Mac OS X on Intel Revisited
November 5th, 2005I don’t like to depend on what rumor sites say, although they are sometimes right. And when Mac developers violate their nondisclosure agreements with Apple and tell you things they shouldn’t be discussing, you want to both listen and tell them to keep their mouths shut. What a dilemma!
But one fact is emerging from the hushed discussions, and that is you will see a very different user experience when you buy the first MacIntels, even the Mac mini. For one thing, the operating system will seem to have, as one technology writer has suggested, “afterburners.” It’ll fly in places where it now seems to run out of steam. No, it’s not just because Intel processors will be faster, and they will of course. But why should this be so? Why should moving Mac OS X away from its “native” processor to a new one suddenly yield such a surprising speed gain? Well, if you can believe what some are saying, and it makes perfect sense, Mac OS X is really better optimized for Intel, and that getting it to run in a satisfactory fashion on the PowerPC required a lot of work under-the-hood. That’s why, for example, Mac OS 10.0 was so sluggish.
Now I suppose you could suggest that it was meant to be that way, that Steve Jobs knew all along he’d one day move to Intel, but was only awaiting the right excuse. Of course, the present chip suppliers, Freescale Semiconductor and IBM, gave him that excuse and he was only too happy to oblige. Now from the standpoint of the typical Mac user, you shouldn’t blame Jobs for engaging in some sort of conspiracy theory. Whether or not he really wanted to switch to Intel, developing Mac OS X for two types of processors was a smart idea. It gave him the options he needed when his existing suppliers let him down.
In the scheme of things, maybe Jobs is too demanding of his suppliers, although we get better products as a result. Why settle for mediocrity when you can get something better by pushing the right buttons and making what some regard as unreasonable demands? From my perspective, it doesn’t really matter how Jobs chooses to control his suppliers or even his troops over at Apple. They choose to work there and they know what they’re getting into when they joined the company. I’m concerned with the results, and I’m mostly pleased with what Apple has done in recent years.
So if Jobs tells us that the PowerPC has, to his way of thinking, run out of steam as a viable part for Macs, I am inclined to cut him a little slack. While I am intrigued by the Power Mac G5 Quad, the fact that it still hasn’t reached 3GHz shows that Jobs was right to cut and run. He is not engaged in some grand Machiavellian plot. He simply wants to sell as many Macs as he can, and if it takes a wholesale processor switch to keep up with the state of the art, so be it.
Now the other part of the equation is how quickly software developers can move their applications to Universal Binaries, the magic bullet that allows them to run on Power PC and Intel. Well, if you check the info on some recent applications, you’ll find that some have already made this grand switch. Of course you wouldn’t notice the difference unless you had one of those special developers computers with a Pentium 4 that are being leased by Apple.
This augers well for the forthcoming switchover, although I concede that some of those sprawling productivity applications, such as Adobe Photoshop and Microsoft Word, won’t change until the next major paid upgrade is out. Why pay what will be a lengthy and possibly difficult development project, when they can just add some extra features and get you to pay instead?
Many of the Universal Binary versions, however, will be free upgrades. If a developer has used Apple’s own programming software, it’ll just be a natural evolution. Check a box to add Intel support, tweak the code a little bit here and there, and just make it happen.
So when you buy your first MacIntel, you’ll find a healthy amount of software is ready to run in native mode. As to the rest, don’t fear Apple’s Rosetta emulation technology too much. True, it won’t run Mac OS 9 applications, unless something changes drastically over the next year or two. But performance will probably be a lot better than you expect. The company that supplies the technology to Apple, Transitive Corporation, says it can deliver 70% to 80% of native processor speed. In the real world, you will see little if any difference unless you are running software that taxes the capabilities of the fastest available chips. Instead, things will appear to run quite normally. Also bear in mind that the processor involved will be a lot faster than the one in today’s Macs, so even a speed hit of 30% may still yield a faster result. More to the point, if Mac OS X’s graphical user interface seems supercharged, you won’t care whether it’s native or emulated except for that small subset of software where it really matters.
You see, I’m really optimistic about the transition to Intel. I don’t care much of Apple is forced to include that infamous Intel “stinger” in its broadcast ads to get those precious co-marketing dollars. I’m interested in results, and if version 1.0 bugs can be minimized, I may indeed be near the front of the line to buy the first PowerBook powered by Intel.
Print This Post
