You had me believing you and trying to figure out what on earth might be behind the decision -until you mentioned that Intel's chips run cooler and offer a better roadmap.
They don't: unless you're comparing a Pentium M to an overclocked G4, but that's where the roadmap comment comes from too, isn't it? It's not that Intel has a better roadmap - it's that IBM's doesn't include you except as a subsidiary.
Unfortunately Intel doesn't have one either. In fact if they don't have a rabbit hiding in the hat, they're not going to last five more years as an independent - and you may think that's a BS prediction, but I'm not just the only pundit to call Cell right, something I wrote for Linuxinsider last year should seem a little more compelling today:
This will get more interesting if, as reported on various sites including Tom's hardware, IBM has been burning the candle at both ends and will also produce a three way, 3.5Ghz version of the PowerPC for use on Microsoft's X-Box. Whether that's true or not, however, my belief is that IBM chose not to deliver on its commitment to Apple because doing so would have exacerbated the already embarrassing performance gap between its own server products and the higher end Macs. Right now, for example, Apple's 2Ghz X-serve is a full generation ahead of IBM's 1.2Ghz p615, but costs about half as much.
Unfortunately this particular consequence of Apple's decision to have IBM partner on the G5 is the least of the company's CPU problems. The bigger issue is that although the new cell processor is a PowerPC derivative and thus broadly compatible with previous Apple CPUs, the attached processors are not compatible with Altivec and neither is the microcode needed to run the thing. Most importantly, however, the graphics and multi-processor models are totally different. As a result it will be relatively easy to port Darwin to the new machine, but extremely difficult to port the MacOS X shell and almost impossible to achieve backward compatibility without significant compromise along the lines of a "fat binary" kind of solution.
In other words, what seemed like a good idea for Apple at the time, the IBM G5, is about to morph into a classic choice between the rock of yet another CPU transition or the hard place of being left behind by major market CPU performance improvements
Good thing you had a Plan B, Steve; too bad it was the wrong one.
You know what the right one is? Ya, I know, it's disgusting but I'm going to quote myself again, and from the same article too:
So what can Apple do? What they should have done two years ago: hop into bed with Sun. Despite its current misadventure with Linux, Sun isn't in the generic desktop computer business. The Java desktop is cool, but it's a solution driven by necessity, not excellence. In comparison, putting MacOS X on the Sun Ray desktop would be an insanely great solution for Sun while having Sun's sales people push SPARC based Macs onto corporate desktops would greatly strengthen Apple.
Most importantly, SPARC is an open specification with a number of fully qualified fabs. In the long term Apple wouldn't be trapped again and in the short term the extra volume would improve prospects for both companies. Strategically, it just doesn't get any better than that.
Niagara rocks. You want low power use for a laptop? How about an eight way 1.4Ghz SMP core with TCP/IP and cryptography done in hardware - at 65 watts flat out. There are some serious software issues, but get past them and you've got eight to ten Xeons in the box - at 65 watts.
Sun's president, Jonathan Schwartz, put a nice invitation for you in his blog last Sunday. Maybe you should think about it, go for a drink with him, talk about the threat IBM poses to both of you. Get him to tell you about interval math on a software embedded array processor -you might like what you hear. And remember one thing: what you just did? Nobody at IBM is going to call a board meeting over it - but a Sun alliance? That would be different.