Nanotech - The Circuits Blog

Globalfoundries, the manufacturing concern spun off from Advanced Micro Devices, plans to announce Wednesday that it has signed up STMicroelectronics--its first true outside customer.

STMicroelectronics supplies the accelerometer for the iPhone 3GS

(Credit: Apple)

Globalfoundries was created last year in order to eliminate the crippling overhead that AMD was incurring to manufacture its processors. Globalfoundries now conducts business as a contract chip manufacturer, commonly referred to as a foundry.

AMD owns 34.2 percent of the company, while Advanced Technology Investment Co. owns the rest. ATIC is an investment company wholly owned by the government of Abu Dhabi, which is part of the United Arab Emirates.

Last week, Globalfoundries broke ground on a $4.2 billion facility in Malta, N.Y., that is expected to put it among the elite chipmakers of the world. Currently, Globalfoundries manufactures chips for AMD at facilities in Dresden, Germany.

STMicroelectronics will commission Globalfoundries to make low-power chips using a 40-nanometer process, which "is ideal for the next generation of wireless applications, handheld devices, and consumer electronics," according to a statement. Production is slated for 2010.

Currently, the most advanced manufacturing processes in the chip industry are at the 34-nanometer level--which Intel uses to manufacture flash memory chips. More typically, chips are made on advanced processes ranging between 40 and 50 nanometers. Generally, the smaller the chip's geometries, the faster and more power efficient it is.

Globalfoundries would not say what specific types of chips it will make for STMicroelectronics, whose product portfolio includes memory, microcontrollers, power management silicon, and MEMS or Micro-Electro-Mechanical Systems.

One of the highest profile products STMicroelectronics supplies today is the accelerometer for the iPhone 3GS. Based on MEMS technology, the accelerometer allows the 3GS to determine device orientation or inclination.

STMicroelectronics was ranked among the top 5 chipmakers in the world based on revenue in 2008, according to market researcher iSuppli.

Micron Technology is entering the graphics memory business, going up against heavyweights Samsung and Hynix.

Micron is targeting its memory at the upper mid-range of the graphics chip market

(Credit: Nvidia)

Micron, which recently vaulted to the No. 3 spot in global sales of dynamic random-access memory (DRAM), is now aiming at the market for DRAM chips used with graphics processors from Nvidia and Advanced Micro Devices' ATI graphics unit.

The market for DRAM used with graphics processors is about 4 percent of the bits shipped into the DRAM market, according to Micron. DRAM is typically used as the main memory in PCs. This type of DRAM is also referred to as Synchronous DRAM, or SDRAM.

"Our upcoming 50-nanometer technology is very competitive when it comes to power consumption and performance," Robert Feurle, Micron's VP of DRAM marketing, said in a phone interview Thursday.

"I think it's a good point in time to begin discussions with big enablers Nvidia and AMD and get started with some design-ins," Feurle said.

Micron is making its debut with Double Date Rate 3 (DDR3) memory. This is the same type of memory used for the main memory of currently shipping PCs, which have gravitated from DDR2. In the future, Micron will look at making more proprietary graphics memory, referred to as GDDR3 and GDDR5. "No decision has been made yet but we're looking into that very seriously," Feurle said.

Initially, Micron is targeting the "upper mid-range" of the graphics processor market.

Micron says its DDR3 has a distinct power consumption advantage over GDDR3: standard DDR3 can go down to 1.35 volts. "GDDR3 is still running a 1.8 volts. We have a giant power savings advantage," he said.

Micron is targeting memory with speeds of 1600MHz "to get started with and going up from there," Feurle said.

The DRAM market overall has seen sliding sales, falling 20 percent in the first quarter from the fourth quarter and 44 percent from the year-earlier period, according to iSuppli. The problem is overcapacity, which has most notably brought Taiwan memory makers to their knees. In that country, some manufacturers have faced possible bankruptcy.

"Micron now has renewed its competitive vigor, mainly due to its acquisition of a 300mm fab from Inotera in Taiwan," iSuppli said recently. Fab refers to fabrication facility or factory.

ATI graphics drivers will now be delivered in one tidy package for both Windows Vista and Windows 7.

Advanced Micro Devices said Wednesday that it has released Catalyst 9.3, a set of new graphics drivers that constitute a "unified" driver installation package, with support for both Windows Vista and Windows 7.

One of the biggest changes for Windows 7 is support for the Windows Display Driver Model 1.1, an update from WDDM 1.0 used in Vista, according to Andrew Dodd, a software product manager at ATI. "(Catalyst 9.3) is one single binary and if you install the driver under Windows 7, it automatically supports all the WDDM 1.1 features. If you install it under Windows Vista, it supports the WDDM 1.0 features," he said Wednesday.

Dodd said ATI is trying to get out in front of the official Windows 7 release. "We're trying to show the world where we are with our Windows 7 driver development. From now on, every single monthly Catalyst driver that comes out will include Windows 7 support (in addition to Vista)," he said. Catalyst 9.3 can also be used with the Windows 7 beta, Dodd said.

So, what kind of improvements will users see? "With Vista, if you had really high resolution and you had a number of windows open, you'd get to the point where you'd just run out of memory. In Windows 7, this will become much more efficient as far as memory usage and you'll never run into that situation. You can have as many windows as you like open at any resolution," he said.

Other features that are seeing changes are the Windows desktop and the Aero interface. Under Vista, DirectX 9 was required. "For Windows 7, in order to have the WDDM 1.1 driver certified (by Microsoft), you have to have DirectX 10 hardware," according to Dodd. "The (Windows) desktop was actually designed using the DirectX 10.1 API (application programming interface)."

The ATI Radeon 3000 and 4000 series of graphics chips have DirectX 10.1 support, Dodd said. The Radeon 2000 series will also be "fully operable under Windows 7," he said, though Radeon 2000 won't support DirectX 10.1. One of the biggest advantages of having 10.1 hardware is its readiness for DirectX 11, which will be an extension of 10.1, according to Dodd.

Gamers will also see performance improvements, Dodd said. "In our internal testing we're actually, in a number of cases, seeing that games are running faster under Windows 7 versus Windows Vista."

In a news release Wednesday, ATI also listed other improvements, including support for the Direct2D API introduced in Windows 7, giving third-party applications the ability to improve ClearType text rendering and hardware-accelerated vector graphics.

A new report says the longstanding integrated graphics chip market will disappear--a market that Intel currently dominates.

In a report entitled "Integrated graphics chip market to disappear by 2012," Tiburon, Calif.-based Jon Peddie Research forecasts the end of the market for "the popular integrated graphics processor chipset...after 15 years of stellar growth."

Many low-end and mainstream consumer and business laptops sold over the last five years use Intel graphics built into the chipset, a low-performance but cheaper alternative to discrete graphics chips from Nvidia and ATI. That's made Intel, ironically enough, the market leader in graphics chip market share despite the performance deficit. In 2008, Intel's market share, by quarter, was typically over 50 percent, while Advanced Micro Devices (ATI's parent company) and Nvidia bounced around in the 20 to 30 percent range.

Overall, in 2008, 67 percent of the graphics chips shipped were integrated. In 2011 this will drop to 20 percent, and by 2013 it will be less than 1 percent, according to Jon Peddie.

Intel is moving graphics into the same package as the processor.

(Credit: Intel)

So what will take its place? "Embedded" graphics built into the same chip package--and later onto the same piece of silicon--as the main processor, according to the report. In the fourth quarter of this year, Intel will bring out a 32-nanometer mobile processor code-named Arrandale that integrates graphics silicon into the same chip package as the main processor, or CPU. Advanced Micro Devices has plans to "fuse" the CPU (central processing unit) and GPU (graphics processing unit), probably sometime in 2011.

All of this has already happened in the smartphone market. Processors from Qualcomm, Texas Instruments, and Nvidia put graphics on the same piece of silicon as main processor.

And what about Nvidia and ATI and their high-octane graphics chips? "This will not, as many believe, impact the discrete graphics and add-in board market. In fact, with hybrid configuration, embedded graphics will enhance the discrete GPU sales," the report said.

Between 2010 and 2012 there will be three choices for graphics, according to Jon Peddie: traditional discrete GPUs mounted on add-in boards and/or the motherboard, integrated graphics processor chipsets, and processors with embedded graphics. One or more of these devices will be employed in PCs.

A short history lesson, courtesy of Jon Peddie: the first integrated graphics controller was Sun Microsystems' Legos, which came out in 1989 for the company's SPARC processor. The first integrated graphics controller for the PC was introduced by Silicon Integrated Systems for Intel processors in 1997.

Updated at 3:45 p.m. PST with correction of Motorola, IBM executives' names.

Even the biggest chip companies churn out their share of flops. But the hype that surrounds these chips is more fascinating than the failures.

It's been almost a year since I posted A brief history of chip hype--and flops (part 1). Consider this Part 2.

Itanium
First, I have to revisit Intel's Itanium. Simply because it's still around and still missing production target dates.

Intel's Itanium has been relegated to obscurity if not practical oblivion

(Credit: Intel)

The hype: "This design philosophy will one day replace RISC and CISC. It is a gateway into the 64-bit future." This copy was, at one time, posted on Hewlett-Packard's Web site. And analysts were drinking the Kool-Aid too. "I expect Itanium to replace Xeon, but not until 2003," one analyst said back in 2001. (Xeon is Intel's successful, lucrative line of server processors that doesn't include Itanium.)

The reality: Yes, Itanium is still warm, still breathing in the rarefied very-high-end server market--where it does have a limited role. But its architecture will never live in a desktop or laptop or even 99 percent of the servers as once thought. And it certainly hasn't remade the computer industry. And it is still chronically late. This time it's Tukwila that's tardy. The quad-core version of Itanium is late because Intel had to make "some engineering enhancements to the Tukwila platform," according to an Intel statement earlier this month. I can only guess that one day Intel will finally let this failed research project go cold and die quietly.

PowerPC
IBM's original PowerPC platform never lived up to the hype. Even when Motorola and IBM processors populated Apple computers.

The hype: "The PowerPC G5 changes all the rules. This 64-bit race car is the heart of our new Power Mac G5, now the world's fastest desktop computer," said Apple CEO Steve Jobs back in 2003. Jobs, a master of hype (also referred to as a Svengali-like reality-distortion field), continued with this precious quote. "IBM offers the most advanced processor design...and this is just the beginning of a long and productive relationship." (Emphasis added.)

The reality: Apple dumped IBM, Motorola, and the PowerPC in 2005 and it was revealed later that the Mac operating system had been leading "a secret double life" for about five years. But the PowerPC platform had really failed long before 2005. Look no further than these comments from an IBM marketing manager in this 1997 Electronic News article: "Many business school case histories will be written about this failure," Jesse Parker, marketing manager at IBM Micro, said at that time. "No one of the three companies involved in PowerPC executed on their plans. IBM didn't. Motorola didn't. And Apple didn't," he said.

The original PowerPC project was conceived by John Sculley, president and CEO of Apple, and Jack Kuehler, vice chairman of IBM. Phil Hester, an IBM manager at the time, and David Mothersole, a Motorola executive, where also instrumental in starting the project, known initially as "Somerset." But as the PowerPC came to market, Mr. Sculley was pushed out of Apple (and) Mr. Kuehler retired. Their replacements did not have the same enthusiasm for the PowerPC alliance, dooming the project.

In short, the PowerPC failed to challenge Intel in the PC market in a big way. (Though it has been reincarnated as IBM's Cell processor that powers Sony's PlayStation and the architecture still powers IBM servers.)

And I have my own vignette to relate that illustrates one reason why Apple eventually dropped the PowerPC. When Apple first began to crow about the dual-processor Power Mac (circa 2003), a neighbor of mine at the time bought into the hype and purchased an Apple Power Mac tower with two IBM G4 processors (this preceded the dual-processor G5 tower that followed soon after). This thing was a furnace. It quite literally raised the temperature in the room it was in, had about five fans too many, and was deafening, to boot. That was the first time I fully understood the magnitude of Apple's fabrications about IBM's "superior" PowerPC designs. (IBM's less-than-impressive--at that time--chip manufacturing process that was used for PowerPC processors also contributed to the problem.)

AMD Puma
Lastly, turning to Advanced Micro Devices, I'll try to look beyond the botched Barcelona launch in September of 2007 (as I've already covered this in Part 1) and focus instead on AMD's mobile "Puma" platform. Though I can't leave Barcelona entirely out of the discussion because there are some disturbing parallels. (Note: AMD's upcoming Yukon and Congo platforms offer some hope for mobile redemption.)

The hype: Like Barcelona, AMD had too much to say about Puma too long before it was real. AMD started pumping Puma back in April 2007 when the company did one of its many (infamous) soft launches (a PR strategy that it has thankfully ditched). This prompted some editorializing from me as well as other publications. "It is questionable whether...Puma will meet the hype AMD is currently trying to generate though these early announcements," according to a rare editorial from DigiTimes in 2007.

Things got even more dicey at the financial analyst day in December 2007 when AMD said Puma would be delayed until the second quarter of 2008.

It's not so much that Puma (aka Turion X2 Ultra coupled with ATI graphics) is a failure of epic proportions like Itanium, it's that the CPU component (separate from the ATI GPU component) fell so far short of the long, ballyhooed build-up it got.

And it is beaten consistently by Intel in the mobile marketplace. Here's an October 2008 CNET review of laptops with AMD's Turion X2 Ultra. "Turning to AMD's 2.0GHz Turion X2 Dual-Core RM-70 CPU might be an option if you're looking to keep costs down and have only basic computing needs," the review said. "The Acer Aspire 5735-4624 costs only $499 and uses a 2.0GHz Intel Pentium Dual Core T3200; it completed our multitasking benchmark test in one-third the time the HP G60 did. The HP G60 wasn't the last-place performer in our mainstream midprice holiday retail laptop roundup. That dubious distinction goes to the Toshiba Satellite L355D-S7825, also an AMD-powered system."

And there are more unfavorable comparisons. This review at Hexus.net of a Toshiba Satellite 300D with AMD Turion X2 Ultra ZM-80 said the Turion X2 Ultra CPU was "found wanting when compared to Centrino 2." The one bright spot was the ATI Mobility Radeon HD 3650 graphics card--but this speaks more about solid ATI technology than AMD's shaky Turion processor.

Here's the silicon scuttlebutt of the weekend, if not the week: Sony will use Intel's Larrabee graphics chip in its upcoming PlayStation 4. (Let's not forget the other tantalizing piece of speculation this week: the Nvidia-powered Microsoft smartphone rumor, which Microsoft apparently put to rest.)

We know for a fact that Jeffery Katzenberg at DreamWorks likes Larrabee--a lot. That apparently was one of the reasons DreamWorks dropped Advanced Micro Devices.

So, chalk that up as one big win for Intel's somewhat-murky next-generation graphics chip due late this year or 2010. Now Sony? A report this week in the U.K.-based technology Web site The Inquirer claims Sony favors Larrabee over Nvidia for its PlayStation 4. (The other major piece of silicon used in the current PlayStation is a Cell processor developed jointly by IBM, Sony, and Toshiba.)

For the record, an Intel spokesperson said the company "cannot comment on rumor or speculation." Sony in Europe reportedly didn't mince words, however, comparing the report to some of the 20th century's great fiction. Though another reported comment from Sony is more insipid and PR-like.

The U.K. report claims Intel paid to play. The report also hinges on the premise that Sony doesn't like Nvidia anymore. (And claims there are others that feel the same way about Nvidia.) Even if there is some special hatred there (as the reporter claims), that's not news--and applies to just about any acrimony-ridden hardware relationship in Silicon Valley. (Just peruse some of the tender exchanges between Intel and Microsoft in court records over the years.)

Anti-Nvidia bias (which is palpable in the report) aside, if there is a broader truth to this, that is, that game box makers are considering Larrabee, the chip would become a serious contender and take its place with GPUs from Nvidia and AMD's ATI graphics unit. But we won't know this for a while since no one (that I know of) has actually put Larrabee through the paces (though DreamWorks has hinted at this). And the PlayStation 4 isn't due, reportedly, until 2012.

Jon Peddie Research said Thursday that estimated graphics chip shipments in the fourth quarter plummeted 34 percent from the third quarter--a very atypical month-to-month decline.

AMD, Intel, Nvidia 2008 laptop graphics chip market share by quarter

(Credit: Jon Peddie Research)

"The fourth quarter is usually a positive quarter for the computer industry. There has obviously been some inventory problems in the quarter," said Jon Peddie, president of Tiburon Calif.-based research firm, in a statement.

And it will get worse. "We're...bracing for what will probably be the worst Q1 and Q2 decline we've seen since the Internet bubble pop of 2000," he warned.

Total shipments of GPUs for the fourth quarter of 2008 were 72.35 million, down 28.5 percent from 100.5 million GPUs shipped in the same quarter last year. And in the third quarter of 2008, 111.26 million units shipped, representing a 34 percent sequential quarter-to-quarter drop. (GPU stands for graphics processing unit.)

"Vendors were bracing for a slower than usual quarter due to economic factors, but performance this quarter was surprisingly low," Peddie said. He attributed it to the usual suspects: the worldwide financial market decline and sapped consumer spending.

In the desktop segment, Nvidia has held a slight edge with a market share of 37.9 percent, gaining on Intel, Peddie said. AMD also gained market share on the desktop, going from 20.3 percent in the third quarter to 21.4 percent in the fourth quarter. "Given the tough circumstances of this quarter, AMD's increase is no mean accomplishment," Peddie said.

Discontent with Intel graphics goes back a few years. But the unsealing of 3-year-old e-mail exchanges between Intel and Microsoft reveals something about the present, too.

Intel 915 chipset

(Credit: Intel)

First some background. Intel makes integrated graphics silicon--that is, graphics functionality that is built into its chipsets. Performance is not the name of the game for Intel. Delivering power-efficient, adequate graphics that can handle everyday tasks and do basic gaming is the goal. Anything beyond this is left to the high-octane discrete chips from ATI and Nvidia.

"We've always been consistent that high-end gamers should use discrete graphics," said Intel spokesman George Alfs. Intel graphics is also inexpensive and comes virtually free on some PCs.

But Intel graphics silicon is everywhere. It ships in tens of millions of PCs every year. And herein lies the issue. The silicon becomes the lowest common denominator that Microsoft and game developers must write to because it's so ubiquitous.

This is the root of the Intel 915 integrated graphics and the "Vista Capable" controversy. As widely reported, Intel's 915 (which shipped as standard in many PCs) was not up to running Vista's Aero Glass interface (among other features). So, Microsoft dropped this as a requirement.

Reams of material have been released according to this Seattle Times blog documenting the infighting that took place trying to resolve the 915 issue. The documents stem from a lawsuit that alleges Microsoft misled consumers by lowering the requirements so a 915-based PC could be designated as "Vista Capable."

According to an unsealed motion citing e-mail and internal Intel and Microsoft documentation released by U.S. District Court Judge Marsha Pechman, Microsoft objected to an internal Intel link "positioning the 915 GM as optimum for Windows Vista on mobile PCs." The motion states that Microsoft viewed this as "misleading" and "egregious" and that Microsoft asserted that the 915 chipset "should not even be in the list of recommended hardware for Windows Vista" and further opined that the "higher end of the chipset choices" from Nvidia and ATI were more suitable.

But that may not be the whole story. According to an article on Channel Web, Microsoft did not "cave" to Intel and the 915, but rather "it was Microsoft, led by Poole, that initiated that change all on its own." Will Poole at that time was a Microsoft senior vice president.

"We are seriously confused. We believed that 915 is NOT vista ready as it will never have WDDM drivers," according to an e-mail from Intel Vice President Renee James, cited in the Channel Web article. (WDDM stands for Windows Display Driver Model.)

Whatever the case, Intel integrated graphics was so commonplace that it was a big issue.

(For the record, Nvidia had issues with its drivers and Windows Vista too.)

Intel targets graphics
Fast-forward to September of 2006 and the Intel X3000 and X3100 (G965/GM965) graphics. With this silicon, Intel decided it was going to provide a better graphics experience for gaming in particular. The 965 started shipping in September of 2006, but it took Intel nearly a year to write the drivers needed to unlock better performance.

"New drivers for the company's 965GM chipset, found in many notebooks and midrange desktops, still don't deliver the uniform performance increases promised earlier this year, according to testing by CNET Labs," CNET News' Tom Krazit wrote in October 2007.

Intel documentation (here) says that "Intel recently introduced the 15.6 and 14.31 Windows Vista and Windows XP graphics drivers that enables Shader Model 3.0 including support for hardware vertex shader and HW TnL on the Intel G965, GM965, and G35 Express Chipsets."

The document continues: "This capability has shown enhancements in game compatibility as well as game play" and concludes the "Introduction" by saying: "The end result is that Intel is able to deliver the highest possible frame rates by leveraging Intel's world class processors."

Now fast-forward to the present and the MacBook Air. The first version of the MacBook Air was rolled out in a show of great camaraderie with Intel CEO Paul Otellini. Intel silicon all around: not only a special version of the Intel mobile Core 2 Duo was used, but Intel X3100 graphics, too. At that time, Apple CEO Steve Jobs heaped praise on the Core 2 Duo processor.

Then came the MacBook Air update. Intel graphics out, Nvidia 9400M graphics in.

This time Apple stressed the graphics capability of the Air.

Gains and compromises
To reiterate, the issue is not that Intel graphics are horrendous. It's simply that Intel's graphics silicon is so widespread that it becomes an issue for people, for example, who buy a laptop and later decide they want to play games at a certain level or do more high-level graphics.

What do analysts think about the X3100? Jon Peddie says Intel graphics has improved, but he is cautious. (Note that the X3100 has recently been superseded in laptops by the Intel GMA 4500MHD.)

"Whereas it would never be used by a real gamer (of which I like to consider myself) it will allow someone with a tighter budget to have some experience (with gaming on a PC)," Peddie said in response to an e-mail query. Peddie does research and testing of graphics products from Intel, Nvidia, and ATI.

Peddie: "Based on early tests we have run on the X3100, we found it ran all the games we tried, i.e., Spore, Stalker Clear Sky, Crysis, and Far Cry Warhammer, but "mind you we had to use lower resolution than we would normally, and if the game didn't automatically turn off some of the special features, we had to in order to get a descent frame rate."

He continues: "But the fact that it ran at all is I think a major slap on the back for Intel. Turning features off and reducing resolution is a reasonable compromise considering the costs."

But Intel (to state the obvious) is not Nvidia. "Now having said that I also have to say that the Nvidia mGPU 9400 (now used in the MacBook Air) is much more capable and you can run at higher resolutions with more features turned on," Peddie said.

The conclusion. Intel graphics is adequate and probably does more than enough for most users. But the issue will never go away because integrated graphics set itself up as a low-watermark benchmark for competitors (that offer higher-end discrete cards) to surpass. Meanwhile, it forces multimedia and game developers to make their games and applications run in a less-than-stellar way on millions of PC worldwide.

Advanced Micro Devices' new graphics chips are taking market share from Nvidia, a report issued Wednesday confirmed.

"AMD has by all accounts exceeded expectations with its Radeon HD 4000 series," according to report issued by market researcher Jon Peddie Research (JPR).

Aggressive pricing by AMD's ATI graphics unit made the difference, bringing down prices on add-in graphics boards. "Priced aggressively yet delivering solid performance, AMD's new line not only took back some market share--jumping up to 40 percent from 35 percent the quarter prior--it forced Nvidia (and partners) to cut prices on its recently released GTX 200 series product," JPR said.

Prices for graphics boards based on Nvidia GTX 260 and 280 graphics chips were slashed back in July. This happened only a few weeks after Nvidia launched the chips.

"Discounts cut into ASPs (average selling prices), taking a toll on revenue for both Nvidia and the market overall, the latter down 27 percent (year-over-year) to $3.8 billion," JPR said.

Graphics add-in board unit shipments were up sequentially but down year-to-year

(Credit: Jon Peddie Research)

The sequential growth for add-in boards tracked the growth in the aggregate market for graphics chips (which also includes motherboard-integrated products). The latter saw an increase (sequentially) in unit shipments of 17.8 percent in the third quarter. Unit shipments for add-in boards increased 11 percent to 21.9 million sequentially but fell 15 percent year-to-year.