AMD Athlon-II X4-640 Propus Processor Review

In September 2009, AMD unleashed a torrent of personal computing potential by introducing the market to the first sub-$100 quad-core processor, the Athlon-II X4-620. Since then, like it has with the entire Athlon-II series, AMD has been releasing newer versions of their inexpensive quad-core CPUs. In this article, Benchmark Reviews takes a look at AMDs latest release in the Athlon-II X4 line, the Athlon-II X4-640 ADX640WFK42GM AM3 Processor. Built on the Propus die, the Athlon-II X4 series cuts costs by eliminating any L3 cache and limiting the L2 cache to 512KB per core. In looking at the performance of the Athlon-II X4-640, we will compare it to other recent AMD offerings in a similar price range, the Athlon-II X2-260 and the Athlon-II X3-445.

AMD is quickly moving into the leader position in the low to mid-range computing world. Their firm grasp on the sub $200 market is expanding rapidly. The lower end of their processor line, the Athlon-II line, has expanded from just X2 (dual core) CPUs last year to the X3 (triple core) and X4 (quad core) processors like the Athlon-II X4-620 which brought quad-core processing to under $100. AMD is also breaching the high end of gaming PCs with their Phenom-II line. The black edition series of processors, including the Phenom-II X4-965BE which won an editor's choice award here at Benchmark Reviews, can be overclocked to extreme highs, making them great gaming CPUs. They can't beat the raw power of the i7 series, but with the 965BE coming in at only $179, the bang for the buck is appealing to computer enthusiasts everywhere.

The Athlon-II series is built to be a less expensive alternative, while still offering a lot of great features. The chips are designed without any L3 cache at all, allowing for those lower prices. Many computer enthusiasts, myself included, often wait a long time after the purchase of a computer before considering an upgrade. I know many of you reading this are the same way. According to the Steam Hardware Survey for April 2010, almost 17% of gamers (remember, the hardware survey is based on Steam users) are still using single core processors in their systems. Quad core use is up, but still only amounts to 27.5% of users. The bulk of the users use dual core processors with speeds between 2.0 and 2.6GHz. Considering the lowest end of new Athlon-II dual core processors are now at 3.2GHz, this leaves a lot of room for upgrade.

In this article, Benchmark Reviews takes a look at the new Athlon-II X4-640 Quad-Core Processor and compares it's performance to that of the Athlon-II X3-445 and X2-260 processors. All three have similar clock speeds, with the X4-640 at 3.0GHz, the X3-445 at 3.1GHz and the X2-260 at 3.2Ghz. With the release of the Athlon-II X4-640 as AMDs flagship Athlon-II processor, the price of the 2.9GHz Athlon-II X4-635 will fall below $100. The X4-640 was released at an MSRP of $122, with the X3-445 and X2-260 comparison processors at $87 and $76, respectively. Here at Benchmark Reviews, we have recently tested both the X3-445 and the X2-260 and found them to be excellent performers for the entry-level prices they represent. We will be quite surprised if the Athlon-II X4-640 provides us with a different view than these others, as AMD has proven to be quite consistent with the price to performance ratio of its recently released products.

About Advanced Micro Devices, Inc (AMD)

"Advanced Micro Devices (NYSE: AMD) is an innovative technology company dedicated to collaborating with customers and partners to ignite the next generation of computing and graphics solutions at work, home, and play.

Over the course of AMD's three decades in business, silicon and software have become the steel and plastic of the worldwide digital economy. Technology companies have become global pacesetters, making technical advances at a prodigious rate - always driving the industry to deliver more and more, faster and faster.

However, "technology for technology's sake" is not the way we do business at AMD. Our history is marked by a commitment to innovation that's truly useful for customers - putting the real needs of people ahead of technical one-upmanship. AMD founder Jerry Sanders has always maintained that "customers should come first, at every stage of a company's activities."

We believe our company history bears that out."

Athlon-II X4 AM3 Features

AMD Direct Connect Architecture

An award-winning technology designed to reduce bottlenecks that can exist when multiple components compete for access to the processor bus. Competing x86 systems use a single front-side bus (FSB) which must carry memory access, graphics, and I/O traffic. Eliminate the FSB, and you can reduce delays that competing access requests can cause.

45 nm Process Technology with Immersion Lithography

More efficient process technology with cutting-edge lithographic performance puts more transistors in less space.

AMD Wide Floating Point Accelerator

Doubles processor bandwidth from 64- to a full 128-Bit Floating-Point math processing pipeline that can double many of the bandwidth paths that help keep it full.

AMD Digital Media XPress 2.0 Technology

Provides support for SSE, SSE2, SSE3, SSE4a, and MMX instructions for digital media applications and security.

CPU Architectural Features

True Multi-Core Processing

The extensive AMD64 architectural optimizations and features enable thorough integration of multiple cores within the same processor, with each core having its own L1 and L2 caches.

AMD Dedicated Multi-Cache

Each core has its own dedicated L2 cache, which enables simultaneous independent core access to L2 cache, eliminating the need for cores to arbitrate for cache access. This helps reduce latency on L2 cache accesses.

AMD Virtualization (AMD-V) Technology

Silicon feature-set enhancements designed to improve the performance, reliability, and security of both existing and future virtualization environments.

AMD PowerNow! 3.0 Technology

The latest power management technologies that deliver performance on demand when you need it, and power savings when you don't.

HyperTransport 3.0 Technology

Third-generation HyperTransportTM interface improves performance, supporting transfer speeds up to 4.4GT/s.

Simultaneous 32-bit and 64-bit Computing

AMD64 technology enables a breakthrough approach to 64-bit computing that doubles the number of registers in the processor and allows PC users to use today's 32-bit software applications while enabling them to also use the next generation of 64-bit applications.

Cool'n'Quiet 3.0 Technology

Up to eight different performance states help enhance power efficiency. Simplified performance state transitions can reduce latency and the software overhead of performance state changes.

AMD Dynamic Power Management

Each processor core, and the integrated memory controller and HyperTransportTM controller, is powered by dedicated voltage planes.

Integrated Dual-Channel Memory Controller

Directly connects the processor to memory for optimum performance, low latency, and high throughput.

Multi-Point Thermal Control

The next-generation design features multiple on-die thermal sensors with digital interface. Automatic reduction of p-state when temperature exceeds pre-defined limit. Additional memory thermal management interface.

AMD CoolCore Technology

Coarse and fine transistor control that can automatically reduce processor energy consumption by turning off unused parts of the processor.

ADX640WFK42GM Specifications

• Model Number & Core Frequency: X3-445 = 3.1GHz
• TRAY OPN# ADX640WFK42GM
• PIB OPN# ADX640WFGMBOX
• L1 Cache Sizes: 64K of L1 instruction and 64K of L1 data cache per core (384KB total L1 per processor)
• L2 Cache Sizes: 512KB of L2 data cache per core (1.5MB total L2 per processor)
• Memory Controller Type: Integrated 128-bit wide memory controller *
• Memory Controller Speed: 2.0GHz with Dual Dynamic Power Management
• Types of Memory Supported: Support for unregistered DIMMs up to PC2-8500 (DDR2-1066MHz) -AND- PC3-10600 (DDR3-1333MHz)
• HyperTransport 3.0 Specification: One 16-bit/16-bit link @ up to 4.0GHz full duplex (2.0GHz x2)
• Total Processor-to-System Bandwidth: Up to 37.3 GB/s bandwidth [Up to 21.3 GB/s total bandwidth (DDR3-1333) + 16.0GB/s (HT3)]
• Packaging: Socket AM3 938-pin organic micro pin grid array (micro-PGA)
• Fab location: GLOBALFOUNDRIES Fab 1 Module 1
• Process Technology: 45-nanometer DSL SOI (silicon-on-insulator) technology
• Approximate Die Size: 169 mm2
• Approximate Transistor count: ~ 300 million
• Max Temp: 71 Celsius
• Core Voltage: 1.05-1.4V
• Max TDP: 95 Watts
• AMD Codename: "Propus"
• MSRP: $122

  *Note: configurable for dual 64-bit channels for simultaneous read/writes

Closer Look: Athlon-II X4-640

Let's get in close for a look at the new Athlon-II X4-640. The X4-640 is built on the Propus die, a 169mm2 die used for the Athlon-II X4 quad-core Processors. It's built on a 45nm process and has around 300 million transistors. The TDP (thermal design power/point) of the Athlon-II X4-640, as well as all Athlon-II X3 and X4 processors, is 95 Watts. Just to be clear here, the TDP isn't the amount of power the CPU is capable of pulling. It is the maximum amount of power dissipated by the processor under normal circumstances. It isn't the max amount of power that the processor can consume, as overclocking and other circumstances might cause the CPU to draw more power that it's TDP. The 95W TDP on the Athlon-II X4-640 is higher than the 65W TDP for the Athlon-II X2 series, but it is still much lower than the up to 140W TDP found on the high-end Phenom processors.

Now let's get down to the core issue here. The Propus die is really a completely new creation for AMD, rather than a cut off the Deneb die or two Regor dies put together. That being said, the Propus die was based off the Deneb model, but makes some changes in order to get the size and the price right. First off, as we mentioned before, and as you can see below, the Propus chips have no L3 cache. This saves a lot of space and helps cut down on costs. Additionally, while the Regor chips come loaded with 1MB of L2 cache per core, the Propus chips have only 512KB of L2 cache per core for the same total of 2MB of L2 cache for the entire chip. While it may hurt performance somewhat to cut the cash to 512KB per core, you can see that there really isn't any wasted space on the Propus die here. Besides, 512KB of L2 cache per core fits right in with how AMD designed the Phenom-II quad-core processors as well.

Now, AMD doesn't like to waste processors. When we tested the Athlon-II X3-445 triple-core processor, we explained the process that AMD uses to recycle CPUs that may not hold up completely under rigorous testing. During the testing of the Propus chips, one of the cores on a quad-core CPU may fail a test. All the other cores function properly and work just fine. Rather than toss the CPU into the incinerator and watch it burn, the folks over at AMD disable the bad core and send it off to market as a triple-core processor instead of a quad-core. Just because the core failed during AMDs stress testing doesn't necessarily mean that the core won't be fine for normal use. On occasion, some of the Phenom chips will have the L3 cache fail under the testing. When this happens, as long as the L1 and L2 cache pass the testing, AMD will simply disable the L3 cache and ship the processor as an Athlon-II X4 quad-core processor. So there is a slight chance that you might receive an Athlon-II X4 CPU that is built on a Deneb die.

This means that there is a slight possibility that the quad-core processor, such as the Athlon-II X4-640 that we are reviewing in this article, might actually have a disabled L3 cache that is perfectly fine, or at least good enough for our purposes. In their SB700 series southbridge chipsets, AMD built in a feature called the Advanced Clock Calibration (ACC). This was integrated into the BIOS by a number of motherboard manufacturers and it gave the user the ability to try and unlock the disabled cores found on their processors. With the release of the 890GX chipset and the SB850 southbridge, this feature was kept out. However, many motherboard manufacturers have still found a way allow users to attempt to unlock the disabled cores. ASUS has integrated a switch onto some of their 890GX motherboards that allows for core-unlocking, Biostar has added a feature it calls Core UnlocKING into its BIOS, and ASRock has begun releasing 890GX boards with its UCC (Unlock CPU Cores) technology. ASRock has also stated that they have found a way to integrate this technology on to NVIDIA chipsets and will soon be using it there as well.

So, like the enthusiasts that we are here at Benchmark Reviews, we promptly tried to unlock a disabled L3 cache on the Athlon-II X4-640 we received for testing. When I was looking into this possibility, I wondered if there was a way to look at the chip and tell if it was built on a Propus or a Deneb die. I found out that you actually can. If you take a look at the Athlon-II chip, under the model number there will be a series of numbers starting with AA. If the number is AAD** AD, it means it is a Propus die. That means there will be no unlockable L3 cache, as there never was one. If the number is AAC** AC, it means you have a Deneb die. If this is the case, you can potentially unlock the L3 cache. Our sample is AADHC AD, meaning that it is built on the Propus die. No hidden L3 cache will surprise us today.

Using the ASUS M4A785TD-M EVO AM3 motherboard, you can activate the ACC either in the BIOS, or by pressing "4" when the ASUS Logo posts. Even though the chip was definitely a Propus Die, I tried both anyway, and each time I was greeted by a weird flash of green on the screen and an immediate reboot of the system. Upon rebooting, the computer wouldn't post and just turned off. I had to reset the CMOS using the jumper on the motherboard before I could get the system to start up again. So all that really proves is that the Propus Die really doesn't have an L3 Cache.

The Athlon-II X4-640 hasn't changed from its predecessors, the X4-630 and the X4-635. The only real difference is that it ships out at 3.0GHz, 100MHz faster than the X4-635 at 2.9GHz. Also, the X4-640 is made with AMD's revision C3 silicon, making for better yields and hopefully better overclocking ability. Each of the four cores on the Athlon-II X4-640 comes with 64KB of L1 instruction and 64KB of L1 cache as well as 512KB of L2 cache. That makes for a grand total of 384KB of L1 cache and 1.5MB of L2 cache.

The Athlon-II X4-640 becomes the fastest, and thereby the flagship, Athlon quad-core processor at 3.0GHz. AMD has made a pattern of releasing the Athlon-II and Phenom-II series chips in this fashion, increasing the clock speed by 100MHz and the product number by 5. It is likely that we will see an Athlon-II X4-645 usurp the position of the X4-640 in the near future. The X4-640 supports DDR2 RAM up to 1066MHz and DDR3 RAM up to 1333MHz. The X4-640 retails now for $121, and the release of this processor pushes the price of the X4-635 and previous processors down, making the X4-635 AMDs newest sub-$100 quad-core processor.

The stock bus speed for the Athlon-II X4-640 is 200MHz with a 2000MHz HT Link speed. The memory controller for the Athlon-II X4-640 matches the HT Link at 2000Mhz and can be configured as either one 128 bit channel or two 64 bit channels. As I mentioned before, the supported memory is listed as DDR3-1333, but just as previous Athlon-II processors, I'm sure it will easily support the DDR3-1600 memory in our test system. Also like its predecessors, the Athlon-II X4-640 comes with full virtualization support through AMD-V technology. This will be important if you plan on using XP mode in Windows 7.

Testing and Results

Before I begin any benchmarking or overclocking, I thoroughly stress the CPU and memory by running Prime95 on all available cores for 12 hours. If no errors are found, I move on to a gaming stress test. To do this, I use Prime95 again to stress the processor, while running an instance of FurMark's stability test on top of this. If the computer survives this test for 2 hours without lockup or corruption, I consider it to be stable and ready for overclocking. After achieving what I feel is stable overclock, I run to these tests again for certainty. The goal of this stress testing is to ensure the clock speeds and settings are stable before performing any benchmarks. I adopted this method from another writer here at Benchmark Reviews and it seems to do a great job of flushing out what only seem to be stable overclocks.

Once the hardware is prepared, we begin our testing. Each benchmark test program begins after a system restart, and the very first result for every test will be ignored since it often only caches the test. This process proves extremely important in the many gaming benchmarks, as the first run serves to cache maps allowing subsequent tests to perform much better than the first. Each test is completed five times, with the average results displayed in our article.

In this review, we will be comparing the performance of the Athlon-II X4-640 against two other Athlon-II processors, the X2-260 and the X3-445. These processors run at very similar clock speeds to the Athlon-II X4-640. We have overclocked the X2-260 from its original 3.2GHz to 3.8GHz, and we have also overclocked the Athlon-II X3-445 to 4.0GHz for the testing. In addition to these, we have tested an Intel Core i7 920 using almost all of the same test equipment, only changing the motherboard and RAM. All of the tests except for one set of the gaming tests were performed using the NVIDIA GTX 285 video card. This was to ensure that the performance of the CPU wasn't bottlenecked at any time by the GPU. The gaming tests were run using both the GTX 285 and the on-board Radeon HD 4290 video card.

Intel X58 Test Platform

• Processor: 2.66GHz Intel Core i7 920 Bloomfield/Nehalem BX80601920 ($280)
• Motherboard: MSI X58 Pro LGA1366 Intel X58 ATX
• System Memory: Kingston 6GB (3 x 2GB) KVR1333D3K3/6GR DDR3 1333MHz (PC3 10666) (CL7-7-7-20)
• Video: MSI N285GTX-T2D1G-OC NVIDIA GTX 285 1GB
• Disk Drive 1: OCZ Agility Series OCZSSD2-1AGT30G 30GB SSD
• Disk Drive: SEAGATE Barracuda ST31500341AS 1.5TB SATA
• Optical Drive: ASUS DRW-24B1ST DVD Burner
• PSU: Corsair CMPSU-850TX 850W
• Enclosure: NZXT GAMMA
• Operating System: Windows 7 Home Premium x64

AMD Socket AM3 Test System

• Processor: 3.0GHz AMD Athlon-II X4-640 ADX640WFK42GM (MSRP $122)
• Processor: 3.2GHz AMD Athlon-II X2-260 ADX260OCK23GM (MSRP $76)
• Processor: 3.1GHz AMD Athlon-II X3-445 ADX445WFK32GM (MSRP $87)
• Motherboard: ASUS M4A785TD-M EVO 785G Motherboard with 128MB Sideport Memory
• System Memory: 2x2GB Patriot Gamer Series DDR3 (1333MHz@7-7-7-21)
• Video: MSI N285GTX-T2D1G-OC NVIDIA GTX 285 1GB; Onboard ATI Radeon HD 4290
• Disk Drive 1: OCZ Agility Series OCZSSD2-1AGT30G 30GB SSD
• Disk Drive: SEAGATE Barracuda ST31500341AS 1.5TB SATA
• Optical Drive: ASUS DRW-24B1ST DVD Burner
• PSU: Corsair CMPSU-850TX 850W
• Enclosure: NZXT GAMMA
• Operating System: Windows 7 Home Premium x64

Benchmark Applications

• EVEREST Ultimate Edition v5.30.1900 by Lavalys
• Passmark PerformanceTest v6.1
• PCMark Vantage v1.0.2.0 by Futurmark Corporation
• SiSoft Sandra 2010.1.16.26
• Cinebench v11.5
• Resident Evil 5 Benchmark

• Devil May Cry 4 Benchmark

EVEREST Benchmark Tests

Lavalys EVEREST is an industry leading system diagnostics and benchmarking solution for enthusiasts PC users, based on the award-winning EVEREST Technology. During system optimizations and tweaking it provides essential system and overclock information, advanced hardware monitoring and diagnostics capabilities to check the effects of the applied settings. CPU, FPU and memory benchmarks are available to measure the actual system performance and compare it to previous states or other systems. Furthermore, complete software, operating system and security information makes EVEREST a comprehensive system diagnostics tool that offers a total of 100 pages of information about your PC.

All of the benchmarks used in our test bed rely on basic x86 instructions and consume very low system memory while also being aware of HyperThreading, multi-processors, and multi-core processors. While the EVEREST CPU tests really only compare the processor performance more than it measures platforms, it still offers a glimpse into what kind of power each platform possesses.

Queen and Photoworxx tests are synthetic benchmarks that operate the function many times and over-exaggerate by several magnitudes what the real-world performance would be like. The Queen benchmark focuses on the branch prediction capabilities and misprediction penalties of the CPU. It does this by finding possible solutions to the classic queen problem on a chessboard. At the same clock speed theoretically the processor with the shorter pipeline and smaller misprediction penalties will attain higher benchmark scores.

The four cores of the Athlon-II X4-640 really push it above the other two Athlon-II processors here in the Queen tests. At stock speeds the X4-640 improves performance over the triple-core X3-445 by 28%. The X4-640 even ekes out a small gain over the overclocked X3-445 at 4.0GHz. That fourth core really gives it an advantage. When overclocked, the X4-640 takes performance in the Queen tests to a whole new level, even managing to come out ahead of the i7-920 processor by a tiny margin. Not bad for something that costs less than half of the i7-920.

Like the Queen benchmark, the Photoworxx tests for penalties against pipeline architecture. The synthetic Photoworxx benchmark stresses the integer arithmetic and multiplication execution units of the CPU and also the memory subsystem. Due to the fact that this test performs high memory read/write traffic, it cannot effectively scale in situations where more than two processing threads are used. The EVEREST Photoworxx benchmark performs the following tasks on a very large RGB image:

• Fill
• Flip
• Rotate90R (rotate 90 degrees CW)
• Rotate90L (rotate 90 degrees CCW)
• Random (fill the image with random colored pixels)
• RGB2BW (color to black & white conversion)
• Difference
• Crop

Photoworxx tells a story similar story to the Queen tests, and the Athlon-II X4-640 climbs ahead of the Athlon-II processors by a healthy margin. The increased performance is not as dramatic as in the Queen tests, but the quad-core certainly shows it's muscle in these multi-threaded applications. A $35 difference in price isn't just a drop in the bucket, but it seems to be made up for if you are planning on using any programs that utilize those four cores. The Athlon-II X4-640 really shines when overclocked, showing a large increase in performance over any of the other Athlon-II processors. In the Photoworxx suite, the i7-920 handily outdoes the X4-640 with its lack of L3 cache.

The Zip Library test measures combined CPU and memory subsystem performance through the public ZLib compression library. ZLib is designed as a free lossless data compression library for use on virtually any computer hardware and operating system. The ZLib data format is itself portable across platforms and has a footprint independent of input data that can be reduced at some cost in compression.

As far as file compression performance is concerned, the X4-640 quad core once again shows the power of a fourth core by increasing stock performance by 28% over the Athlon-II X3-445. The huge performance gain speaks for itself and begins to further justify the extra expenditure. When overclocked, the X4-640 once again squeezes out a narrow win over the i7-920. While this is not the trend, it looks like file compression really benefits from four cores running at high clock speeds. If you are spending a lot time compressing files as a website administrator or because you email a lot for work, you will definitely benefit from the four cores of the Athlon-II X4-640, especially if you boost the performance by overclocking it a little.

The AES integer benchmark measures CPU performance using AES data encryption. It utilizes Vincent Rijmen, Antoon Bosselaers and Paulo Barreto's public domain C code in ECB mode and consumes 48 MB of memory.

While less noticeable, there is still a very significant increase in performance from the Athlon-II X4-640 processor. As encryption becomes more and more popular in the personal computing environment with increased online interaction, AES performance will become that much more important for the individual user and consumer. If encryption is a concern for you, the Athlon-II X4-640 provides good quality performance at an entry level price.

Once again, the floating point tests show the same trend. Performance using the Athlon-II X4-640 is enhanced throughout all of the tests. The 128-bit encrypted Sin Julia tests show a 29% increase over the triple-core at stock speeds and a 28% increase when both processors are overclocked. The Mandel test operates at 64-bit encryption and also realizes impressive gains by using the Athlon-II X4-640 processor, to the point where an overclocked X4-640 outpaces a stock i7-920 by a little bit. In the 32-bit Julia tests, stock performance was increased by 29% and overclocked performance was increased by 30%. The results of the Everest tests all show us right off the bat that the ADX6400WFK23GM provides a very viable option for users looking to upgrade older computer systems. The price difference of $35 begins to look more reasonable when we see how well the X4-640 performs in the Everest tests against the less expensive and core-deficient Athlon-II line-up. Let's move on and see if further testing continues to show this trend.

Passmark Performance Test

PassMark Performance Test is a PC hardware benchmark utility that allows a user to quickly assess the performance of their computer and compare it to a number of standard 'baseline' computer systems. The Passmark Performance Test CPU tests all benchmark the mathematical operations, compression, encryption, SSE, and 3DNow! instructions of modern processors.

In our tests there were several areas of concentration for each benchmark, which are combined into one compound score. This score is referred to as the CPU Mark, and is a composite of the following tests: Integer Math, Floating Point Math, Find Prime Numbers, SSE/3DNow!, Compression, Encryption, Image Rotation, and String Sorting. For this review, we've also decided to run the memory benchmark, which results in a composite score based on the following tests: small block allocation, cached read, uncached read, write performance, and large block allocation.

The Passmark performance tests show a slightly different result than the Everest tests. Passmark individually tests each core that is present, and we can see that from our results. The slower the clock speed of the Athlon-II processor in question, the lower the score. That is why the Athlon-II X4-640 scored lower than both other Athlon-II processors at stock speeds. Overclocked, the X4-640 is able to retake the top position as far as the Athlon processors are concerned.

The memory scores are really here just for reference, as the memory tested was the same throughout all the AMD tests. The i7 system had different RAM, of course. What is interesting to note here is that the memory bandwidth tests fall exactly in line with the clock speeds of the processors and their HT link speeds. At the standard 2000MHz HT link, the slower clock speeds decrease bandwidth and the X4-640 falls to the back of the pack. When overclocked, with the HT Link increased alongside the bus speed, the bandwidth changes to reflect it. Even though both the X2-260 and the X4-640 are both running overclocked at 3.8GHz, the reason the X4-640 pulls ahead is because the bus speed was increased more in order to reach that overclock. This is due to the fact that the stock multiplier for the X4-640 is at x15 while it is at x16 for the X2-260.

PCMark Vantage Benchmark Tests

PCMark Vantage is an objective hardware performance benchmark tool for PCs running 32- and 64-bit versions of Microsoft Windows Vista or Windows 7. PCMark Vantage is well suited for benchmarking any type of Microsoft Windows Vista/7 PC: from multimedia home entertainment systems and laptops, to dedicated workstations and high-end gaming rigs. Benchmark Reviews has decided to use a few select tests from the suite to demonstrate simulated real-world processor usage in this article. Our tests were conducted on 64-bit Windows 7, with results displayed in the chart below.

TV and Movies Suite

• TV and Movies 1 (CPU=50%, RAM=2%, GPU=45%, SSD=3%)
  • Two simultaneous threads
  • Video transcoding: HD DVD to media server archive
  • Video playback: HD DVD w/ additional lower bitrate HD content from HDD, as downloaded from net
• TV and Movies 2 (CPU=50%, RAM=2%, GPU=45%, SSD=3%)
  • Two simultaneous threads
  • Video transcoding: HD DVD to media server archive
  • Video playback, HD MPEG-2: 19.39 Mbps terrestrial HDTV playback
• TV and Movies 3 (SSD=100%)
  • HDD Media Center
• TV and Movies 4 (CPU=50%, RAM=2%, GPU=45%, SSD=3%)
  • Video transcoding: media server archive to portable device
  • Video playback, HD MPEG-2: 48 Mbps Blu-ray playback

Gaming Suite*

• Gaming 1 (CPU=30%, GPU=70%)
  • GPU game test
• Gaming 2 (SSD=100%)
  • HDD: game HDD
• Gaming 3 (CPU=75%, RAM=5%, SSD=20%)
  • Two simultaneous threads
  • CPU game test
  • Data decompression: level loading
• Gaming 4 (CPU=42%, RAM=1%, GPU=24%, SSD=33%)
  • Three simultaneous threads
  • GPU game test
  • CPU game test
  • HDD: game HDD

Music Suite

• Music 1 (CPU=50%, RAM=3%, GPU=13%, SSD=34%)
  • Three simultaneous threads
  • Web page rendering - w/ music shop content
  • Audio transcoding: WAV -> WMA lossless
  • HDD: Adding music to Windows Media Player
• Music 2 (CPU=100%)
  • Audio transcoding: WAV -> WMA lossless
• Music 3 (CPU=100%)
  • Audio transcoding: MP3 -> WMA
• Music 4 (CPU=50%, SSD=50%)
  • Two simultaneous threads
  • Audio transcoding: WMA -> WMA
  • HDD: Adding music to Windows Media Player

* EDITOR'S NOTE: Hopefully our readers will carefully consider how relative PCMark Vantage is as "real-world" benchmark, since many of the tests rely on unrelated hardware components. For example, per the FutureMark PCMark Vantage White Paper document, Gaming test #2 weighs the storage device for 100% of the test score. In fact, according to PCMark Vantage the video card only impacts 23% of the total gaming score, but the CPU represents 37% of the final score. As our tests in this article (and many others) has already proven, gaming performance has a lot more to do with the GPU than the CPU, and especially more than the hard drive or SSD (which is worth 38% of the final gaming performance score).

The PCMark Vantage test further solidify the trend of increased performance by the Athlon-II X4-640. As we look at the TV and Movies scores it is important to take into consideration that the testing is only 50% reliant on the CPU in 3 out of 4 of the tests in the suite. In the third test, the performance is reliant solely on the SSD. By using the same SSD and GPU in all of the test systems, we can hopefully isolate the differences in the CPU for the TV and Movies suite. As you can see, the Athlon-II X4-640 improves upon the performance of the X3-445 processor, with stock performance gaining about 10% from the triple-core performance. The gains are increased even more when the Athlon-II X4-640 is overclocked.

The gaming suite shows also relies heavily on the SSD and in some tests the GPU to measure the performance. RAM is also a slight consideration, which is why we ensured that all of our RAM timings were the same across the board. As we expected, the Athlon-II X4-640 climbs to reach the highest scores among the Athlon processors, above the Athlon-II X3-445, but only by a small margin. The gains are extended slightly when the two processors are overclocked. In the gaming test it appears that the number of cores plays a decent role, since the triple and quad-core processors easily outpace the dual-core processor. The cache may also play an important role, as we see the i7-920 far outreaching the performance of the Athlon-II processor.

The PCMark music suite tests not only audio transcoding, but also webpage rendering using popular web-based music store formats. These tests rely heavily on the CPU and it appears that clock speed takes the advantage here. When the Athlon-II X2-260 is overclocked, it actually exceeds the score of the stock Athlon-II X4-640. However, when overclocked to 3.8GHz itself, the Athlon-II X4-640 regains a lead over the X2-260, but can't quite keep up with the 4.0GHz of the overclocked X3-445. In rare form, the overclocked Athlon-II X4-640 and the X3-445 both actually outperform the i7-920, adding credit to the thought that clock speed plays more of a role than the amount of cores or the lack of L3 cache in the music suite. When looking at a single core clock speed, we are pitting a 3.8GHz core against a 2.66GHz core and the 3.8GHz is barely winning.

SiSoftware Sandra

SiSoftware Sandra (the System ANalyser, Diagnostic and Reporting Assistant) is an information & diagnostic utility. It should provide most of the information (including undocumented) you need to know about your hardware, software and other devices whether hardware or software.

It works along the lines of other Windows utilities, however it tries to go beyond them and show you more of what's really going on. Giving the user the ability to draw comparisons at both a high and low-level. You can get information about the CPU, chipset, video adapter, ports, printers, sound card, memory, network, Windows internals, AGP, PCI, PCI-X, PCIe (PCI Express), database, USB, USB2, 1394/Firewire, etc.

The SANDRA DhryStone and Whetstone tests are CPU tests that run completely within the CPU + cache memory itself. These tests are perfect for seeing general efficiency per processing core. Dhrystone is basically a suite of arithmetic and string manipulating programs and is an older CPU tests. Even so, it remains a simple and accurate way to show RAW CPU processing performance. The whetstone benchmark primarily measures floating-point arithmetic performance.

The SiSoft Sandra Arithmetic Suite further exemplifies the performance gain from an extra core. At stock speeds, the Athlon-II X4-640 shines above the X3-445 by 27% and extends the gains to 29% when the two processors are overclocked. Clearly the advantage is held by the Athlon-II X4-640 processor, and I would say a nearly 30% performance increase is well worth the extra $35. In the Whetstone tests, the Athlon-II X3-445 secures nearly the same differences as in the Dhrystone tests, with 28% at clock speeds and 29% when overclocked. Not only does this show superiority for the Athlon-II X4-640, but it helps lend credibility to the testing.

Cinebench R11.5 Benchmarks

Maxon Cinebench is a real-world test suite that assesses the computer's performance capabilities. Cinebench is based on Maxon's award-winning animation software, Cinema 4D, which is used extensively by studios and production houses worldwide for 3D content creation. Maxon software has been used in blockbuster movies such as Spider-Man, Star Wars, The Chronicles of Narnia and many more. Cinebench Release 11.5 includes the ability to more accurately test the industry's latest hardware, including systems with up to 64 processor threads and the testing environment better reflects the expectations of today's production demands. A more streamlined interface makes testing systems and reading results incredibly straightforward.

The Cinebench R11.5 test scenario uses all of a system's processing power to render a photorealistic 3D scene, "No Keyframes" the viral animation by AixSponza. This scene makes use of various algorithms to stress all available processor cores. The OpenGL graphics card testing procedure uses a complex 3D scene depicting a car chase with which the performance of your graphics card in OpenGL mode is measured. During the benchmark tests the graphics card is evaluated by way of displaying an intricate scene that includes complex geometry, high-resolution textures, and a variety of effects to evaluate the performance across a variety of real-world scenarios.

Its readily apparent that the single core tests are right in line with the actual processer clock speed. The scores consistently reflect that throughout all of the Athlon-II processors. The i7-920, at 2.66GHz, doesn't follow that same trend, but we are focusing on the Athlon-II X4-640 and the other AMD processors here. When the Cinebench suite focuses on multi-core functionality, a good representation of their differences becomes apparent. At stock speeds, the Cinebench performance increase from the $35 more expensive X4-640 processor is right in line with the previous PCMark Vantage tests, at 29%. When overclocked, the two processors again follow the trend set by PCMark with the X4-640 outperforming the X3-445 by 30%. Again, these consistent results lend credit to the testing procedures used by both test suites.

Video Game Benchmarks

PC-based video games can depend heavily on the CPU if the attached GPU (Graphics Processing Unit) is less powerful, or the graphics settings are configured so low that they create no strain on the video card and must rely purely on system processing speed; a phenomenon known as CPU-dependence. The opposite is true when the video game has a powerful video card installed, and can handle all graphical demands without receiving assistance from the CPU. Benchmark Reviews has proven consistently that, with a high end GPU in use, frame rates are not often noticeably impacted by changes in processor or RAM. Since it is unlikely that someone spending enough money to buy a top-of-the-line graphics card would settle for the Athlon-II X4-640 as their gaming processor, we have decided to use Radeon HD 4200 with 128MB of DDR3 SidePort Memory, the ATI on-board video solutions provided with the ASUS M4A785TD-M EVO motherboard for these gaming tests.

It is important to realize, however, that the Athlon-II X2, X3 and X4 processors can all be used to play modern games, and even at very high settings. The only way to do this, however, is through the purchase of a high end graphics card. While the most likely purchasers of the Athlon-II X3-445 will not be high end gamers, it is feasible that some people might want to play the latest games, but have a budget that limits them to either a high end processor, or a high end graphics card. For this reason, we have included here the results of the gaming tests with the MSI NVIDIA GTX 285 GPU. This card runs about $370. There are other cards, i.e. some of the Radeon HD 5xxx cards, that will provide enough power to play these games even with an entry level processor. In these cases, the GPU is doing most of the work for the game, and the processor is much less involved.

Built upon an advanced version of Capcom's proprietary MT Framework game engine to deliver DirectX 10 graphic detail, Resident Evil 5 offers gamers non-stop action similar to Devil May Cry 4, Lost Planet, and Dead Rising. The MT Framework is an exclusive seventh generation game engine built to be used with games developed for the PlayStation 3 and Xbox 360, and PC ports. MT stands for "Multi-Thread", "Meta Tools" and "Multi-Target". Games using the MT Framework are originally developed on the PC and then ported to the other two console platforms.

On the PC version of Resident Evil 5, both DirectX 9 and DirectX 10 modes are available for Microsoft Windows XP and Vista Operating Systems. Microsoft Windows 7 will play Resident Evil with backwards compatible Direct3D APIs. Resident Evil 5 is branded with the NVIDIA The Way It's Meant to be Played (TWIMTBP) logo, and receives NVIDIA GeForce 3D Vision functionality enhancements. NVIDIA and Capcom offer the Resident Evil 5 benchmark demo for free download from their website, and Benchmark Reviews encourages visitors to compare their own results to ours.

Regardless of processing power, with the on-board graphics, none of these CPUs can give us playable frame rates in Resident Evil 5. The tests were completed with the lowest settings possible at the most likely low-end gaming resolution of 1280x1024. In order to play the game, you will need to invest in a discrete GPU. The Radeon HD4200 just doesn't have what it takes to play this game. The game seems to favor clock speed over cores when dealing with the X2 and X3 processors, but the X4-640, with its fourth core, seems to pull a few more frames out of it.

The results from the Devil May Cry 4 tests are consistent with the Resident Evil 5 benchmark scores. None of the processors can muster enough frame rates to play the game without trouble. We are not surprised by the results of the gaming tests. While the frame rates to fluctuate generally based on what seems to be the clock speed of the CPU being used, none of the processors can make up for the lack of power in the Radeon HD4200. Even if you were to use the fastest Phenom-II chips available, it is doubtful that frame rates would reach decent playable levels. Playable levels are generally agreed to be over 30 FPS. So in Scene 4, some powerful CPUs might make it. Gameplay overall, however, would be less than viable with an onboard solution. Again, if you were using the fastest Phenom-II CPUs, you wouldn't likely be using the on-board graphics. These graphics are meant for media playback in an HTPC or for routine everyday uses, not for gaming.

If you have kept up with Benchmark Reviews articles in the past, it will come as no surprise to you that we have continually proven that, when using a high end graphics card, CPU speed and RAM speed and timings have very little to do with a noticeable increase in Frames Per Second, even at the highest settings. In order to show that the Athlon-II X4-640 actually IS a viable processor for use in gaming, when paired with a high end video card, we have included benchmark testing of the same two games using the NVIDIA GTX 285 video card. This card will set you back close to $370, so it is unlikely that it will often be paired with an Athlon-II processor. However, if your budget allows for only one high end item, and you want to play the latest videos games, these tests clearly show that you can do so with a good video card.

Like I said before, playable frame rates are at least above 30 FPS. I have used the V-Sync here to prove a point. At 60 FPS, you are seeing just about as much of the game as you can. Undoubtedly, if I were to take off the V-Sync, the i7-920 would provide a higher margin of success than the Athlon-II processors, with their lack of L3 cache and hyperthreading. Even so, the Athlon-II X4-640 can easily get up over 100 FPS when paired with a high-end video card. Again, the point is that the games are much less dependent on the CPU than some may think.

We can see that when using the much better GPU, the games are easily playable with the Athlon-II X4-640 and even the X2 processors. Due to the 60Hz refresh rate at 1920X1080 on the Acer X233H used for testing, and the V-Sync feature, the results are capped at 60 FPS. I did bump the refresh rate up to 75Hz at a much lower display setting, and the results were similarly maxed at right near 75 FPS. The real purpose of these tests is just to show how little difference the processor really makes when it comes to playing high end games. The GPU is the real factor here.

AMD Athlon-II Overclocking

Now let's get into overclocking. Since the Athlon-II X4-640 is not a black edition CPU, I couldn't increase the multiplier to overclock it. Still, while the multiplier is stopped out at x15, it is important to note that it can still be decreased down as far as x4. Because of this, it is possible to decrease the multiplier and increase the bus speed and voltage and still maintain and stable overclock. To start off, however, I wanted to see how far I could get with the overclock of the Athlon-II X3-445 by just increasing the bus speed. Using the ASUS M4A785TD-M EVO, I started at 225MHz on the bus speed and increased by 5 until I couldn't boot into Windows. I was able to reach 250MHz before the Auto voltage setting didn't give me enough power to boot the system, but the overclock wasn't stable enough to withstand stressing. After increasing the voltage to 1.55V, I was able to get to 255MHz. This time it withstood the stressing like a champ, giving us our most stable overclock at 3.8GHz.

With the Athlon-II X4-640 at 3.8GHz, we achieved a 21% increase in core speed using just air to cool it. I tried getting even higher by dropping the multiplier down as far as x13.5 and increasing the bus speed above 300MHz, but I wasn't ever able to achieve a stable setting higher than the 4.1GHz I had already reached. All of my overclocking was done with air cooling only using the Scythe Mugen II CPU Cooler.

Before ending up at the 255MHz bus speed magic number, I was able to make it into windows and begin stressing the CPU at even higher speeds. Originally, I thought I had found a stable overclock as high as 3.99GHz with the bus speed set to 266MHz. Running the stressing procedure I mentioned above usually weeds out any unstable overclocks and I am able to perform testing after that. Even so, with the bus speed at 266MHz and the voltage at 1.54V, as you can see in the screenshot below, I got through all the stressing and was ready to start testing.

When I began the testing with the X4-640 at 3.99GHz, everything went fine. It wasn't until about halfway through the PCMark Vantage tests that I started having problems. The computer crashed at that point and restarted. I was a little confused by that, so I ran the stress testing procedures again. Within an hour, the overclocked failed stressing. I pushed up the voltage as high as 1.60V to see if that would help the CPU withstand stressing at the same high bus speed. It didn't work. At 3.99GHz, the processor could not again hold up to stressing. I moved the bus speed down 1MHz at a time until I ended up back at 255MHz and the processor once again withstood the stressing.

Even when using two different processors of the same make and model, or two of the same motherboard, they will likely result in different abilities as far as overclocking is concerned. While I was able to reach an overclock of 3.8GHz with air cooling, I highly encourage anyone to try out their own methods, especially with an entry-level quad-core CPU such as the X4-640. The mechanics of overclocking have more to do with the amount of voltage I can safely send to the CPU or through the northbridge. With the testing motherboard being an entry-level product itself, it doesn't have a MOSFET heat sinks or even that large of a heatsink on the northbridge. Overall, I was not disappointed by the overclock I was able to achieve at 3.8GHz. I hope you will all take the time to (carefully) test your own hardware to see how far you can safely stretch it. It certainly is a lot of fun.

AMD Athlon-II X4 Final Thoughts

The launch of the Athlon II brand marked a complete 45nm refresh of AMD's mainstream and enthusiast processors. On the high end, we have the Phenom-II X4's and X3's. Below that comes the recently launched Phenom-II X2 series, although some of these may actually be faster than the low end X3's in single-threaded applications. That leaves the Athlon-II's one notch lower, sitting directly in the mainstream market segment. These budget processors bring quad-core processing to under $100. The latest releases, just 100MHz speed bumps on the older versions of the same processors, will help fill in the gaps in the AMD processor line, providing a perfect fit for nearly any user.

Our tests have been focused on looking at the Athlon-II X4-640 as a potential purchase over the Athlon-II X3 models that sit at just $35 less expensive. The quad-core processor market is quickly building its consumer base, and many games and programs are taking notice and designing their applications to take advantage of multiple cores. Even so, dual-core usage is still the highest on the Steam hardware survey. Throughout our testing, we have proved very consistently that the Athlon-II X4-640 can outperform the X3-445, especially in multi-core applications. While this doesn't come as a huge surprise, it is very interesting that the improvement from X2-260 to the X3-445 was consistently near 40% to 50%. The improvement from the X3-445 to the X4-640 was still impressive, but only about 30%. With the price difference from X2 to X3 at only $11 and the difference from X3 to X4 at $35, it seems that the upgrade to the quad-core might be less valuable then the upgrade to the triple-core.

Of course, for users of entry-level processors, there isn't as large a need for multi-core usage as there is amongst enthusiast and high-end gamers who are going to opt for a quad-core processor anyway. Only the latest, most resource intensive games even utilize more than one core at this point. This is rapidly changing, and most new games coming out take advantage of extra cores, but it is still a new trend. As the trend continues, I think we will see more entry-level users opting for quad-core processors over low-end dual-core processors with the prices being as enticing as Athlon-II X4-640 at $122.

With the Athlon-II X4-640 priced as it is, it offers a great option for upgraders or individuals on a budget looking to get a new computer for basic home uses. AMD suggests using the newly released Athlon-II processors with the 890GX chipset. The new chipset really isn't all that different from the 790G chipset, in use now for quite a while. The main difference is the addition of built-in compatibility with SATA 6Gb/s. Given the extremely limited amount of hardware available that takes advantage of the new SATA standard, the compelling reasons to move to the 890GX chipset at this point are few. Look out for our upcoming reviews of new 890GX motherboards for more information on this.

That being said, I tested the Athlon-II X4-640 in a 785G motherboard. I did this because I had more luck overclocking and better overall test results with the 785G over the 890GX. While this may be limited to only the two motherboards I used, and other 890GX motherboards might perform better, I think I can safely conclude that the benefits of upgrading to the 890GX chipset right now are not necessary. I highly recommend using the Athlon-II X4-640 in the 785G chipset. It works particularly well in the ASUS M4A785TD-M EVO motherboard. I would definitely look for a motherboard that offers the 128MB of DDR3 sideport memory if you are planning on using the onboard GPU. The Athlon-II X4-640 processor offers an excellent entry point for any budget minded user.

ADX640WFGMBOX Conclusion

The Athlon-II X4-640 performed very well against the lower end Athlon-II X3 and X2 processors that we tested it against. When overclocked to 3.8GHz it even got close and overcame to the performance of the i7 920, mostly in single threaded or low-end environments. As I mentioned during the testing, that is pitting a 3.8GHz quad-core processor against a 2.66GHz quad-core processor, and the slower still generally comes out ahead, but that really is to be expected from a CPU costing more than double the amount of the Athlon-II X4-640. What it does show us, however, is that the Athlon-II X4-640 processor can perform at the level of much higher priced Intel processors. We proved months ago that the Athlon-II X2s can outperform similarly priced Intel Dual-Core processors, and the trend continues. AMD has a vice grip on value for the low-end, entry-level processor market, and with its back to back releases of 100MHz increases, it continues to solidify that position.

The Athlon-II X4-640 withstood rigorous testing like a champ. I pushed the processor past the limits multiple times while trying to discover the best overclocking scenario in two different motherboards with two different chipsets. After literally days of stress testing to ensure stability, the processor still ran strong overclocked to 3.8GHz. Anyone who has ever desired to experiment with overclocking now has an entire line of very inexpensive processors to try out without spending a lot of money. All this stability is a testament to the high quality and increasingly high yield of Athlon-II processors.

AMD has made it possible for their Athlon-II line to run in AM2+ motherboards as well as AM3 motherboards. This allows the door to be wide open for the choice of motherboard with which the Athlon-II X3-445 will function. The motherboard we used to test the Athlon-II X4-640 is the ASUS M4A785TD-M EVO AM3 motherboard using the 785G chipset. We also tested the processor in the Biostar TA890GX HD AM3 motherboard using the 890GX Chipset, but the results were not included in this article. The Athlon-II X4-640 functions extraordinarily well with either of these motherboards. Both boards maintain the entry-level expectations associated with the Athlon-II line and are under $100. I would expect that the X4-640 be used with a lower priced motherboard such as this in order to maintain the budget appeal of the processor. Even so, we saw conclusive evidence that the Athlon-II X4-640 paired with a budget level motherboard can still be used to play high end games if a high end GPU is used. Entry level users need not necessarily wait until they can afford all the of the highest end equipment to play the latest games. The Ahtlon-II X4-640 will play them if used in tandem with a high enough performing video card.

The Athlon-II X4-640, with its 45nm process and somewhat low TDP of 95W, is an excellent overclocker. This is common amongst Athlon-II processors, especially the higher yield ones that being released now at faster clock speeds and with newer revisions of silicon being used by AMD. The X4-640 is not a black edition processor, so we were working with a locked multiplier of x15. We could lower the multiplier through the BIOS, but not increase it. Even so, just using the bus speed and voltage, we were able to get the Athlon-II X4-640 to 3.8GHz cooled only with air using the Scythe Mugen II CPU cooler. (For more information on this cooler, check out Benchmark Review's 1st Quarter 2010 CPU Cooler Performance Review.) This represents a 21% increase in the clock speed of the X4-640. We were able to reach a semi-stable 3.99GHz as well, but it eventually failed stressing. With some more patience and time, this processor might be able to make it stably to close to 4.0GHz. A quad-core processor running at 4.0GHz and priced at only $121 is a force to be reckoned with. In fact, the entire Athlon-II line has proven to be very good at overclocking and represents a great starting point for any would-be enthusiast.

AMD's ADX640WFGMBOX retail kit is available at NewEgg for $120.99, which means the Athlon-II X4-640 is priced to sell. While enthusiasts and hard-core gamers will find that the X4-640 doesn't offer the L3 cache that their high end games and programs need, any user working with a computer that is over a couple of years old will find that the Athlon-II X4-640 offers an amazing bang for the buck. Even without the L3 cache offered by higher end, and therefore higher priced, processors the Athlon-II X4-640 still provides great performance for the everyday user. Even most games will be easily playable with the addition of a high-end video card. Expected to be priced at just $35 more than its Athlon-II X3-445, the X4-640 has shown nearly a 30% increase consistently over the X3-445.

Pros:

Cons:

- Not as high performance increase as seen from the X2-260 to the X3-445

Ratings:

• Performance: 9.00
• Construction: 9.50
• Functionality: 9.50
• Overclock: 9.25
• Value: 9.00

Final Score: 9.25 out of 10.

Excellence Achievement: Benchmark Reviews Golden Tachometer Award.

Questions? Comments? Benchmark Reviews really wants your feedback. We invite you to leave your remarks in our Discussion Forum.

Related Articles:

• Seagate Barracuda 3TB Hard Drive ST3000DM001
• AMD Phenom-II X6-1090T Black Edition Processor
• Rosewill Ranger Mid-Tower Computer Case
• Intel BOXD945GCLF2D Atom 330 mITX Motherboard
• NZXT Cryo LX Aluminum Notebook Cooler
• D-Link DNS-323 SATA RAID Gigabit NAS
• Thermalright Silver Arrow CPU Cooler
• Antec Mini P180 mATX Computer Case
• Pivos XIOS DS Media Player PTGMCXDALU
• IOCELL NetDISK 351UNE Network Storage Device