Intel executives on Wednesday talked 3-D transistors, performance leaps on lower power and usage in a wide range of computing devices. However, the elephant in the room remained ARM and Intel’s prospects in the smartphone and tablet market.
Those prospects may still rest with Apple and a foundry deal.
Simply put, Intel’s announcement that it will add “fins” to transistors to make them 3-D eliminates clock speed as the measuring stick for performance. Performance “won’t be necessarily measured by frequency,” said Dadi Perlmutter, executive vice president of Intel’s Architecture Group.
Sounds like a recipe for smartphones and tablets right? When asked about when Intel’s 22nm 3-D chips will hit smaller devices, Perlmutter said: “We have a date for transistors in smartphones and tablets, but we’re not going to tell you.” In fact. Perlmutter promised more on Intel’s mobile plans at its May 17 investor meeting.
From there the conversation between reporters and Perlmutter revolved around ARM. Perlmutter downplayed ARM architecture vs. x86, but at the same time made it clear that Intel was going to be competitive with performance on tablets and smartphones.
The big issue here is that Intel’s Atom chip today doesn’t have the ecosystem or graphics capability to compete with ARM. The other reality is that Intel’s move into smartphones and tablets with Atom would require an industry to change architectures.
Add it up and Intel may have a tough road even with 22nm, 3-D transistors and a three year lead on rivals. What could change the equation? Apple. Related: The case for Intel making Apple’s mobile chips: A Wintel of the post-PC era
Analysts have been arguing that Apple could use Intel for a foundry deal. In other words, Intel may be an option for Apple’s A6 chips, but today’s announcement is really about the A7, A8 or A9. Can Apple and Intel team up and grab a massive lead—Mactel if you will—and then force the industry to move to Atom? Or conversely, could Intel go to ARM and leverage its breakthrough on 3-D transistors in volume?
It’s possible either way. Clearly, Apple and Intel could create a win-win chip partnership. Intel on Wednesday talked about transistor gains not necessarily architecture. There’s no reason why Intel couldn’t use its 22nm manufacturing lead and couple it with 3-D transistors on an ARM architecture.
If you read between the lines, Intel dangled a few nice carrots for Apple. Perlmutter said “there are a few things we’re trying to penetrate” in the smartphone and tablet market. You can read that to mean an Apple foundry deal.
