Mobile

Chips for tablets: The silicon land grab is on

As we look past the iPad, the tablet market is poised to become a significant market for chipmakers. And that’s an exciting opportunity for the likes of Nvidia and Qualcomm to seize this new ground from Intel.

ARM-based processors power most of the smartphones and tablets now in the market. And Apple’s A4 processor (an ARM-licensed design) has led the pack in tablets, given Apple’s commanding market lead with the iPad.

But Apple is the sole consumer of the proprietary A4, and with an explosion of tablets coming to market in the next 6 to 12 months, it’s unlikely that Apple and its A4 will continue its total domination.

Tablet device vendors will choose from an array of primarily ARM-based chip suppliers who could be rewarded handsomely if the market takes off as predicted (100 million-plus tablets in the next 1-2 years, 200 million-plus in 3-4 years), and Android-based systems eclipse iPads in sales volumes. (We expect non-Apple tablets to outsell iPads by a factor of three to four within 2-3 years). So the silicon land grab is on!

Nvidia with its ARM-based, dual-core Tegra chip is chalking up some wins for next-generation tablets, and even some high-end smartphones. (LG recently announced Tegra will power its soon-to-be-released tablet running a new version of Android.) Qualcomm, with its Snapdragon chips powering many leading-edge smartphone designs, will be a contender. Intel has set it sights on this market with its Atom chips. The chipmaking giant lacks a substantial vendor base for smartphones, but should never be counted out.

Nvidia, long known for its graphics processors, gambled on the emergence of new mobile devices requiring both processor and graphics capability. It licensed the ARM core and built a graphics engine into a chipset it promoted as being the most advanced low-power, graphics-enabled processor available.

However, its efforts for the past year have gone largely without major success in the smartphone arena. (It had something that momentarily looked like a win with Microsoft’s Kin smartphones, but the devices failed to sell and were pulled off the market within weeks of launch. Windows Phone 7 devices use Snapdragon.) But tablets have larger screens and are much more media-centric than phones, and Nvidia’s recent move to dual-core and its enhanced graphics capability means it is finding new respectability.

Its primary competitor, Qualcomm, has a long presence in the phone business and has been very successful with the Snapdragon in smartphones. But it has fallen behind Nvidia in graphics, relying on a licensed graphics core from Imagination Technologies, which licenses its graphics cores to most ARM producers, including Apple and Intel. That licensing deal means Qualcomm does not completely control its own destiny. Nonetheless, with a who’s who of smartphone vendors buying its chips, it has an advantage winning new business in tablets since all of the major smartphone vendors are expected to have tablet devices to market in the next 6-12 months.

The real turf battle in tablets is between ARM and Intel architectures. Intel’s Atom chips own the market for any Windows-based tablets, since Windows does not currently support anything but Intel. (Windows Phone 7, Microsoft’s rewritten mobile operating system, does run on ARM and could eventually make its way to tablets.) But other than Apple’s iPad, the vast majority of tablets will run Android on ARM-based chips. And graphics performance will increasingly become a product differentiator in tablets since media aptitude is such a compelling requirement.

Of course there is no shortage of contenders in the chip wars. Besides Nvidia and Qualcomm, Marvell has made some gains with its ARM-based processors, a business it bought from Intel several years ago, but so far has had less impact in the tablet market than in phones. Samsung’s own chips power its successful Android devices. TI, Freescale, and others have done well in phones but have minimal presence in tablets so far. However lower-end, lower-cost devices may give them an opportunity.

Completely absent so far is longtime Intel rival AMD, which does not have a competitive product for this class of device, and is unlikely to have one in the short term. When it does, it will likely be based on Intel’s architecture, like Intel’s own Atom — a nod to AMD’s long history of providing compatible alternatives to Intel’s designs. With AMD’s graphics prowess, it could potentially rival Intel’s Atom, although I’d expect AMD to have a hard time catching Atom in power-to-performance ratios at this point.

So, this will be a knock-down, drag-out fight. Nvidia needs to be successful in tablets, and 2011 may indeed be the long-promised, long-unrealized year of the Tegra if Nvidia can demonstrate performance and price advantage over Qualcomm’s Snapdragon. The market for standalone graphics chips is shrinking as an increasing share of PCs have central processing units with integrated graphics.

Nvidia needs to find a new high-volume, high-margin market. Tegra in mobile devices just may be the market that provides this opportunity. It will depend on how many chips it can sell and if it can keep margins high in an increasingly competitive tablet market. Nvidia stands to gain handsomely if it can stay ahead of Qualcomm (and Intel) in design wins.

But Tegra will have to work hard to displace Snapdragon as the king of smart devices, and Qualcomm is not standing still with its technology. And Intel has an all-out blitz going to capture market share with newer more competitive Atom designs. This battle is going to get graphic.

Jack Gold is the founder and principal analyst at J.Gold Associates, an information technology analyst firm based in Northborough, Mass., covering the many aspects of business and consumer computing and emerging technologies.


Mobile developer or publisher? VentureBeat is studying mobile marketing automation. Fill out our 5-minute survey, and we'll share the data with you.