What's new
  • Please do not post any links until you have 3 posts as they will automatically be rejected to prevent SPAM. Many words are also blocked due to being used in SPAM Messages. Thanks!

AMD Richland Review; A10-6800K & A10-6700 Benchmarked

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
13,410
Location
Montreal
AMD has some catching up to do and Richland is their transitional solution between Trinity and Kaveri. It is also meant to shore up their performance metrics now that Intel has released their new Haswell architecture. In a way, this is AMD trying to steal a march on Intel since the Trinity to Richland refresh took just nine months while Intel’s cycles typically take 12 months or more.

Richland is actually a bit of a surprise since AMD’s product roadmaps never showed it until a short time ago. Simply put, GlobalFoundries’ 28nm manufacturing process delays and intrinsic revisions to upcoming architectures have pushed back the APU and CPU roadmap. This has necessitated the introduction of gap-filler products like Richland which are based on proven architectures.

While it may be a slight departure from earlier roadmap predictions, Richland still very much adheres to AMD’s Heterogeneous System Architecture or HSA. This amalgamation of CPU and GPU processing onto a single die will be the central focus for AMD far into the future but, judging from sales, potential customers haven’t quite embraced this approach. Luckily, the software is quickly catching up to hardware capabilities and soon we should see a painless transition towards more adaptable programs in the near future.


With Kaveri on the horizon, Richland may not be what many expected but it does give AMD a slightly better chance of competing against Haswell’s lower-end SKUs than Trinity did. Just don’t expect these new processors to go toe to toe against the i7 4770Ks and i5 4670Ks of this world in the performance or efficiency categories.

With that being said, this refresh is being accomplished through a revised feature set and higher clock speeds. Expect good performance per dollar ratios against Intel’s lower-end Ivy Bridge for now and dual core Haswells when they eventually hit the market.


So what makes Richland tick? Mostly the same things as its predecessor, but there has been some fine tuning going on behind the scenes which will allow Richland APUs to compete against Intel’s processors on a more level footing.

While Trinity mated up to four Piledriver cores with a Northern Islands class GPU, and AMD’s next generation Kaveri APUs will use Steamroller CPU architecture paired up with GCN-based graphics, Richland takes a path between these two. It uses the same architectural bones as Trinity, but rolls in several next generation power management optimizations which directly affect clock speeds. As a result, its GPU and CPU cores operate at higher frequencies without negatively impacting TDP.

Regardless of the HD 8000-series moniker used for the GPU core, Richland still uses the same VLIW4 Northern Islands design as Trinity did. Once again, AMD has rebranded theses units in order to properly reflect the additional performance benefits derived through their higher operating frequencies.

The memory controller on unlocked A10 APUs has also been revamped with support for DDR3-2133 modules, though lower-end APUs don’t receive the same treatment. Meanwhile, the use of an FM2 socket retains platform compatibility, allowing Richland to become a simple drop in solution for existing motherboard designs.


The Richland lineup follows in the exact footsteps of AMD’s outgoing Trinity APUs, literally replacing them on a 1:1 basis, though the newcomers are slightly more expensive. The only odd man out is the dual core A4-5300 which will stick around until stocks are depleted or it gets discontinued since there’s no replacement.

Headlining this new APU lineup is the $142 A10-6800K, an unlocked APU which has received a significant Base Clock increase coupled with a healthy 200Mhz boost in Turbo frequencies. The 384 core GPU also gets a new HD 8000-series designation (though it still uses an older, pre-GCN architecture) alongside slightly higher clock speeds. This has been accomplished without a corresponding TDP increase.

The A10-6700 uses the same core as the 6800K but includes lower clock speeds across its x86 cores and GPU in order to achieve a lower TDP value. For many, this will become a great companion for an HTPC system or low power desktop.

Much like their higher priced siblings, the A8-6600K and A8-6500 are cut from the same cloth with the K-series part having higher clock speeds and an unlocked multiplier while consuming more power. These are the lowest-end quad core parts as the A4-6400K is simply an unlocked 65W dual core part with a cut-down GPU and clock speeds that are higher than some other quad core SKUs. It is meant to target the entry level, desktop all in one systems.


These new APUs are compatible with AMD’s Dual Graphics technology. Dual Graphics is essentially Crossfire for APUs, pairing up the internal GPU with a discrete card, in this case, the HD 6670, HD 6570 and HD 6450. The performance increase is impressive to say the least but buying an outdated $60 to $80 graphics card likely won’t appeal to many people considering Richland’s already-low price. This technology does however ensure that higher performance can be achieved should someone want to do more serious gaming on an APU-based system.


In this particular we will be looking at two APUs which are quite similar in the performance end of the spectrum but they diverge quite drastically in other respects. The A10-6000K is AMD’s flagship APU and will remain so until Kaveri is released. It features an unlocked multiplier for easy overclocking and boasts the highest clock speeds of any APU to date. The A10-6700 on the other hand isn’t unlocked and operates at slightly lower frequencies but its 65W TDP will likely appeal to those running more efficient systems who don’t need the extra performance boost overclocking allows.

All in all, Richland looks like an interesting addition to AMD’s product stack but its underlying architecture has already been proven to be deficient in some areas. Luckily, the performance increases should shore things up in some areas that lagged behind while the boost in graphics capabilities (albeit minor) will put some more distance between Richland and the latest Haswell entrants.
 
Last edited:

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
13,410
Location
Montreal
Richland; Created Through Hybrid Boost

Richland; Created Through Hybrid Boost


Richland may only represent a slight refresh to the original Trinity design, but AMD claims it delivers between 10% and 20% more performance than its predecessor for CPU and GPU related tasks respectively. Whether or not that’s performance per watt or raw benchmark data will be up for debate, but it is nonetheless promising to see what can be achieved through a few modifications.


Much of Richland’s potential performance improvements are derived through the optimized clock speed responsiveness of AMD’s new Hybrid Boost technology. Much like Trinity, it uses a 32-bit microcontroller within the APU which dynamically tracks the temperature of every on-die component. The major difference in this implementation is its ability to effectively parallelize its calculations in order to balance clock speeds in relation to temperatures within different regions of the APU, power needs, regional loads and other factors.

The entire point of this exercise was to add some granularity to Turbo speeds. As it stood, Trinity could accomplish many of these same tasks but relied on a strictly limited number of P-states, limiting effectiveness when it tried to find an optimal frequency point. Hybrid Boost meanwhile provides additional “sub-states” which lead to more fine grain control over the operating points for frequencies and power consumption. Compare this real-time responsiveness to Intel’s Turbo Boost which estimates many of these factors and you can begin to comprehend the amount of engineering which goes into AMD’s new solution.


This approach also allows Richland’s architecture to find a balanced operating frequency between GPU and CPU performance when both are loaded. Basically, AMD has determined “happy medium” points to ensure that neither component is starved for data at any given time. As a result, workload aware power management coupled with temperature aware TDP balancing can properly manage operating cycles of both functional architectural blocks in a concurrent, on-the-fly manner.

Richland’s use of Hybrid Boost Higher leads to more consistently high clock speeds and better overall efficiency than previous architectures without any large scale changes to the silicon. In plain English, this allows Richland to run faster more often while maintaining a TDP of 65W to 100W.

Another important addition to this architectural refresh is a set of scalable TDP points. These can be configured by the OEMs based on the needs of a given platform, opening up the possibility of using higher end quad core APUs within ultra-thin systems. Since AMD’s Hybrid Boost calculation is also based upon temperature status, we could also see potentially higher performance if OEMs or end users utilize more efficient cooling assemblies.
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
13,410
Location
Montreal
New “Elite” Features Come to the Desktop

New “Elite” Features Come to the Desktop


Alongside Richland, AMD is also launching a new top-to-bottom branding policy for their upcoming APU designs by moving away from the sometimes confusing “Vision” designation. In its place will be a standardized typeface and logo along with different “modules” that are supposed to designate separate feature sets.

The new approach isn’t quite as convoluted, but the features’ implementation is anything but guaranteed across AMD’s entire APU product lineup.


The best way to dive right into this explanation is an overview of the new branding we’ll see from AMD over the course of 2013. While the usual A10, A8, A6 and A6 will remain in place, certain product lines will receive the “Elite” feature set which includes a wide range of items designed to enhance a user’s experience. These SKUs will have “Dual Core” or “Quad Core” added in order for an end-user to distinguish the number of processing threads present in a given APU design. In addition, for certain designs, dual graphics will once again be an option for increased graphics performance. For example, a Richland A10 can be combined with Dual Graphics making it an AMD A10 Elite Quad Core with Dual Graphics. That’s a mouthful, isn’t it?

Meanwhile, the “Mobility” line of Temash ultra-low voltage APUs doesn’t quite follow the same naming scheme since dual core implementations will likely be utilized in the lion’s share of devices. It does however include the Elite feature set and there will be a Quad Core option so both monikers have been added in one way or another.


Richland’s expanded feature set includes a number of old favorites which have been brought up-to-date alongside some new additions. These all put a focus on leveraging GPU acceleration in order to benefit the user’s overall experience with an AMD-equipped system. However, since many of the features have been implemented through ISV partnerships, OEMs can pick and choose their inclusions so they may not be available on all systems.

We’ve all come to know and love AMD’s Quick Stream, Steady Video and Picture Perfect HD and they have been ported over to Richland largely unchanged. There have been some new hardware encoder optimizations but don’t expect any drastic departures in this area.


The new kids on the block are the “Elite” features which will only be available on select APU products and take the next step towards integrating mobile devices into everyday life. Once again, they are all GPU accelerated and have been brought about by AMD’s close working partnership with software developers.

Gesture Control allows a user to navigate through certain applications without actually touching the screen. It does this by enabling the system’s onboard webcam and tracking recognized hand movements while interacting with Windows 7 or Windows 8. According to AMD, there are a limited number of supported applications but they include Windows Media Player, Windows Photo Viewer, Microsoft PowerPoint and Adobe Acrobat Reader along with Microsoft Photos, Microsoft Music, Microsoft Reader and Kindle within the MS Store.

AMD’s Face Login is exactly what it sounds like: by using several distinguishing points, it can use your face for login access to a computer or certain websites.

The final Elite addition is Screen Mirror which uses the Miracast Wireless Display standard to stream onscreen content (from your desktop to videos to games) to DLNA-equipped devices. AMD also uses GPU acceleration to effectively reduce lag times. Unfortunately though, it won’t stream HDCP or other protected content so using your notebook as a wireless Blu Ray player isn’t possible.
 
Last edited:

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
13,410
Location
Montreal
Test Setups & Methodology

Test Setups & Methodology


For this review, we have prepared a number of different test setups, representing many of the popular platforms at the moment. As much as possible, the test setups feature identical components, memory timings, drivers, etc. Aside from manually selecting memory frequencies and timings, every option in the BIOS was at its default setting.


For all of the benchmarks, appropriate lengths are taken to ensure an equal comparison through methodical setup, installation, and testing. The following outlines our testing methodology:

A) Windows is installed using a full format.

B) Chipset drivers and accessory hardware drivers (audio, network, GPU) are installed.

C)To ensure consistent results, a few tweaks are applied to Windows 7 and the NVIDIA control panel:
  • UAC – Disabled
  • Indexing – Disabled
  • Superfetch – Disabled
  • System Protection/Restore – Disabled
  • Problem & Error Reporting – Disabled
  • Remote Desktop/Assistance - Disabled
  • Windows Security Center Alerts – Disabled
  • Windows Defender – Disabled
  • Screensaver – Disabled
  • Power Plan – High Performance
  • V-Sync – Off

D) Windows updates are then completed installing all available updates

E) All programs are installed and then updated.

F) Benchmarks are each run three to eight times, and unless otherwise stated, the results are then averaged.

G) All processors had their energy saving options / c-states enabled
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
13,410
Location
Montreal
System Benchmarks: AIDA64 / Cinebench r11.5

System Benchmarks


In this section, we will be using a combination of synthetic benchmarks which stress the CPU and system in a number of different domains. Most of these tests are easy to acquire or are completely free to use so anyone reading this article can easily repeat our tests on their own systems.

To vary the results as much as possible, we have chosen a selection of benchmarks which focus upon varied instruction sets (SSE, SSE3, 3DNow!, AVX, etc.) and different internal CPU components like the floating point units and general processing stages.



AIDA64 Extreme Edition


AIDA64 uses a suite of benchmarks to determine general performance and has quickly become one of the de facto standards among end users for component comparisons. While it may include a great many tests, we used it for general CPU testing (CPU ZLib / CPU Hash) and floating point benchmarks (FPU VP8 / FPU SinJulia).


CPU ZLib Benchmark

This integer benchmark measures combined CPU and memory subsystem performance through the public ZLib compression library. CPU ZLib test uses only the basic x86 instructions but is nonetheless a good indicator of general system performance.




CPU Hash Benchmark

This benchmark measures CPU performance using the SHA1 hashing algorithm defined in the Federal Information Processing Standards Publication 180-3. The code behind this benchmark method is written in Assembly. More importantly, it uses MMX, MMX+/SSE, SSE2, SSSE3, AVX instruction sets, allowing for increased performance on supporting processors.




FPU VP8 / SinJulia Benchmarks

AIDA’s FPU VP8 benchmark measures video compression performance using the Google VP8 (WebM) video codec Version 0.9.5 and stresses the floating point unit. The test encodes 1280x720 resolution video frames in 1-pass mode at a bitrate of 8192 kbps with best quality settings. The content of the frames are then generated by the FPU Julia fractal module. The code behind this benchmark method utilizes MMX, SSE2 or SSSE3 instruction set extensions.

Meanwhile, SinJulia measures the extended precision (also known as 80-bit) floating-point performance through the computation of a single frame of a modified "Julia" fractal. The code behind this benchmark method is written in Assembly, and utilizes trigonometric and exponential x87 instructions.



RESULTS: Richland's first results follow a very familiar scene, with relatively poor performance in many of these tests. Granted, there is a measurable performance improvement over its predecessor but that isn't enough to compete head to head against similarly priced Intel processors.


CineBench r11.5 64-bit


The latest benchmark from MAXON, Cinebench R11.5 makes use of all your system's processing power to render a photorealistic 3D scene using various different algorithms to stress all available processor cores. The test scene contains approximately 2,000 objects containing more than 300,000 total polygons and uses sharp and blurred reflections, area lights and shadows, procedural shaders, antialiasing, and much more. This particular benchmarking can measure systems with up to 64 processor threads. The result is given in points (pts). The higher the number, the faster your processor.


RESULTS: Cinebench shows us similar results with the A10-6800K meeting the performance of an upper-echelon Llano part. As we have mentioned again and again, the Piledriver CPU cores jsut aren't that great when used in programs which have been compiled using older methods.
 
Last edited:

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
13,410
Location
Montreal
System Benchmarks: Civ V / PCMark 7

System Benchmarks (pg.2)



Civilization V: Gods & Kings Unit Benchmark


Civilization V includes a number of benchmarks which run on the CPU, GPU or a combination thereof. The Unit Benchmark simulates thousands of units and actions being generated at the same time, stresses multi core CPUs, system memory and GPU We give the non-rendered score below as it is more pertinent to overall CPU performance within the application.


RESULTS: Considering the Richland APUs being tested in this review are merely quad core units, their results here aren't all that bad.



PCMark 7


PCMark 7 is the latest iteration of Futuremark’s system benchmark franchise. It generates an overall score based upon system performance with all components being stressed in one way or another. The result is posted as a generalized score. We also give the Computation Suite score as it isolates the CPU and memory within a single test, without the influence of other components.



RESULTS: PCMark shows some very good numbers for the A10-6800K and A10-6700. They are able to compete with or surpass similarly-priced Intel processors.
 
Last edited:

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
13,410
Location
Montreal
System Benchmarks: 3DMark (CPU) / WPrime

System Benchmarks (pg.3)



3DMark06 CPU


While 3DMark06 may be a slightly older synthetic benchmark, its CPU test still allows for multi threaded performance evaluations within a gaming environment. It effectively removes the CPU from the equation, generating a CPU-centric score.




WPrime


wPrime is a leading multithreaded benchmark for x86 processors that tests your processor performance by calculating square roots with a recursive call of Newton's method for estimating functions, with f(x)=x2-k, where k is the number we're sqrting, until Sgn(f(x)/f'(x)) does not equal that of the previous iteration, starting with an estimation of k/2. It then uses an iterative calling of the estimation method a set amount of times to increase the accuracy of the results. It then confirms that n(k)2=k to ensure the calculation was correct. It repeats this for all numbers from 1 to the requested maximum. This is a highly multi-threaded workload. Below are the scores for the 32M and 1024M benchmarks.



RESULTS: The two benchmarks on this page illustrate Richlands strengths and shortcomings quite well. One one hand, it does well in 3DMark's CPU test which uses a regimen of optimized instruction sets. WPrime meanwhile uses slightly older coding and we see Richland drop back into the lower areas of the charts.
 
Last edited:

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
13,410
Location
Montreal
System Benchmarks: Single Thread Performance

System Benchmarks: Single Thread Performance


Even though most modern applications have the capability to utilize more than one CPU thread, single threaded performance is still a cornerstone of modern CPU IPC improvements. In this section, we take a number of synthetic applications and run them in single thread mode.




RESULTS: Single thread performance is one area where AMD has struggled with their newer architectures and these benchmarks display that shortcoming in stark detail. Once again though, the A10-6000 APUs do improve upon Trinity's results by between 10%.
 
Last edited:

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
13,410
Location
Montreal
Productivity Benchmarks: 7-Zip / MediaCoder

Productivity Benchmarks


In this section, we will avoid generalized synthetic benchmarks and instead concentrate upon CPU performance within real-world applications and standard usage patterns. Every one of the programs included here has functions that many professionals and everyday users utilize in their day to day computing lives.


7-Zip


At face value, 7-Zip is a simple compression/decompresion tool like popular applications like WinZip and WinRAR but it also has numerous additional functions that can allow encryption, decryption and other options. For this test, we are avoiding its built-in benchmark and once again only focus upon real world testing by compressing a 2.6GB folder of various files and adding an AES-256 encryption layer for good measure. The test is timed until it is complete.


RESULTS: Now that we have moved onto real-world applications, it looks like AMD's latest APUs have received a second breath of life. They are able to play up their AES encryption abilities to return reasonably good results, especially when placed next to some Intel processors which lack this feature.



MediaCoder x64


Due to the varying compatibility of certain mobile devices, video transcoding performance has become something of a big deal. Transcoding allows one type of video / audio file to be converted into a different format and it typically takes up a huge amount of system resources. The MediaCoder application brings multi format transcoding to an accessible level with numerous options and acceleration for Intel’s QuickSync and NVIDIA’s CUDA technologies. In addition, its CPU support allows for full multi core utilization. In this test, we use the MediaCoder i-devices edition to convert a 600MB AVCHD file to an iPhone 4S friendly MPEG-4 format.


RESULTS: With the latest iteration of MediaCoder x64, AMD's performance has drastically increased. It may still not be up to the level of some older Sandy Bridge processors but as you can see, certain Ivy Bridge SKUs lack the feature set necessary to deliver adequate performance in this application.
 
Last edited:

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
13,410
Location
Montreal
Productivity Benchmarks: Photoshop CS6 / POV Ray 3.7

Productivity Benchmarks (pg.2)



Adobe Photoshop CS6


For the image editing portion of this section, we use Photoshop CS6 in coordination with a custom benchmark script. This script automates the application of 20 different image manipulation functions to a 120MB PNG image, acting as an excellent test of CPU power and memory bandwidth. For this test, we have disabled GPU acceleration so it won’t play a factor in the areas where it would typically be used. We use Photoshop’s built-in timing feature to provide a result at each test stage.


RESULTS: Adobe Photoshop certainly isn't one of Richland's high points as it trails similar Intel processors by a significant amount. However, once again we are seeing a notable improvement over


POV Ray 3.7 RC6


POV Ray is a complex yet simple to use freeware ray tracing program which has the ability to efficiently use multiple CPU cores in order to speed up rendering output. For this test, we use its built-in benchmark feature which renders multiple passes of a high definition scene. In order to get the most accurate results, the second pass of the first test is logged, resulting in a benchmark score showing the average amount of pixels rendered per second.


RESULTS: Rendering speeds have improved but POV Ray doesn't play to AMD's strengths.
 
Last edited:

Twitter

Top