Intel is betting on the notion that their process lead will grow and that others won't be able to keep up without having to spend said "disproportionate amounts of money."
Yet each new node gets more expensive. Intel's bet is that they can push through ever stronger headwinds and that everyone drafting behind them can't catch up? I wouldn't take that bet...
Well see whether that process lead can actually translate into better components that customers would pay a premium for. It might take a while, but that's the kind of long-term strategy that plays to Intel's strengths. Then Intel can just integrate everything but the kitchen sink into their silicon and gain advantages that the "roll your own IP" folks can't match.
Absolutely, that's Intel's plan.
Long-term strategies like this are great as long as the market is predictable. OTOH, long-term strategies are vulnerable to short-term changes in the market...
And if that doesn't work, maybe Intel has a bright future as a premium foundry, LOL ...
Maybe... The "premium" has to be good enough to keep pushing through those headwinds though.
30% of Qualcomm's revenues are from licensing, which makes sense given that QCOM used to be an IP-only company. Take that away and Qualcomm's margins should be more comparable to Intel's.
Certainly, not having all their eggs in one basket is working in QCOM's favor.