Will Intel Make Broadwell An Open Mobile Platform?

| About: Intel Corporation (INTC)
This article is now exclusive for PRO subscribers.

In the history of semiconductors, when a company is behind it needs to open up its closed system to allow an industry to coalesce around a new standard. Intel's (NASDAQ:INTC) miscues in the mobile market are a direct result of not having all the SOC pieces and a leading edge wireless solution together on time for the complete platform to be competitive. Moreover they have started from a laggard position trying to overwhelm Qualcomm (QCOM) and others with process technology. It is time they pulled out all the stops and opened the platform as they did in the early PC days. Winning the processor piece and being first in line with the leading operating systems is the most important objective these days. Intel can then target the rest of the platform when they are ready.

In the early days of the 286 and 386, startup chipset companies swarmed the platform with highly integrated, lower cost solutions that allowed the clone makers to jump in and feed off of the IBM (NYSE:IBM) PC carcass. Intel allowed this chaos to continue until they developed their own chipsets for the Pentium Processor family in 1992. They then honed the platform performance and capabilities that the chipset companies couldn't match. Today they face a threat to their entire business because they at first have chased the wrong mobile market at the wrong time (i.e. smartphones instead of tablets) and then they bet their mobile business on a custom design philosophy that is not keeping pace with other competitors [e.g. nVidia (NASDAQ:NVDA)].

It is time to re-frame what the PC and mobile market is and where Intel can thrive given the company's strengths. The tablet market is closer in capability and mechanicals to the mobile PC than the smartphone. In the marketplace, in many cases, it is turning into a substitution purchase in price points that overlap. What most people miss is that the room for the silicon building blocks of a tablet of 7" screen size or more is adequate for using notebook PC components with room left over for the optional baseband chip. Highly integrated SOCs are not required and so for Intel to sweat over highly integrated Atom SOCs is a waste of an effort that is guaranteed to be low margin.

Intel's Broadwell processor finally solves the TDP power issue that constrained earlier chips like Ivy Bridge and Haswell from being designed into very thin formfactors. At the right die size, with the right number of processor cores and SRAM cache size, it can be cost effective as well.

Like the PC, the tablet platform can be a three chip solution of processor, chipset and wireless. Under this configuration, Intel can be free to stay on a regimented Tick-Tock sequence with its processors while allowing innovation to occur on a different cadence with the chipset and wireless solutions, the areas that it is having trouble keeping up.

This can only occur however if the folks at Intel are willing to be truly open with the interface specifications between the processor and the chipset and is willing to let others thrive in co-opetition with the processor. If Intel allows a wider range of performance/cost with the Broadwell processor, then most assuredly will we see a wider breadth of chipset and wireless solutions with competition driving costs to the floor. Increased hardware options will allow more tablet price points for OEMs. More importantly it increases the likelihood that Google's (NASDAQ:GOOG) Android for x86 will take more precedence relative to the ARM (NASDAQ:ARMH) based Android.

In summary, the tablet market is an easier nut to crack than Intel is making it. I suspect that because Intel re-arranged the whole company to chase what was a 3.5" smartphone market first, which does require an SOC, the company is embarrassed to go back to the public and analysts to tell them Atom and SOCs are not where it's at. A large mobile market with phablets, tablets, ultrabooks, etc., is a high growth market that could fill a lot of fab capacity. Intel's domination of the PC market with its 22nm Haswell gives it freedom to slow down the high performance, large die treadmill in order to build a very low power, much smaller die for tablets. For Intel to win though they need to open up a platform that every other supplier treats as closed.

Disclosure: I am long AAPL, QQQ. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.