Kind of a surprising move, really. At a point where the key players in the console wars are all moving to PowerPC-based technologies, Apple is moving away from it?

Maybe Apple has finally realized that it’s competitive pricing, not testimonials from half-stoned high school students, that will convince John Q. Dell to switch? The CPU was the primary reason for the increased price of Macs over their similarly-equpped x86 competitors, wasn’t it?

My big question is, if they embrace a specific set of hardware for the core computing tasks, will manufacturers be able to build motherboards so that hobbyists/dorks will be able to build their own Mac-compatible boxes for OS X? Because, man, that would be sweet.

It’s all about the notebooks; specifically, IBM couldn’t deliver a mobile G5 soon enough. Right now, the fastest Powerbook is bested by an entry-level iMac G5, which is annoying in the extreme for those of us wanting to upgrade.

As sweet as it sounds, I doubt that Apple will be remove the licensing restrictions from the platform anytime soon, because Apple wants to control the entire computing experience (not just the OS). However, it does present them with an exit strategy for the future, should things not go so well on the hardware side of the business.

I’m worried more that Apple may suffer from the fate that befell Osborne computer, where a preannounced product kills sales for current stock (i.e. who’s going to buy a G4 Powerbook or iBook now when in a year they can get a laptop for the same price at twice the clock speed)? In Osborne’s case, this killed their company. Apple is not Osborne (they have tons of cash and other successful products) but it can’t help in the meantime.

And no, the x86 will not make things cheaper, per se. I personally don’t think Apple gives much of a rat’s ass about capturing the lowest of the low end (Mac Mini aside, they are not looking to be another Dell). It’s in the brand as much as it is about the design, the experience. That’s why the iPod sold so well in spite of the premium cost. After all, you don’t see Porsche and BMW clamoring to get a piece of the Toyota Corolla market, do you?

I’ve always been a little wary of automobile analogies when discussing computers and other technology. In this case, the analogy breaks down because of the breadth of the envy factor. Porsche and BMW don’t have that many detractors (save for those who think that all luxury items are bad-evil), whereas one can scarcely find a technology-related forum, BBS, USENET froup, or IRC channel that hasn’t at some point seen a message or five of the “OMG APPLE IS TEH GAY” sort.

Regardless… my concern was not so much with catering to the lowest common denominator, but more with Macs finally matching their competition on a price:performance basis for once. Even if they only touch the high-end market, it would be beneficial. I’m reluctant to drop $3k on a computer, that I’ll admit. But I might be less hesitant if that computer wasn’t outclassed by one of a competitor’s midrange products at half the price.

Or to return to the car analogy, why would I buy a Porsche if I could spend half as much on a new Honda that would outperform it?

You might put more money on a slower product if it were less likely to explode during usage. Meaning: it’s not just the speed that counts, but the whole package. That is one of the reasons I’m still using my little iBook; it is hardly fast, but I like using OS X so much that I tolerate the suck-ass performance. And, since I’m no longer into gaming, I don’t need a the s00per-1337 setup that some in the PC gaming world demand.

That said, it sucks that the fastest G5s are currently only in desktops, and only are “touching the competition” when run in a dual CPU configuration. As far as laptops are concerned, anything offered by Intel today blows the doors off of the fastest Mac notebook. This is lamentable and a cause of much consternation for people like me, who want to upgrade, but aren’t going to do it twice in as many years.

So, I anxiously await the first Macintel Powerbook. My little G3 iBook needs to hold on for one more year. Until then, I’ll start saving my money.

Alms?

A couple comments on other comments here:

IBM couldn’t deliver a mobile G5 soon enough

IBM didn’t supply the mobile chips, Freescale (a spinoff of Motorola) did.

You might put more money on a slower product if it were less likely to explode during usage.

You’re really comparing operating systems here, not computer makers or chipsets. A Windows box running on x86 might indeed not be stable enough for your needs but if you format your hard drive and install Linux, perhaps it is.

The previous comment raises a good point.

Most users who aren’t tied to Windows by a specific software app, and who are the sorts who don’t mind learning a new OS on their own time, and who give two craps about stability/security have probably already converted to Linux.

So really, the only people who might be enticed to the Mac side without a reduction in price are going to be those who want something more stable/secure, but who are tied to Windows by one or more apps that are already available on both Win and Mac platforms (Reason, any number of Adobe products, etc.)... in other words, the current base of potential converts.

That being the case, the switch to Intel hardware would do nothing to broaden their potential user base. I can’t imagine that this switch is going to be cheap for Apple, so I can’t understand why they’d pursue such a course of action unless it would enable them to do something that would help increase their userbase (like lowering prices), or they foresaw some sort of dead end in sticking with the PowerPC.

Bill,

A couple of things:

IBM didn’t supply the mobile chips, Freescale (a spinoff of Motorola) did.

This is true; however, Freescale was making the G4 chip for mobile use. It is not clear to me that they had the ability to produce the PPC 970 (G5) at all, even if they wanted to. That was up to IBM, and from all indications, it seems that IBM did not (or could not) make good on this.

You’re really comparing operating systems here, not computer makers or chipsets. A Windows box running on x86 might indeed not be stable enough for your needs but if you format your hard drive and install Linux, perhaps it is.

Mixing metaphors is harmful to one’s health. I should probably be careful. :)

Andy,

So really, the only people who might be enticed to the Mac side without a reduction in price are going to be those who want something more stable/secure, but who are tied to Windows by one or more apps that are already available on both Win and Mac platforms (Reason, any number of Adobe products, etc.)... in other words, the current base of potential converts.

How about those of us who are willing to pay more for a better user experience than any Linux distribution in existence today can provide? That user experience is in part provided by the OS, but also by a stable of very nice “Apple” only applications (the iLife series, NetNewsWire, Quicksilver, etc.). Some of us feel if you pay more, you get a little more, clockspeed non-withstanding. That said, it sucks that the current Apple laptops lag in performance over cheaper Wintel versions. Hopefully, that will be rectified in about a year.

If it were purely about speed and low cost for me, I’d go and grab a barebones PC kit and build a Pentium 4 + hyperthreading machine and slap a tricked compile of Gentoo Linux on it. But the results would be fast and cheap, not necessarily great.

Considering the extra cost to Apple and the extra hardship (no matter how trivial) imposed on developers, moving to x86 had to have been a last resort. I fully believe that if the PPC 970’s (and its eventual successor’s) roadmap looked better, the announcement would have never been made.

A good analysis (one of many) on all of this can be found at Daring Fireball.

I guess it all depends on how much more you’re willing to spend. I mean, OS X upgrades are already more expensive than home-user Windows upgrades typically are ($129 for Tiger, $99 for XP Home). I could deal with that kind of delta.

But I have a huge problem with paying out a lot more for a box containing basically the same hardware as a Wintel box just so that I can turn around and pay more for the OS.

And it’s really hard to argue that the “Apple style” is worth the price hike. Every so often they turn out a decent design, such as the iPod, but their design history is also riddled with things that should have the designers hanging their heads in shame. Like the “Fisher-Price My First Computer” iBook, or every iMac design ever. And even the iPod is pretty flawed in design. Super fragile, and a fscking magnet for scratches and fingerprints. Brilliant.

Don’t get me wrong… I love my iPod. I could have certainly gotten something with higher capacity, possibly even cheaper, but I was willing to pay the premium for iTunes support, because iTunes is a brilliant piece of software.

Likewise, I can see the value in OS X over Windows or Linux. However, I cannot help but separate the OS from the hardware, and paying twice as much for hardware just because it happens to have some little flag in the BIOS that lets it run OS X makes no sense to me… especially considering that every piece of software I currently own will be useless, except maybe through emulation.

I suspect that I’m not the only person who thinks like this.

Smerp,

You wrote:

This is true; however, Freescale was making the G4 chip for mobile use. It is not clear to me that they had the ability to produce the PPC 970 (G5) at all, even if they wanted to. That was up to IBM, and from all indications, it seems that IBM did not (or could not) make good on this.

I also don’t really understand what the dependency was between Freescale’s chips and IBM’s lower-level PowerPC technology. I’ll ask some folks over in Server and Technology Group to find out.

Bill,

I don’t think Freescale had the ability or capacity to fab G5s (i.e. I think that was the exclusive domain of IBM Microelectronics). I’d love to know if that was true or not.

Even if Freescale did, someone has to design a low-power version. VLSI designs don’t just spring fully formed from the heads of engineers. It’s unclear that IBM or Freescale had the ability (or desire, for that matter) to spend the time and money necessary on getting the power + heat down for a mobile version of the PPC 970. All indications is that if it could have been done, neither company was willing or able to make it a priority.

So really, what we’re looking at is this scenario:

IBM is drooling at the prospect of having Microsoft, Sony, and Nintendo sticking PowerPC tech in their consoles, which combined will easily ship millions within the next several years, which leads them to the conclusion that high-power applications, without the constraints of heat or efficiency concerns, are Where It’s At.

So Apple calls up and says, “Hey, we’d like to put something more powerful in our laptops, but without giving our users scorched laps or sucking down batteries like they were hors d’oeuvres at a compulsive eaters’ conference.”

And IBM says, “yeah, we’re not really interested in doing that. click.”

I asked a product team lead in PowerPC to explain the dependency between Freescale on IBM and he said the following:

Freescale has an architecture license to design and manufacture PowerPC processors. The G4 is a Freescale processor. The G5 is an IBM processor. We each manufacture our own processors. Freescale does not depend on IBM for their next generation PowerPC.

Hey guys, thought you might enjoy this post by Irving Wladawsky-Berger, IBM’s VP of Technology and Strategy, on the ecosystem we’re trying to build around the Power architecture.

http://irvingwb.typepad.com/blog/2005/06/humanizing_info.html