Home AMD’s Athlon X2 BE-2350 processor
Reviews

AMD’s Athlon X2 BE-2350 processor

Scott Wasson
Disclosure
Disclosure
In our content, we occasionally include affiliate links. Should you click on these links, we may earn a commission, though this incurs no additional cost to you. Your use of this website signifies your acceptance of our terms and conditions as well as our privacy policy.

AS I WRITE THESE words, I’m comfortably reclined in an overstuffed chair in my living room, laptop perched on my lap, sipping on a homemade cafe latte. Sunlight streams in through a window across the room, and every so often, I can hear the shuffle caused by my oldest child turning the page in the book he’s reading. All is well, or so it would seem. But in the background, just above the sound of the air conditioning system forcing air through the vents, I can hear it: the ever-so-slight but unmistakable whir of the fans spinning in my home theater PC, piercing the silence like a faint whisper.

I’m sure you’re aghast. Why, you ask, should a computing device be audible in one’s living room? Good question. The short answer, in my case, is that our HTPC is based on an Athlon 64 X2 4200+ processor that requires a little more relief than passive cooling or inaudibly low fan speeds will allow.

To help others—especially the children, who will think of them?—in living rooms everywhere avoid this tragic fate, AMD has just introduced a new CPU aimed at home theater PCs, small form factor systems, and small-footprint corporate desktops. Dubbed the Athlon X2 BE-2350, this chip has a confusing new alphanumeric amalgamation attached to its name, and what could be cooler than that? Perhaps a 45W thermal/power rating for the processor. The BE-2350 sips power like a mobile CPU but carries a wallet-friendly price tag of under 100 bucks, which might make it an attractive prospect for your next system build.

Especially if you care about the kids.

If not, your cold, calloused heart may be warmed by the news that our BE-2350 sample also overclocks like a mofo. Read on to see how we used the BE-2350 as a low-power processor and then abused it as a high-power one, to the delight of all involved.

A tale of low-power desktop chips
The story of the Athlon X2 BE-2350 begins with another low-power processor from AMD, a version of the Athlon 64 X2 3800+ with a TDP rating of only 35W, which we reviewed last summer. This processor, known fully as the Athlon 64 X2 3800+ Energy Efficient Small Form Factor, fared well in our power consumption testing, besting anything Intel had to offer at the time and promising good things for builders of quiet PCs everywhere. In fact, at 35 watts, this CPU was essentially the exact same thing as a Turion 64 X2 processor, but ensconced in a desktop-style package and ready to drop into practically any Socket AM2 motherboard. What’s not to like?

Well, apparently, not much at all. Big PC makers like Dell and HP liked it so much, they ordered up gobs of these chips, handing AMD the kind of success that leads to problems—supply problems. With only so many 35W CPUs to go around and important markets like the mobile space demanding quite a few chips, the 35W Athlon 64 X2 3800+ never did make it into regular supply channels where folks like you and I could buy them. In fact, to this very day, the X2 3800+ EE SFF remains on AMD’s price list with a set of asterisks where the price ought to be—mocking us.

AMD is seeking to remedy this situation with a pair of new low-power desktop processors, the Athlon X2 BE-2300 and BE-2350. Rather than giving them a 35W TDP equivalent to many Turions, AMD has backed off just slightly to 45W. The thinking here is that chips capable of operating at lower voltages that would bring them inside of a 35W TDP can go to the mobile market, while others that can’t quite fit into that thermal envelope can still serve well as low-power desktop CPUs. With this arrangement, AMD expects to be able to supply ample quantities of 45W BE-series CPUs to PC makers and other channels, including retail boxed processors.

The BE-2300 and 2350 will have some help fitting into their thermal envelope courtesy of AMD’s new 65nm fab process. In fact, these CPUs are essentially the same as the Athlon 64 X2 “Brisbane” 65nm processors we’ve already reviewed, save that those chips come with a higher 65W TDP. Just like them, these BE-series processors have dual cores with 512K of L2 cache per core and are intended for Socket AM2 motherboards. The BE-2300 is clocked at 1.9GHz, and the BE-2350 at 2.1GHz.

More intriguingly, these CPUs are bargain priced. The BE-2300 lists for $86, and you can add 200MHz for another five bucks with the $91 BE-2350. That’s cheaper than any variant of the Core 2 Duo, including the E4300 at around $115, despite the fact that the E4300 has a 65W TDP rating. (For what it’s worth, we have included an E4300 in our testing for comparison.)

What’s with that funky name?
So all of this sounds pretty good so far, but you’re probably wondering: what the heck is up with these names? Athlon X2 BE-2350? Is that a motor oil?

Turns out these products are the first fruits of AMD’s new processor naming scheme. The old “true performance initiative” numbers attached to current Athlons and Semprons was getting to be more than a little threadbare in this age of multicore processors and new microarchitectures, so AMD finally decided to scrap it. The new scheme is intended to provide more information about a processor at a glance and to confuse AMD’s enemies, bringing it victory on the field of battle.


Pay no attention to the “64” on the cap!

Notable by its omission in the new naming scheme is the “64” after “Athlon.” Our review sample CPU came with “Athlon 64 X2” emblazoned across its cap, but we’ve been told to expect the shipping product to have this extraneous number excised. Now that the whole world has seen the wisdom of adding 64-bit extensions in hardware and continuing to use only 32-bit software, AMD figures its work here is done. Thus the simpler “Athlon X2” series is born. This change lines things up with future product names like “Phenom X4,” as well.

The series of letters and numbers after that moniker is not entirely random, either. The first two letters indicate the class and power rating of the CPU, with the “E” in “BE” signifying a sub-65W TDP rating. The four numbers after that are divvied up into groups of one and three. The first digit “reflects major increments in processor attributes,” according to AMD, and all 2xxx-series CPUs are presently in the Athlon X2 family. The last three digits are intended to indicate relative performance within a given product family.

At least, I believe they indicate relative performance. Here’s what AMD says about it: “Increasing numbers within a class series indicates increments in processor attributes.” You figure it out.

Incidentally, if the BE-2300 were named according to AMD’s old scheme, I believe it would be considered a low-power version of the Athlon 64 X2 3600+, while the BE-2350 would be a variant of the Athlon 64 X2 4000+.

One obvious advantage of the new naming scheme is that it more closely matches what Intel is now doing, especially the final four digits of the model names. This setup may allow AMD to cozy up next to Pentiums and Core Duos with product numbers that suggest similar or better performance. That may be part of what’s happening with the BE-2300 and 2350, whose model numbers are just a tad higher than the recently introduced Pentium E2140 and E2160 dual-core CPUs. We probably won’t know how fully AMD will deploy this tactic until it announces more details about its naming scheme or introduces more products that use it. For now, the company intends to retain the “plus” model numbers on its existing products rather than renaming its whole lineup.

Now, on to our test results. We have a full slate of performance results for the BE-2350, but we’re going to move through them quickly, since the BE-2350 is a low-end, low-power processor with few true competitors. I’ll keep my performance commentary to a minimum. We’ll then focus more on our energy efficiency tests and overclocking efforts, since those are worthy of some additional attention.

 

Our testing methods
As ever, we did our best to deliver clean benchmark numbers. Tests were run at least three times, and the results were averaged.

In some cases, getting the results meant simulating a slower chip with a faster one. For instance, our Core 2 Duo E6600 and E6700 processors are actually a Core 2 Extreme X6800 processor clocked down to the appropriate speeds. Their performance should be identical to that of the real thing. Similarly, our Athlon 64 FX-72 results come from an underclocked pair of Athlon 64 FX-74s, our Athlon 64 X2 4400+ is an underclocked X2 5000+ (both 65nm), and our Athlon 64 X2 5600+ is an underclocked Athlon 64 X2 6000+.

Our test systems were configured like so:

Processor Core 2 Duo E6300 1.83GHz
Core 2 Duo E6400 2.13GHz
Core 2 Duo E6600 2.4GHz
Core 2 Duo E6700 2.66GHz
Core 2 Extreme X6800 2.93GHz
Core 2 Quad Q6600 2.4GHz
Core 2 Extreme QX6700 2.66GHz
Core 2 Extreme QX6800 2.93GHz
Athlon 64 X2 3600+ 1.9GHz (65nm)
Athlon 64 X2 4400+ 2.3GHz (65nm)
Athlon 64 X2 5000+ 2.6GHz (65nm)
Athlon 64 X2 5000+ 2.6GHz (90nm)
Athlon 64 X2 5600+ 2.8GHz (90nm)
Athlon 64 X2 6000+ 3.0GHz (90nm)
Athlon 64 FX-70 2.6GHz
Athlon 64 FX-72 2.8GHz
Athlon 64 FX-74 3.0GHz
Core 2 Duo E4300 1.8GHz Athlon X2 BE-2350 2.1GHz
System bus 1066MHz (266MHz quad-pumped) 1GHz HyperTransport 1GHz HyperTransport
Motherboard Intel D975XBX2 Asus M2N32-SLI Deluxe Asus L1N64-SLI WS
BIOS revision BX97520J.86A.2618.2007.0212.0954 0903 0205
BX97520J.86A.2747.2007.0523.1129 1004
North bridge 975X MCH nForce 590 SLI SPP nForce 680a SLI
South bridge ICH7R nForce 590 SLI MCP nForce 680a SLI
Chipset drivers INF Update 8.1.1.1010
Intel Matrix Storage Manager 6.21
ForceWare 15.00 ForceWare 15.00
Memory size 2GB (2 DIMMs) 2GB (2 DIMMs) 2GB (4 DIMMs)
Memory type Corsair TWIN2X2048-6400C4
DDR2 SDRAM
at 800MHz
Corsair TWIN2X2048-8500C5
DDR2 SDRAM
at 800MHz
Crucial Ballistix PC6400
DDR2 SDRAM
at 800MHz
CAS latency (CL) 4 4 4
RAS to CAS delay (tRCD) 4 4 4
RAS precharge (tRP) 4 4 4
Cycle time (tRAS) 12 12 12
Audio Integrated ICH7R/STAC9274D5 with
Sigmatel 6.10.0.5274 drivers
Integrated nForce 590 MCP/AD1988B with
Soundmax 6.10.2.6100 drivers
Integrated nForce 680a SLI/AD1988B with
Soundmax 6.10.2.6100 drivers
Hard drive Maxtor DiamondMax 10 250GB SATA 150
Graphics GeForce 7900 GTX 512MB PCIe with ForceWare 100.64 drivers
OS Windows Vista Ultimate x64 Edition
OS updates

Our Core 2 Duo E6400 processor came to us courtesy of the fine folks up north at NCIX. Those of you who are up in Canada will definitely want to check them out as a potential source of PC hardware and related goodies.

Thanks to Corsair for providing us with memory for our testing. Their products and support are far and away superior to generic, no-name memory.

Also, all of our test systems were powered by OCZ GameXStream 700W power supply units. Thanks to OCZ for providing these units for our use in testing.

The test systems’ Windows desktops were set at 1280×1024 in 32-bit color at an 85Hz screen refresh rate. Vertical refresh sync (vsync) was disabled.

We used the following versions of our test applications:

The tests and methods we employ are generally publicly available and reproducible. If you have questions about our methods, hit our forums to talk with us about them.

 

The Elder Scrolls IV: Oblivion
We tested Oblivion by manually playing through a specific point in the game five times while recording frame rates using the FRAPS utility. Each gameplay sequence lasted 60 seconds. This method has the advantage of simulating real gameplay quite closely, but it comes at the expense of precise repeatability. We believe five sample sessions are sufficient to get reasonably consistent results. In addition to average frame rates, we’ve included the low frame rates, because those tend to reflect the user experience in performance-critical situations. In order to diminish the effect of outliers, we’ve reported the median of the five low frame rates we encountered.

For this test, we set Oblivion‘s graphical quality to “Medium” but with HDR lighting enabled and vsync disabled, at 800×600 resolution. We’ve chosen this relatively low display resolution in order to prevent the graphics card from becoming a bottleneck, so differences between the CPUs can shine through.

Notice the little green plot with four lines above the benchmark results. That’s a snapshot of the CPU utilization indicator in Windows Task Manager, which helps illustrate how much the application takes advantage of up to four CPU cores, when they’re available. I’ve included these Task Manager graphics whenever possible throughout our results. In this case, Oblivion really only takes full advantage of a single CPU core, although Nvidia’s graphics drivers use multithreading to offload some vertex processing chores.

Rainbow Six: Vegas
Rainbow Six: Vegas is based on Unreal Engine 3 and is a port from the Xbox 360. For both of these reasons, it’s one of the first PC games that’s multithreaded, and it ought to provide an illuminating look at CPU gaming performance.

For this test, we set the game to run at 800×600 resolution with high dynamic range lighting disabled. “Hardware skinning” (via the GPU) was disabled, leaving that burden to fall on the CPU. Shadow quality was set to very low, and motion blur was enabled at medium quality. I played through a 90-second sequence of the game’s Terrorist Hunt mode on the “Dante’s” level five times, capturing frame rates with FRAPS, as we did with Oblivion.

The BE-2350 is a competent enough processor for running both of these games, but it’s clearly not one of the faster CPUs around for gaming. Its closest competitor from Intel, the Core 2 Duo E4300, easily has it beaten here.

 

Supreme Commander
This game is multithreaded and can actually take advantage of more than two processor cores, making it a rare commodity indeed. We ran into some snags when we first tried to test this game with FRAPS. Getting consistent results proved difficult, and the sound didn’t want to work on our Intel D975XBX2 motherboard, whose Vista x64 audio drivers may not yet be up to snuff. I was also developing the first signs of extreme RTS addiction—a grave condition indeed. I found myself analyzing unit types and lusting after level-two engineer bots. Fortunately, we were able to overcome these problems by using Supreme Commander‘s very nice built-in benchmark, which plays back a test game and reports detailed performance results afterward. We launched the benchmark by running the game with the “/map perftest /nosound” options. (Normally, we prefer to test games with audio enabled, but we made an exception here.) We tested at 1024×768 resolution with the game’s default quality settings.

Supreme Commander’s built-in benchmark breaks down its results into several major categories: running the game’s simulation, rendering the game’s graphics, and a composite score that’s simply comprised of the other two. The performance test also reports good ol’ frame rates, so we’ve included those, as well.

Supreme Commander shows us more of the same. The BE-2350 is competent, but really only faster than the Athlon 64 X2 3600+ (whose performance, incidentally, should be identical to that of the Athlon X2 BE-2300).

 

Valve Source engine particle simulation
Next up are a couple of tests we picked up during a visit to Valve Software, the developers of the Half-Life games. They’ve been working to incorporate support for multi-core processors into their Source game engine, and they’ve cooked up a couple of benchmarks to demonstrate the benefits of multithreading.

The first of those tests runs a particle simulation inside of the Source engine. Most games today use particle systems to create effects like smoke, steam, and fire, but the realism and interactivity of those effects are limited by the available computing horsepower. Valve’s particle system distributes the load across multiple CPU cores.

Valve VRAD map compilation
This next test processes a map from Half-Life 2 using Valve’s VRAD lighting tool. Valve uses VRAD to precompute lighting that goes into its games. This isn’t a real-time process, and it doesn’t reflect the performance one would experience while playing a game. It does, however, show how multiple CPU cores can speed up game development.

Valve’s multicore-capable benchmarks demonstrate how much performance can be had from quad-core systems, and it’s an impressive sight. That puts the performance of the BE-2350 in context for us.

 

3DMark06
3DMark06 combines the results from its graphics and CPU tests in order to reach an overall score. Here’s how the processors did overall and in each of those tests.

Our subject just trails the Core 2 Duo E4300 in 3DMark’s overall score and, in a photo finish, edges out the E4300 in the composite CPU score.

 

The Panorama Factory
The Panorama Factory handles an increasingly popular image processing task: joining together multiple images to create a wide-aspect panorama. This task can require lots of memory and can be computationally intensive, so The Panorama Factory comes in a 64-bit version that’s multithreaded. I asked it to join four pictures, each eight megapixels, into a glorious panorama of the interior of Damage Labs. The program’s timer function captures the amount of time needed to perform each stage of the panorama creation process. I’ve also added up the total operation time to give us an overall measure of performance.

You can save yourself 30 seconds a pop in creating panoramic photos by going with a quad-core system over the BE-2350. Still, it gets the job done in less than a minute.

picCOLOR
picCOLOR was created by Dr. Reinert H. G. Müller of the FIBUS Institute. This isn’t Photoshop; picCOLOR’s image analysis capabilities can be used for scientific applications like particle flow analysis. Dr. Müller has supplied us with new revisions of his program for some time now, all the while optimizing picCOLOR for new advances in CPU technology, including MMX, SSE2, and Hyper-Threading. Naturally, he’s ported picCOLOR to 64 bits, so we can test performance with the x86-64 ISA. Eight of the 12 functions in the test are multithreaded, and in this latest revision, five of those eight functions use four threads.

Scores in picCOLOR, by the way, are indexed against a single-processor Pentium III 1 GHz system, so that a score of 4.14 works out to 4.14 times the performance of the reference machine.

The BE-2350’s performance is as expected in picCOLOR, right between the X2 3600+ and 4400+. The Core 2 processors are stronger here, with the E4300 outpacing the X2 4400+.

 

Windows Media Encoder x64 Edition
Windows Media Encoder is one of the few popular video encoding tools that uses four threads to take advantage of quad-core systems, and it comes in a 64-bit version. For this test, I asked Windows Media Encoder to transcode a 153MB 1080-line widescreen video into a 720-line WMV using its built-in DVD/Hardware profile. Because the default “High definition quality audio” codec threw some errors in Windows Vista, I instead used the “Multichannel audio” codec. Both audio codecs have a variable bitrate peak of 192Kbps.

LAME MP3 encoding
LAME MT is a multithreaded version of the LAME MP3 encoder. LAME MT was created as a demonstration of the benefits of multithreading specifically on a Hyper-Threaded CPU like the Pentium 4. Of course, multithreading works even better on multi-core processors. You can download a paper (in Word format) describing the programming effort.

Rather than run multiple parallel threads, LAME MT runs the MP3 encoder’s psycho-acoustic analysis function on a separate thread from the rest of the encoder using simple linear pipelining. That is, the psycho-acoustic analysis happens one frame ahead of everything else, and its results are buffered for later use by the second thread. That means this test won’t really use more than two CPU cores.

We have results for two different 64-bit versions of LAME MT from different compilers, one from Microsoft and one from Intel, doing two different types of encoding, variable bit rate and constant bit rate. We are encoding a massive 10-minute, 6-second 101MB WAV file here, as we have done in many of our previous CPU reviews.

The X2 BE-2350 shadows the Core 2 Duo E4300 in our all-important (for an HTPC chip) media encoding tests.

 

Cinebench
Graphics is a classic example of a computing problem that’s easily parallelizable, so it’s no surprise that we can exploit a multi-core processor with a 3D rendering app. Cinebench is the first of those we’ll try, a benchmark based on Maxon’s Cinema 4D rendering engine. It’s multithreaded and comes with a 64-bit executable. This test runs with just a single thread and then with as many threads as CPU cores are available.

POV-Ray rendering
We’ve finally caved in and moved to the beta version of POV-Ray 3.7 that includes native multithreading. The latest beta 64-bit executable is still quite a bit slower than the 3.6 release, but it should give us a decent look at comparative performance, regardless.

Quad cores (or more) are the path to excellence in 3D rendering work. Still, the BE-2350 pulls more than its own weight here, outperforming the E4300 across the board.

 

MyriMatch
Our benchmarks sometimes come from unexpected places, and such is the case with this one. David Tabb is a friend of mine from high school and a long-time TR reader. He recently offered to provide us with an intriguing new benchmark based on an application he’s developed for use in his research work. The application is called MyriMatch, and it’s intended for use in proteomics, or the large-scale study of protein. I’ll stop right here and let him explain what MyriMatch does:

In shotgun proteomics, researchers digest complex mixtures of proteins into peptides, separate them by liquid chromatography, and analyze them by tandem mass spectrometers. This creates data sets containing tens of thousands of spectra that can be identified to peptide sequences drawn from the known genomes for most lab organisms. The first software for this purpose was Sequest, created by John Yates and Jimmy Eng at the University of Washington. Recently, David Tabb and Matthew Chambers at Vanderbilt University developed MyriMatch, an algorithm that can exploit multiple cores and multiple computers for this matching. Source code and binaries of MyriMatch are publicly available.

In this test, 5555 tandem mass spectra from a Thermo LTQ mass spectrometer are identified to peptides generated from the 6714 proteins of S. cerevisiae (baker’s yeast). The data set was provided by Andy Link at Vanderbilt University. The FASTA protein sequence database was provided by the Saccharomyces Genome Database.

MyriMatch uses threading to accelerate the handling of protein sequences. The database (read into memory) is separated into a number of jobs, typically the number of threads multiplied by 10. If four threads are used in the above database, for example, each job consists of 168 protein sequences (1/40th of the database). When a thread finishes handling all proteins in the current job, it accepts another job from the queue. This technique is intended to minimize synchronization overhead between threads and minimize CPU idle time.

The most important news for us is that MyriMatch is a widely multithreaded real-world application that we can use with a relevant data set. MyriMatch also offers control over the number of threads used, so we’ve tested with one to four threads.

STARS Euler3d computational fluid dynamics
Charles O’Neill works in the Computational Aeroservoelasticity Laboratory at Oklahoma State University, and he contacted us to suggest we try the computational fluid dynamics (CFD) benchmark based on the STARS Euler3D structural analysis routines developed at CASELab. This benchmark has been available to the public for some time in single-threaded form, but Charles was kind enough to put together a multithreaded version of the benchmark for us with a larger data set. He has also put a web page online with a downloadable version of the multithreaded benchmark, a description, and some results here. (I believe the score you see there at almost 3Hz comes from our eight-core Clovertown test system.)

In this test, the application is basically doing analysis of airflow over an aircraft wing. I will step out of the way and let Charles explain the rest:

The benchmark testcase is the AGARD 445.6 aeroelastic test wing. The wing uses a NACA 65A004 airfoil section and has a panel aspect ratio of 1.65, taper ratio of 0.66, and a quarter-chord sweep angle of 45º. This AGARD wing was tested at the NASA Langley Research Center in the 16-foot Transonic Dynamics Tunnel and is a standard aeroelastic test case used for validation of unsteady, compressible CFD codes.

The CFD grid contains 1.23 million tetrahedral elements and 223 thousand nodes . . . . The benchmark executable advances the Mach 0.50 AGARD flow solution. A benchmark score is reported as a CFD cycle frequency in Hertz.

So the higher the score, the faster the computer. I understand the STARS Euler3D routines are both very floating-point intensive and oftentimes limited by memory bandwidth. Charles has updated the benchmark for us to enable control over the number of threads used. Here’s how our contenders handled the test with different thread counts.

Ok, so the BE-2350 isn’t exactly a scientific computing powerhouse—probably not a deal-breaker.

 

Folding@Home
Next, we have another relatively new addition to our benchmark suite: a slick little Folding@Home benchmark CD created by notfred, one of the members of Team TR, our excellent Folding team. For the unfamiliar, Folding@Home is a distributed computing project created by folks at Stanford University that investigates how proteins work in the human body, in an attempt to better understand diseases like Parkinson’s, Alzheimer’s, and cystic fibrosis. It’s a great way to use your PC’s spare CPU cycles to help advance medical research. I’d encourage you to visit our distributed computing forum and consider joining our team if you haven’t already joined one.

The Folding@Home project uses a number of highly optimized routines to process different types of work units from Stanford’s research projects. The Gromacs core, for instance, uses SSE on Intel processors, 3DNow! on AMD processors, and Altivec on PowerPCs. Overall, Folding@Home should be a great example of real-world scientific computing.

notfred’s Folding Benchmark CD tests the most common work unit types and estimates performance in terms of the points per day that a CPU could earn for a Folding team member. The CD itself is a bootable ISO. The CD boots into Linux, detects the system’s processors and Ethernet adapters, picks up an IP address, and downloads the latest versions of the Folding execution cores from Stanford. It then processes a sample work unit of each type.

On a system with two CPU cores, for instance, the CD spins off a Tinker WU on core 1 and an Amber WU on core 2. When either of those WUs are finished, the benchmark moves on to additional WU types, always keeping both cores occupied with some sort of calculation. Should the benchmark run out of new WUs to test, it simply processes another WU in order to prevent any of the cores from going idle as the others finish. Once all four of the WU types have been tested, the benchmark averages the points per day among them. That points-per-day average is then multiplied by the number of cores on the CPU in order to estimate the total number of points per day that CPU might achieve.

This may be a somewhat quirky method of estimating overall performance, but my sense is that it generally ought to work. We’ve discussed some potential reservations about how it works here, for those who are interested. I have included results for each of the individual WU types below, so you can see how the different CPUs perform on each.

Our Folding benchmark relies on executables for each WU type that it downloads from Stanford’s servers. Unfortunately, Stanford has recently updated its Gromacs 3.3 core, and it now produces fewer points per day than previous versions. Among these results, only the Athlon X2 BE-2350 and the Core 2 Duo E4300 were tested with the new Gromacs 3.3 core, and as a result, they are at a bit of a disadvantage compared to the rest of the field.

Even so, the BE-2350 comes out ahead overall versus the Core 2 Duo E6300 on the strength of its performance with the Tinker and Amber work units. I don’t believe those WU types are as common as the Gromacs WUs these days, though.

 

SiSoft Sandra Mandelbrot
Next up is SiSoft’s Sandra system diagnosis program, which includes a number of different benchmarks. The one of interest to us is the “multimedia” benchmark, intended to show off the benefits of “multimedia” extensions like MMX, SSE, and SSE2. According to SiSoft’s FAQ, the benchmark actually does a fractal computation:

This benchmark generates a picture (640×480) of the well-known Mandelbrot fractal, using 255 iterations for each data pixel, in 32 colours. It is a real-life benchmark rather than a synthetic benchmark, designed to show the improvements MMX/Enhanced, 3DNow!/Enhanced, SSE(2) bring to such an algorithm.

The benchmark is multi-threaded for up to 64 CPUs maximum on SMP systems. This works by interlacing, i.e. each thread computes the next column not being worked on by other threads. Sandra creates as many threads as there are CPUs in the system and assignes [sic] each thread to a different CPU.

We’re using the 64-bit version of Sandra. The “Integer x16” version of this test uses integer numbers to simulate floating-point math. The floating-point version of the benchmark takes advantage of SSE2 to process up to eight Mandelbrot iterations in parallel.

Oh, well. Let’s look at power consumption.

 

Power consumption and efficiency
We’ve tested the BE-2350’s power consumption in several different configurations, to give you an idea how it compares with the other chips we’ve tested on a common platform—our Asus M2N32-SLI Deluxe mobo—and to show you how low its power consumption can go when coupled with a motherboard based on AMD’s 690G chipset, the Asus M2A-VM HDMI. Thus, you’ll see several sets of results for the BE-2350 below. Those marked “690G chipset” came from a system based on—wait for it—the 690G chipset and include the same GeForce 7900 GTX graphics card we used for the rest of our tests. The results marked “690G chipset/IGP” come from the same config with no discrete graphics card, relying only on the 690G’s integrated graphics processor.

To make things even more interesting, we also tested the BE-2350’s spiritual predecessor, the (big breath) Athlon 64 X2 3800+ Energy Efficient Small Form Factor (phew), on the 690G platform. This chip has a 35W TDP rating, and makes a good foil for the new coolness.

Our Extech 380803 power meter has the ability to log data, so we can capture power use over a span of time. The meter reads power use at the wall socket, so it incorporates power use from the entire system—the CPU, motherboard, memory, video card, hard drives, and anything else plugged into the power supply unit. (We plugged the computer monitor and speakers into a separate outlet, though.) We measured how each of our test systems used power during a roughly one-minute period, during which time we executed Cinebench’s multithreaded rendering test. All of the systems had their power management features (such as SpeedStep and Cool’n’Quiet) enabled during these tests.

You’ll notice that I’ve not included the Athlon 64 FX-72 here. That’s because our “simulated” FX-72 CPUs are underclocked versions of faster processors, and we’ve not been able to get Cool’n’Quiet power-saving tech to work when CPU multiplier control is in use. I have included test results for genuine Athlon 64 X2 4400+ and 5600+ chips, though. I’ve also included our simulated Core 2 Duo E6600 and E6700, because SpeedStep works fine on the D975XBX2 motherboard alongside underclocking. The simulated processors’ voltage may not be exactly the same as what you’d find on many retail E6600s and E6700s. However, voltage and power use can vary from one chip to the next, since Intel sets voltage individually on each chip at the factory.

The differences between the CPUs are immediately obvious by looking at these plots of the raw data, and the BE-2350 looks quite nice, especially on the 690G platform. We can slice up the data in various ways in order to better understand them, though. We’ll start with a look at idle power, taken from the trailing edge of our test period, after all CPUs have completed the render.

The BE-2350’s idle power use is nice and low, as expected, but so is the idle power use in the majority of the systems tested. Low-power chips rarely distinguish themselves at idle. Chipsets can, though. The idle power draw on our BE-2350 test system drops by 14W simply by swapping out the motherboard, replacing the relatively power-hungry nForce 590 SLI with the AMD 690G. We can shave off another 43W by dispensing with the discrete graphics card, too. Notice, however, that the 35W version of the X2 3800+ draws even less power at idle than the BE-2350.

Next, we can look at peak power draw by taking an average from the five-second span from 10 to 15 seconds into our test period, during which the processors were rendering.

Under load, the BE-2350-based system once more pulls about 5W more than the one based on the X2 3800+ EE SFF. That’s more or less in keeping with expectations. On the M2N32-SLI Deluxe motherboard, the BE-2350 draws 12W less under load than the Athlon 64 X2 3600+, despite its higher clock speed. Only on the 690G mobo does the BE-2350 pulls less power while rendering than the Core 2 Duo E4300, though.

Another way to gauge power efficiency is to look at total energy use over our time span. This method takes into account power use both during the render and during the idle time. We can express the result in terms of watt-seconds, also known as joules.

The BE-2350 remains in league with the 35W X2 3800+ when we slice things up this way. Coupled with the 690G, the BE-2350 can be very energy efficient—more so than anything we tested, other than its predecessor.

We can quantify efficiency even better by considering the amount of energy used to render the scene. Since the different systems completed the render at different speeds, we’ve isolated the render period for each system. We’ve chosen to identify the end of the render as the point where power use begins to drop from its steady peak. There seems to be some disk paging going on after that, but we don’t want to include that more variable activity in our render period.

We’ve computed the amount of energy used by each system to render the scene. This method should account for both power use and, to some degree, performance, because shorter render times may lead to less energy consumption.

More cores equals shorter render times, so quad-core systems generally do well here, despite their relatively high peak power draw. The BE-2350 on the 690G motherboard with integrated graphics takes a different path to the upper reaches of this graph, simply by drawing less power throughout its relatively long render time.

Update 7/15/07: We have now tested the Core 2 Duo E4300 with integrated graphics, as well, and the results are intriguing. The full details are here.

 
Overclocking
This processor draws slightly more power than the Athlon 64 X2 3800+ EE SFF, but it has a huge edge in one area: overclocking. AMD’s 90nm low-power chips intentionally sacrificed clock speed headroom for lower power use via manufacturing tweaks. As a result, those CPUs were never good overclockers. I’m here to testify that the Athlon X2 BE-2350 is nothing like that. In fact, I believe it was probably made using the exact same fab process as any of AMD’s desktop CPUs. I expect they just cherry-picked the ones capable of running at lower voltages and designated them as BE-class parts.

You know what that means….

This CPU has a stock clock speed of 2.1GHz, and I started out trying to get it to reach what I thought would be a respectable 2.4 to 2.5GHz. But this thing wasn’t nearly done when it reached that speed, so I kept pushing it. When the dust had settled, the BE-2350 was running relatively stable at 3003MHz with a 286MHz HyperTransport clock. I had to bump up the voltage to 1.425V in order to reach that speed. As is often the case once you’ve hit a wall, raising the voltage beyond that point didn’t help much.


Holy moly.

Still, ladies and gents, we’re talking about a nearly 50% overclock out of this $91 CPU. That’s into the realm of fabled chips like the Celeron 300A and Athlon XP-M 2500+. That clock speed also puts the BE-2350 on nearly equal footing with AMD’s fastest dual-core processor, the Athlon 64 X2 6000+.

Here are a couple of benchmark results, just to show how performance scales.

More or less as expected, I’d say. This may be a low-power marvel, but it really sings once you turn up the voltage, too.

Oh, and my attorney reminds you that overclocking performance is never guaranteed, your mileage may vary, and you may contract the gout just by thinking of raising the HT clock speed, for all we know. I’ll be interested to see whether other folks get similar results out of other BE-series chips. If they do, this could be one heckuva purchase for under a hundred bucks.

 
Conclusions
AMD intends to pair up the Athlon X2 BE-2350 with its 690G chipset as part of a push into smaller form factors, including systems built around its proposed small-footprint DTX specification. The BE-2350 should serve quite well in that role. It’s not the fastest CPU around, but it should perform well enough for use in a home theater PC, office desktop system, or basic email/surfing/word processing box. Those of you looking to use this chip in a home theater PC may be interested to know that I tried playing back an H.264-encoded HD DVD (the movie Babel) on our test system, and the processor handled it quite well, with CPU utilization ranging between about 48 and 63%. CPU utilization was even lower with a VC-1 encoded disc. This system’s GeForce 7-series video card can’t accelerate the most processor-intensive stages of H.264 decoding, yet playback was consistently fluid, so the BE-2350 looks to be very much up to the task.

Power-wise, the BE-2350 lives up to its billing and is a worthy successor to the 35W version of the Athlon 64 X2 3800+—or it will be, provided AMD can actually supply these chips in sufficient volume. Our power consumption tests put the BE-2350 very close to its predecessor—just about 5W higher—both at idle and under load. Truth be told, the measured power is only part of the story. The power draw numbers we saw from the Core 2 Duo E4300 weren’t bad, either, but its 65W TDP rating remains a sticking point. The fact that PC makers and chassis designers can count on a 45W TDP will make the BE-2350 very attractive to them. The BE-2350’s low-power cred is enhanced by the availability of AMD’s 690G chipset, as well. We’ve found the idle power consumption of Intel’s G965 Express chipset to be substantially higher than the 690G. All of these factors, combined with its $91 price tag, make the Athlon X2 BE-2350 a stand-out value in a low-power processor.

Then there’s the fact I was able to get our review unit cranked up to 3GHz with nothing but a stock air cooler and a slight voltage bump. The low-power aspect goes out the window when the BE-2350 is flipping bits at 3003MHz and 1.425V, but then this puppy morphs into a high-performance processor the equal of anything AMD has to offer. I don’t want to overstate things on this front; the Core 2 Duo E4300 is also known for ample overclocking headroom, and a 3GHz Athlon X2 isn’t likely to match a Core 2 at 2.5GHz in overall performance. Still, for a cheap CPU that you may not mind frying in an evil overclocking experiment, the BE-2350 is hard to match, if most of them end up overclocking like ours did. 

Latest News

Gold Miner Nilam Resources Shares Surge 22x Amidst Bitcoin Buying Announcement
Crypto News

Gold Miner Nilam Resources Shares Surge 22x Amidst Bitcoin Buying Announcement

BlackRock CEO Goes Bullish on BTC as Spot Bitcoin ETF Crosses $17 Billion
Crypto News

BlackRock CEO Goes Bullish on BTC as Spot Bitcoin ETF Crosses $17 Billion

Spot Bitcoin exchange-traded funds (ETFs) continue to receive massive inflows as investor demand skyrockets, with BlackRock’s IBIT at the forefront. Notably, IBIT has hit a whopping $17 billion in AUM...

Elliott Wave Pattern Indicates Ripple (XRP) Might Surge to $13
Crypto News

Elliott Wave Pattern Indicates Ripple (XRP) Might Surge to $13

Renowned market analyst Tony Severino, also called “The Bull,”  has unveiled a captivating analysis of XRP’s potential price trajectory. His findings suggest that XRP could be poised for an extraordinary...

XRP ETF Premium May Record 100x to $500 Chad Steingraber
Crypto News

XRP ETF Premium May Record 100x Growth Chad Steingraber

NFL
Streaming News & Events

NFL Discloses Moving Two NFL Games Into Streaming in 2024

Crypto News

Unveiling the Most Popular Crypto Presales in March Among Americans

Apple Users Are Being Spammed with Unwanted Password Reset Requests as Part of ‘MFA Bombing'
News

Apple Users Are Being Spammed with Unwanted Password Reset Requests as Part of ‘MFA Bombing’