MacBook Air processor situation gets explained

via Engadget by Donald Melanson on 1/18/08

We already knew the basic details about the processor at the heart of Apple’s MacBook Air, but those itching to know exactly how Apple and Intel managed to cram everything into that oh so small package may want to head over to AnandTech, which has pieced together a fairly thorough report on the matter. As the site reports, the processor is based on Intel’s 65nm Merom architecture and packs an 800MHz bus, yet it uses the significantly smaller chip package that Intel had originally only planned to debut with the launch of its Montevina laptop platform later this year. That combination, along with the Intel 965GMS chipset with integrated graphics, allowed for a 60% reduction in total footprint size, and a TDP rating of just 20W, as opposed to 35W from the regular Core 2 Duo processor. If that’s still not enough MacBook Air minutia you, you can hit up the link below for the full rundown.

[Via AppleInsider]

Intel demos iPhone-like MID of the future

Intel just keeps banging out the hits from IDF. After the handful of McCaslin “next-quarter” and “coming-soon” UMPCs we saw from the chipmaker (and associates), Intel started busting out prototypes from its forthcoming Menlow chipset, using smaller, 45nm Silverthorne CPUs, and the 2009/2010 offering Moorestown… which is the bad-boy you’re looking at in these photos is based on. In a rather obvious homage to the iPhone, the chip-kingpin presented this do-anything, go-anywhere MID (provided you can cram this French-bread-sized device into a pocket). The device will feature a 45nm CPU as well, plus all kinds of goodies like integrated WiFi and WiMAX, and apparently 24 hours of battery life on a single charge. Obviously, this product will probably never see the light of day (at least not in this form factor), but then again — you never really know.

Read — Intel shows concept iPhone running on Moorestown platform
Read — Intel’s iPhone clone, we’re not joking
Read — Intel Details Next Generation “Menlow” MID, UMPC Platform

[originating url]

Intel Launches Cheaper Intel Quad-core While AMD Still Looks Dumbfounded


In addition to their mobile Extreme CPU, Intel has also announced its 3.0GHz Core 2 Extreme processor, the 65-nm QX6850 with four cores and dual 4MB Level 2 cache. The QX6850, touted as the fastest consumer processor now available, is the flagship of their new 1,333MHz Front Side Bus CPU family, which includes the Core 2 Duo E6850, E6750 and E6550, all of them with cheaper prices than the previous generation.

Intel Core 2 Extreme QX6850
3.00GHz 1333 4MBx2 $999
Intel Core 2 Duo E6850
3.00GHz 1333 4MB $266
Intel Core 2 Duo E6750
2.66GHz 1333 4MB $183
Intel Core 2 Duo E6550
2.33GHz 1333 4MB $163


[originating url]

Intel readies massive multicore processors

Ants and beetles have exoskeletons–and chips with 60 and 80 cores are going to need them as well.

Researchers at Intel are working on ways to mask the intricate functionality of massive multicore chips to make it easier for computer makers and software developers to adapt to them, said Jerry Bautista, co-director of Intel’s Tera-scale Computing Research Program.

These multicore chips, he added, will also likely contain both x86 processing cores, similar to the brains inside the vast majority of Intel’s server and PC chips today, as well as other types of cores. A 64-core chip, for instance, might contain 42 x86 cores, 18 accelerators and four embedded graphics cores.

Some labs and companies such as ClearSpeed Technology, Azul Systems and Riken have developed chips with large numbers of cores–ClearSpeed has one with 96 cores–but the cores are capable of performing certain types of operations.

The 80-core mystery

Ever since Intel showed off its 80-core prototype processor, people have asked, “Why 80 cores?”

There’s actually nothing magical about the number, Bautista and others have said. Intel wanted to make a chip that could perform 1 trillion floating-point operations per second, known as a teraflop. Eighty cores did the trick. The chip does not contain x86 cores, the kind of cores inside Intel’s PC chips, but cores optimized for floating point (or decimal) math.

Other sources at Intel pointed out that 80 cores also allowed the company to maximize the room inside the reticle, the mask used to direct light from a lithography machine to a photo-resistant silicon wafer. Light shining through the reticle creates a pattern on the wafer, and the pattern then serves as a blueprint for the circuits of a chip. More cores, and Intel would have needed a larger reticle.

Last year, Intel showed off a prototype chip with 80 computing cores. While the semiconductor world took note of the achievement, the practical questions immediately arose: Will the company come out with a multicore chip with x86 cores? (The prototype doesn’t have them.) Will these chips run existing software and operating systems? How do you solve data traffic, heat and latency problems?

Intel’s answer essentially is, yes, and we’re working on it.

One idea, proposed in a paper released this month at the Programming Language Design and Implementation Conference in San Diego, involves cloaking all of the cores in a heterogeneous multicore chip in a metaphorical exoskeleton so that all of the cores look like a series of conventional x86 cores, or even just one big core.

“It will look like a pool of resources that the run time will use as it sees fit,” Bautista said. “It is for ease of programming.”

A paper at the International Symposium on Computer Architecture, also in San Diego, details a hardware scheduler that will split up computing jobs among various cores on a chip. With the scheduler, certain computing tasks can be completed in less time, Bautista noted. It also can prevent the emergence of “hot spots“–if a single processor core starts to get warm because it’s been performing nonstop, the scheduler can shift computing jobs to a neighbor.

Intel is also tinkering with ways to let multicore chips share caches, pools of memory embedded in processors for rapid data access. Cores on many dual- and quad-core chips on the market today share caches, but it’s a somewhat manageable problem.

“When you get to eight and 16 cores, it can get pretty complicated,” Bautista said.

The technology would prioritize operations. Early indications show that improved cache management could improve overall chip performance by 10 percent to 20 percent, according to Intel.

Like the look and feel of technology for heterogeneous chips, programmers won’t, ideally, have to understand or deliberately accommodate the cache-sharing or hardware-scheduling technologies. These operations will largely be handled by the chip itself and be obscured from view.

Heat is another issue that will need to be contained. Right now, I/O (input-output) systems need about 10 watts of power to shuttle data at 1 terabit per second. An Intel lab has developed a low-power I/O system that can transfer 5 gigabits per second at 14 milliwatts–which is less than 14 percent of the power used by current 5Gbps systems today–and 15Gbps at 75 milliwatts, according to Intel. A paper outlining the issue was released at the VLSI Circuits Symposium in Japan this month.

Low-power I/O systems will be needed for core-to-core communication as well as chip-to-chip contacts.

“Without better power efficiency, this just won’t happen,” said Randy Mooney, an Intel fellow and director of I/O research.

Intel executives have said they would like to see massive multicore chips coming out in about five years. But a lot of work remains. Right now, for instance, Intel doesn’t even have a massive multicore chip based around x86 cores, a company spokeswoman said.

The massive multicore chips from the company will likely rely on technology called Through Silicon Vias (TSVs), other executives have said. TSVs connect external memory chips to processors through thousands of microscopic wires rather than one large connection on the side. This increases bandwidth.

[originating url]

Intel’s super-thin laptop sets the bar a little higher


Now this — this is a sexy laptop. A mere 0.7 inches thick, it’s nearly as thin as a Motorola RAZR phone yet still manages to pack in an impressive set of features. Inside the 2.25-pound body are all Intel components, including always-on connectivity that keeps you hooked up to Wi-Fi hotspots, EV-DO networks, or even WiMax connections, depending on what’s available.

In addition to the insanely thin form factor, there will be magnetically attached skins to give it different looks, and it even has an external screen on top to let you see information without opening the computer.

So the real question is: When can I get one? Maybe as soon as the end of the year. It’s merely a concept design by Intel, but the company seems serious about making it a reality. Don’t expect the price to be anything but astronomical, however. If the price manages to get down to MacBook levels, Intel will have a serious contender on its hands.

Business Week, via Gizmodo

[originating url]

New AMD Desktop Processors Trigger Price Drops

The introduction this morning by AMD of new dual-core desktop processors, including the Athlon 64 X2 6000+, and the company’s subsequent lowering of prices of existing processors, may have nearly or completely equalized the price/performance balance between AMD and Intel processors, according to an updated performance model using newly published data balanced against this morning’s average street prices for CPUs.

Last July’s introduction by Intel of Core 2 Duo processors enabled that company to effectively wrest the price/performance crown from AMD, which it had previously held for several years. In the intervening months, AMD has held a slim lead in the value segment – meaning that for about the same $100, you’d be likely to get a slightly better performing AMD-brand processor than an Intel. But Intel’s mid-range Core 2 Duo E6600 proved a better value than comparable AMD products by as much as $200.

Today, that gap may have been erased, as AMD’s introduction of the 6000+ at a suggested retail price of $464 in the original retail box (OEMs will pay less) has been followed up by a slash of $171 in the suggested retail price of the Athlon 64 X2 5200+, down to $232.

In fact, Intel’s very slight lead in price/performance over AMD based on our computer models could dissipate by the end of this week, as artificially high prices for pre-ordered 6000+ models, according to figures supplied by Froogle, will probably plummet to just above suggested retail.

Here’s the current situation: Based on recently updated performance figures from Tom’s Hardware Guide, our computer model suggests that, in an average of five benchmarks from varying categories, an Intel Core 2 Extreme X6800 processor delivers exactly twice the performance of a single-core Pentium 4 520. This remains the best performing processor tested thus far, with Intel’s Core 2 Quad QX6700 performing slightly lower overall.

AMD’s best performing processor tested thus far is its top-of-the-line Athlon FX-74 pair, which uses the company’s new Quad FX architecture. Still, there’s a performance gap, with the FX-74 performing only about 69% better than the P4 520, versus the Quad FX’s 100%.

Recent performance tests show the new 6000+ performing 69% better than the P4 520 – indeed, better overall than the midrange pair of Quad FX processors put together. The 6000+’s processor-in-a-box (PIB) price of $464 helps push the street price of its recently introduced Athlon 64 X2 5600+ down to $325.

When you bring Intel back into the picture, the nearest Intel processor in performance to the 5600+ is the E6600, now priced at $314 on average. That $9 represents the remainder of the price/performance gap between Intel and AMD for now.

Shoppers should take note of the fact that the 6000+ single dual-core CPU is a better performing processor in many categories than AMD’s own Quad FX double dual-core CPUs. That may raise some eyebrows for enthusiasts considering investing in a high-power platform. But for customers who are looking not so much for power but for the proper balance between power and performance, there’s a familiar knock on their door once again.

[originating url]

Moore’s Law Rescuscitated by New Intel Chip Tech


Intel and IBM have independently developed new ways to make transistors, keeping Moore’s Law going for at least two more generations of chips—down to 22 nm. Both methods involve new insulators made out of hafnium, which can be made thicker to reduce current leakage without reducing the electric charge.

If you want to skip to why you should care, basically it means faster chips that run cooler and consume less power. (Think more cores—lots of ’em—and mobile devices.) It also means that Intel maintains a nine-month lead over everybody else, with their chips made with the new tech dropping later this year, while IBM’s don’t roll out til ’08.

P.S. Neither article is written very clearly, so I recommend reading both of them.

Intel Says Chips Will Run Faster, Using Less Power [NYT]
Moore’s Law seen extended in chip breakthrough [Reuters]

If seeing chip factories from the inside is your thing, Robby Scoble’s got some videos here.

[originating url]

AMD: Go to ‘Barcelona’ over ‘Clovertown’

When it comes to quad-core chips, good things come to those who wait, Advanced Micro Devices believes.

After years breathing AMD’s dust, Intel beat its rival to the punch by releasing its quad-core Xeon 5300 “Clovertown” processor for servers in November. But AMD’s “Barcelona” quad-core chip, due to arrive midway through 2007, will be a significant notch faster than the Clovertown chips expected to be on the market at that time, said Randy Allen, AMD’s corporate vice president for server and workstation products.

“We expect across a wide variety of workloads for Barcelona to outperform Clovertown by 40 percent,” Allen said. The quad-core chip also will outperform AMD’s current dual-core Opterons on “floating point” mathematical calculations by a factor of 3.6 at the same clock rate, he said.

Trumpeting the performance of unreleased products is not a strategy unique to AMD. When launching the Xeon 5100 “Woodcrest,” chip Intel said its chips would beat AMD’s by at least 40 percent.

Having a quad-core product has helped restore Intel’s fortunes, and not just by buffing the chipmaker’s image, Mercury Research analyst Dean McCarron said. “Early indications are that Clovertown is contributing a meaningful amount of business to Intel in a surprisingly short period of time,” McCarron said. “It’s not marketing fluff.”

For its part, Intel points to Sun Microsystems’ announcement on Monday that it would sell Intel-based servers, ending its reliance on AMD as its exclusive x86 server chip supplier. “We’ll let our competitors talk the talk while we walk the walk,” spokesman Bill Kircos said, adding that Intel will improve performance and energy efficiency compared to today’s products.

In the third quarter of 2006, Intel’s restored competitiveness helped Intel reclaim server processor market share lost to AMD. Mercury Research figures for the fourth quarter aren’t yet available, but AMD warned earlier this month that its chip-selling prices were “significantly lower.”

Clovertown “has allowed Intel to put some pricing pressure on AMD. Intel can tout a lower price per core, given that it’s pricing much of the quad-core Xeon 5300 line the same as its dual-core Xeon 5100 chips,” said Technology Business Research analyst John Spooner.

“AMD has to respond to that by offering lower Opteron prices to satisfy those customers who might look at switching between now and when Barcelona comes out,” Spooner said. “Discussing Barcelona performance is another way to help stave off Intel for the time being.”

AMD stands by its Barcelona engineering decisions, though, including its choice to build a single chip with four cores rather than employing Intel’s dual-core, dual-chip package approach. AMD calls its approach “monolithic” or “native” quad-core.

Customers don’t care whether chips are monolithic or combine separate processors, Allen said, but they do care about performance. “We came to the conclusion that, given the capabilities and performance with the monolithic design, it was clearly the right answer,” Allen said.

Barcelona employs several features to improve performance, Allen said. Among them:

• It’s AMD’s first chip with a built-in level-three cache. Cache memory can respond faster than main memory, and Intel has relied on large amounts of cache to improve its processors’ performance. Each Barcelona core has its own 64-kilobyte first-level data cache, 64KB first-level instruction cache and 512KB second-level cache; and the four cores together share a 2MB third-level cache, though AMD has said that size can be increased.

• AMD redesigned the Barcelona core, marking the biggest changes since the company made its 2003 transition from its 32-bit Athlon chips to the current 64-bit lineup. The magnitude of the transition is about halfway between the small tweaks AMD has made to Opteron over the years and the clean-sheet redesign Intel employed in moving from NetBurst to its current Core design, Allen said.

• A faster floating-point engine performs mathematical calculations–long an Opteron strong suit, though not as important a part of the chip as that for integer operations. At a given clock frequency, a Barcelona core outperforms a current Opteron core by a factor of 1.8. By going quad-core, a Barcelona chip overall will provide a boost factor of 3.6, Allen said.

Not all things are better, though. Specifically, Barcelona’s clock frequency will be lower than that for the company’s dual-core chips. That’s a common situation because quad-core chips require more circuitry, and more circuitry means more power consumption and waste heat, unless the chips run slower.

AMD is moving its manufacturing from a 90-nanometer process to a 65-nanometer process, permitting more circuitry to fit in a given amount of chip real estate. Even with that change, the quad-core chips will run more slowly, Allen said. He argued that it’s worth the tradeoff, though, since the additional cores can run more jobs simultaneously, even if an individual job isn’t completed as swiftly.

“The slight degradation for frequency with quad-core will be overwhelmed by the increase in performance from dual-core to quad-core,” Allen said. He declined to mention the chips’ frequency.

[originating url]

Intel sez Penryn’s done, lookout for 45nm Wolfdale / Yorkfield

If you’re desperate for some positive Intel news after hearing those less-than-inspiring margin forecasts earlier today, the chipmaker is once again keeping itself on track in regard to pumping out its forthcoming 45-nanometer processors. While we were briefed on the dual-core Wolfdale and quad-core Yorkfield just a few weeks back, Intel is now claiming that its Penryn-based chips are “complete” and will play nice with Windows Vista, Mac OS X, and Linux operating systems when they hit the shelves. Both chips are (still) slated to hit production during the second half of this year, with manufacturing to hit full stride during 2008. Penryn is supposed to “extend the Core 2 architecture” by playing host to the next set of Intel’s Streaming SIMD Extensions (SSE), and will also lend a hand in the future development of Montevina, and just in case you’re the (way) forward-looking type, you can expect Penryn’s successor — dubbed Nehalem — to roll out in late ’08.

[originating url]

Vaio VGX-TP1 Media Center PC

Here’s the Vaio VGX-TP1 Media Center PC at the CES. This round PC is powered by Intel T5600 Core 2 Duo processor at 1.83 Ghz running on Intel 845 GM chipset (Viiv technology). It comes with 2GB of DDR-RAM, 300GB Hard Drive, an Intel GMA 95 chipset for handling graphics and it runs on Windows Vista Home Premium edition. This PC also includes a matching wireless keyboard, a remote control and Wi-Fi antenna which support IEEE 802.11b/g (there’s no built in Wi-Fi). The I/O port panel is concealed by a magnetic cover, open it and a host of connections such as HDMI, RGB, Ethernet, AV input, USB etc are unveiled. It also comes with a TV tuner allows you to record you favorite show.

[via New Launches]