And some people out there still naively think that Apple isn't planning MacBook Airs with ARM CPUs.
There's no hurry, of course. Intel-based MBAs are selling like hotcakes and the Ultrabook herd have no response. "Flummoxed," as Steve would say. But Apple has been at the mercy of Motorola and IBM for PowerPC CPUs before. And they've been burned badly. Twice burned, thrice shy.
Depending on a single 3rd party for CPUs, or any crucial component, means that Apple doesn't control their destiny. Motorola dragged their feet fixing PowerPC bugs and ramping up clock speeds. Motorola and IBM ran into the clock speed and operating temperature walls. Apple suffered through that for years, and Steve had to admit in the middle of one of his MWSF keynotes that the PowerPC didn't reach the 3GHz that he had promised. Two years after that, they still hadn't reached 3GHz. That was just before Apple switched Macs to Intel.
There was a brief Apple-Intel honeymoon, when Intel built custom chips for Apple (ironically for the original MacBook Air) and gave Apple early exclusivity on speed-bumped CPUs and chipsets. But the honeymoon was over by the time Apple picked ARM for iPhone and all iOS devices. There were far too many advantages to using ARM SoCs over the power-hungry Intel Atom CPUs. Intel was left behind in the mobile space, and they may never catch up.
ARM-based MacBook Airs would also benefit from using ARM chips. The MBA would extend its technical and competitive advantages: lower cost, lower power consumption, increased battery life, and cooler and quieter fanless operation. All of which Intel and their partners will be very hard-pressed to replicate.
What's that? You say ARM MacBook Airs won't run Windows? Well exactly how many consumers run Windows on their MacBook Airs right now? And would it be worth delaying the ARM transition for that tiny minority? I say no.
The vast majority of MBA owners use their MBAs for surfing, email, tweeting, texting, iLife, and iWork. (It is, after all, Apple's cheapest all-in-one Mac line, and therefore the mass market line.) An ARM-based MBA will be able to handle that easily with a little more development. A quad-core ARM chip of sufficient clock speed should be enough. And guess what. The iPad 3 just might have a quad-core ARM SoC. And if not this year, next year for sure. Remember: Apple isn't in a big hurry.
I'm sure Tim Cook has done the math. There's a Keynote slide, on an iPad somewhere in Cupertino, with a graph that shows exactly when ARM chips all be powerful enough to run OS X, and its suite of apps, on MacBook Airs. It's just a matter of time.
