Review Index:
Feedback

AMD A10-6800K and A10-6700 Review: Richland Finally Lands

Author:
Subject: Processors
Manufacturer: AMD

Trinity... but Better!

Richland.  We have been hearing this name for a solid nine months.  Originally Richland was going to be a low end Trinity model that was budget oriented (or at least that was the context we heard it in).  Turns out Richland is something quite different, though the product group does extend all the way from the budget products up to mainstream prices.  We have seen both AMD and Intel make speed bin updates throughout the years with their products, but that seems like it is becoming a thing of the past.  Instead, AMD is refreshing their Trinity product in a pretty significant matter.  It is not simply a matter of binning these chips up a notch.

View Full Size

Trinity was released last Fall and it was a solid product in terms of overall performance and capabilities.  It was well worth the price that AMD charged, especially when compared to Intel processors that would often be significantly slower in terms of graphics.  The “Piledriver” architecture powers both Trinity and Richland, and it is an improved version of the original “Bulldozer” architecture.  Piledriver included some small IPC gains, but the biggest advantage given was in terms of power.  It is a much more power efficient architecture that can be clocked higher than the original Bulldozer parts.  Trinity turned out to be a power sipping part for both mobile and desktop.  In ways, it helped to really keep AMD afloat.

It turns out there were still some surprises in store from Trinity, and they have only been exposed by the latest Richland parts.  AMD is hoping to keep in front of Intel in terms of graphics performance and compatibility, even in the face of the latest Haswell parts.  While AMD has not ported over GCN to the Trinity/Richland lineup, the VLIW4 unit present in the current parts is still very competitive.  What is perhaps more important, the software support for both 3D applications and GPGPU is outstanding.

Click here to read the entire review on the AMD A10-6800K and A10-6700.

Digging into Richland

Richland is not a totally new product.  It is in fact still based on Trinity, but it is a slightly different revision.  A few hardware changes were made, but nothing that would radically change overall performance or characteristics.  Instead, AMD activated parts of Richland that were not doing anything of great import in Trinity.  The Trinity die was originally designed with quite a few sensor sites and a pretty robust controller that handled power needs depending on both workload and temperature.

View Full Size

Firmware and software both were heavily tweaked to enable greater overall turbo performance while keeping TDPs and power consumption in check.  Tweaks were also made to the 32 nm HKMG/PD-SOI process to get a little more performance.  Finally Richland is a new, minor revision of Trinity that enables the hidden functionality and helps to clock the APU up without eating a whole lot more power or creating more heat than previous parts at lower clockspeeds.

Richland comes in 100 watt and 65 watt TDPs for the desktop.  Previously we saw the latest Richland mobile parts which could go down to 15 watts TDP in some circumstances.  The high end unit is the A10 6800K.  This part has a base clock of 4.1 GHz (up 300 MHz from the A10 5800K) and a max turbo clock of 4.4 GHz.  It has a full 384 shader cores running at a maximum speed of 844 MHz.  This again is up from the previous 800 MHz of the 5800K.  This chip also increases the memory speed from previous models.  The new APUs can now run officially at DDR3-2133 MHz speeds, which should give a nice little boost to most graphics workloads.  The only downside to this speed is that AMD only officially supports 2133 speeds with one DIMM per channel.  So those RG2133 4 x 4 GB packs that AMD is selling seem sort of useless when they may not entirely be stable with 4 DIMMS occupying 2 channels at 2133 speeds.

View Full Size

Motherboards will need to have their BIOS updated to be able to fully utilize the Richland design.  There is a tremendous amount of firmware work that went into to make Richland what it is.  The hardware is now much more aware of where potential bottlenecks are, and it will dynamically speed up and slow down different parts of the design to adjust for performance issues.  Power, temperature, and utilization are all measured throughout the core by many sensors.  If the CPU is bottlenecked, it will raise the voltage and clockspeed on that unit until a thermal limit is reached, then it will start clocking that part down.  At the same time the APU will clock the GPU unit down and lower the voltage, so that it will act more as a heatsink for the CPU portion.  It is not as efficient as what Kabini showed us in terms of providing a “heatsink” for the other portion, but it does make a small difference.  A couple of extra seconds at a high turbo state might help to complete the work quicker, and allow the APU to be clocked down until it needs to be under heavy load again.

Richland does seem to stay under turbo mode longer than previous versions.  This can be a blessing or a curse.  In small chunks it is a fast processor, but the longer it is under load the lower the clockspeed goes.  For example, when running Cinebench and watching the watts being measured, it would start out at near the highest recorded wattage and about 30 seconds later we would see power being lowered over time.  There can be a good 5 to 7 watts difference between the peak at the beginning of the benchmark to when the test ended.  From top to bottom there are also more frequency and voltage levels available to the processor.  These again are entirely dictated by power consumption, load, voltage, and temperature.

June 5, 2013 | 12:20 AM - Posted by Tim Verry

Really good review Josh! It looks like AMD has another solid APU on its hands, all things considered.

EDIT: Wow, those are some nice overclocking results, especially for the GPU!

June 5, 2013 | 02:10 AM - Posted by SPBHM

I think the CPU OC is a lot more impressive, 4.8GHz and using 140w under load? that's pretty good

what software was used for the (stock) load entire system test?
it's strange that the difference between the 100 and 65w TDP CPUs would only be 15w (6700 vs 5800K), but still, performance is the same, so... not bad.

the ability to use the IGP + any other card would be quite good.

June 5, 2013 | 04:51 AM - Posted by Tim Verry

Well that too of course :).

June 5, 2013 | 07:29 AM - Posted by Josh Walrath

I used Cinebench 11.5 to load up the CPU cores for power testing.

Real world TDPs are... interesting.  AMD sets a level for the worst case scenario for their parts, so there is going to be some variances chip to chip.  So while the 6800K is officially a 100 watt TDP part, it most likely consumes less power than that rating implies.

There are certainly some interesting wrinkles to this release.  It is certainly not a world beater, but the overall product is still pretty impressive.  AMD must be extremely happy that Intel decided to forego putting GT3 and GT3e graphics on the desktop.  If Intel had done that, AMD would have an even tougher time selling chips.

June 5, 2013 | 02:04 AM - Posted by Anonymous (not verified)

Doesn't matter what AMD releases for the next month or two.

Like it or not, unless it comes with a supermodel serving you beer, it will be hard to drown out all the buzz around Haswell.

On the other hand, so far, I've been very unimpressed with Haswell for the desktop enthusiast.

June 5, 2013 | 05:10 AM - Posted by Anonymous (not verified)

the ability to use the IGP + any other card, is disallowed for higher end cards! Is this in the hardware or BIOS? Can Lucid's MVP Virtu software get around, AMD's obvious antiHSA move (not allowing higher end discrete cards and IGPU), the AMD restriction? Why is AMD, the founder of the HSA foundation, doing something so antiHSA, by not allowing the user access to the full computational ability of their APUs, in conjuction with Higher End discrete GPU cards! AMD do NOT let your marketong folks ruin you! This whole thing reeks of milking the user base, Just like Intel does with it's less than adequate, planed incremental performence/technological increases! Oh let's give the user just enough advancment to keep them buying! I wish there was a third company out there with an x86 license, and some GPU IP, to keep both AMD and INTEL on the straight and narrow!

June 5, 2013 | 06:36 AM - Posted by Anonymous (not verified)

It's not because AMD is trying to be anti-HSA, it's because if you have a higher end card it would actually not help you process data any faster. This is due to the fact that the current way they an IGPU + Discrete card interact is by "Crossfire X". The methodology is that both GPUs"combine" to appear as one card. This eliminates a lot of other issues that AMD saw with "Hybrid Crossfire". If you read around there are plenty of people that have messed with trying to get an HD 7870 to work with an 5800K APU for example and come to the same conclusion as AMD. Why hobble the powerful card to get comparably marginal performance from the IGPU? I understand wanting to use it since it's there but the way the software and hardware are built right now, it's not worth the worry.

June 5, 2013 | 07:17 AM - Posted by Anonymous (not verified)

Yes but do not shut the IGPU off for Opencl workloads, while the higher end descrete GPU crunches graphics, The one thing AMD, and Nvidia have that can give them leverage, against Intel x86 CPU might, is using the IGPU as an extra processing resource (GPGPU) for gaming physics/ray tracing boost! Read up on Lucid mvp technology, and tell me if you think this offers a solution to the problem!

AMD Like INTEL, both, if given the chance, will slow CPU/APU technological improvements if there is insufficient competition in the market place to fore them to move forward! Just look at AMD VS NIVIDA in the descrete GPU market!

June 5, 2013 | 06:48 AM - Posted by Spigzone (not verified)

You're con-fusing AMD (desperately trying to stay in the processor race) with Intel (800 lb. gorilla cash cow milking it's lead).

AMD's marketing does not drive this process, the step by step design and engineering learning process and system evolution leading up to a full HSA implementation does.

AMD would love nothing more than to CURRENTLY have a full array of HSA APU processors on the market able to additively couple with it's full array of current GPU boards.

In short, AMD flat out cannot afford to leave performance advantages on the table to 'milk the consumer'. It has never been in that position. That is what Intel routinely does.

June 5, 2013 | 07:31 AM - Posted by Anonymous (not verified)

AMD is not in the x86 race on the high end, Intel won, and AMD said forget the high end, but AMD can and does still compete with Intel on the low power Mobile x86 market, and AMD integrated GPU IP, EATS Intel's IGPU IP for lunch! AMD could easily use its better GPU IP to put Intel Haswell/Broadwell GPU IP in the rear view mirrior, for a few years, at least! Why do you thik AMD's lack of progress with AMD's IGPUs vs Intel's IGPUs has a back and forth feel to it, when AMD could, if they so desired, Blow Intel's IGPU's so completely out of the water, and be done with it!

June 5, 2013 | 08:10 AM - Posted by Anonymous (not verified)

Oh And the real con- here is anyone trying to paint AMD as the Underdog, and Intel As The Overdog, the truth is both AMD and Intel are both just plain Dogs, and the consumer wins when they are fighting each other for survival! I hope APPLE takes ARM technology and soups it up and combines it with an equally souped up IGPU, to compete with AMD and Intel's mobile offerings, Apple has a wad-o-cash, and has allready shown it can take the ARM instruction set, and pair it with some GPU IP and go to market with it! Apple is another DOG with deep pockets, the more dogs in the competition, the more the consumer becomes winner!

June 8, 2013 | 05:46 PM - Posted by deowll (not verified)

Right now there really isn't an AMD Intel fight. Intel could finish crushing AMD and put it out of business but it doesn't want to because if it did then it might become a monopoly and get attacked as such the way MS, Google, and others have been attacked. Much cheaper to allow AMD just enough room to survive but not enough room to prosper.

June 5, 2013 | 11:01 AM - Posted by D1RTYD1Z619

So what would you do Josh? I was going to upgrade the video cards (HD4850 512mb Crossfired) in my 12yr old daughters computer, she plays Minecraft, TF2, Sims 3 and L4D2, to a HD7770 because they are under a 100 bucks or should I just dump the system and build her one off the A10-6800K with no discrete card.

Current System specs:
Antec 300 Case
Asus P5K-E WIFI-AP Motherboard
Intel Core 2 Quad Q6700 @ 3.4 Processor
Corsair Dominator CM2X1024-8500C5D2GB 2GB (2 x 1GB) PC2 8500 1066MHZ DDR2 Memory X 2
Hittachi-LG HL-DT-ST DVDRAM GH20NS15 Optical Drive
Maxtor MaXLine 7H500F0 500GB Hard Drive
Asus EAH4850/TOP/HTDI/512M HD4850 512MB DDR3 in Crossfire
Antec TruePower New TP-750 750 Watt Power Supply

This system puts out way too much heat and I bet draws too much power for its performance benefits.

June 5, 2013 | 11:29 AM - Posted by Josh Walrath

If you are worried about heat, power, and age then yes... upgrade and a 6800K will be a great replacement that will enable her to play those games without issue.

That being said, a Q6700 at 3.4 GHz is still a really fast processor. It likely is slower overall than the 6800K, but it is still fast. Getting a $100 card will provide quite a few more years out of it, plus cut down on heat because the old 4800 series could really push it out. For the price though, you might want to consider the 7790... could get some really nice mileage out of that combination.

June 7, 2013 | 07:26 AM - Posted by Yamaeda (not verified)

I'd just buy one 7790 and be done with it. :)

It's a very good processor, especially overclocked, but i understand if your rig gets warm. Though it's more due to teh graphics cards than the cpu.

/Y

June 5, 2013 | 11:59 AM - Posted by w_km (not verified)

QUESTION: Could I use the A10 GPU to run one 1440p monitor with an additional 7xxx series GPU to run another 1440p monitor?

Or would the discrete 7xxx GPU take over both monitors with the APU disabled???

June 5, 2013 | 02:28 PM - Posted by Josh Walrath

I guess I will have to check.  I am not sure, but I believe that when a standalone card which is not a 6670 and below is inserted it disables the GPU portion of the APU.  If not, then I believe Win7 does support multiple, non-identical graphics adapters as long as they utilize the same drivers.

June 6, 2013 | 05:27 AM - Posted by pyjamarama (not verified)

For building a low end system that does minor gaming, if we pared the Intel low end Pentium G2020 with a 100$ to 150$ graphics card would we get a better gaming performance?

Or putting the question another way if we put a G2020 or a Core I3 what graphics card we would have to buy to have the same gaming performance?

June 7, 2013 | 07:25 AM - Posted by Josh Walrath

I believe a G2020 is actually going to be far slower than Richland in both CPU and GPU performance.  A person would have to buy around a $75 card to match the graphics performance of the A10 6800K in most instances.  If you were dead set on buying the cheapest, decent processor and adding a $150 graphics card... then that will still make a great gaming machine anymore.  The FX-6350 is around $139 (you might be able to find the FX-6300 for cheaper) and seems like a nice combination of performance, thermals, and price.  There are also plenty of i3s in the $100 to $175 range that will perform well.

As for graphics, I wouldn't go with anything less HD 7790 anymore.  That and the GTX 650 Ti Boost seem pretty solid.

June 8, 2013 | 10:14 AM - Posted by Anonymous (not verified)

This is a very good review.

By the way, I am a fan of pcper.com.

June 11, 2013 | 01:44 AM - Posted by Anonymous (not verified)

Rly nice CPU.

June 12, 2013 | 02:42 AM - Posted by Blair (not verified)

I'd build a new pc around it and give my old Q6600 to a friend.

June 12, 2013 | 05:53 PM - Posted by Poci

Wow im impressed, these little apus can score close to what my old 5970 did in 3dmark????

And people complain about AMD?????

June 14, 2013 | 12:12 AM - Posted by Panta

i think in-light of recent time intel CPU's heat & OC fails
its a good time for AMD pick themselves up & start fight intel again.

June 14, 2013 | 08:16 AM - Posted by Josh Walrath

Yeah, it seems like Intel is so comfortable in their lead, that they aren't pushing things in terms of enthusiast level products.  How long has Sandy Bridge E been out?  Still any sign of Ivy Bridge E?  Also, Intel is really pushing into mobile and they probably think power desktop computing is dead.  Hard to say where exactly that will go, but I hope power desktop computing isn't becoming a niche market (even though it sorta is...).

June 20, 2013 | 03:25 AM - Posted by Anonymous (not verified)

Can we have a windows 7 aéro test screen pleaz
Cause the A10-5800k has 6.9 aéro and 6.9 3D in windows 7 and I whant to know if the A10 6800k has better performance (win7 test) with 2133 memory

September 10, 2013 | 05:43 AM - Posted by ezjohny

good read Josh, AMD needs to make a good solid APU for the desktops that utilizes AMD latest 7000 series graphic cards. Maybe HSA is a better way to move data around faster, we will have to see when it comes out. Hope AMD does not disappoint! I'm thinking of getting an APU set up!

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Lines and paragraphs break automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd> <blockquote><p><br>
  • Web page addresses and e-mail addresses turn into links automatically.

More information about formatting options

By submitting this form, you accept the Mollom privacy policy.