Intel opened a CPU design facility down the street from Motorola's PowerPC operation and bought away key parts of the team in '98-'99 setting Motorola back enough that they could no longer be ahead in the horse race. (For years, PPC had beaten Intel in some categories when the new PPC architectures came out, then Intel would pull ahead until the next generation. Apple always found something good to put on a Keynote slide. After the brain drain that wasn't going to happen.)
My take is that it's all about the execution of creating a chip. x86 for example is a much worse ISA, but there is some research (no source on this assertion) that says at the end of the day it doesn't matter too much since the machinecode is just translated into some internal RISC code.
So, it isn't that PowerPC is doomed from a technical standpoint. Instead it's all about money, business cycle stuff. Less sales means less R&D. Less R&D means you fall behind of the competition. IBM doesn't really have the heavy hitters they used to in the chip business (Relative to Intel/ARM/TSMC). If you want the newest flashiest tech, you can't really use their fab - that sort of stuff.
Every chip technology node is getting more expensive for foundries, which means the chip market will likely naturally converge to a small number of players.
I think it's truer to say that PowerPC stalled in performance per watt. The high end Power7/Power8 chips are massively powerful but that comes with Power and Cooling requirements that aren't going to work in a laptop.
Kindof true. I'd forgotten how much of a Mhz bump the P6 was -- It came out 12 months after the first intel Mac Pro (and ran at 4.7Ghz at the high end).
The P5 was a contemporary of the G5 (although the G5 was really a P4). But it wasn't quite the beast the P6 was.
And the G5 (the PPC 970 in IBM's nomenclature) was an "ultra-light" POWER4 — the POWER5 was ~30% quicker than the G5. POWER never really fell behind at the high end — there was just no real focus on anything except the high-end.