Intel inside the Mac: Questions, questions…
Posted by Pierre Igot in: Macintosh, TechnologyJune 6th, 2005 • 10:43 pm
So for once the rumours were true… Apple is actually switching to Intel processors. This is a big decision and a major shock.
In theory, it shouldn’t matter to the end user much, because the processor (CPU) is not something that he has to deal with directly in his everyday interaction with his machine. But in reality, the switch will have major consequences on a number of aspects of personal computing that the user does experience directly.
The first one is speed. Interestingly, while Steve Jobs is busy announcing and explaining the switch from the PowerPC to the Pentium, as noted by a reader on the MacInTouch web site yesterday, Apple’s own website still states quite clearly:
The PowerPC G5 out-shoots the Pentium 4 in a battery of tests. But it’s in the rough-and-tumble of real-world performance that the G5 really shines — shredding the PC’s reputation in the process.
I don’t doubt that this particular blurb will soon be removed from Apple’s website. And any seasoned consumer of Apple technology knows that this type of statement is more marketing propaganda than actually relevant information.
But still… It does make you wonder whether we are switching from a more powerful CPU to a more mainstream one. In truth, however, like most Mac users, I don’t know enough about processor technology to make a sound judgement here. Most important, we don’t know much about what’s in store for the next 10 years. So I am afraid that we have to trust Steve Jobs’ judgement on this one. He knows more than we do about what’s coming from both IBM and Intel in terms of CPU technology and, according to him, “Intel’s technology will help us create the best personal computers for the next ten years.”
It also seems to me that, in the past few years, raw CPU power has become less and less important (remember the Megahertz Wars?) and other speed bottlenecks have come to the fore. For example, I cannot help but notice that the raw specs of computer hard drives have not evolved much. Most of us are still using and buying computers with hard drives running at 7,200 r.p.m. Some have them have bigger caches and I am sure that there are a number of improvements other than raw speed that makes today’s hard drives faster than hard drives from a few years back.
But it seems to me that the speed (or lack thereof) of hard drives is more important to the end user at this point in time than the speed of CPUs, or video cards, or RAM. In that respect, Apple’s switch from the PowerPC to the Pentium probably matters less today than it would have a few years ago.
Another significant issue that end users will be faced with because of this switch is, of course, software upgrades. While Apple took pains to enlist Microsoft and Adobe in their press release on the switch and to stress that both software vendors will develop versions of their flagship software titles for both the PowerPC and the Pentium, the press release still leaves many questions unanswered.
One of them is: How much is this going to cost us? Will Adobe and Microsoft release Pentium versions of their existing Mac titles, or only of the next major revision? In other words, if I buy a copy of Adobe Creative Suite 2 (PowerPC) today, and a new Mac next year, will I have to buy a new version of Adobe Creative Suite (Pentium) or will I be able to get from Adobe, for a nominal fee, a copy of my current software that will run on my new Mac?
According to the information released by Apple, current PowerPC software will run on the Pentium machines, with an emulator. It is quite possible that the performance will be acceptable — but I don’t see how it can be great. (See John Gruber’s note on what will not run in that emulator. It’s quite a lot.) That is certainly not going to entice people to buy newer machines.
If recent history is any indication, what will happen is that, in order to really get the desired performance, you’ll have to buy expensive software upgrades, especially with Adobe and Microsoft. Smaller software vendors will probably be much more user-friendly and provide registered users of their current software with inexpensive upgrade paths if they transition to a Pentium-based Mac. But Adobe and Microsoft? No way. Adobe will probably force us to buy Adobe Creative Suite 3, which will combine the PowerPC-to-Pentium update with new features — as they did for the transition from the classic Mac OS to Mac OS X.
And Microsoft will probably use this as an excuse to force Mac users to buy yet another new version of their Office software with almost no new features and no bug fixes, just so that we can run it on a Pentium-based Mac. Remember Office v. X? It was, for the most part, a version of the software that was identical to Office 2001, except that it was Mac OS X-native. And Microsoft’s own literature stressed that their developers had to rewrite “millions of lines of code” — as if this was of any interest to end users. A decent software company would have taken this opportunity to clean up their mess and fix outstand bugs etc. Microsoft did nothing of the sort. They just made Office 2001 Mac OS X-native, and made us pay the full price of an upgrade for it.
I am pretty sure it will be exactly the same this time, for both Adobe and Microsoft. I sincerely hope to be proven wrong — but I also doubt very much that things will be any different.
Even without counting required software upgrades, the Mac-buying consumer will be faced with crucial and difficult choices in the next couple of years. Do I buy a PowerPC Mac or a Pentium Mac? What will be the best time to buy a new machine? It’s going to be tough, and lots of Mac users are going to adopt a wait-and-see approach, which will probably cause a significant lull in unit sales for Apple.
The good news is that, thanks to the success of the iPod, Apple is currently in the best possible financial shape to face such a situation. They have large reserves of cash, and are generating healthy profits from the sales of iPods, which can help compensate for slower Mac sales. If this switch is a risk that they had to take, now is probably the best possible time for it.
Other interesting issues include the fate of other systems running within the Mac OS X environment. The switch to Intel clearly marks the beginning of the end for the Classic environment. The classic Mac OS simply cannot run on a Pentium processor, either as a stand-alone system or as part of the Mac OS X architecture. Fortunately, Classic is becoming less and less necessary.
But my current employer, for example (a provincial French-language learning resource centre), is currently using a bunch of flat-panel iMacs running Mac OS X 10.3 with Classic, because they are using a Telnet emulator that only runs under Classic. And this is not expected to change any time soon, because the library management system that they are using still requires this Telnet emulator, and this library management system itself is not expect to change in the near feature.
The concern is therefore about what will happen in a couple of years when one of the flat-panel iMacs dies and we have to replace it with a new machine. If this new machine cannot run Classic, we’ll be stuck! I guess we’ll have to cross that bridge when we get to it.
On the other hand, this switch could be good news for Virtual PC users. Surely, with Mac OS X running on Pentium processors, Microsoft will be able to develop a version of Virtual PC that runs much faster than the current one, right? After all, Virtual PC will no longer have to emulate a Pentium processor — there will be one right there in the machine on which it’s running! But then, it’s Microsoft we are talking about. They’ll probably find a way to screw this one up just the same.
In conclusion (for now), we are probably entering a period of significant turbulence for the Mac platform. Yes, Apple did successfully transition from the old CPUs to the PowerPC back in the early 90s, and then from the classic Mac OS to Mac OS X in the early 2000s. But it was not all rosy all the time and the end users did have to deal with a number of technological and financial hurdles along the way.
However, I suppose that we just have to trust that Steve Jobs knows what he’s doing and that this is for the best.
June 7th, 2005 at Jun 07, 05 | 9:09 am
Correct, nowadays processor speed is generally not an issue any more, for some applications it of course is but for the most applications and users it’s more than enough.
I’m assuming the biggest reason for the switch was heat. Apple’s current powerbooks are not particularily fast (especially the memory bus is very slow) but they run very hot, at least my 12″ does. Intel’s Centrino technology for laptops is very good compared to G4 and it seems to be developing a faster pace than G4 chips from Freescale and apparently IBM was not interested in developing a low-power chip but instead did some game console chips. At the same time the laptops are selling more and more compared to desktops, so it is strategically important to have good offerings there.
One thing that is important to see, if IBM or Freescale get their act together soon and produce a great chip that outperforms Intel offerings there is nothing that would prevend Apple using that, the dual chip strategy which is now planned to last about two years could be extended for a short while. Of course nothing indicates that freescale or IBM would get their act together.
June 8th, 2005 at Jun 08, 05 | 12:15 am
The speed concern here is the emulation part. If too many important things are emulated, there is going to be a significant performance hit for people with their existing software. Will game developers release new Pentium versions of their existing games, for example? It’s a major concern, as users have no reason to repurchase a game that they already own.
The heat issue is indeed probably at the core of the decision.