There have been rumors circulating lately that Apple (AAPL) was planning to switch processors from Intel (INTC) to ARM Holdings (ARMH) for its Mac lineup. Was this simply a tactic by Apple to achieve leverage in pricing negotiations with Intel, or is it a real possibility? I think it was both.
Click to enlarge images.
It is no secret that Apple is using ARM processors in all its iPhones and iPads. For the Mac platform, Apple was using Intel after the infamous switch from the Motorola/IBM-supplied PowerPC some years back. There would be obvious synergies within Apple if all its products were using a single processor family. It certainly makes more sense to gravitate toward ARM rather than to Intel given ARM's stranglehold in the mobile sector. Running only ARM code across the entire Mac/iPhone/iPad lineup would surely make life easier for the software guys and gals.
From a hardware perspective, one should not overlook design team productivity improvements from focusing on a single processor architecture. One issue many analysts omit is a processor's design system ecosystem. This involves all the nitty-gritty tasks of code development, emulation, debugging, and performance modeling. Both Intel and ARM do a great job in this area, but certainly there would be advantages of scale and simplicity for Apple were it to consolidate its two differing processor families into one.
But what Apple really wants is lower processor prices. Going to ARM would potentially open the door to multiple silicon suppliers. In the electronics world, competition brings down prices. Currently, Apple is hoping just the rumor of competition will bring down prices.
But can ARM actually deliver the performance required by the Mac platform? Can the code be migrated from Intel to ARM? These seem to be the two big questions. I'll address the second one first.
With respect to code migration, I think it's no big deal. If Apple could migrate from PowerPC to Intel successfully (it obviously did), there should be no problem migrating from Intel to ARM. Although the PowerPC had a bit in its Machine State Register to run in either Big Endian or Little Endian mode ("Endianess" has to do with byte ordering with a word), it powered up in Big Endian mode. So there were some issues there when moving from PowerPC to x86. Going from Intel to ARM should be much simpler as both of these architectures are Little Endian from the get-go.
I have heard that much of iPhone and iPad product development within Apple is done primarily on Macs. I don't know for certain if this is true, but it makes sense. That being the case, you would basically have an ARM processor model running on the Mac's Intel processor. Enough said.
Still, there are those who say that Intel-based Macs' big advantage in the marketplace is their multiplatform capability: Macs are the only machines in the world that can natively run OS X, Windows, and Linux. While this is an important consideration by any enterprise or business user of the Mac, and the capability is thought to be dependent on the Intel processor, does this necessarily mean it thwarts Apple's ability to use ARM in any product offering? I think not. There are many customers out there who simply want a decent performing laptop (or netbook) with great battery life that easily integrates their iPhone/iPad/iPod lives and don't give a hoot about running Linux or Windows. But if they want to, almost anything can be done in software these days -- it just depends on how much processing power (i.e., battery life) you want to use doing it.
So the bigger issues come down to processing power. Many of the analyst I have read seem to doubt Apple's willingness to sacrifice processing power in its Mac lineup by switching from Intel to ARM. While that may arguably be true today, the future is not so clear. ARM's Cortex-A series is a shot across Intel's bow. These processors combined with the MPCore Technology enable multi-core system on a chip (SoC) designs. The ARM MPCore technology allows for design-configurable processors supporting between one, two, three, or four CPUs to operate in an integrated cache-coherent manner.
Obviously, Intel also has multi-core processors. Indeed, I am writing this article on a PC powered by an Intel dual core processor. But that's not the point. The point is that ARM is making inroads into the higher processing power arena. It has its eye on the power-hungry server market, where cooling costs are a big factor and low-power cores are definitely part of the solution. The server market also will drive ARM to deliver higher performance solutions ,and that takes it right into the PC market.
Price negotiations will proceed as usual, with Intel wanting to keep its margins intact and Apple wanting a price break. Next year it will be the same. Rumors will only buy Apple so much in the long run. At some point, it will need to back it up with facts. My guess is that Apple will offer a MacBook with an ARM processor by the end of 2014 or early 2015.
As a former engineer at Motorola who once worked on a competing processor to ARM, I know for a fact the company should not be underestimated. It has some excellent engineering talent and outstanding marketing. Anyone who has ever attended an ARM conference or architectural meeting can attest to this. That said, Intel is also staffed by quality engineers and is very profitable today, even with the stagnating Windows market. It has some of the best, if not the best, silicon processing technology in the world.
But Apple must play on both sides of the fence. Intel processors will continue to be used in Apple products for a long time to come. After all, if Apple were to go solely ARM, at some point down the road it may face the same pricing issues it has today with Intel. From a strategic perspective, if ARM was successful at pushing Intel aside, what's to say ARM licensing fees and royalty rates (currently kept low to gain market share) could not be dictated at will? Apple's best strategy is to play Intel against ARM and keep them both guessing. That cannot be done on rumor alone.
- ARMH: $36.70
- Market Cap: $16.88 billion
- EPS (TTM): $0.52
- P/E (TTM): 70.09
- Div (Yield): $0.08 (0.5%)
So is my expectation of Apple shipping a MacBook with an ARM processor in a couple years' time worth making an investment in ARM today? No, not on that merit alone. But let's face it: ARM owns the mobile market. With a P/E of 70 and earnings per share of $0.52/share, the stock is very richly priced. Any hiccup in Apple or the market in general could see this stock swoon. That said, it's hard not to want to own a few shares in a company that is dominate in its field of low-powered, high-performance processor systems for the mobile market -- and, perhaps soon, the MacBook market.