News

Intel promises 3G phones at 2G prices with WiFi and camera

Intel has not yet managed to ship the Bulverde super-phone chipset it announced last summer - but Paul Otellini is already promising delegates at 3GSM that for 2005, he'll have a phone that will save 3G, and integrate it with WiFi. Oh, and it will have a two megapixel camera built in.

Otellini, Intel's chief operating officer, did a keynote here in Cannes where he focused on a wide range of technology promises. But the mobile comms thrust was paramount, and he climaxed his presentation with a demo of a reference design which Intel hopes to have ready in 18 months, more or less.

"It will be a chip set, using our XScale Application Processor, normally known by its Bulverde code name - but the new reference design will be called Hermon," sources inside Intel revealed before his presentation.

Hermon and Bulverde can work alone, or together. On its own, Hermon (pronounced "hair-moan") will bring 3G technology down to 2G pricing and features, Intel hopes. Working together with Bulverde, however, it will be possible to build a high-end machine far outstripping anything on the market today.

Otellini showed a demo of Bulverde plus a GPRS module, plus integrated 802.11b and Bluetooth. "It's in a handset," executives emphasised, "not a PDA with phone features, but a genuine phone handset, with genuine phone handset features, and aimed at the mainstream price level.

Exactly what the mainstream will be in mid 2005, when Intel is hoping to have sorted all the development problems it still faces in designing this machine, was something Otellini - understandably, perhaps - was vague about.

Certainly, what he showed would be an astonishing bit of equipment if he could sell it today.

With the two applications processors working together, you'd have a product that included a 4 megapixel camera, or even two cameras, each with its own "quick capture" port, one of four and one of two megapixels. You'd have full UMTS capability on the Wideband CDMA encoding scheme. You'd have a battery life of "multiple days" - average of about four. You'd have seamless roaming from ordinary GSM to 3G to WiFi. It would be costly, yes; but it would be a device capable of astonishing performance. It would be equally happy as a Symbian, or a Windows Mobile, or a Palm or a Linux platform.

But you could also design a phone around the Hermon chip alone. That would have a smaller camera resolution, probably only two megapixels - perfectly adequate for holiday snaps. And you'd have a notional price tag of around two hundred pounds, which could be subsidised down to nothing.

And it would also be a boon for 3G operators, because it would have wireless noise reduction techniques which would allow far more users to log onto the same cell (no promises on exactly how many, but "far bigger cells" would be possible, said executives). This noise reduction would also mean far fewer dropped calls, far better quality of service, they added. in handset, not wireless PDA, but handset.

This phone is due out before WiMAX becomes mainstream, but it looks to the future of ubiquitous wireless data. Otellini is predicting a future where digital wireless pervades everything, and probably, expecting to see this vision real in a 2006 timeframe.

But while Intel appears to have impressed phone operators like Orange enough to be working with them on XScale based handsets, rival chip makers won't be giving up just yet.

Many rival companies will, by then, have roughly equivalent technology, with two-chip solutions being common. Intel does hope to have the best single-chip solution, but it can't guarantee the failure of rivals to match them.

And several of the features Otellini is promising are based on technology which is "under development," and for which there is only a reasonable expectation that the problems will be solved.

For example, executives refused to discuss what technology they would hope to use to achieve the mainstream battery performance that they are talking about. Nothing available in today's silicon would run for a full day at this sort of speed and power, and Intel is well aware that by mid 2005, people will expect top end phones to have the same sort of power life that you get from mid-range phones today - that is, several days without a recharge on normal talk patterns.

There are also issues relating to co-working with WiFi and mobile phone networks, which depend on industry working groups and new standards before they can be addressed. Otellini is, understandably, assuming that these problems will be solved, but his team admits that they aren't in control of all these matters.

"We are working on better battery life: it's a major area of development," was the only comment available from Frank Bryan, Intel's 3G boss. "We're trying to achieve 'best in class' talk times, and data and standby times out there in the industry - but giving much better performance and user experience," he added. "But we can't disclose details of this until much later in the project."