At an investor event in London, remarks by Intel CFO Stacy Smith gave a boost to the recent rumor that Intel may be in talks with Apple to fabricate the ARM-based chips that go into iPhones and iPads.
"If Apple or Sony came to us and said 'I want to do a product that involves your IA (Intel architecture) core and put some of my IP around it,' I wouldn't blink. That would be fantastic business for us," said Smith in a Q&A session with journalists.
The remarks, captured by Reuters, then continued, with Smith entertaining the possibility of fabbing an ARM core: "Then you get into the middle ground of 'I don't want it to be a IA core, I want it to be my own custom-designed core,' and then you are only getting the manufacturing margin, (and) that would be a much more in-depth discussion and analysis."
Clearly Smith is quite publicly leaving the door open for Intel to be a fab for Apple's A4 and A5, and the conclusion that can be drawn from that is fairly straightforward. As we suggested in our previous coverage, if Intel really is open to the idea of fabbing an ARM-based part for Apple, then it suggests a serious lack of confidence in the x86-supported iOS alternatives—namely Meego, Windows, and Android.
We've heard some negative rumblings about Intel's Android port from a source close to the matter, and the fact that Intel declined to talk it up or attempt a demo at the recent investor event is further evidence that the port isn't in very good shape. MeeGo, of course, is on life-support after being publicly kicked to the curb by Nokia. And Windows is far from being interesting on a touchscreen device; indeed, the company probably won't get its act together in that department until the ARM port comes out anyway.
So if Intel wants to bite off a chunk of the tablet market, iOS unfortunately is currently the only game in town.
If the chipmaker and Apple do reach an agreement to fab Apple's SoCs on Intel's 22nm process, then the resulting part could give Apple a truly potent performance and power edge over the competition. Right now, the main advantage that Apple gets from having its own tailored SoC (versus buying an off-the-shelf ARM SoC from a third-party vendor) is a slight power advantage. With an off-the-shelf SoC, Apple would have to take all the hardware that's built-in, even if the iPad or iPhone wasn't using it (e.g., support for multiple USB ports). But with its own custom SoC, Apple can make exactly the chip that it needs, and no more.
However, a 22nm A5 would give Apple the added advantage of not only process leadership versus the rest of the ARM ecosystem, but it would also get a further boost from Intel's recently announced tri-gate transistor. This would put the A5 or its successor far out in front of the rest of the ARM pack in terms of power and performance, and give Apple the kind of fundamental advantage in hardware that no amount of software effort from rivals can overcome.
Intel, for its part, would be able to keep its 22nm fabs busy making a popular family of mobile products while it waits for one or more of its three OS partners to mature into some sort of credible iOS competitor.
Note that that Smith's stance doesn't actually conflict with recent remarks made by Intel CEO Paul Otellini at investor event here in the US, where Otellini was clear that the company will not be getting into the ARM business by making its own line of chips (as it did with XScale a few years back). Otellini told the audience that Intel has an ARM architecture license, so it could do so if it wants to, but it doesn't want to.
No, fabbing Apple-designed ARM parts is a much better way for Intel to get in on the ARM action than if the company were to throw its hat into the ring with yet another ARM SoC.
