The Wall Street Journals story that Apple is in talks with Intel over the possibility of using its processors in the Mac represent arguably the most solid confirmation yet of long-term reports that the chip maker has been angling for one of the few remaining markets that have resisted its charms, and that Apple has been not entirely faithful to its PowerPC partners.
There have been sightings of an Intel port of Mac OS X before.
In August 2002, eWEEK.com published a long and detailed report on Marklar, a secret project lurking in the Apple labs to maintain an Intel-compatible Mac OS X in step with the current PowerPC version—at the time, the recently released OS X 10.2 Jaguar.
However, little has been heard of Marklar since and some had assumed that the adoption of the IBM PowerPC 970—AKA G5—meant that the project had either been canned entirely or relegated to a sideline, acting as insurance policy in case IBM lost confidence in the PowerPC.
On its release, the performance boost delivered by the PowerPC 970 lifted the Mac from the doldrums into the big league of desktop performance, and ensured that Apples power-hungry customers in publishing and video continued to buy its high-end products.
As the chip has migrated down the product line into the consumer desktops in the shape of the iMac G5, it has also helped to ensure that Apple has remained competitive against low-cost versions of the Pentium 4 and AMD Athlon processors.
Since its release, Apple has faced two problems with the G5.
The first is that development of the chip has been slower than expected, with Steve Jobs forced to backtrack from his promise at 2003s Worldwide Developer Conference that there would be a desktop Mac clocked at 3GHz within 12 months.
In fact, although its been widely perceived that IBM has been slow to upgrade the PowerPC 970, the company has delivered decent speed boosts.
Since June 2003 it has effectively managed to keep pace with Intel, moving from 2GHz to 2.7GHz while the Pentium 4 has been upgraded from 3.4GHz to 3.73GHz.
Yet, with Intel itself moving away from pushing clock speed ever higher toward a future that emphasizes a multi-core approach, its hard to see why Apple would choose to take the risk and expense of decamping from PowerPC to Intel on the basis of whats happened with the G5 so far.
Unless Apple believes that IBMs commitment to PowerPC is wavering—something that Big Blue has given no indication of—then jumping ship while there is still plenty of room for development of the G5 series appears an unlikely move.
The Fate of Apples
What may be concerning Apple more is the fate of its portables.
Here, the company has been forced to rely on the aging G4 series, which is clearly coming to the end of its life cycle—and, thanks to the huge power requirements of the G5, there appears to be little chance of that chip making it into a PowerBook anytime soon.
Meanwhile, Intel has been pressing forward with the development of its Pentium M series so successfully that it has, in many ways, outshone the development of the Pentium 4.
The importance of the PowerBook and iBook lines to Apple shouldnt be underestimated.
Revenue contributed by the PowerBook in Apples last financial quarter was second only to the iMac, and the company sold significantly more PowerBooks than Power Macs.
If the future plans of IBM no longer include a low-power chip in the G5 series, it would make sense for Apple to look elsewhere for its portable processors—and, if Freescale (Motorolas chip business, now spun out as a separate entity) is no longer interested in the PowerPC except in the lucrative embedded market, Apples only option would be to look to Intel for the PowerBook line.
However, even if the reports prove entirely accurate, it should be remembered that there is little chance that Apple will produce a version of Mac OS X capable of running on the average Windows-laden PC.
Supported hardware will be strictly limited, and its likely that the company would put hardware roadblocks into place that would make it more difficult to use an Intel version of OS X with a generic machine.
Just as it took a major reverse engineering effort to clone IBMs original PC-BIOS and create the market for PC clones, so it would need a similar effort to recreate the inevitable ASICs on an Intel-equipped Mac motherboard.
Whether such an effort would be worthwhile is a moot point.
There are many other issues that would have to be resolved. For example, what degree of compatibility would there be between PowerPC applications and the Intel-equipped Mac? Would there be a PowerPC emulation layer, perhaps based on technology from a company like Transitive, whose Web site claims to be able to “allow any software application binary to run on any processor/operating system”?
Or would applications need to be recompiled? This is not necessarily an onerous task, but still one that would cut the appeal of an Intel-equipped Mac.
Either way, the Wall Street Journals report will undoubtedly build a huge level of excitement for Junes Worldwide Developer Conference, which would be an ideal place to announce this potential future direction.
After the introduction of OS X 10.4 Tiger, it was thought that WWDC would be relatively quiet: Perhaps this will, in fact, turn out to be one of the momentous conferences in Mac history.
Check out eWEEK.coms for the latest news, reviews and analysis on Apple in the enterprise.