While we were on our honeymoon, a strange story arrived in our cabin on the m.s. Statendam. According to the New York Times Digest—a source of dubious trustworthiness—Steve Jobs announced during the keynote at the Worldwide Developers Conference (WWDC) that Apple would begin a switch to Intel chips in 2006 and complete the transition in 2007.
Well, dubious source or not, it turns out to be true. Apple’s ten year commitment to the PowerPC processor line (manufactured my IBM and Freescale [formerly Motorola's semiconductor division]) is coming to an end. Intel—whose chips run most Windows computers—will effectively own the desktop and notebook computer markets in their entirety.
Apple, as usual, has gone a long way toward making this easy for average consumers (the transition from 68k chips to PPC in the 1990s and the transition from Mac OS Classic to Mac OS X in the early 2000s both had similar technical challenges—so they have experience). Most developers will have an easy time making their applications run on both existing PPC Macs (like mine) and on the upcoming Intel Macs, and most software that isn’t made compatible with both will run (albeit slower than usual) through a PPC emulator called Rosetta that will come with the Intel Macs.
Developers of Mac applications that were originally written for Mac OS Classic—most notably the entire line of Mac software products by Microsoft and Adobe—will need to take more time and effort, but the major players (including those two) have already committed to making their software fully compatible.
As for Mac OS itself, its Unix core (called Darwin) has always been available for both PPC and Intel architectures, and Steve Jobs revealed that Apple has been maintaining Intel versions of each version of Mac OS since the release of OS X. It’s no surprise, really, since OS X’s predecessor (NeXT’s NextStep) was available for Intel processors.
It is not hard to see why Apple is making the jump, and—while I admit to being apprehensive at first—I find I agree with the reasoning more with time. IBM has been unable (or unwilling) to develop a G5 processor that can be used in a notebook, and they have failed to deliver on speed increases in the desktop version. What this means is that Apple’s premier Power Mac model, which debuted two years ago as the ‘world’s fastest personal computer,’ now lags behind the PC competition. Apple’s premier notebook, the PowerBook, lags ever further behind with Freescale’s G4 processor. Mac OS X alone is the most compelling reason to get a Mac, but the hardware must remain competitive. PowerPC, for all its potential, has not kept up (thanks to IBM and Motorola/Freescale) and it doesn’t look like that will change.
Now before anybody makes rash assumptions, this shift is unlikely to change Apple’s mode of operation in the short term. Even though Macs and PCs will both run the same processor inside, Apple has already said that Mac OS will not be able to install on any old PC. To use Mac OS, you’ll still have to buy a Mac. That said, nothing would preclude a user from running Windows on a Mac (assuming Microsoft includes the correct drivers) either in a dual-boot setup or through virtualization software (probably a port of the Windows version of Virtual PC).
I run Windows on my system today (on rare occasion) through the Mac version of Virtual PC, but since it’s a processor emulator it runs very slowly. With computers running Mac OS and Windows operating on the same platform for the first time in history (and no, Microsoft’s brief flirtation with Windows NT for PPC doesn’t count), users will be able to do the same thing without the speed hit.
From a user perspective, the Intel switch will mean very little (aside from new Macs maintaining speed parity with Windows machines). It’s some work for developers, but your average user will just continue doing what they’ve always done—buy computers when they need them and use them. Apple seems to have gone to great lengths to make this seamless and painless for normal people.
Where this switch poses a real problem is when it comes to nerds (like me) making buying decisions for the long-term. I had planned to replace both my primary desktop Mac (Nadia; Power Mac G4) and my notebook (Kitka; iBook G3) with one high-end PowerBook later this year. Previous to the Intel announcement, my worry was that if Apple was still on the G4 processors I would be a generation behind-the-times from the get-go (as compared to the desktop Power Mac and iMac G5s). Now, I have to take platform shifts into account—not only would I be getting a generation-old PPC, but I’d be buying one of the last rounds of PPC Macs altogether.
While this poses little short-term risk (since there will still be new PPCs on the market into 2007, which will presumably have full OS, software, and peripheral support), what happens when the transition is complete? How long after that will software developers continue compiling their applications to work on both platforms? How quickly will the users of then-outdated PPC Macs be locked out of OS updates, new software, and new peripherals because the new versions are only compatible with Intel Macs? At this point, there is no real way to know.
Also, because Apple has not given specific dates for specific models, it is next-to-impossible to decide when to buy if you care about what processor is inside. If Intel Powerbooks are slated for release in January 2005, I would wait to get one then (or after its first revision). If the PowerBook is slated to be one of the last models to transition in 2006, I’d bite the bullet and buy a PPC PowerBook soon like I’d originally planned to. But without knowing when Intel PowerBooks are coming, I’m forced to either risk buying a last-run PPC or risk waiting a long time to get a new computer—and I don’t want to do either.
While I’m talking about one particular model, in-the-know consumers considering buying any Mac will be faced with these same dilemmas and questions. When is model x transitioning to Intel? How long will a current PPC Mac maintain compatibility with new software? When will PPC Macs stop being able to run the newest Mac OS? And so on.
All-in-all, I think the Intel switch will be good for Apple—as the switch to PPC from the Motorola 68k processors and the switch to Mac OS X were ultimately good for Apple. For most users, this transition will be as relatively-painless as Apple’s last two. But when it comes to power users looking for long-term compatibility in a new machine, the picture gets a little muddy. Until we have more explanation from Apple, we’re shooting in the dark on what Mac we should buy and when we should buy it. As a user who is considering a computer purchase soon, that’s especially frustrating.
That said, Apple characterizes this switch as a long-term, gradual transition. There is always a risk, when buying a computer, that something better will come out a few months later. So it goes. That this improvement involves something as major as a change in processor architecture doesn’t really change that fact, given the pains Apple has gone through to make the Mac experience compatible with both the PPC and Intel architectures. I will continue to digest this news for the next few months, and will decide on whether to buy when the time comes.
As much as I’d like to wait for an Intel Mac, my systems are getting old and ought to be replaced (I can’t run the new versions of the X-Plane Flight Simulator, my system gets a bit laggy with high-end audio work, I want to combine my home- and portable-computing environments, and new Macs come with a newer OS and newer iLife suite that run much better on new machines—among other reasons). Unless there’s good evidence that an Intel PowerBook is coming in the first part of 2006, don’t be surprised if I come home with a shiny PPC PowerBook before then.