The Apple/Intel Report: Did Apple Rush to Judgment?
July 9th, 2005The way Steve Jobs tells it, Apple decided to switch to Intel because its processor road map is superior to IBM’s. On the surface, it does appear that IBM was unable to deliver the parts Apple needed when it needed them. Since then, there’s been plenty of speculation as to just what chips will make their way into the next generation of Macs, which some of us call Macintels just to distinguish the new from the old. Or just to be silly, or a little bit of both.
Now comes news from Japan of new versions of IBM’s PowerPC 970, affectionately known as the G5, and perhaps you’re wondering if maybe Apple was a little too quick on the trigger. The first, the 970MP, is a dual core line, meaning two processors reside on a single chip. Sounds impressive, but the maximum speed is listed at 2.5GHz; in other words, slower than today’s 2.7GHz. Unless the chips are cheaper, and don’t run near as hot, there’s no real upside for Apple unless it puts two of them in a future Power Mac, delivering the equivalent of four processors. I suppose we’ll find out when the chips go into production later this year, but it’s clear that 3GHz still remains a pipedream.
The other chip is the 970FX, a line of low power processors topping out at 1.6GHz, and using just 16W of power. Assuming it runs reasonably cool, and that may be a stretch, does that mean a PowerBook G5 is finally in our future? But more important, does that 1.6GHz chip deliver substantially better performance than today’s 1.67GHz G4? If not, there is no reason, other than psychological, for Apple to use it.
On the other hand, with these new developments, you have to wonder where IBM will go from here. Will the 970MP and 970FX, assuming they go into production on schedule, which isn’t certain by a long shot, scale up sufficiently to justify their use by Apple? What is IBM going to deliver in 2006 and 2007?
Well, you can be sure there will be updates to Apple’s PowerPC computers until the switch is complete. Those computers will indeed be faster than the ones you buy today, and Steve Jobs said as much during his WWDC keynote last month. But remember, it’s Intel’s long-range road map, not just IBM’s inability to deliver the right parts, that influenced the switch. I do not for a moment believe Steve Jobs did it strictly because he got mad at IBM. A long-range strategy is at work here and Intel chips that haven’t been discussed in the media may well be the deciding factors, not just short-term advantages. Apple wouldn’t make a huge investment in moving to a new processor, and take a possible sales hit, simply as a result of pique with its present suppliers. Sure, Steve Jobs can become emotional about things when they don’t go his way, but it takes more than emotion to influence such a crucial business decision. Lest we forget, Steve Jobs is one smart businessman.
In the end, I don’t think those new IBM chips are cause for Apple to change its mind. Apple knew those chips were under development, that they would be announced and delivered sometime this year. They also know what IBM and Freescale Semiconductor will deliver next year and in 2007, or at least promise to deliver. As I said this is a forward looking decision that impacts Apple’s prospects for long-term success.
At the same time, I’m certain that some commentators will seize on IBM’s announcement as evidence that Apple screwed up big time, that it rushed to judgment. In addition, IBM is surely putting a different spin on the situation, that it really could deliver the processors Apple needs, despite public pronouncements to the contrary. For a concise summary on IBM’s posture, you might want to check out a recent article in eWeek on the subject.
My take on the whole thing is that promises aren’t enough. IBM promised Apple a 3GHz G5 by the summer of 2004. IBM was late in delivering chips for last year’s introduction of the new iMac G5 line and Apple stopped production of the previous models prematurely as a result. The promise and the reality were quite different, and there’s certainly no indication that IBM learned its lesson, or has resolved the problems with the 970, make that G5, to keep it from happening again. Of course, I’m not privy to the behind-the-scenes discussions among Apple, IBM and Intel that into that final decision announced last month.
Clearly Apple isn’t going to turn back at this late date. IBM had its chance, and stumbled at the plate. And when new PowerPC Macs appear, you shouldn’t pass them by simply because even better products are in the pipeline, because that would have been true even if IBM was able to get its chip act together.
Now let the debates begin.
Print This Post
