Yea it definitely felt like at the time Intel was in a tough spot.. while the article is stating that the narrative about the margins being narrow for arm chips so Intel was nervous about the area, one thing I can attest to is that even if that narrative was false when it comes to Apple, competitors of Apple at the time definitely would be assuming that if Intel went into that space that they gave Apple a good price, so just the perception of them selling a good chip to Apple could have hurt Intel's fat profit margins. Like Apple was selling phones for the same price Intel was selling chips for -- so maybe the higher cost could be justified by Apple who knows, but the perception of Intel's profit margin they are willing to live with impacts their negotiations with other customers dramatically as well.
The point here is that there is tons of money to be made in manufacturing those chips that Intel lost out on by ignoring mobile. That “Apple makes their own chips” does not alone mean Intel could not have profited from this — as indeed TSMC, ASML, and ARM are.
This analysis misses kinda a major point. The nice thing about making mobile SOCs is that the yields are more forgiving. If you have 1000mm^2 of silicon that you're slicing into 10mm^2 dies, each defect (of which there will be many on early leading nodes), will only cost you a 10mm^2 chip, instead of a 50mm^2 chip. And because you're making them in volume, you get many many cycles to improve your process before you try to make bigger chips with it, ie: CPUs/GPUs. And because Apple wanted the increased performance/battery life, they reserved capacity from TSMC on these leading nodes in advance, helping to finance their development while providing a garunteed customer that was going to buy in volume. This gave TSMC a huuuge advantage over Intel that materialzied around ~2018.
That is actually a good point - the fab process would turn out to become another weak point (besides the missing moat of ARM) of Intel. There is no way Apple would watch TSMCs leadership year after year and still stick with Intel fabs.
Apple hates being beholden to one company. It’s bitten them many times in the past.
I suspect if anyone else could keep up with TSMC Apple would still dual source.
But between their lead and the benefits of being such a big customer they get first crack at the cutting edge stuff, single sourcing manufacturing probably makes the most sense.
And their interests are aligned, unlike their dependencies on MS/Adobe/Motorola/IBM/Intel.
Well volume is an issue. Even a low margin part which has very high volume sales can help you afford to build leading edge Fabs, and keep process technology leadership.
Apple can only do that now, because of the billions they have made on the iPhone, plain and simple.
And - more importantly - they were well geared to do that, since the beginning of computing.
There being no money in SoC's, is because they're all being made across the other side of the planet, mostly, from the intended final users.
If Intel were really 'leading edge', they'd have made desk-side custom fabrication a thing in the makerspace already. Such that I can, as a computer user, print 10 or 20 or X little chips, for my own specific purposes, non-mass-market.
This would be a truly revolutionary adventure from a 'grandfather of computing' style company.
Alas, the x86 is, indeed, everywhere. Grandfather Intel has a massive garden.
If only the SoC battles were truly localized, and a real computing revolution can happen (before its too late).
You should have a locally-built device in your hand.
Qualcomm market cap: $193 billion
TSMC market cap: $888 billion
Seems designing and fabbing SoCs is plenty valuable