Intel lost the tablet war -- is the desktop next?

With meager sales of Windows-based tablets predicted till 2017, the door is open for ARM to establish a beachhead on your desk

1 2 Page 2
Page 2 of 2

By contrast, iOS has thousands of developers who are used to developing cheap apps; to lesser extent, so does Android. What's more, developers want to work with a platform that has critical mass, and it doesn't look like Windows Phone will be there for some time. The DisplaySearch forecast shows a combined iOS/Android share of 92 percent of the tablet market in 2017, which doesn't leave much for Windows. You'll notice that the Windows share is a bit bigger than the x86 share I mentioned above. That's because Microsoft will support ARM in Windows 8, but the expectation is that ARM-based tabets won't account for many Windows 8 devices because the ARM versions won't be able to run traditional Windows apps.

Fundamentally, the issue comes down to the nature of the chips themselves. The ARM architecture offers much better performance per watt than does the x86. That difference becomes even more pronounced when you realize that mobile chips are becoming more and more integrated, and the system-on-a-chip (SOC) is where the future lies. "SOCs can power just about anything," says McGregor. "ARMs and its partners are very well positioned. It's the richest ecosystem in the industry," he adds, noting it includes companies like Samsung, Nvidia, Freescale Semiconductor, and Texas Instruments -- oh, and Apple.

ARM has a processor-partitioning technology it calls Big Little. Using that technology, processing tasks can be switched from a lower-performance, but more power-efficient, core to a higher-performance core using an interconnect fabric called CCI-400. ARM demoed that technology in London recently using an unmodified Android OS on its Cortex-A7 and A15 cores. "The mobile OS will not know, does not need to know, which processor is being used for specific tasks," ARM CEO Warren East told ElectronicsWeekly.

That's a huge step that could well raise the ceiling on the kind of tasks a tablet or smartphone could perform. After all, even fairly large and processor-intensive applications need the CPU to crank at full power only intermittently. If that happens on one core, and the other lower-power core runs the rest of the time, you still have plenty of power left to keep the device ticking.

McGregor points out, correctly I think, that ARM will probably never replace x86 at the high end. And Intel is working hard to develop lower-power Core i5 and Atom chips and SOCs for tablets and even smartphones.

But McGregor is skeptical that Intel can reverse the trend: "It's difficult for Intel to move down and for ARM to move up." Speaking of moving up, ARM today introduced its first 64-bit architecture, an advance that will position the company and its partners to move into the data center.

Ultimately, if most of our work can get done on a less powerful mobile platform, there's a real likelihood that Intel will be a much smaller presence in our computing life.

I welcome your comments, tips, and suggestions. Post them here (Add a comment) so that all our readers can share them, or reach me at bill.snyder@sbcglobal.net. Follow me on Twitter at BSnyderSF.

This article, "Intel lost the tablet war -- is the desktop next?," was originally published by InfoWorld.com. Read more of Bill Snyder's Tech's Bottom Line blog and follow the latest technology business developments at InfoWorld.com. For the latest business technology news, follow InfoWorld.com on Twitter.

Copyright © 2011 IDG Communications, Inc.

1 2 Page 2
Page 2 of 2
How to choose a low-code development platform