better performance than any other APU across the board. That was never in question, and that is true regardless of the RAM speed.

Those numbers I posted above were from AMD's own testing and not one of those tests shows a performance increase of more than 6.9 FPS between the DDR3-1600MHz RAM and the DDR3-2133MHz RAM.

If there are any lingering doubts, the answer is yes, 2133MHz RAM will increase performance for the A10-6800K, but not dramatically enough to call testing with DDR3-1600MHz RAM a waste of time.
Report Comment
 
 
# RE: about the RAMOlin Coles 2013-06-05 07:10
It's easy to claim "this was proven false" in a comment. It's much harder to prove the claim is true. It would be helpful to link to a report from a reputable source.
Report Comment
 
 
# OCflanker 2013-06-05 01:47
What?! Your OC is totally wrong. AMD is not Intel as HK SOI is other. You can use without problems 1.5V for 24/7 at 15h Family. 1.378V is for AMD nothing, some chips has this voltage as VID
Report Comment
 
 
# RE: OCHank 2013-06-05 06:49
My OC is not wrong. 1.3875v is not high for AMD, however, whenever I increased voltage above that, the system crashed within seconds. I couldn't increase it above 1.3875v.
Report Comment
 
 
# Two QuestionsKrisnatharok 2013-06-05 03:21
1. Why did you pick two very new games that still are causing Nvidia/AMD to release regular driver updates to increase performance? Would it not have been better to use some games that have been out a bit longer, say Battlefield 3, Crysis 3, or Far Cry 3?

2. What is the performance of matching the APU + discrete GPU against an i3 or i5 + with the same discrete GPU? The iGPU is great, but most gamers are going to wonder if they can recommend the A10-6800K as a viable alternative to the i3-3225/i5-3570K/4670K as a gaming CPU with single GPU. With the ability to run two GPUs in x8/x8, it would be nice to see Crossfire tests with two discrete GPUs, if that is possible...

Thanks,
Kris
Report Comment
 
 
# RE: Two QuestionsHank 2013-06-05 07:03
I chose new games because those will be the considerations of someone looking to purchase a new APU or a new PC. If they needed a new APU or new PC to play BF3, they would have looked at buying when BF3 was new. I also used those games because they are the games AMD used in their testing, albeit with much higher results that I achieved. Either way, any of those games would have to be played on the very lowest settings to be playable with the APUs. Of course, they wouldn't be playable at all on the Intel IGPs.

As for your second question, I would have loved to run some CrossfireX tests. Unfortunately, the Richland APUs only support CFX up to Radeon HD 6670. I didn't have a 6670, 6570, or 6470 on hand to test with. I tried to use a 7790 and a 7850, but they don't work. As for using two discrete GPUs, those results aren't going to factor in the APU in any measurable way. You'd get close to the same results regardless of the setup with two discrete GPUs in CFX. I think I showed pretty well that, when paired with a Radeon HD 7850 or a GTX 660Ti, the A10-6800K is more than viable as a gaming CPU. Those scores will be consistent with any CPU.
Report Comment
 
 
# RE: RE: Two QuestionsKrisnatharok 2013-06-05 11:05
Hank,

Firstly, thank you for the getback!

So if I understand you correctly, even though they support x8/x8 configs, APUs do not support disabling the iGPU and running, say, two 7850s or 660 Ti's?

I am wondering how the APU compares directly to the i3-3225 and i5-3570K/4670K as *just* a gaming CPU, where a discrete card will be present and the iGPU will not be utilized for gaming.

A game like Supreme Commander, with lots of units on screen, would be a great way to test the CPU's capability in the game.
Report Comment
 
 
# RE: RE: RE: Two QuestionsHank 2013-06-05 11:50
Krisnatharok,

You can definitely use two video cards in CF or SLI configuration in an FM2 motherboard and bypass using the IGP altogether. My point was that, for nearly all high-end games, the CPU wouldn't make a significant difference in that case. The FPS using an A10-6800K would be about the same as if you were using an i5-4670K or even an i7-4770K. With CPU intensive games, of which there are a few, the faster CPU would obviously provide better performance. I ran tests using the GTX 660Ti as the sole GPU and the on both the i3-3220 machine and the A10-6800K machine, the actual gaming performance was always within 3%, generally considered to be the margin of error.
Report Comment
 
 
# DnaAngelDnaAngel 2013-06-05 04:21
My answer to the comment question: "How often do you upgrade your PC?". For my that depends on where the benchmarks are at, and what technology's are presenting themselves. On average every 3-4 years, but if I built an over the top system maybe 4-5.

I have little experience with the desktop APU's, as most of my customers rarely ask for those. I can however comment on the A10 in laptops. I bought a Trinity (A10-4600m) based laptop for school around this time last year and I was floored at the graphics performance. AMD is a clear winner if your shopping for a laptop on a budget and dont want or cant afford one with a discrete GPU in it. I am partial to Intel/Nvidia in the desktop market, but next time Im on the market for a Laptop without a discrete GPU, I know what I will be getting.
Report Comment
 
 
# RE: AMD A10-6800K APU Richland ProcessorSi 2013-06-05 05:41
Here is a youtube link showing 3dmark11 and windows index rating for the 4770k 3.5ghz Haswell. Not overclocked.

This is around 10-30fps slower than the 6800k in most games. And almost twice the price!!

Youtube link:
#youtu.be/k7Yo2A__1Xw
Report Comment
 
 
# APU's like fast DRAMFastersticks 2013-06-05 10:08
The Trinity 5800's had good gains in FPS going from 1600 to 1866 to 2133 and even more to 2400 as demonstated in numerous reviews i.e.

#hexus.net/tech/reviews/cpu/46073-amd-a10-5800k-trinity-needs-faster-ram/?page=3

and many more

See no reason why the 6800 shouldn't also
Report Comment
 
 
# RE: AMD A10-6800K APU Richland ProcessorTerrance Earle 2013-06-16 19:27
This review is bull#. I've played all those games on my Lanbox, with just the iGPU 7660D. My FPS for all those games were well over 30fps and the settings were medium-high. 1600x900. The newer one should perform margainly better. This review is fake and bought for by Intel.
Report Comment
 
 
# Another glimmering visitorOlin Coles 2013-06-16 21:55
I'm going to keep your comment online, so that everyone who visits this article can see exactly how ignorant Terrance Earle () truly is. I'm not even going to tell you why your comment is so full of poor comparison and incomprehension, but I'll give you a hint: 44%.
Report Comment
 

Comments have been disabled by the administrator.

Search Benchmark Reviews

Like Benchmark Reviews on FacebookFollow Benchmark Reviews on Twitter