To print: Click here or Select File and then Print from your browser's menu

This story was printed from silicon.com, located at http://www.silicon.com/

Story URL: http://comment.silicon.com/0,39024711,11036836,00.htm

The ultra personal computer - the PC industry's only hope?
Good news for users and vendors - apart from Microsoft?

By Tony Hallett

Published: Tuesday 17 December 2002

Could the Windows XP-based 'ultra PC' be the form factor vendors, IT departments and end users are crying out for? Tony Hallett explains why it is being hyped, and what the component companies, analysts and Microsoft think its effect might be on licensing...

The argument for modular computing and, in particular, the ultra PC is compelling. Today, many users have a work PC, a home PC, quite possibly a laptop, almost certainly a mobile phone and maybe even a personal digital assistant (PDA) such as a Palm or iPaq. For them, a single, hybrid device may make life simpler. It may also prove a boon to a number of hardware companies - as long as software issues don't make adoption problematic.

First, imagine a device about the size of two cigarette packets placed side-by-side. It has a high-resolution touch screen, at least three USB ports, and Bluetooth and 802.11 wireless networking capability built in. It is based on Windows XP, has a power-efficient 1GHz processor, around 10GB of disk space, at least 256MB of SDRAM memory and a lithium polymer battery that will last up to eight hours.

What you've just imagined is a replacement for that work PC, that home PC, that laptop, that PDA and maybe even your mobile - all for under E1,000/$1,000. How? Connected to this central device are displays, keyboards, mouses, headsets, even laptop shells and, of course, various networks.

Welcome to the ultra PC. This is the dream, the new form factor that may shake up the hardware industry over the next few years.

Rob Enderle, Giga research fellow, is among the experts who has been watching this space. In a white paper earlier this year he identified three key reasons for it standing a chance, and wrote: "This could be the most important change in personal computing technology."

Why? First, margins in the PC industry are now non-existent for most OEMs. Dell is the only company making real money. Its large rivals are looking at alternative form factors and smaller players who don't adapt will drop by the wayside.

Then, there are the IT departments. As PCs have become more complicated and ubiquitous the deployment and management headaches have increased. Oh for a single device per employee.

And lastly, individuals are getting sick of trying to manage several devices, typically with several address books. Hot syncing - wirelessly or otherwise - doesn't always cut it.

One of the companies backing the ultra PC - one the cynics may say is championing it because its very future rests on its acceptance - is Transmeta. Still known as the once secretive Santa Clara start-up with Linus Torvalds in its employ, Transmeta designs efficient, low power consumption x86 processors. Giga's Enderle reckons its 1GHz and higher Crusoe chips will be integral to the ultra PC's success, in the same way Toshiba's tiny 10GB hard drive will.

Matthew Perry, Transmeta CEO, said: "We're talking about your one and only computer. Total cost of ownership (TCO) should be substantially less and there is an intriguing business model there for OEMs."

And then there are the OEMs, the companies that will make devices. OQO, a San Francisco start-up founded by the team behind Apple's Titanium notebook is perhaps the best-known advocate of the ultra PC and has a product on the market. But others are following, most notably IBM as well as some innovative players from Japan - Transmeta's most successful market to date - and Taiwan, now known for its device making prowess.

Yet two things Transmeta's Perry says stick out. First, the business model. Everyone recognises that while PC makers generally aren't making any profit on each unit sold, others offering related products are. Logitech, maker of PC accessories such as webcams, keyboards and mouses, is doing fine, and the idea is that a modular model will offer incentives for hardware companies to innovate - because there are margins.

Then there is the TCO argument. One of the sub-texts here is that a single device means a single operating system. So does this mean one OS more for Microsoft and four or five fewer?

When OQO launched a trailblazing ultra PC in the spring, Microsoft hardware platform boss Jim Allchin said Microsoft was "impressed with the direction that OQO is showing". Gates and co were sponsors of the device.

However, Brian Gammage, Gartner principal analyst, thinks things aren't so straightforward.

"There's no doubt this form factor is evolving and we're in a period of flux," he said. "But this is not driven by the processor. Vested interests in the OS and application licence spaces will have as much say as the hardware companies."

The warning here is that Microsoft is unlikely to let a new form factor enter the market and threaten its major revenue streams.

But Microsoft says it doesn't see a threat. Allchin is bullish about the model but Neil Laver, Windows product manager, points out the ultra PC's success would be unlikely to mean no home PCs, laptops and so on.

"It's interesting as a new technology but still unproven," he said. "I doubt we'll see any contraction in licensing because of a new form factor."

Microsoft loses money on every device sold with a PocketPC or Smartphone operating system right now and Laver rightly points out that many companies are likely to want their staff to keep work and home IT separate by giving them a laptop to use in addition to, not instead of, a family PC.

He also points out another weapon in Microsoft's arsenal - the company can always offer organisations per user licensing.

Clive Longbottom, senior analyst at Quocirca, describes the ultra PC as a "not next year but a year after" technology and points out: "Microsoft can always be more flexible on licensing."

Perhaps most worryingly for supporters, he thinks people may realise they actually like carrying around a separate laptop, PDA, phone and so on, according to what they need at a given time, and the challenge is to connect them well.

And let's not forget that the two-headed Wintel monster has been central to most mainstream hardware developments in the last 20 years. Intel isn't going to be as big a hardware sponsor as IBM or Transmeta, and Microsoft has other, potentially conflicting interests, most notably Mira screen technology and the Tablet PC OS.

Even Giga's Enderle, optimistic about changes to the PC industry, ends up warning: "It is, however, all about execution, and in a world defined by the lack of co-operation between vendors, even a product with phenomenal potential may simply not reach expectations because of the lack of key vendor support."

It wouldn't be the first time IT departments and individual users are promised something better, only to be denied by vendor politics.