When Apple Inc. switched from using IBM processors to Intel chips in its venerable Macintosh computer line almost seven years ago, it was a big deal.
The late Steve Jobs and Paul Otellini, president and CEO of Intel, appeared on stage together at Macworld 2006 to make the surprise announcement—with Jobs' typical flair—that iMacs with Intel's Core Duo processors were available immediately, six months ahead of schedule. Two of Silicon Valley's most famous and successful companies were finally getting together.
But now it appears that that run may be winding down. On Monday (Nov. 5), the Bloomberg news service reported that Apple is exploring ways to replace the Intel processors it uses in Macs with some version of the A5 and A6 SoCs it uses in its iPhones and iPads. The Bloomberg story, which cited unnamed sources "familiar with the company’s research" said Apple engineers have grown confident that the chips it uses for its mobile devices will one day be powerful enough to run desktop and notebook PCs.
Recall that last week Apple showed the door to two senior executives and reshuffled responsibility for several of its divisions. The company seemed to take pains to include in that announcement the cryptic statement that Apple's semiconductor teams—which now report to Senior Vice President Bob Mansfield—"have ambitious plans for the future."
Retooling Apple's ARM-based A6 to power PCs would certainly qualify as ambitious. It would also be consistent with Apple's modus operandi in recent years—moving toward vertical integration and greater control over the chips it uses in its products. It would, of course, require a great deal of work on both the hardware and software sides.Such a move, however, would not occur overnight--it would likely take Apple several years.
For Intel, it is unclear how large of a blow such a move by Apple would be. Apple—with its wild success and iconic products—is a prestigious logo to include on the customer slide in a Powerpoint presentation. But Intel's biggest customers are Hewlett-Packard Co. and Dell Inc. Apple sold 4.9 million Macs in its most recently concluded quarter; HP shipped 13.6 million PCs in the third quarter, while Dell shipped 9.2 million. Related stories:
Integration is really the key challenge and it's unclear that Apple can or wants to do that. Intel's recent Digital RF on CMOS Atom SOC ("Rosepoint") is the first of its kind; no one else has demonstrated that level of integration on a modern process tech node.
The node advantage is expected to reduce going forward if not go away. Smaller technology shrinks (22nm-14nm) are not providing the dramatic reduction in power or increase in frequency anymore. So if the Apple design team can get close to intel performance with better integration of other functions, the management would take that and get a huge improvement in margins.
Remember that Apple is already pretty much vertically integrated already.
The "architecture" (x86 vs ARM) is irrelevant. x86 is actually a disadvantage, but a minor one, as all Intel chips since the Pentium immediately translate x86 into an ARM like RISC ISA.
What's left is microarchitecture and circuit technology. Of these, Intel currently leads in absolute performance per core, while ARM leads in performance per watt, which translates into throughput performance.
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.