Maximum PC technology RSS Feed

Another Day, Another Promising New Battery Technology

Paul Lilly — Tue, 24 Nov 2009 08:30:50 -0600

It seems like we're constantly hearing about promising battery technologies that could ultimately lead to longer battery life, more power, and smaller units, but as of yet, that big breakthrough hasn't occurred. Maybe nanotechnology, which is the current hot topic in the battery innovations field, will prove to be different.

Right at this moment, a ton of research is being put into carbon nanotubes (CNTs) for a bunch of uses, including electronics and batteries. Researchers are drawn to CNTs because, according to them, carbon nanotubes are near perfect. That has paved the way for a professor and a UC San Diego graduate student to discover a breakthrough that involves introducing purposeful defects into CNT structures. By doing so, the 'defective' CNTs actually work better for the development of super capacitors, DailyTech reports.

"While batteries have large storage capacity, they take a long time to charge; while electrostatic capacitors can charge quickly but typically have limited capacity. However, super capacitors electrochemical capacitors incorporate the advantages of both," Professor Prabhakar Bandaru said.

The duo also discovered that other methods, such as bombarding CNTs with argon or hydrogen, could also increase or decrease the charge capacity. In the end, the two researchers believe that their discovery could ultimately lead to electronics that charge faster and last longer than what's available today.

Image Credit: jacobsschool.ucsd.edu

"Unfriend" is Oxford's Word of the Year

Bart Salisbury — Mon, 16 Nov 2009 16:53:11 -0600

The modification of the English language, due to advances in technology, continues apace. The latest addition to the language, and “word of the year” in fact, is “unfriend.”

Unfriend, a verb that means “to remove someone as a ‘friend’ on a social networking site such as Facebook” is the New Oxford American Dictionary’s word of the year for 2009. A simple example of usage, according to the Dictionary: “I decided to unfriend my roommate on Facebook after we had a fight.”

Christine Lindberg, a Senior Lexicographer for Oxford’s U.S. dictionary program says the word “has both currency and potential longevity. Lindberg notes “most “un-” prefixed words are adjectives (unacceptable, unpleasant), and there are certainly some familiar “un-” verbs (uncap, unpack), but “unfriend” is different from the norm. It assumes a verb sense of “friend” that is really not used (at least not since maybe the 17th century!).” Unable to resist the pun, Lindberg adds: “Unfriend has real lex-appeal.”

Other new technology-generated words in competition for the award were “hashtag,” “intexticated,” “netbook,” “paywall,” and “sexting.”

Image Credit: Oxford University Press

2010 Technology Preview

The Maximum PC Staff — Fri, 16 Oct 2009 10:00:00 -0500

We've seen the future and it's full of new and exciting hardware for power users

So much in life is unknowable. Will the economy rebound? Hard to say. Will oil prices skyrocket? Maybe, maybe not. Will Brangelina add to their brood? Frankly, we don’t care. But one thing’s for sure: Technology is ever-changing and each year guarantees new advances for the PC user.

As we do every year around this time, we got on the horn with our industry contacts—experts in their respective fields—and pressed them for details about what new and exciting hardware power users can look forward to in 2010. Some of what we learned was expected (SATA speeds will double), some came from out of left field (six 30-inch panels on a single videocard?!), and some just plain make sense (like a Nehalem chip for the masses).

Read on to find out how your personal computing landscape stands to be altered in the year ahead.

Core i7 Goes Mainstream

Intel's latest troika of new CPUs brings Nehalem goodness to the masses

Nehalem for everyone! That simple sentence best explains Intel’s brand-new series of CPUs, which is sure to please budget users everywhere while confounding power users.

Why would a new CPU that gives you the best bang for the buck in town be greeted nervously? Because Intel’s new CPU brings with it a new socket as well as a new infrastructure. This new infrastructure is essentially a fork in the road that forces users to make a difficult choice: Save money today but get locked out of the high-end, or splurge today knowing that the budget CPU is damn near as good as the top-end part.

For the details on Intel’s new budget monster, savor our full report, consume the specs, and then digest the benchmarks to see just which path your next PC should take.

AMD Pins Hope on 32nm Parts

New Orochi core, based on Bulldozer, will see the light in 2011

Even before Intel released its first Core i7 last year, AMD didn’t really have a part that could compete in the high end. Now the company says it will get back in the ball game—but not before 2011.

That’s when its enthusiast Orochi chip is expected to begin shipping. Based on the new “Bulldozer” modular microarchitecture, Orochi will have four or more cores and 8MB of cache, and dump support for DDR2 in favor of DDR3. The company originally hoped to have Bulldozer out much sooner using its existing 45nm process, but has had to postpone the chip so it can be built on a 32nm process. AMD’s fab partner, Global Foundries, won’t actually have that up and running until next year.

For mainstream users, AMD plans to release a version dubbed Llano. Llano will have 4MB of cache and DDR3 support, and some versions will feature integrated graphics cores.

Until Orochi and Llano arrive, however, AMD will have to rely on its existing Deneb and Propos cores. The big question that’s up in the air is whether Orochi will introduce a new socket design or not. AMD isn’t saying and observers seem split on whether AMD can continue to use the AM3 socket for the redesigned chip. Reusing AM3 would make the loyalists happy, and AMD has been far more careful not to force its users to buy new motherboards, so AM3-compatibility wouldn’t surprise us.

Still, with a new budget Core i5 part already faster at far lower clocks, 2011 is a long time away for the AMD faithful.

PCI Express 3.0

New spec removes bottlenecks and improves throughput, but when will we see it?

PCI Express 3.0 will offer a substantial increase in both bandwidth and efficiency over the existing PCI-E 2.0. A good thing, as bandwidth requirements are being pushed ever higher by the increasing capability of graphics cards, with frame buffers now at two gigabytes on high-end cards; increasing graphics features in DirectX 11; and demands made by multiple GPUs on a single card.

PCI Express 3.0 pushes the peak bandwidth from five to eight gigatransfers per second. The new standard will also use 128- and 130-bit encoding, rather than the current 8/10-bit encoding, which should improve efficiency. Additional features include optimizations for improved signaling and better data integrity, while maintaining backward compatibility with PCI Express 1.0 and 2.0 hardware.

However, it’s an open question as to when we’ll actually see PCI Express 3.0–capable hardware. The PCI-SIG, the standards body responsible for PCI Express 3.0, has pushed back the final definition for the new version until mid-2010, with hardware availability pushed to mid-2011. The strong need to ensure backward compatibility as well as a high degree of reliability have been cited as the reasons for the delay.

Even the new generation of DirectX 11 graphics cards, due to hit the streets this fall, will only be PCI-E 2.0 cards. The recently announced Intel P55 motherboards are also built with PCI-E 2.0 slots. The good news is that bandwidth limits aren’t likely to hit a wall with the new generation of GPUs. So, while it’s disappointing to see PCI Express 3.0 pushed back, we won’t suffer performance bottlenecks in the near term.

Graphics

Hang onto your wallets—a new generation of videocards is about to arrive

The speculation about AMD's Direct3D 11 graphics card offering has come to an end. AMD's Radeon 5870 is the fastest single-GPU videocard we've ever tested, packing 2.15 billion transistors and 1600 shader units into an RV870 chip built on the 40nm process.

What about Nvidia? The company has been uncharacteristically quiet about consumer graphics, instead touting design wins with its Tegra and Ion mobile platforms. It's next-generation Fermi chip has a focus on compute functionality (given Nvidia's heavy bet on CUDA), and we've yet to see performance numbers in consumer products like PC games, let alone any definitive release timeframe for Fermi graphics cards outside of the workstation market. This means that AMD has a substantial lead in the push to the next GPU generation.

Trailing behind AMD has got to hurt Nvidia’s pride, and allows AMD to play the pricing game—charging a little more for cards early on, before the competition can ship an equivalent GPU.

Nvidia should be racing to get its new GPUs out because the 900-pound gorilla that is Intel is readying its own GPU, code-named Larrabee. Based on a modified, multicore version of the venerable x86 architecture, with significant vector extensions, Larrabee is unlikely to ship until late Q1 or Q2 of 2010.

In some respects, these new GPUs will actually bring PC graphics hardware just a little closer to game consoles, as DirectX 11 builds on and enhances the tessellation features built into the Xbox 360 GPU. Hardware tessellation is a substantial departure from previous generations of DirectX, which used triangles and vertices as key graphics primitives. Instead, patches are passed to the tessellation pipeline, which contain control points that define areas within which triangles can be automatically generated by the hardware. This is different than previous approaches, which required the application to generate the triangles passed to the GPU.

What this means is that geometry can be automatically generated where it’s needed, allowing smoother curved surfaces. Using hardware tessellation also reduces the number of steps required by game artists to create the artwork, since they only have to create one representation of an object, instead of multiple versions for different levels of detail.

DirectX 11’s other major feature is compute shaders. Using graphics processors for general computing tasks has become a hot topic in the graphics world. Applications like video transcoding, certain Photoshop filters, and scientific applications lend themselves well to the massively parallel floating-point engines built onto graphics chips.

While the main target for DirectX 11 will be Redmond’s newly minted Windows 7 operating system, the new API will run on Windows Vista, as well (but not on Windows XP). Full DirectX 11 support will require new hardware, but a few of the features—particularly multithreading—will run on existing DX10-, DX10.1-, and DX9-capable hardware. So, even owners of older GPUs may see some performance improvements once DirectX 11 actually ships, late in 2009.

USB 3.0

Like USB 2.0, but 10 times as fast

The maximum data transfer speed of USB 2.0 is 480Mb/s, which was fine when it was invented. Now that you have to fill dozens of gigabytes of apps, music, and movies on your cell phone or iPod, it seems kind of pokey. Enter USB 3.0, dubbed SuperSpeed USB (2.0 is officially Hi-Speed USB). The new spec boosts transfer speeds 10x to 4.8Gb/s, which means in the real world you might see transfer speeds up to around 400 megabytes per second. It also operates in full-duplex mode, meaning the USB host can send and receive data simultaneously. All previous USB specs are half-duplex.

SuperSpeed USB ports will be backward compatible with Hi-Speed USB—of course, you won’t get the additional speed. You’ll notice that the ports and cables used for SuperSpeed mode are a little different, though. The heads are a little longer, with the additional pins for the SuperSpeed mode data extending beyond the usual USB plug.

Other nice additions to the spec include new power management modes and an increase in the base power load, so charging your USB 3.0–compatible devices may be 50–80 percent faster than with USB 2.0. The best part? Motherboards with USB 3.0 ports should start rolling out by the end of this year—if we’re lucky.

DisplayPort

Smaller, simpler, faster

DisplayPort is not so much an “upcoming” technology as an “already here” one. AMD, Dell, and Apple already ship a few products with DisplayPort support, for instance. This new VESA digital display connection standard is essentially a replacement for DVI for external monitors and LVDS for internal connections to notebook displays.

What’s so special about it? Well, the connector is smaller, simpler, and doesn’t have those annoying thumb screws that catch onto every cable like a grappling hook, for starters. The cables are slimmer, and a DisplayPort-only monitor could itself be slimmer—and cheaper.

Primarily, DisplayPort provides more data per wire than DVI. You know how you need a dual-link DVI cable to use a monitor with a resolution over 1920x1200? A “single-link” DisplayPort cable should provide enough bandwidth for 2560x1600, or deeper color modes. There’s also an auxiliary 1Mb/s bidirectional data channel that could be used to carry touch-screen data, data for a built-in microphone, etc. The spec supports HDCP content protection, but don’t expect it to replace HDMI on consumer electronics. Each will serve its own market. DisplayPort might pick up traction fastest in notebooks to replace LVDS to drive the display with fewer wires. Hinge space is already at a premium and crammed with wires, so less is more.

Touch

It's not just for your cell phone anymore

It seems like touch-screen technology is everywhere these days. The resistive touch screens seen on old Windows Mobile devices and the Nintendo DS are quickly being replaced by more finger-friendly capacitive multitouch technology (iPhone, Zune HD). It seems like every smartphone in the world and half the portable media players these days are built around the idea that you’ll operate them entirely by smearing your grubby fingers all over the screen.

Touch might be poised to enter the main computing world, too. Sure, you can get an HP TouchSmart all-in-one or a Tablet PC today, but those aren’t exactly the norm. Microsoft is desperately interested in touch technology these days, and where Microsoft goes, the PC industry often follows. Witness the Surface computer and Windows 7. The latest OS out of Redmond incorporates native touch controls throughout and a multitouch API for developers. Windows 7 is clearly designed primarily for a mouse, but the seeds have been planted.

All we need now is a proliferation of touch-screen PC hardware. We need desktop monitors that are touch-enabled, and notebooks with touch screens (that aren’t necessarily Tablet PCs). Building this kind of support into devices is getting cheaper all the time, but the push these days is to lower-cost PCs, not premium features. Will touch for mainstream PCs and notebooks take off? It’s hard to say, but it’s definitely worth keeping an eye on.

Storage

Bigger, faster, solid-state-ier drives await in 2010

To say that in 2010, hard drives will get more capacious, faster, and cheaper is to state the obvious. Shelves have been stocked with 5,900rpm 2TB 3.5-inch drives for months, and Hitachi’s 7,200rpm 2TB drive comes out in September. But how will 2010 improve on that? Henry Fabian, executive director of marketing for Seagate, says, “We’ll see 3TB drives, probably even higher, as everyone’s vying in the areal density race.” But the more data you have, the harder it is to back up.

Three terabytes is way more storage than we estimate most desktop users will need—but then again, Maximum PC readers aren’t most users. Video editing takes up a lot of space, and those of us who back up our movies to hard drive will quickly find that 3TB holds only about 120 Blu-ray movies.

In the solid state market, expect capacity to go up quickly as prices come down—but maybe not as quickly. Troy Winslow, director of marketing at Intel’s NAND Solutions Group, says he expects solid state drives to double in capacity—at least. Already, we’re seeing lots of gaming PC vendors ship rigs with speedy SSDs for the OS and games, and terabyte-plus drives for storage; expect this to become even more mainstream as 2.5-inch SSDs approach 320GB—or even 500GB. But don’t expect them to match magnetic-drive prices any time soon.

We’ll also see widespread adoption of the TRIM command, which helps keep solid state drives performing at their fastest by informing the controller of empty blocks before a write cycle, so writing files to blocks containing deleted data goes faster. The command is implemented in Windows 7 and in the Linux kernel, and will be available in new drives as well as old drives (with a firmware update).

Is 2010 the year that solid state drives overtake standard hard drives in any sector?

It all depends who you ask. Troy Winslow says that in 2010 SSDs will “continue to displace high-rpm hard drives in enterprise applications, and standard HDDs in corporate and consumer laptops and enthusiast desktops.” But Henry Fabian doesn’t think so. “We don’t see flash today overtaking hard drives, in enterprise or anywhere else, until costs come down. Early adopters will have them, but they’re not ready for prime time,” says Fabian. He cites other concerns besides costs, saying solid state drives won’t replace magnetic enterprise drives until they can match the durability and reliability of enterprise drives. Intel and Hitachi, however, are betting that that happens in 2010, when they jointly release a line of SAS and fiber-channel SSDs.

What about magnetic hard drives with a large solid state cache? Several manufacturers released hybrid drives in 2007 and 2008, but the lines have been allowed to languish, leading many to wonder if the market segment is dead. We’d love to see terabyte-plus hard drives married to a few gigabytes of NAND for speed. Seagate’s Fabian wouldn’t tell us whether Seagate has any hybrid drives in the works, saying merely, “It’s a capacity game, so hybrids could definitely have a role. You get your capacity, with a little boost of speed. It makes perfect sense.”

6Gb/s SATA Will Give SSDs Some Growing Room

Expect 2010 to be the year of SATA 3.0, the 6Gb/s follow-up to the current 3Gb/s SATA spec. High-speed SSDs are already starting to bump up against the 3Gb/s ceiling with their read speeds, so SATA 3.0’s doubled speed gives SSDs some much-needed breathing room. SATA 3.0 also adds greater support for Native Command Queuing and better power management. Drives, motherboards, and adapters utilizing the new spec will appear before the end of 2009—expect announcements at the Intel Developer Forum in late September. We expect widespread adoption by the end of 2010.

Mobile Broadband

LMDS is dead, LTE and WiMax are coming

LMDS

Not too many years ago, it looked like LMDS (Local Multipoint Distribution Service) was going to solve the "last mile" problem and bring broadband to rural areas, while enabling urbanites to roam around with their laptops at broadband speeds anywhere they go. It never really caught on. There are a few providers, but in practice the technology never really sees the 8-mile range or 1.5Gb downstream, 200 Mb upstream connection rates it promised. In reality, this relatively expensive and high-power technology is limited to less than 2 miles and rarely achieves high bandwidth rates. It's going to be all but killed by WiMAX and LTE.

WiMAX

Intel spearheaded the development of WiMAX, now ratified by IEEE in the 802.16d and 802.16e specifications. With speeds of over 100 Mbit/s downlink and 30 Mbit/s uplink, WiMAX has been poised as both a "last mile" solution for bringing broadband to rural areas and a 4G solution for mobile phones and notebooks. Sprint has fallen way behind on their promised WiMAX rollout, and other carriers are opting for LTE as their 4G technology of choice. It's far from a dead technology, though, especially since it's so technically similar to LTE. Clearwire, which is mostly owned by Sprint, plans to roll out WiMAX service to quite a few major metropolitan areas over the next year.

LTE

The 3GPP Long Term Evolution or "LTE" technology looks very hot...on paper. As with WiMAX, this "4G" technology is an all-IP based, low latency network that could truly be called mobile broadband. With enough antennas, bandwidth should be 2-3x what WiMAX offers. Early devices won't show off the full speed, but you could still see notebooks and smartphones getting 10+ Mbit/s with low latencies. Best of all, it seems as if every cell company (save Sprint) has jumped on the LTE bandwagon for their 4G rollout. Verizon, AT&T, T-Mobile, MetroPCS, Rogers Wireless and Telus in Canada, and many others around the world are all adopting LTE and rolling out service over the next year or two. Fortunately, the fundamental radio technology and communications protocols between LTE and WiMAX are so similar that some chipmakers are simply building one chip that can be configured to work with either standard.

Multiscreen Madness

If you think a 30-inch monitor insufficient, how about gaming on six?

Today’s graphics cards can barely handle one 30-inch monitor in gaming. Pushing around 2560x1600 pixels is a challenge for current-generation GPUs. While it’s true that each new generation of graphics cards can push performance, we weren’t quite prepared for the preview AMD gave us of its upcoming DirectX 11–capable graphics hardware.

AMD ushered us into its Sunnyvale, CA, test lab, where it had a high-end system set up with a single graphics card. AMD would only disclose that the card had a single GPU, and was one of the company’s upcoming DirectX 11–capable chips—nothing about the amount of video RAM, clock speeds, or anything else. This particular graphics card also sported six DisplayPort connectors. Attached to each DisplayPort connector was a 30-inch Dell display. The whole affair was configured as a single, 7680x3200 monitor. That's 24.6 megapixels!

Sure, you say, you can hook up six monitors and run Windows… but can it do 3D?

The short answer: yes, in spades. We witnessed the flight sim XPlane 9 running at full resolution, as well as Far Cry 2. Also shown was the flying ship scene from 3DMark 2006, running at a full 7680x3200, at between 12 and 20 frames per second. Dubbed Eyefinity, the tech demo was an amazing tour de force, and we can’t wait to get our hands on one of those cards.

Optical Computer Performs First Mathematical Operation

Paul Lilly — Tue, 08 Sep 2009 11:01:00 -0500

We've been hearing about optical computing and quantum computers for some time now, but for the most part, it's been theoretical talk and 'gee-whiz wouldn't that be cool?' chatter. No longer the case, an optical computer has performed its first ever calculation.

The calculation in question consisted of finding the prime factors of 15. It did this by coupling four photons into and out of the chip using optical fibres. The photons carried the input for the calculation and then implemented a quantum program called Shor's algorithm to complete and output the answer (3 and 5, if you're playing along at home). That might not sound very impressive, but it marks a significant step towards creating a quantum computer.

"This task could be done much faster by any school kid, but this is a really important proof-of-principle demonstration," said PhD student Alberto Politi from the University of Bristol.

WiTricity CEO Believes in a World without Power Cords

Paul Lilly — Mon, 07 Sep 2009 08:26:17 -0500

As we get ready to put 2009 in our rear view mirror, so too will we look back at power cords as a thing of the past. Or at least that's how Eric Blier, CEO of WiTricity, predicts the near future. According to Glier, laptops, phones, and other electronic devices will start operating without a power cord within a year, and be mainstream not long afterward.

"Five years from now, this will seem completely normal," Glier said.

We've already seen wireless pads and other similar de-tethering contraptions, but Glier, whose company is a spin-off of a Massachusetts Institute of Technology research group, is hard at work on a technology called "magnetically coupled resonance." Put simply, this technology sends a magnetic field through the air at a specific frequency, which is then received by a compatible phone or other electronic device and turned back into electricity.

Glier says the technology won't significantly impact the price of MP3 players and other affordable gadgets, and the environmental benefits are pretty straightforward. But what health risks might be involved are yet to be seen.

Image Credit: pastdeadline.com

BMW Shows Off Augmented Reality Concept for Mechanics

Paul Lilly — Mon, 07 Sep 2009 08:24:03 -0500

Remember when virtual reality was the hottest concept on the tech block? These days, augmented reality has become the new go-to fad for future tech talk, and even BMW is getting in on the action. Researchers for the German automobile manufacturer have developed a pair of augmented reality glasses designed to "assist BMW Service staff in their highly demanding technical work."

We'll admit the concept seems cool enough, but there's also something unsettling about a mechanic donning a pair of augmented tech-specs and being fed step-by-step instructions as he pokes around under the hood. We'd feel the same way if we spied a 'Car Repair for Dummies' book sitting in the Service center garage.

Proposed Cybersecurity Bill Would Give the President Unprecedented Control Over the Internet

Justin Kerr — Sun, 30 Aug 2009 15:46:40 -0500

Back in April we reported on new legislation which, if passed, would give the president the authority to take control of the Internet. Over four months later it appears that not only has this bill continued to be worked on, but it is now closer to fruition than ever before. Revisions to the legislation made by the office of Sen. Jay Rockefeller, a West Virginia Democrat, remains “vague” according to Larry Clinton, president of the Internet Security Alliance. “It is unclear what authority Sen. Rockefeller thinks is necessary over the private sector. Unless this is clarified, we cannot properly analyze, let alone support the bill.

The legislation which is now up to 55 pages in length isn’t all controversial, in fact the only section that is being hotly debated at the moment is Section 201. In this section the President is permitted to “direct the nations response to the cyber threat” if necessary for “the national defense and security.” This would allow the White House to engage in “periodic mapping” of private networks that are determined to be critical, and those companies will “share” requested information with the federal government. In plain English, this simply means that if your company is deemed “critical”, regulations determine who you can hire, what information you can disclose, and under what conditions the government can take control over your companies computers or network.

“The language has changed but it doesn’t contain any real additional limits,” according to the Electronic Frontier Foundation. “It simply switches the more direct and obvious language they had originally to the more ambiguous version. The designation of what is a critical infrastructure system or network as far as I can tell has no specific process”.

CNET’s Declan McCullagh spoke with White House deputy communications director Jena Longo who issued the following written statement:

“The president of the United States has always had the constitutional authority, and duty, to protect the American people and direct the national response to any emergency that threatens the security and safety of the United States. The Rockefeller-Snowe Cybersecurity bill makes it clear that the president's authority includes securing our national cyber infrastructure from attack. The section of the bill that addresses this issue, applies specifically to the national response to a severe attack or natural disaster. This particular legislative language is based on longstanding statutory authorities for wartime use of communications networks. To be very clear, the Rockefeller-Snowe bill will not empower a "government shutdown or takeover of the Internet" and any suggestion otherwise is misleading and false. The purpose of this language is to clarify how the president directs the public-private response to a crisis, secure our economy and safeguard our financial networks, protect the American people, their privacy and civil liberties, and coordinate the government's response”.

Do you think this is a serous threat? Or do legislators need to stop watching late night re-runs of Die Hard?

PCI Express 3.0 Hits Setback, Products Delayed Until 2011

Paul Lilly — Fri, 21 Aug 2009 13:22:36 -0500

Don't worry about your swank new motherboard soon being outdated by new models boasting PCI-E 3.0 support, the new specification is running into some unexpected snags, Fudzilla reports.

The main issue boils down to backwards compatibility and getting the PCI-E 3.0 specification to play nice with current PCI-E standards. Before the third gen PCI-E can get a stamp of approval, PCI SIG needs to verify products in the lab, and this is taking longer than expected.

"In this particular case, with pushing the technology so hard, and with PCI gen 3 providing so much more capabilities but with the need to be still backwards-compatible, we had to do the diligence required to move the date," said Al Yanes, president of PCI SIG.

The PCI-E 3.0- specification was originally supposed to be released this year, but now it looks like the second quarter of 2010 at the earliest. This would push shipments of products based on the new spec to 2011.

Image Credit: Flickr