Intel phone boss: 'Multi-core detrimental to Android mobes'
But if you want Windows 8, you need to be rolling in cash
Intel’s head of mobile has dissed handset-makers that have already adopted multi-core processor architectures, saying that most implementations so far are actually “detrimental”.
Mike Bell, general manager of Intel’s Mobile and Communications Group, also said that smartphone users who buy into Chipzilla's platform can expect to see their devices running the Ice Cream Sandwich build of Google's mobile operating system sometime this year.
However, anyone looking for an Intel-based handset running Windows 8 better be ready to pony up enough cash for their own production line.
Bell, who before joining Intel last year had stints at Apple and Palm, where he worked on the iPhone and Pre respectively, was in London to provide some heavyweight backup to the launch of Orange’s San Diego phone. This, apart from the Orange logo, is effectively an all-Intel phone - designed by the chip giant around its Medfield platform and run up by a contract manufacturer. While an Intel-based smartphone was launched in India earlier this year, the San Diego is a more upmarket proposition aimed at a highly developed market.
Asked why Intel was trying to crack the handset market again, and what it could actually bring to it, Bell insisted, “Quite a lot.” More to the point, he claimed “no one has a market advantage right now” – meaning Intel has a chance of mirroring its dominance of the somewhat mature PC market in the comparatively immature handset (and tablet) market.
Rather than focus purely on silicon, Bell looked at Intel’s software capabilities, and claims that its experience with Windows, including compiler work, was easily transferable to the Android platform.
“We have an end-to-end solution [that] no one can touch,” he said. Bell added that Intel was capable of bringing in software resources that no other companies could, or even the next five ARM-based outfits.
The reference to Wintel begged the question: why isn't Intel working closely with Microsoft so that they can build the same domination of the mobile market they [once] enjoyed in the PC market? Windows is indeed Intel’s prime focus on tablets, though one wonders whether ARM’s foray into Intel in that form factor hasn’t helped focus the chip giant’s mind there.
Bell’s short answer was “Google’s great to work with” adding that having “access to source code where you can go in and really work with it” is a massive advantage.
Bell was more tight-lipped on the open source dogmatic aspects of working with Android. Asked if it was feeding back its innovations on the platform, he said that it was doing so "where it [was] required", but added: “[I] don’t like doing R&D for my competitors.”
Bell went on to say when it came to handsets, customers – handset manufacturers, operators and users – were asking for Android. “If someone came to us and said, ‘We want a Windows 8 phone,’ we’d work with that,” he said.
But while Intel is happy to produce “iconic” reference designs like the San Diego, leaving partners little to do except tweak the interface and choose a logo, we suspect it will have to do a lot more legwork when it comes to a Windows phone. Luckily Microsoft has Nokia to play with.
Not that Google is hanging around waiting for Intel to catch up. The San Diego runs on Gingerbread, and timing issues meant that Intel couldn’t hold off for Ice Cream Sandwich. Asked if San Diego buyers could expect an Ice Cream Sandwich upgrade this summer, Bell replied: “The next quarter is my guess.”
He was similarly non-time-specific about when Intel would go multi-core with its mobile platform. He said, “When it makes sense, we’ll put it out.”
Android multi-core processing slammed
But Bell was scathing of existing implementations, which he partly put down to software issues with Android itself. “Android doesn’t make as effective use of multi-core as it could,” he said.
“In some of the use cases we’ve seen, [the] second core is detrimental because of scheduling.” Having looked at the multi-core options on the market, he said, the performance didn’t justify “the size and cost of putting in that part".
This was the sort of instance where Intel could bring its experience to bear on Android, he said – though how much this will benefit other handset-makers is, of course, open to question.
Any shift to multi-core would also benefit from Intel’s relentless shrinking of its chips. And perhaps this will give some indication of how well Intel’s handset and tablet effort is panning out. Medfield is built on 32nm Atom technology, and will shift to 22nm next year.
“The big core side of the house does it before,” Bell said, explaining how process shrinks happen. While there has been a lag in the past between Atom getting onto the newer lines, Bell said: “The goal is to be coincidental.” Asked if it was possible that the mobile chips would eventually get first crack at new production lines, he said: “When we have higher revenues and profits...” In the meantime, he said, “The PC processor side of the house has great growth.”
Which perhaps outlines the challenge that Intel has set itself: break into a market that still has plenty of growth, while your existing market still has the growth to sustain your effort. And perhaps by watching who gets first crack at 2015’s geometry shift, we’ll know whether Intel has declared the effort a success. ®
We don't have a decent product on the market, so we'll slag off the competition until we do.
There, a lot less marketing bullshit.
Did I hear that before?
“In some of the use cases we’ve seen, [the] second core is detrimental because of scheduling.”
Sounds like PC dual core CPUs when Intel was behind AMD and working to duct tape two Pentium cores together.
Once they had the Core2Dual out everyone needed dual core.
“no one has a market advantage right now”
So, that almost universal dominance of the arena by ARM processors isn't a market advantage? Maybe he means that no single manufacturer of ARM chippery has an advantage over the others?
That's a good thing, it's called competition. I know it's an alien concept to Intel, but they should probably have looked it up before wading into this field.