ARM will own the smartphone market, Intel will own the desktop and high-end laptop market, and the space in between the two will likely become more confused.
If you had any doubt that Intel has the ability to turn out the finest core processors in the world, then its Ivy Bridge announcement Monday should have put those doubts to rest. Unless real-world testing turns up surprises that Intel does not expect, the new processors will delight power users--gamers, graphics professionals, video editors, stock traders--and anyone else who wants a ridiculously fast desktop system with great graphics, fast I/O, inherent security capabilities, and just about anything else you can do with a few billion transistors.
That Intel builds the Ivy Bridge chip on a brand new 22-nm process (which will soon turn into a 14- and then 10-nm process) should give power users the sense that they'll be well taken care of through the end of the decade. Yes, even if you just need good speedy desktop systems, this week's announcement should have left you quite satisfied. Moore's Law is alive and well, and Intel is using it to its best advantage for the desktop.
If you're an IT pro charged with setting the path for your end users' devices, the announcement was still a good one--with one caveat. On the up side, Intel is over-delivering. It says video performance on the new chips is better than it had predicted, and it's delivering the new core processors a bit ahead of schedule. But what Intel won't deliver, at least until June, are dual-core versions of the Ivy Bridge chip. That's because there's an inventory of Sandy Bridge chips--it seems ultrabooks aren't selling quite as fast as expected. The obvious culprit is the iPad.
Now it's certainly not news to anyone that the popularity of the iPad would have a slowing effect on the sales of other portables. It's not that the iPad can easily replace a laptop, but having one could make it reasonable to keep the laptop you have longer than you otherwise would have. Regardless of what that means for Intel and Apple, it means the IT pros will absolutely have two platforms (or more) for end users as far as the eye can see.
It seems clear that ARM is going to own smartphone market, Intel will own the desktop and high end laptop market, and the space in between the two is likely to become more confused.
That's because the smart engineers at ARM Holdings aren't sitting still either.
ARM is pushing into 64-bit technology and adding virtualization features. Both of these developments will be useful on phones, but more critical on tablets and laptops--and appealing to enterprise IT shops uncomfortable with the current mobile device management options.
We've even seen vendors demoing laptops and tablets sporting both ARM and Intel processors. That's probably not something that will catch on, but it's an interesting proof of concept. The question is where ARM and Intel will split the end user device market. That'll take a while to work itself out. But as things look now, neither company will get the whole market.
There's been some talk about how Intel perfecting its 22-nm technology could be an opening back into the smartphone market. Indeed, the company has at least one smartphone design win--the Xolo X900, which debuts in India. But at this point, unless ARM stumbles in some unimaginably bad way, engineering alone won't be enough for Intel to win over the smartphone market. There's too much ARM-specific software, and the device makers prefer ARM's licensing model to Intel's.
Art Wittmann is director of InformationWeek Reports, a portfolio of decision-support tools and research reports. You can write to him at email@example.com.
At this interactive Enterprise Mobility Virtual Event, experts and solution providers will offer detailed insight into how to bring some order to the mobile industry innovation chaos. When you register, you will gain access to live webcast presentations and virtual booths packed with free resources. It happens May 17.
Server Market SplitsvilleJust because the server market's in the doldrums doesn't mean innovation has ceased. Far from it -- server technology is enjoying the biggest renaissance since the dawn of x86 systems. But the primary driver is now service providers, not enterprises.