AMD Athlon X2 BE-2350: Mainstream X2s with 45W TDP

Those hoping for nail biting, teeth clenching battles should apply elsewhere – the CPU war these days is a one horse race. If reports out of Taiwan are to be believed, initial performance results from AMD’s Barcelona fail to impress and we’ve got at least a quarter before the race can even potentially get competitive. But as we’ve seen lately, you don’t need chart topping performance to bring excitement to the game.

By aggressively cutting prices, AMD actually made most of its product lineup below $300 competitive with equivalently priced Intel offerings. Granted that AMD won’t be making a tremendous amount of money by doing this, but the end user stands to benefit, especially those with Socket-AM2 motherboards looking for faster CPUs.

Today AMD follows with yet another affordable CPU introduction; priced at $91 and $86 respectively, the Athlon X2 BE-2350 and 2300 aren’t designed to take the performance crown from Intel, but rather they are decent dual core performers with a mere 45W TDP.

These two 65nm processors run at 2.1GHz and 1.9GHz, and are architecturally no different than the Athlon 64 X2s we’ve been reviewing for a while now. Carefully selected as chips that can run at lower operating voltages and thus lower TDPs, these two processors are designed to be a more available version of the Energy Efficient Small Form Factor X2s that were introduced last year. While they have a higher TDP than the 35W Athlon 64 X2 3800+, AMD insists that availability of these 45W parts won’t be a problem. At the time of publication we couldn’t confirm AMD’s claims, so we’ll just have to wait and see. Prior to today’s introduction, the lowest TDP widely available from an X2 was 65W, so the release of 45W parts is designed to fill a gap in AMD’s product lineup.

At the price points AMD is targeting with the BE-2350 and 2300, Intel doesn’t really have a good competitor. While you can get older Pentium 4s for less than $100, you wouldn’t really want to from a power and performance standpoint. The closest Intel has to offer is the Core 2 Duo E4300, which we’ve been able to find online for $113.50, thus making it the best competition we can find. Note that both of these chips are more expensive than the X2 3800+, currently priced at $83 and running at 2.0GHz. The slight premium comes from the lower yield on these chips, and resultant lower TDP.

Keep in mind that Intel’s closest competition is more expensive throughout the course of this review, because after AMD’s latest price cuts you can truly get some powerful CPUs for less than $100.

Eeech, Model Numbers

Although all current AMD processors retain their original names, the two being introduced today are the first to use AMD’s new model numbers. As announced during its Phenom introduction, AMD is dropping the 64 from its product names – the new chips are simply Athlon X2s. The 64-bit race is over now that both AMD and Intel have 64-bit support on a majority of their processors, and now it’s time to move on. All previous X2s will still be called Athlon 64 X2s and AMD isn’t changing the logo just yet, but eventually it will phase out the old names/model numbers in favor of the new system.

What exactly is the new system? It’s a slightly more complicated version of Intel’s model number system. Here’s the explanation of the new system straight from AMD:

The introduction of the AMD Athlon^TM X2 dual core processor BE-2350 and BE-2300 brings the first opportunity to learn about AMD’s new model methodology. The goal with the new system was to better inform processor choice and utilize a methodology that be long lasting. Existing products will retain their current model numbers. Our customers are familiar with the current models and we will continue to utilize that system until it is phased out over a period of time by new product introductions.

Let’s look at a sample model number: BE-2350 (This is the AMD Athlon X2 dual-core 45-watt desktop CPU you have for review)

Format:

The new AMD desktop processor models have an alpha numeric format of A A – # # # #.

First two characters: BE-2350

The first and second alpha indicate the processor class. The second alpha character indicates the TDP of the processor. The “BE” class is comprised of sub-65W processors. This chip’s TDP is 45 watts. As additional products are introduced, new classes will also be introduced and these new classes will distinguish between key attributes of the processors.

First numeric digit: BE-2350

The first numeric digit after the dash is the processor series and indicates reflects major increments in processor attributes. The “2XXX” series is currently contained within the AMD Athlon X2 family of processors.

Note that we have dropped the “64” from the Athlon X2 name. AMD pioneered simultaneous 64/32-bit x86 processing. Now that 64-bit processing is ubiquitous and AMD is recognized for its leadership, maintaining a “64” in our desktop product naming methodology is not necessary, and the shortened name simplifies product references.

Last three numeric digits: BE-2350

The last three numeric digits after the dash indicate the relative position of the CPU within its class series. Increasing numbers within a class series indicates increments in processor attributes.

In summary:

Please note that the actual assignment of letters and numbers are intentionally arbitrary, but these digits are combined in such a way as to avoid confusion between models while indicating major and minor processor increments. Just by reading the “BE-2350” model number, you know that it is a mainstream desktop CPU. You know its power consumption level is below 65 watts. You know that it is in the Athlon X2 family. And you know its position relative to other CPUs. As new processors are introduced, the combination of class and models should be of increasing value in identifying and distinguishing AMD processors. Previously, our model numbers indicated relative performance but were unable to capture the step function performance multi-core processors in many usage scenarios and were unable to capture additional processor features or attributes.

Normally we don’t quote manufacturer emails to us verbatim, but this one just seemed so appropriate. To break it down for you, we’ll compare AMD’s new naming scheme to Intel’s.

The first letter in Intel’s naming system indicates processor class, for example the E6600 vs. X6800. With AMD’s new system, we have two letters that describe the class, with the second one being used to indicate TDP. The following four digits in Intel’s system simply indicate performance of the processor relative to others in its class; e.g. an E6600 is faster than an E6320, the first digit indicating major performance differences between chips (e.g. E6600 has 4MB L2 cache 1066MHz FSB, while the E4300 has a 2MB L2 cache and 800MHz FSB). AMD’s system is similar, the first digit is reserved for major differences in performance, while the latter three digits are used for minor differences (think speed bins).

All in all, AMD’s system is a response to Intel’s system, neither of which is perfect. We liked Intel’s naming system on the Core 2 lineup back when it was simple and each model was separated by increments of 100. The introduction of the E6420 and E6320 made the system a bit more messy and the upcoming 1333MHz FSB CPUs will only further complicate the lineup. AMD appears to be starting in a period of disarray and if recent articles on the forthcoming lineup are correct, we’ll absolutely hate talking about CPUs from both manufacturers.

The Test

Note that SYSMark 2007 only runs on 32-bit versions of Windows, so all of the tests for this article were conducted under Windows Vista Ultimate 32-bit.

CPU:	AMD Athlon 64 X2 3800+ (2.0GHz/512KBx2) AMD Athlon X2 BE-2350 (2.1GHz/512KBx2) AMD Athlon X2 BE-2300 (1.9GHz/512KBx2) Intel Core 2 Duo E6320 (1.86GHz/4MB) Intel Core 2 Duo E4300 (1.80GHz/2MB)
Motherboard:	ASUS P5B Deluxe (P965) ASUS M2N32-SLI Deluxe (nForce 590 SLI)
Chipset:	Intel P965 NVIDIA nForce 590 SLI
Chipset Drivers:	Intel 8.1.1.1010 (Intel) Integrated Vista Drivers (NVIDIA)
Hard Disk:	Seagate 7200.9 300GB SATA
Memory:	Corsair XMS2 DDR2-800 4-4-4-12 (1GB x 2)
Video Card:	NVIDIA GeForce 8800 GTX
Video Drivers:	NVIDIA ForceWare 158.18
Desktop Resolution:	1600 x 1200
OS:	Windows Vista Ultimate 32-bit

General Performance

After years of waiting we can finally move away from SYSMark 2004 as BAPCo has just released SYSMark 2007, its latest benchmark suite that boasts full Vista compatibility (only 32-bit however). As always, SYSMark is divided into a number of individual performance categories, which together provide an overall performance score for the system.

AMD’s Next-Generation Mobile Architecture Revealed: Griffin

Several years ago Intel discovered surprisingly enough that its NetBurst architecture was not very good for the mobile space.  As wonderful as the idea of battery powered space heaters was, Intel quickly discovered that to build the perfect mobile platform you had to start from scratch and design a CPU that was built for the mobile space.  By doing so Intel could make tradeoffs that it wouldn’t normally make, performance for power reduction, many of which we diagrammed in our first Centrino articles.

Intel also discovered the power of the platform; by bundling a good CPU with a good chipset and wireless controller, three independent Intel products were transformed into a marketing powerhouse.  The Centrino brand simplified notebook purchasing and quickly became a mark associated with a notebook you wanted to buy.

It took AMD a bit longer to get on the bandwagon, putting marketing first and worrying about architecture last.  We had heard rumors of a mobile-specific AMD microarchitecture, but nothing ever surfaced until now.  AMD’s design team out of Massachusetts worked on the project, and today we’re finally able to tell you about it.  The processor is called Griffin, and the platform is called Puma, both are codenames; AMD will undoubtedly come up with a phenomenal name for the final product (sorry we couldn’t resist).

When Intel started development on the first Centrino processor, Banias, it had to go back to the P6 for a starting point.  The Pentium 4’s NetBurst architecture was hardly suitable and the design team was intimately familiar with the P6 core at the time.  The end product hardly resembled a P6 and if you look at what the architecture evolved into today, you would be hard pressed to say it was similar at all to a Pentium III. 

AMD didn’t make the misstep of a Pentium 4, it made a solid evolutionary step to K8 from K7.  Griffin’s execution core and underlying architecture is based on the current generation 65nm K8 design, not Barcelona/Phenom.  You can take everything you are looking forward to being in Phenom and throw it out the window, as AMD is starting from the same K8 core that launched in 2003.

By no means is it a bad starting point, but thankfully AMD did toss in some enhancements.  Griffin gets a new North Bridge, a new memory controller, a power optimized Hyper Transport 3 interface and a 1MB L2 cache per core.  Griffin will still be built on a 65nm process as AMD will have, at best, only begun its 45nm production by the time Griffin debuts. 

Right off the bat you see a disparity between AMD’s approach and Intel’s approach; while the K8 is arguably a better starting point for a mobile-specific architecture than the P6, the K8 was heavily designed for servers and scaled down.  But as we’ve seen, the K8 is quite power efficient, with 35W TDPs easy achievable for dual core versions, so the race isn’t over before it has started.

Griffin will go into production at the end of this year, and AMD is targeting availability in the first half to middle of 2008.  Given the launch timeframe, much like Phenom, AMD won’t be competing with today’s Core 2 processors but rather tomorrow’s Penryn based notebooks.  Penryn does have some mobile-specific power improvements that even Griffin does not, but the opposite is also true as you will soon see.  AMD quoted a maximum TDP of 35W for dual core Griffin CPUs.  AMD hopes that notebooks based on Griffin can offer beyond 5 hours of battery life, but do keep in mind that battery life will vary greatly based on OEM implementation.

Truly Independent Power Planes

While architecturally Griffin is no different than today’s Athlon X2s, it will draw noticeably less power in normal use.  AMD is the first to announce the next step in multi-core power management: independent voltages and frequencies for each core.  While Phenom splits the North Bridge and CPU cores into two separate voltage planes, Griffin goes one step further and puts each individual CPU core onto an independent voltage plane.  The benefit is that not only can each core run at its own frequency, but it can also run at lower voltages giving you significant reductions in power consumption.

Dynamic power of a CPU can be determined by the following equation:

Power = ∝ * C * V^2 * F

Simply reducing the frequency of a processor (F) will result in a linear reduction in power consumption, but as you can see voltage (V) has an exponential relationship to power.  Reducing both is ideal, and that’s exactly what Griffin does. 

Each core can run at one of 8 frequency steps and five voltage levels, independently of one another.  Deep and deeper sleep states are supported, however AMD is currently looking into the possibility of implementing a C6 sleep state similar to what Intel announced for mobile Penryn.  AMD wouldn’t commit to whether or not we’d see a C6 state in Griffin, leading us to believe that it simply wasn’t implemented at the time of Intel’s Penryn announcements and there may not be enough time to add it in before launch.

New Memory Controller

Although the underlying architecture of Griffin is K8 based, the memory controller takes a lot of cues from Barcelona/Phenom.  There’s a new DRAM prefetcher, similar but not identical to what will be in Phenom, but many of the efficiency improvements in the new desktop core will make their way to Griffin as well.  Taken from our Barcelona architecture article:

“One strength of Intel’s FB-DIMM architecture used in Xeon servers is that you can execute read and write requests to the AMB simultaneously. With standard DDR2 memory, you can do one or the other, and there’s a penalty for switching between the two types of operations. If you have a fairly random mixture of reads and writes you can waste a lot of time switching between the two rather than performing all of your reads sequentially then switching over to writes. The K8’s memory controller made some allowances for preferring reads over writes since they take less time, but in Barcelona the memory controller is far more intelligent.

Now, instead of executing writes as soon as they show up, writes are stored in a buffer and once the buffer reaches a preset threshold the controller bursts the writes sequentially. What this avoids is the costly read/write switch penalty, helping improve bandwidth efficiency and reduce latency.”

AMD did not make it clear whether Griffin also featured two independent 64-bit DDR2 memory controllers or a single 128-bit one.  And, of course, as the memory controller is a part of the North Bridge it operates at a separate, lower voltage than the rest of the CPU cores.

Mobile Specific HT3

Much like Phenom, Griffin will support HyperTransport 3, offering more bandwidth between the CPU and the outside world.  A major change to Griffin’s HT interface however is that it is highly power optimized. 

By default, a Griffin CPU has two x16 HT3.0 links (one inbound and one outbound); depending on power and bandwidth requirements, those links can be dynamically scaled down to x8, x4, x2 or completely turned off.  The inbound and outbound lanes can dynamically change independent of one another (e.g. inbound could be scaled to x8 while the outbound could be turned off).  Each link width change requires a HT disconnect, meaning it can’t be done as frequently as a CPU frequency change, but the power savings should be substantial. 

Better Thermal Control

Griffin adds better thermal control than its predecessor.  The mobile K8 cores simply had a single on-die analog thermal diode that would report CPU temperature, while Griffin features two thermal sensors per core for more accurate thermal monitoring. 

The current generation Turion CPUs communicate processor thermal data over the SMBUS, however an external thermal monitoring circuit is used requiring additional board real estate.  Griffin supports an integrated SMBUS interface directly to the chipset, so there’s no extra chip required.

Griffin also allows monitoring of DRAM temperature (through an external temperature sensor placed near the DRAM modules) and based on pre-configured thermal limits, it can now throttle memory frequency if the modules get too hot.

AMD Names Athlon’s Successor: Phenom

We’ve been calling it Barcelona or Agena for several months now, but AMD has finally named its next-generation desktop processors; and the new name is Phenom. Not quite as odd sounding as the Athlon was when it was launched, but still very much an AMD product name. The new AMD line up is as follows:

At the top of the product lineup we have the Phenom FX processors (codenamed Agena FX). These processors will be quad-core only and run at the highest clock speeds in AMD’s lineup, much like the current Athlon 64 FX. At the Quad FX introduction, AMD indicated that FX processors would be Socket-1207 only, simplifying its product lineup. Unfortunately, AMD has once again reversed its decision and Phenom FX processors will be available in both Socket-1207 and Socket-AM2 flavors.

The Phenom X4 and X2 processors are the sensible versions of the Phenom, these are the ones we will most likely be recommending out of AMD’s lineup if history holds true. The X4 and X2 will be Socket-AM2/AM2+ only and are 100% backwards compatible with current AM2 motherboards.

The current Athlon 64 X2 has been renamed to the Athlon X2; given that both AMD and Intel offer 64-bit processors, dropping the 64 from the name makes sense. At the bottom of the list is AMD’s Sempron, which is the only single core brand in the product lineup.

AMD hasn’t updated us on other details for Phenom, which is a bit odd given how much more forthcoming AMD has been privately about other aspects of its microprocessor plans. Either AMD is doing its best to hide a rough ramp of Phenom or it is sitting on a very powerful weapon to combat Intel with, and we can’t get any indication of which it is.

Architecturally, we’ve already said all there is to say about Phenom. The core is an evolution of the current K8 architecture, with a lot of attention placed on SSE performance and other general IPC enhancements. Phenom will also be the first monolithic quad-core x86 processor to hit the market, which does have some technical advantages but we’re not sold on whether or not we’ll see any tangible real world benefits over Intel’s multiple die approach (e.g. we didn’t see any benefit with monolithic dual core vs. multi-die dual core).

Phenom will work in current Socket-AM2/Socket-1207 motherboards with a BIOS update, but it loses the ability to run its Northbridge and CPU cores at separate voltages/clock frequencies. If you buy a new Socket-AM2+/Socket-1207+ motherboard, then the CPU cores and Northbridge can run at separate voltages/frequencies. The benefit of doing this is not only power savings, but AMD has indicated that it can actually run the Northbridge faster than the CPU cores (by 200 – 400MHz) which will improve performance. The L3 cache happens to run on the same voltage plane and at the same frequency as the Northbridge, compounding the performance benefits of using a new “plus-socket” motherboard (Socket-AM2+/Socket-1207+).

The memory hierarchy of Phenom has been improved over the current K8 architecture; there’s now an L3 cache shared by all cores and a higher efficiency DDR2 memory controller, which is needed given that there are now more cores vying for the same amount of bandwidth.

AMD has officially confirmed that Phenom will support up to DDR2-1066, reasserting AMD’s commitment to the memory technology it switched to a year ago.

What about DDR3?

We asked AMD’s Phil Hester when he expects to see DDR3 make sense for the desktop, and he gave us an unusually candid answer. According to Hester, it won’t be until late 2008 or early 2009 that DDR3 will make economic sense to move to. Given that Phenom will be DDR2 from the start, it looks like AMD’s transition to DDR3 will be much like its transition to DDR2: it will complete well after Intel has made the move.

AMD views Intel’s constant switching of memory technologies as adopting new technology for the sake of technology, while it prefers to respond directly to its customers’ needs. Intel’s argument has always been that it needs to make these transitions early to prepare the entire market for them. Clearly we need both approaches; we need a company to take the hit and move the market to a new memory technology so that volumes can ramp up enough to drive the price down. At the same time we need a company like AMD to give those in the know an alternative in the interim, otherwise we would all be paying a lot of money for a new memory technology that today offers no performance improvement.

If AMD moves to DDR3 in the late-2008/2009 timeframe it’s unclear as to whether or not we’ll need a new processor socket. AMD’s current roadmap shows the second generation Phenom core (Shanghai) due out in 2008 but still on Socket-AM2+/1207+. It’s quite possible that AMD will introduce another socket in late 2008 (AM3?) that will offer DDR3 support to begin its transition to the new memory technology.

Shanghai : Barcelona :: Penryn : Conroe?

AMD has quietly introduced Shanghai into its processor roadmaps, and it’s the successor to the Barcelona core. Little is publicly known about Shanghai but you can expect the newer, open AMD to begin revealing details in a controlled manner; the days of a totally silent AMD are over.

Based on AMD’s roadmap, Shanghai will be socket compatible with Barcelona and it is listed as an enhanced version of the Barcelona core. Given that AMD is expected to be at 45nm in the second half of 2008, we’d expect Shanghai to be to Barcelona what Penryn is to Conroe. Unfortunately, Barcelona will most likely compete with Penryn and Shanghai will go head to head with Nehalem, and we have absolutely no indication of how those battles will end up at this point.

Final Words

Today’s announcement is more marketing than substance, but it brings us one step closer to what we want: Phenom performance results. Any noise from AMD at this point is better than nothing. Can you guess its next move? It may be more difficult to predict than you’d think…

AMD – The Road Ahead

A quiet AMD isn’t a good AMD, but unfortunately it’s the AMD we’ve been left with ever since Intel started becoming more competitive. In fact, the more Intel changed for the better, the more it seemed AMD changed for the worse. Intel started bringing out better product, talking more about its plans for the future, and made a whole lot of sense in just about everything it was doing and saying. Meanwhile, AMD just seemed to freeze up; we got no disclosures of upcoming products, no indication of direction, and very little sign of the hungry, competitive AMD that took Intel on and actually won a bout.

Enough complaining, poking, and prodding eventually got us a disclosure of AMD’s Barcelona architecture last year. While we appreciated the depth with which AMD gave us information on Barcelona, the product itself was over a year away when we first heard about it. With no relief in sight for AMD other than a vicious price war, we began to worry not about Barcelona, but about what would come next. Would Barcelona have to tide us over for another three years until its replacement? How will AMD compete in the mobile and ultra-mobile spaces? And how does the ATI acquisition fit into AMD’s long-term microprocessor design philosophy? In fact, what is AMD’s long term microprocessor design philosophy?

You see, we have had all of these questions answered by Intel without ever having to ask them. Once or twice a year, Intel gathers a few thousand of its closest friends in California at the Intel Developer Forum and lays out its future plans. We needed the same from AMD, and we weren’t getting it.

When Intel was losing the product battle late in the Pentium 4’s lifespan, it responded by being even more open about what it had coming down the pipeline. When everyone doubted what Intel’s next-generation micro-architecture would do, Intel released performance numbers months before any actual product launch. AMD’s strategy of remaining guarded and silent while it lost market share, confidence, and sales simply wasn’t working. Luckily, there were a handful of individuals within AMD that saw the strength and benefit of what Intel was doing.

A former ATI employee by the name of Jon Carvill was a particularly staunch advocate of a more open AMD. He fought to bring us the sort of detail on Barcelona that we wanted, and he was largely responsible for giving us access to the individuals and information that made our article on AMD’s Barcelona architecture possible. Carvill got it, and he waged a one-man war within AMD to make sure that others within the company did as well.

We thanked him dearly for helping us get the information we needed to be able to tell you all about Barcelona, but we wanted more, and he wanted to give more. He convinced the CTOs within AMD to come together and break the silence, he put them in the same room with us, and he told them to tell us just about everything. We learned about multiple new AMD architectures, new chipsets, new directions, and nearly everything we had hoped to hear about the company.

Going into these meetings, in a secluded location away from AMD’s campus, we honestly had low expectations. We were quite down on AMD and its ability to compete, and while AMD’s situation in the market hasn’t changed, by finally talking to the key folks within the company we at least have a better idea of how it plans to compete.

Over the coming weeks and months we will be able to share this information with you; today we start with a better understanding of the ATI acquisition and its impact on AMD’s future CPU direction. We will look at where AMD plans on taking its x86 processors and what it plans to do about the ultra mobile PC market. And of course, we will talk about Barcelona; while AMD has yet to let us benchmark its upcoming processors, we can feel that our time alone with the CPU is nearing. We’ve got some additional details on Barcelona and its platform that we weren’t aware of when we first covered the architecture.

AMD Unveils New-Generation Graphics Product Lineup

The head of ATI, graphics product group of Advanced Micro Devices, Rick Bergman, said during a meeting with financial analysts late last week that the company plans to refresh its graphics chip lineup early next year with at least three new product offerings. Unfortunately for AMD, which needs to return to profitability, the bulk of new products will be aimed at sub-$150 market segment with relatively low margins.

“We want to address the enthusiast space [with the ATI Radeon HD 3800]. Early in 2008 we will introduce code-named R680 product that will take two of those [chips] on a single board and will ship to customers with scaling close to 2x the performance of the existing [ATI Radeon HD] 3800 products,” said Rick Bergman, senior vice president and general manager of AMD’s graphics products group.

The code-named ATI R680 graphics card, which is based on two ATI Radeon HD 3800-series (RV670) graphics processing units (GPUs), will be the highest-performing offering by AMD’s graphics product group early in 2008. The card will retail for “over than $300” and will be able to work in pair with another similar one, which will give end-users the power of four GPUs.

“We are also going to have two other products for very important volume segments – mainstream and value – ATI RV620 and ATI RV635. We will introduce those products in January. Both of the products are 55nm and both of them will ship with the first silicon as well,” Mr. Bergman said.

ATI RV620 and ATI RV635 graphics chips are projected to substitute currently available ATI Radeon HD 2400 (RV615) and ATI Radeon HD 2600 (RV630) graphics series in January, 2008. The new GPUs will feature DirectX 10.1, PCI Express 2.0 as well as “second generation UVD” unified video decoding engine. Other details were not revealed and it is unclear whether the new parts for entry-level and mainstream markets will feature higher performance compared to currently available graphics offerings from AMD’s graphics product group.

AMD hopes that the new ATI Radeon HD 3000-series desktop products and their mobile brethren code-named M8x will help the company to win contracts with original equipment manufacturers (OEMs), which guarantees stable revenue streams for several months.

The firm indicated that if in 2007 its discrete graphics chips could be found in about 60 notebook designs, then in 2008 its standalone graphics adapters will be present in about 100 notebook designs. The company did not disclose whether those notebook designs with Mobility Radeon inside are based on AMD Turion 64-series processors, which are found in 20% – 25% of notebooks available in the world, or are powered by Intel Core 2 chips, which command 75% – 80% of the whole mobile microprocessor market.

“In 2008, during the summer, it’s fully my expectation, ATI graphics will return to the No. 1 spot, which [we used to hold] for 7 or 8 years as the leading notebook discrete graphics vendor,” AMD’s Rick Bergman hopes.

While AMD manages to increase its shipments graphics cards, its growth is below that of the market. According to Jon Peddie Research, total shipments of graphics adapters – both discrete and integrated – for the third quarter were 97.85 million units, up 20% in over the second quarter. Compared to the same quarter last year shipments were up 18.2%. During the quarter Nvidia Corp. increased its shipments by 25.1%, Intel Corp. boosted sales of its chipsets with built-in graphics core by 21.6%, whereas graphics product group of AMD only posted 17.8% shipments raise.

ATI CrossFire Technology Remains Open. ATI CrossFireX to Get AMD, Intel, Others’ Chipsets Support

ATI, graphics product group of Advanced Micro Devices, said on Thursday that it remained committed to its open eco-system strategy with multi-GPU technology. In fact,  every company, including Intel Corp., might get the ability to use the “ATI CrossFire” logotype to push their platforms, provided that they had passed a validation process.

“We remain committed to an open ATI CrossFire eco-system,” said Chris Hook, a spokesperson for AMD.

ATI CrossFire and ATI CrossFireX are multi-GPU technologies allow several graphics processing units to work in team to double or even triple the speed of a single high-end graphics accelerator.

Intel Corp. is the arch-rival of AMD in the market of x86 central processing units (CPUs). Besides microprocessors, Intel competes against AMD with its chipsets featuring built-in graphics core. Still, all current Intel chipsets that have two PCI Express controllers as well as future chipsets, such as Intel X48 and P45, that have reportdly passed the validation tests of AMD can use ATI CrossFire or ATI CrossFireX multi graphics chip technology.

“We are constantly working with the third-party companies, including Intel,” said Godfrey Cheng, who is in charge of ATI CrossFire multi-GPU technology at AMD.

There’s no license fee for ATI CrossFire technology, the company said.

ATI CrossFireX as well as its improved version ATI CrossFire multi-graphics processing technology that has features like 3-way, 4-way GPU rendering on a single monitor and so on.

AMD Readies “Black Edition” Phenom Processor, Phenom FX May Face Further Delay.

Advanced Micro Devices said at a presentation of its new platform in Japan that although it will not be able to offer any high-end AMD Phenom FX processors for many months from now, it will unveil “Black Edition” of its AMD Phenom 9600 chip that will allow easy overclocking capabilities already in December.

AMD Phenom processor 9600 Black Edition will operate at 2.30GHz clock-speed, 2MB shared L3 cache and 512KB dedicated L2 cache per core and will have support for dual-channel PC2-8500 (DDR2 1066MHz), just like the typical AMD Phenom 9600 chip. However, the Black Edition processor has unlocked clock multiplier, which allows enthusiasts to clock the new chip higher than 2.30GHz without any issues. If the roadmap that AMD recently unveiled in Japan is correct, then AMD Phenom 9600 Black Edition is due in December, 2007.

The customizable clock multiplier and auto-adjustable memory frequency of the new chip enables amateurs and value-conscious enthusiasts to maximize the performance of their desktop computers without substantial hassles. Typically enthusiasts have to increase base HyperTransport bus speed from default 200MHz, which eventually causes not only CPU clock-speed boost, but also alters HT bus and memory frequency, meaning that other components apart from the microprocessor also should be able to handle increased speeds. Previously customizable clock multiplier were only available on high-end AMD Athlon 64 FX processors that could retail for as much as $1000 per chip.

A slide that AMD demonstrated during AMD Phenom official launch in Japan.
Photo by Ascii.jp web-site

Earlier this month AMD said that it could not release any high-end AMD Phenom FX or even AMD Phenom processors clocked at 2.40GHz and above due to errata that causes system instability under high load. Even though the company said that the errata can be corrected by implementing a fix into BIOS of AMD Phenom-supporting mainboard, AMD itself has no plans to release current revision of Phenom at speeds beyond 2.30GHz. As a consequence, it is unclear whether AMD Phenom 9600 Black Edition provides high overclocking potential without causing system instabilities or not.

AMD did not reveal pricing of the new central processing unit (CPU), but it will hardly be considerably more expensive than $283, the price of AMD Phenom 9600 chip in business quantities.

Separately, AMD reconsidered the release timeframe of its AMD Phenom FX processors as well as AMD Phenom 8000-series triple-core chips. If earlier it was claimed that the novelties will emerge in Q1 2008, then now AMD slides imply that the new CPUs will be available only in the first half of 2008, which typically would mean that the new chips are only projected to emerge sometime in Q2, a massive delay from initially announced Q4 2007 timeframe for the new high-end enthusiast platform called FASN8.

AMD Projects Impairment Charge Due to Weak Graphics, Multimedia Business Performance.

Just one day ahead of its meeting with financial analysts Advanced Micro Devices said that the value it paid for graphics and multimedia chip designer ATI Technologies last year was too high and the actual revenues the company gets and will be able to obtain from its graphics and multimedia businesses are below expectations.

AMD: ATI Is Guilty of Everything

According to a statement with the U.S. Security and Exchange Commission (SEC), AMD concluded that the current carrying value of its goodwill which it had recorded as a result of its October 2006 acquisition of ATI Technologies was impaired. The write down will allow AMD to easily explain financial analysts why its current market capitalization ($4.97 billion at press time) is below the price of ATI it paid last year as well as poor financial results and losses with problems that allegedly existed at ATI before the merge.

“The acquisition took place at the moment, when ATI was not really leading in terms of technology… It is not like we acquired ATI and we lost market share. It was just a consequence of [ATI’s execution]: Nvidia had better graphics than ATI back then and that is why AMD lost market share. We also should consider [hardware] cycles of OEMs: if you are missing their cycles, you are out for a while. You have to be [ready] with the right [product] part at the right moment to get a cycle. When you are in, you are going to stay in for a long time. So, the main reason behind the share loss is missing OEM cycles. But we are regaining them now,” recently said Vincenzo Pistillo, director of consumer business development in EMEA region for AMD.

“This conclusion was reached based on the results of an updated long-term financial outlook for the businesses of the former ATI Technologies as part of AMD’s strategic planning cycle conducted annually during the company’s fourth quarter and based on the preliminary findings of the company’s annual goodwill impairment testing that commenced in the beginning of October 2007,” a statement concerning the write-down reads.

AMD’s Graphics Product Group Performs Below Expectations
Formally speaking, ATI Radeon X1900-series graphics cards were considerably more advanced than Nvidia’s GeForce 7900-series offerings at the time when AMD acquired the Canada-based graphics chip developer. But ATI, which at the time was already AMD’s graphics product group, could not launch its DirectX 10-supporting high-end offering last November and then was also late with mainstream DX10 graphics products.

But market share declines emerged not because ATI could not offer a competitor to Nvidia’s GeForce 8800 GTX, which still retails for $549 and higher. Just after AMD and ATI announced the transaction, partners of Intel and ATI annuled orders onto ATI Radeon Xpress-branded chipsets for Intel processors, which dramatically lowered ATI’s market share from 27.6% to 20.3%, according to Jon Peddie Research (JPR) data. However, already in Q4 2006 the market share of AMD’s graphics product group rebound to 23%, only to gradually decline to 19.1% in Q3 2007.