4 ms·

My uninformed gut says Intel was more screwed by missing mobile than was immediately obvious. Sure, it lost them a bunch of money to not be in cellphones. But

My uninformed gut says Intel was more screwed by missing mobile than was immediately obvious. Sure, it lost them a bunch of money to not be in cellphones. But also, as transistors have gotten smaller and designs have gotten more complex, yield, (which was always important) has become a massive problem. Chiplets are the solution, a defect just destroys one chiplet instead of the whole chip. Designs should use more chiplets, the smaller the better. Intel’s trick of making giant monolithic designs turned into a liability. The smartphone fab, it turns out, is making chips that are perfectly fine for desktops and servers if you jam enough in one package. Epyc was the writing on the wall.

UncleOxidant 2y ago
Certainly getting into mobile would have taught them how to compete in a very different market with lower margins and that would have put them in a better position today. Intel was addicted to the easy money of the Wintel days and has had a hard time transitioning. Otellini's decision to not supply Apple with cpus for the iPhone was a huge mistake. (I seem to recall that the calculations to determine that it wouldn't be profitable might even have been based on a mistake in a spreadsheet, but that's probably apocryphal)