Why do I call this an important foray for Intel? Well, from the way Intel execs described it at the meeting, Intel sees these handheld Web browsers as replacements for Smartphones. If you dig a bit deeper, it's evident that conceptually, Intel doesn't really see its handheld browser platform as all that different from today's Blackberrys and Smartphones. But since Intel isn't the main supplier for Smartphone chips, it's not calling its device a Smartphone. Hence the UMPC name, though that doesn't exactly roll off one's tongue, which is why I'm using the more sensible "handheld Web browser."
Intel's rationale for why people would want to swap their Blackberrys for UMPCs is both self-serving yet, strangely, seems to make sense. (No, it's not those aching thumbs.) It's the fact that browsing the Web on Smartphones sucks. Of course, Intel doesn't put it like that.
"The Internet was developed on Intel architecture (processors), and because it was developed on the Intel architecture, it runs best on the Intel architecture," is how Anand Chandrasekhar, the general manager of Intel's ultra mobility group, described it in New York on May 3. (Well, technically, no, Arpanet was not developed on x86. But if we grant your point that the Internet came to fruition--nay, worldwide dominance--on Intel chips, then we must also point out that that's most because it just happened to be there. But I digress.)
The rub is that this lightweight mobile device on which Intel says you can run Web better seems less like a next-generation Smartphone than a slightly lighter PC crammed into a small (but not as small as a Smartphone) form factor. Not that you'd expect different from Intel.
They're already well along with the design of processors which will power UMPCs. The first one already exists. It's called McCaslin, and it's used in the Apple TV. A more powerful chip--the one which Intel is really pitching for these handheld browers--is code-named Menlow. It'll be ready in the first half of 2008. Interestingly, unlike McCaslin, which only runs Windows, Menlow will run Windows and Linux.
(Not to go off on another tangent, but I don't understand the technical limitations precluding McCaslin from running Linux. Perhaps it's a marketing thing: No one wanted to run Linux on McCaslin, and Apple TV its main destination. On the other hand, Menlow is aiming at other ODMs, a number of smartphones use Linux, so it's a good marketing angle. But I digress again.)
After Menlow, smaller, faster, and less power-hungry processors are planned. (Pretty much the profile of Smartphone chips, huh?)
"This is just the beginning for us," Chandrasekhar said. "We get the base technology elements in place with the Menlow platform and then we iterate on that and get the power down and performance up."
Further along, in the 2009 time frame, is a system-on-chip (SoC) processor called Moorestown. (After a place or Gordon, I wonder?) SoCs allow highly customized chips to be rolled out quickly, in response to different market demands. So, in theory, if UMPCs catch fire, Intel could roll out a whole lotta UMPClet designs.
Still, the big question remains, is Intel really looking to refine the Smartphone or is it firing a volley into the ultra-small-form-factor PC space, with the hope that something will stick?
I ask because Chandrasekhar's handouts pointed to Apple's iPhone as an early indicator of interest in uber-capable wireless handhelds. However, in his discussion, he verbally mentioned the McCaslin-equipped Apple TV is the more direct progenitor of the UMPC.
Interestingly (again), the iPhone doesn't exist (completely) yet. And when it does, it may not quite live up to the vaunted expectations Steve Jobs has seeded for it. (How can it possibly?). Then you come to the observation that the Apple TV is the first successful implementation of another long-suffering Intel vision. That would be its Digital Home, which is a three-year-old group/platform/vision within the company to get people to put PC-like devices into their living room, where they'll serve as digital entertainment hubs. (I should note that Intel is continuing to emphasize Digital Home; it's a separate group, see here.)
So, extrapolating from Intel's past experiences with cellphones, Digital Home, ViiV, and other ventures somewhat off the sweet spot of the PC market, I make the following bold (but really not so bold) prediction: If the UMPC is a success--and it may well be--it won't be in anything like the form Intel envisions today.