Logo   Banner   TopRight
TopUnder
Transparent
AMD A6-3650 Llano APU Performance Review
Transparent
Date: Aug 02, 2011
Section:Processors
Author: Marco Chiappetta
Transparent
Introduction and Specifications

A few weeks back, AMD officially launched their Llano-based A-Series Accelerated Processing Units, or APUs, in both desktop and mobile flavors. At the time, we covered the high-end variants of both A-Series offerings; our coverage of the desktop version A8-3850 APU and its companion chipsets is posted here and the mobile A8-3500M is detailed here. If you’re unfamiliar with Llano, we suggest perusing those two articles as they go into much more detail than we will in this piece.

We won’t rehash many of the architectural details, but just for a quick refresher, Llano is AMD’s mainstream, low-power APU that fuses four x86 cores with a DX11-class graphics processor on a single piece of silicon. The current flagship APU in the desktop line-up is the A8-3850, which hums along at 2.9GHz, with 400 active Radeon cores, that operate at 600MHz. The A6-3650 we’ll be showing you here is built using the same piece of silicon, but this lower-end A6-series part clocks in at 2.6GHz, with 320 active Radeon cores, running at 443MHz. Disregarding its operating frequencies and GPU configuration, current A6 and A8 series APUs are essentially identical, as is evidenced by the specifications below...

AMD A6-3650 "Llano" APU
Specifications & Features
Tech / Package 32nm / FM1 905-pin lidded μPGA, 40x40 mm, 1.27 mm pitch
TDP Configurations 65W and 100W configurations
Processor Core “Stars” 32nm HKMG process core (up to 4 cores), 128 KB L1 Cache(64 KB Instruction, 64 KB Data) 1 MB L2/Core, 128-bit FPUs
Memory Up to DDR3 1866
Graphics Core Up to 400 Radeon Cores, DirectX 11 capable, UVD3
Displays Digital Display I/F DP0: Display Port, HDMI, DVI
Digital Display I/F DP1: Display Port, HDMI, DVI
Graphics Features AMD Dual Graphics
Blu-ray 3D
AMD Steady Video
AMD Perfect Picture
DisplayPort 1.1a, HDMI 1.4a
Power Management Multiple low-power states
32-nm process for decreased power consumption
PCIe core power gating
PCIe speed power policy
GPU power gating of Radeon Cores and video decode (UVD3)
AMD Turbo Core technology on select models

Although there will undoubtedly be more Llano-based APUs released in the future, the initial line-up is as follows:

As we’ve mentioned the A8-3850 sits at the top of the stack, followed by the A8-3800. The A6-3650 we’ll be featuring here is the top-of-the-line A6-series part. And the A6-3600 is the entry level chip. These APUs differ in their frequencies and GPU configuration, their TDPs, and in their support for AMD’s Turbo Core technology. The parts with “50” in the model numbers run at full bore, while the standard parts use Turbo Core to temporarily boost operating frequencies to increase performance only when the workload demands it. The use of Turbo Core and their lower clocks result in the non-"50" part's lower average TDP.

We should point out, however, that at this time it is only the A8-3850 and A6-3650 that are available at retail. The other members of the A-Series have yet to ship.

Transparent
Vital Signs and APU Details

Desktop Llano APUs, like the A6-3650 we'll be featuring here use AMD’s FM1, 40mm x 40mm, lidded packaging as opposed to the FS1, 35mm x 35mm, lidless packaging of the mobile variants.


AMD A6-3650 APU In a Hard Candy Shell

While Llano APUs look very similar to current Phenom II and Athlon II processors from the top, thanks to the similar lid and integrated heat spreader installed on the chips, they use a completely different pin configuration and also a different socket. Desktop Llano APUs have 905 pins on their underside whereas socket AM3 Phenom II and Athlon II processors have 939 pins.

 
AMD A6-3650 CPU-Z and GPU-Z Details

The AMD A6-3650 APU has a default CPU clock speed of 2.6GHz (26 x 100MHz). It has four x86 cores each with 128 KB L1 Cache (64KB Instruction, 64KB Data) and 1MB of L2 per core—no L3 cache is present. Current Phenom II processors have similar L1 configurations, but only half the L2, plus a large 6MB L3. The larger L2 should help mitigate the loss of the L3, along with some other improvements to the cores. The 32mn “Stars” derived cores in Llano have improved schedulers and branch predictions units, along with some other low-level tweaks that result in an approximate 6% improvement in IPC performance over current-gen Phenom II processors.

There’s not much to see with regard to the integrated Radeon HD 6530D GPU core in the GPU-Z screenshot that we haven’t already talked about. To reiterate, the GPU core runs at 443MHz and has 320 shader ALUs arranged in 4 SIMDs. The fifth SIMD that’s active in A8-series APUs is disabled on the A6-3650, and other A6-series parts.

Transparent
Test Systems and SiSoft SANDRA

Test System Configuration Notes: When configuring our test systems for this article, we first entered their respective system BIOSes and set each board to its "Optimized" or "High performance Defaults". We then saved the settings, re-entered the BIOS and set the memory frequency to DDR3-1600 and IGP frame buffer size to 512MB (where applicable). The hard drives were then formatted, and Windows 7 Ultimate x64 was installed. When the Windows installation was complete, we updated the OS, and installed the drivers necessary for our components. Auto-Updating and Windows Defender were then disabled and we installed all of our benchmarking software, performed a disk clean-up, defragged the hard drives, and ran the tests.


G.SKILL RipJawsX 8GB DDR-1866 Memory Kit

G.Skill supplied is with a dual-channel, 8GB (4GB x 2) DDR3-1866 memory kit for the purposes of this review. The kit was tested for compatibility with Llano / Lynx and Sandy Bridge, so it was a perfect match for our test systems here. Please note, while this kit is capable of 1866MHz, it ran at 1600MHz by default, in dual-channel mode with Llano, which is how we ran it on all of the test systems.

HotHardware's Test Systems
Intel and AMD - Head To Head

System 1:
AMD A8-3850
(2.9GHz - Quad-Core)
AMD A6-3650
(2.6GHz - Quad-Core)

MSI A75MA-G55
(AMDA75 Chipset)

2x4GB G.SKILL DDR3-1866
(@ 1600MHz)

Radeon HD 6550D IGP
On-Board Ethernet
On-board Audio

WD150 "Raptor" HD
10,000 RPM SATA

Windows 7 x64

System 2:
Intel Core i3-2100T
(2.5GHz - Dual-Core)
Intel Core i3-2120
(3.3GGHz - Dual-Core)
Intel Core i5-2500
(3.3GHz - Quad-Core)

Asus P8Z68-A Pro
(Z68 Express Chipset)

2x4GB G.SKILL DDR3-1866
(@ 1600MHz)

Intel HD Graphics
On-Board Ethernet
On-board Audio

WD150 "Raptor" HD
10,000 RPM SATA

Windows 7 x64

System 3:
AMD Phenom II X4 980
(3.7GHz Quad-Core)

Asus CrossHair V Formula
(AMD 990FX Chipset)

2x4GB G.SKILL DDR3-1866
(@ 1600MHz)

Radeon HD 6570
On-Board Ethernet
On-board Audio

WD150 "Raptor" HD
10,000 RPM SATA

Windows 7 x64

Preliminary Testing with SiSoft SANDRA 2011
Synthetic Benchmarks

We began our testing with SiSoftware's SANDRA 2011, the System ANalyzer, Diagnostic and Reporting Assistant. We ran three of the built-in subsystem tests that partially comprise the SANDRA 2011 suite with AMD's Llano APU (CPU Arithmetic, Multimedia, and Memory Bandwidth). All of the scores reported below were taken with the APU running at its default clock speeds of 2.6GHz with 8GB of DDR3-1600 RAM running in dual-channel mode on the MSI A75MA-G55 motherboard.


AMD A6-3650 APU @2.9GHz
Processor Arithmetic


AMD A6-3650 APU @2.9GHz
Processor Multimedia


AMD A6-3650 APU @2.9GHz
Memory Bandwidth

Our SiSoft SANDRA benchmarks don't reveal any big surprises. Due to its lower frequency, the A6-3650 falls in with scores just below the A8-3850 in the Processor Arithmetic and Multimedia benchmarks. The Memory Bandwidth score, however, is right on par with its more powerful sibling.

Transparent
Futuremark PCMark Vantage
Next up, we ran our test systems through Futuremark’s total-system performance evaluation tool, PCMark Vantage. PCMark Vantage runs through a host of different usage scenarios to simulate different types of workloads including High Definition TV and movie playback and manipulation, gaming, image editing and manipulation, music compression, communications, and productivity.

Preliminary Testing with PCMark Vantage
Synthetic Benchmarks

Most of the sub-tests used to come up with the final scores in each category are multi-threaded as well, so the tests can exploit the additional resources offered by a multi-core CPU.


** The Phenom II X4 980 System Used A Radeon HD 6570 Discrete GPU

Generally speaking, the AMD A6-3650 trails all its competition in the PCMark Vantage test suite. There are a couple of instances where the A6-3650 pulls ahead of the dual-core Core i3-2100T, but the deltas are small and the Intel CPU leads overall.

 

Transparent
Futuremark PCMark 7
Futuremark's PCMark 7 is the latest version of the PCMark suite, recently released this spring. It has updated application performance measurements targeted for a Windows 7 environment. Here's what Futuremark says is incorporated in the base PCMark suite and the Entertainment suite, the two modules we have benchmark scores for you here.

Futuremark PCMark 7
General Application and Multimedia Performance
The PCMark test is a collection of workloads that measure system performance during typical desktop usage. This is the most important test since it returns the official PCMark score for the system
Storage
  • Windows Defender
  • Importing pictures
  • Gaming

Video Playback and transcoding
Graphics

  • DirectX 9

Image manipulation
Web browsing and decrypting

The Entertainment test is a collection of workloads that measure system performance in entertainment scenarios using mostly application workloads. Individual tests include recording, viewing, streaming and transcoding TV shows and movies, importing, organizing and browsing new music and several gaming related workloads. If the target system is not capable of running DirectX 10 workloads then those tests are skipped. At the end of the benchmark run the system is given an Entertainment test score.

The AMD A6-3650 trailed the pack in the two PCMark7 tests we ran as well. Its fewer active Radeon cores and lower operating frequency don't help matters much, as even the dual-core Core i3-2100T finishes ahead of the A6.

Transparent
HD Media Playback and Encoding
On the video decode side of the equation, we viewed an assortment of HD movie trailer clips and monitored CPU utilization. With a relatively powerful chip like the AMD A6-3650, we didn't expect HD video playback to be an issue but regardless, we looked at CPU utilization playing back 1080p Flash video. We also fired up an I am Legend movie trailer .mov file with Windows Media Player, and used PowerDVD to play other various formats, while taking note of thread activity in Windows Task Manager Performance Monitor.


Core i3-2120: 1080p Flash Video


AMD A8-3850 APU: 1080p Flash Video


AMD A6-3650 APU: 1080p Flash Video


All of the video clips we played back worked flawlessly and exhibited very low CPU utilization in the single-digits or low double-digits. Full-screen 1080p Flash video is notorious for high CPU utilization when using older drivers or Flash Players, but the situation looks good today. All of the platforms represented here performed very well in our HD Media playback tests, although AMD does offer some video enhancements that Intel does not.

HD Video Encoding Performance
Testing With and Without Intel Quick Sync

Cyberlink's Media Show Espresso is a video conversion tool that imports various video media files types and converts them to other standard video formats for publication, distribution and / or streaming. In this test, we take a 184MB high definition 1080p AVCHD video clip and compress and convert it to a iPhone 4 H.264-encoded .MP4 file. Times are measured in minutes:seconds with lower times representing faster throughput in the video conversion process.


                     ** The Phenom II X4 980 System Used A Radeon HD 6570 Discrete GPU

The A-Series APUs got clobbered in our MediaShow Espresso video encoding benchmarks. Notice, performance is lower with GPU acceleration enabled, versus encoding on the CPU cores alone.

Transparent
LAME MT Audio Encoding and Cinebench R11.5

In our custom LAME MT MP3 encoding test, we convert a large WAV file to the MP3 format, which is a popular scenario that many end users work with on a day-to-day basis to provide portability and storage of their digital audio content. LAME is an open-source mid to high bit-rate and VBR (variable bit rate) MP3 audio encoder that is used widely around the world in a multitude of third party applications.

LAME MT
Audio Conversion and Encoding

In this test, we created our own 223MB WAV file (a hallucinogenic-induced Grateful Dead jam) and converted it to the MP3 format using the multi-thread capable LAME MT application in single and multi-thread modes. Processing times are recorded below, listed in seconds. Shorter times equate to better performance.


Audio encoding is most certainly not a strong point for the AMD A6-3650. This test only exercises two cores, so half of the APU's resources remain idle, but even against Intel's dual-core processors, it's not pretty for AMD.

Cinebench R11.5
3D Rendering

Cinebench R11.5 is an OpenGL 3D rendering performance test based on Cinema 4D from Maxon. Cinema 4D is a 3D rendering and animation tool suite used by 3D animation houses and producers like Sony Animation and many others. It's very demanding of system processor resources and is an excellent gauge of pure computational throughput. This is a multi-threaded, multi-processor aware benchmark that renders and animates 3D scenes and tracks the length of the entire process. The rate at which each test system was able to render the entire scene is represented in the graph below.


                     ** The Phenom II X4 980 System Used A Radeon HD 6570 Discrete GPU

Cinebench performance is a mixed bag. In terms of CPU performance, the A6-3650 falls in somewhere between the Core i3-2100T and i3-2120. Its GPU score, however, is far better than anything Intel can muster and trails only the A8 or Phenom, when the latter is equipped with a Radeon HD 6500 series discrete GPU.
Transparent
Low-Res Gaming: Crysis and ET:QW

For our next set of tests, we moved on to some in-game benchmarking with Crysis and Enemy Territory: Quake Wars. When testing processors with Crysis or ET:QW, we drop the resolution to 1024x768, and reduce all of the in-game graphical options to their minimum values to isolate CPU and memory performance as much as possible. However, the in-game effects, which control the level of detail for the games' physics engines and particle systems, are left at their maximum values, since these actually do place some load on the CPU rather than GPU.

Low-Resolution Gaming: Crysis and ET: Quake Wars
Taking the GPU out of the Equation (Well, trying to at least)


** The Phenom II X4 980 System Used A Radeon HD 6570 Discrete GPU


** The Phenom II X4 980 System Used A Radeon HD 6570 Discrete GPU

Although these low-resolution tests are meant to be CPU bound, the relatively low-performing Intel HD Graphics engines in the Core i3 and i5 processors hold the CPUs back. The AMD A6-3650 APU clearly outpaced the Intel chips here and trail only the A8 and Phenom system which used discrete graphics.

Transparent
Gaming: APU and Dual Graphics - 3DMark 06 and FarCry 2
FutureMark states that, "3DMark's score is an overall measure of your system’s 3D gaming capabilities, based on comprehensive real-time 3D graphics and processor tests. By comparing your score with those submitted by millions of other gamers you can see how your gaming rig performs, making it easier to choose the most effective upgrades or finding other ways to optimize your system. 3DMark06 has been downloaded more than any other 3D benchmark and the ORB database now contains over 8.5 million 3DMark06 benchmark scores from around the world."

Futuremark 3DMark06
Synthatic DirectX Tests


** The Phenom II X4 980 System Used A Radeon HD 6570 Discrete GPU

3DMark06 tells the whole story of AMD's design philosophy with Llano. The AMD A6-3650-based system offered strong graphics performance with relatively low CPU scores that trailed Intel's Core i3-2120 dual-core, while the Intel-based rigs had relatively good CPU scores, with much lower GPU-related scores. Also note the Core i3-2100T couldn't even finish the HDR/SM3.0 test.

FarCry 2
DirectX Gaming Performance


FarCry 2
Like the original, FarCry 2 is one of the more visually impressive games to be released on the PC to date. Courtesy of the Dunia game engine developed by Ubisoft, FarCry 2's game-play is enhanced by advanced environment physics, destructible terrain, high resolution textures, complex shaders, realistic dynamic lighting, and motion-captured animations. We benchmarked the systems in this article with a fully patched version of FarCry 2, using one of the built-in demo runs recorded in the "Ranch" map.


** The Phenom II X4 980 System Used A Radeon HD 6570 Discrete GPU

The AMD A6-3650 offered nearly double the performance of Intel's integrated graphics in FarCry 2, with the relatively high-quality settings we used. The A8-3850, however, offers much better graphics performance due to its increased number of active shader cores.

Transparent
Gaming: APU and Dual Graphics - Metro 2033 and ET:QW
Next, we turned up the graphics workload a notch or two, with Metro 2033, a 3D graphics stress test if we ever saw one, albeit using more relaxed settings.

Metro 2033
DirectX Gaming Performance


Metro 2033

Metro 2033 is your basic post-apocalyptic first person shooter game with a few rather unconventional twists. Unlike most FPS titles, there is no health meter to measure your level of ailment, but rather you’re left to deal with life, or lack there-of more akin to the real world with blood spatter on your visor and your heart rate and respiration level as indicators. The game is loosely based on a novel by Russian Author Dmitry Glukhovsky. Metro 2003 boasts some of the best 3D visuals on the PC platform currently including a DX11 rendering mode that makes use of advanced depth of field effects and character model tessellation for increased realism. Since Intel's HD Graphics core only supports up to DX10.1 rendering, we tested the game set to medium quality using the game's DX10 rendering mode with 4X Anisotropic Filtering enabled.


** The Phenom II X4 980 System Used A Radeon HD 6570 Discrete GPU

AMD's A6-3650 APU offered about 53% better performance than the Intel processors here, due to the APU's much more powerful integrated GPU.

Enemy Territory: Quake Wars
OpenGL Gaming Performance


Enemy Territory:
Quake Wars

Enemy Territory: Quake Wars is Based on a radically enhanced version of id's Doom 3 engine and viewed by many as Battlefield 2 meets the Strogg, and then some. In fact, we'd venture to say that id took EA's team-based warfare genre up a notch or two. ET: Quake Wars also marks the introduction of John Carmack's "Megatexture" technology that employs large environment and terrain textures that cover vast areas of maps without the need to repeat and tile many smaller textures. The beauty of megatexture technology is that each unit only takes up a maximum of 8MB of frame buffer memory. Add to that HDR-like bloom lighting and leading edge shadowing effects and Enemy Territory: Quake Wars looks great, plays well and works high end graphics cards vigorously. The game was tested using its "High" quality preset with 4x anisotropic filtering.


** The Phenom II X4 980 System Used A Radeon HD 6570 Discrete GPU

We also saw a big advantage for the AMD A6-3650 in the OpenGL-based ET:Quake Wars. In terms of GPU performance, there really is no comparison--AMD beats Intel hands down.
Transparent
Total System Power Consumption

Before bringing this article to a close, we'll take a look at power consumption of the AMD A6-3650 and some competing platforms. Throughout all of our benchmarking and testing, we monitored how much power this new APU was consuming with a power meter, versus other test systems we used for benchmarks in the previous pages. Our goal was to give you an idea as to how much power each configuration used while idling on the desktop and while under a heavy workload Keep in mind, this is total system power consumption being measured at the outlet and not the the individual CPUs or GPUs alone.

Total System Power Consumption
Tested at the Outlet


** The Phenom II X4 980 System Used A Radeon HD 6570 Discrete GPU

Power consumption, especially when idle, is a strong point for AMD's Llano, and the A6-3650 in particular. We saw idle power consumption of only 51W from the 3650 and peak power consumption of 144W (while taxing the CPU and GPU cores), which was right on par with the Intel Core i5-2500 based system.

With some lower-speed / lower-voltage system memory and a bit of CPU undervolting, it would be interesting to see how low a Llano system could go. Mid 40 Watt territory should be doable.
Transparent
Performance Summary and Conclusion

Performance Summary: Due to its lower operating frequencies, for both its CPU and GPU components, and its fewer number of active Radeon cores, the AMD A6-3650 performs a few notches below its A8-series counterpart across our entire battery of benchmarks. In comparison the Intel’s Sandy-Bridge based parts, the A6-3650 is typically outperformed by the dual-core Core i3-2120, let alone the quad-core i7, in terms of x86 performance, but the A6-3650’s graphics performance is far superior. In some graphics heavy workloads, like 3D gaming, the A6-3650’s integrated Radeon HD 6530 outpaced Intel’s integrated HD graphics by over 50%.

 


AMD Llano Die Shot...

 

The AMD A6-3650 APU is currently available for about $119 on-line. For what is a relatively small investment for a low-power quad-core processor, the A6-3650 offers decent x86 performance with a good performing DX-11 class graphic processor. Pair a chip like this with a low-priced A75 chipset-based motherboard and a nice 8GB memory kit and you’ve got yourself an inexpensive foundation for a casual gaming, home theater, or general purpose PC, for under $250. There’s a lot of value in a Llano-based system currently, but looking back at the numbers it’s clear there is significantly more performance to be had with a modest additional investment in something like a Core i3 with a discrete mid-range discrete GPU.

If you’re looking to build a full-featured, low-power, affordable rig with a DX11-compatible Radeon, though, building around AMD's Llano definitely has some merit. The platform’s low power consumption and graphics capabilities make it a good fit for quiet computing or HTPC applications and overall performance should be “good enough” for a large percentage of users.

  • Low Power
  • Affordable
  • Dual-Graphics Support
  • Integrated DX11 GPU Core

  • x86 Performance Well Below Intel
  • More Performance Available For Minimal Additional Investment



Content Property of HotHardware.com