Now I’m not going to suggest that Apple planted those stories about the switch to Intel processors. Oh well, I’ve said it. In any case, when Steve Jobs let the cat the rest of the way out of the bag, some of the technology pundits managed, as usual, to screw up the facts about the transition.
So let’s just set the record straight, and get on with it. First of all, there is the perception created in some articles that Apple lost a lot of market share as a result of the move from 680×0 to PowerPC back in 1994. In case you forget, that switchover was far more painful for developers than Mac users. Don’t forget the infamous emulator that allows you to run older applications with good compatibility. It wasn’t terribly fast at the outset, but got better over time, particularly as processor speeds rose quickly. The real reason Mac profit share declined was most likely the release of Windows 95, the first really useful version of Microsoft’s operating system. That, and a huge marketing campaign, made some feel the Mac was irrelevant, that its advantages had vanished.
In retrospect, we know that wasn’t true, but you can’t forget how many Mac users and software publishers went to the Dark Side in those days. So no, I don’t think the migration to PowerPC was that serious a factor.
So, of course, with another major transition afoot, the pundits suggest that Apple again is confronted with the potential loss of market share. Now I can see where some of you might hold off buying a new Mac until the model you like offers Intel Inside. Or maybe not, because it shouldn’t matter what kind of processor is used, so long as performance meets your standards. Do you care, for example, what sort of Ethernet or Wi-Fi chips Apple uses? Of course not!
Besides, there will always be newer, faster computers. To take this logic to its ultimate conclusion, you should never buy any computer, ever, because it’ll some day become obsolete.
If there is any genuine pain in the transition process, it’ll be on the part of developers who must begin to update their software as Universal Binaries, so their apps run on both the PowerPC and Intel processors. But since the first Intel-equipped Macs are roughly a year away, there’s plenty of time to tweak and recompile, which Apple claims should be sufficient for most developers to get their software ready. Between now and then, you will see the magic words “Universal Binaries” on some new products, even though it won’t make a bit of difference until the new hardware is out. But it makes for good copy. The real hope is that you and I won’t be gouged with high prices for those updates. If it’s so easy, it shouldn’t involve more work than a routine bug fix update, maybe even less. Of course, some companies will wait for a major revision before switching to Universal Binaries, so the standard upgrade fee will apply. Microsoft no doubt fits into this category, based on their public pronouncements on the matter.
The other big issue is emulation. Just how fast will Rosetta be once the new hardware is ready for sale? It looked all right during the keynote, on a 3.6GHz Pentium 4, but Photoshop took quite a long time to launch. I expect performance will be tweaked between now and then, and it may only make a difference for processor intensive software. Regardless, Rosetta won’t be a panacea in any case. It has its limits, according to Apple’s latest documentation on the subject. It will, for example, only work on applications that include G3 support, and won’t run code designed to harness the power of the G4 and G5’s Velocity Engine. Oh, and software that predates Mac OS X won’t run either, which appears to indicate that Classic is on the way out.
And once again, I want to dispel the myth that you’ll be able to run Mac OS X on a plain, jane Intel box. Apple says no. Of course, hackers will try, perhaps attempt to reverse engineer the technology, so Apple will have to cover all its bases. Of course, their legal staff will be in readiness just in case.
The real problem I foresee is device drivers, so your printers and other peripherals will continue to operate on your new Mac with Intel Inside. I imagine all the remaining issues will be clarified in the months to come.
But don’t forget that a year is an eternity in the computer business. While Apple will clearly not back out of its deal with Intel, I’m curious to see just what sort of problems developers will face in getting those Universal Binaries out the door in the real world. I’m also curious to see what sort of Power Macs will appear in the interim. Will it be more of the same, minor incremental updates, or a last hurrah from IBM before Apple cuts the umbilical cord?