Laptops Desktops Monitors & LCDs Graphics Cards Handhelds Phones Software Networks Printers More »
search
AnchorDesk

Patrick Houston
How Intel plans to make a wireless everything

Patrick Houston
Editorial Director, AnchorDesk
Wednesday, June 18, 2003
TalkBack!Add your opinion
Corporations will tolerate visionaries, but they can't take kooks. Only the finest of lines separates one from the other. That line gets even finer at a place like Intel, populated and managed as it is by people grounded in the pragmatics of engineering and science. When you're stuffing billions of transistors onto little squares of silicon, those lines are measured in nanometers.

Intel's first-ever chief technology officer, a 40-something whippersnapper named Pat Gelsinger, has been straddling that line lately, by articulating a very big idea. It goes something like this:

The handheld of the future?
If Intel succeeds, PDAs with built-in wireless networking--like this one--could become standard.
CHEAP AND UBIQUITOUS wireless connectivity is the key to enabling everyone to compute and communicate anytime, anywhere. Using the same manufacturing capacity that's enabled the company to put microprocessors in anything and everything made by man, Intel would spew out wireless radios so small and so inexpensive they could go into almost anything, from the usual laptops and PDAs to home entertainment systems, kitchen appliances, watches--you name it.

Gelsinger calls his vision "Radio Free Intel." The cute little name doesn't make it any less radical, because it calls for Intel to turn a "boutique" analog technology into a mass-manufactured digital commodity.

Right now, wireless radio chips are relatively expensive, single purpose devices, normally manufactured in relatively small lots. Different wireless networks require their own separate radios--one for cellular nets, another for Wi-Fi, another still for a Bluetooth connection. Stuffing all three capabilities into a single PDA, laptop, or cell phone makes those devices, if not too big, then certainly too expensive. Gelsinger's idea is to create a single, small chip that could connect to all three.

Oh, sure, Intel has become very good--exceptional, in fact--at making microprocessors using its complementary metal-oxide semiconductor (CMOS) processes, the same ones that have allowed the company to maintain Moore's Law longer and more ably than anyone ever could have imagined. But it's never made CMOS radio chips. In fact, no one has.

THAT'S WHY, Intel insiders have hinted, CEO Craig Barrett was none too pleased with Gelsinger's pronouncement. The company brass had no idea whether Intel could make good on such a grand scheme.

Just last week, though, some of Gelsinger's researchers started making him look a little less quirky and a lot more credible. They unveiled some preliminary results of their work on silicon radios, results deemed significant enough to be presented for peer review at an international semiconductor symposium in Kyoto, Japan.

What the Intel scientists achieved is too technical for me to explain here--and for you to care about. Suffice it to say that they succeeded in using Intel's highly evolved CMOS manufacturing processes to create two key components of a wireless radio--an oscillator that determines the frequencies at which signals are transmitted and received, and a synthesizer that enables the radio to switch and tune into signals at impressively fast rates.

Intel researchers themselves admit that they still have a long way to go before they'll fulfill Gelsinger's big idea for Radio Free Intel. But if they succeed, it could have enormous implications for you and me.

IF GELSINGER'S researchers stick to their timetable, they say Intel could be producing cheap, multinetwork chips by 2005. By 2010, they think we could have a generation of "agile radios," capable not only of picking up different kinds of networks, but of seamlessly switching from one to another for the best and most cost-effective connection at any given time.

I'm not underestimating Intel's ability to deliver on this promise. It has the resources--Gelsinger's playing with a $4 billion R&D budget, after all. Plus, it has the resolve. And well it should: Making wireless radios would allow the company to utilize its existing--and ever so expensive--manufacturing capacity to tap a new market and the potential profits therein.

But I'm not overestimating Intel, either. Receiving, sending, and processing electromagnetic radio waves is an analog technology. Converting waves being aired at gigahertz frequencies into digital representations requires extraordinary processing power and sophistication.

Not only that, but wireless technology is developing at, well, the speed of sound--making the target that much harder to hit. And let's not overlook Intel's rivals in this space--giants such as Motorola and Philips. They aren't going to just sit around while Intel tries to steal food from their table.

Still, Gelsinger's got his bosses saying, "Gee, maybe we can do this." Which means that his own company, at least, is beginning to consider him less of a kook and more of a visionary. I, for one, buy into the vision. But I've seen enough by now to know that visions are easy to put on paper. They're very, very hard to put into products that people like you and me will actually buy and use.

What to you think? Would you want all your electronic gadgets to be able to communicate with each other? Do you think Intel will succeed? TalkBack to me! 

Previous Story  Next Story 

Special sponsor stores

Social Networking

advertisement
Click Here