John Stokes over at Ars Technica has brought insider information to bear on Apple's move to Intel, providing what seems a sound and
most insightful perspective on the forces that drove the decision. What Stokes presents makes perhaps more sense than any other analysis that we have encountered and paints an impressive picture of Apple's business savvy as it moves its technologies forward.
While it is true that IBM fell short of their initial promises with regards to the near-term potential of the G5 processor...
First, let's talk about the broken 3GHz promise. It's apparent in hindsight that 3GHz on the 970 was never going to happen on a 90nm process without lengthening the 970's pipeline, which is a fairly significant change. Who knows why IBM promised Jobs 3GHz? All I do know is that despite the objections of some within IBM the company tried to hit that target without the needed pipeline change, and missed it.
...
IBM's announcement last week of the new dual-core PowerPC 970MP and the low power, high performance 970FX shows that, although it tooks some time, an apparent solution to Apple's stated gripes with the PowerPC was just around the corner when the company announced their plans to jump platforms. And Apple new it. So why, then, put themselves and the developer community through the ordeal of a processor architecture shift? Why indeed....
For the real reason behind the switch, you have to look to the fact that it's the iPod and iTMS—not the Mac—that are now driving Apple's revenues and stock price. As I stated in my
previous article on the switch, Apple is more concerned with scoring Intel's famous volume discounts on the Pentium (with its attendant feature-rich chipsets) and XScale lines than it is about the performance, or even the performance per Watt, of the Mac.
It's critical to understanding the switch that you not underestimate the importance of
Intel's XScale to Apple's decision to leave IBM. The current iPods use an ARM chip from Texas Instruments, but we can expect to see Intel inside future versions of the iPod line. So because Apple is going to become an all-Intel shop like Dell, with Intel providing the processors that power both the Mac and the iPod, Apple will get the same kinds of steep volume discounts across its entire product line that keep Dell from even glancing AMD's way.
Could it be that Apple was, in fact, looking beyond the Mac towards a product-wide shift to Intel? The points presented here are compelling. Volume discounts not just on the Pentium line, but also on the powerful Xscale processor (derived from the
StrongARM, used in the
Apple Newton MP2000) which would sit at the heart of future iPods would be a real win for Apple. And the rumored, future "Video iPod" would need more juice than the current iPod's 80MHz
ARM7 core, with which the Xscale is compatible. It all just makes sense.
Since the news of the new PowerPC's from IBM broke, we have seen many criticize Apple for being short sighted and impatient, abandoning IBM when a solution was close at hand. On the contrary, in light of this examination of the forces behind the decision, it seems that Apple has, in fact, worked a real coup.