>What exactly does that say about their corporate strategy before recently?

I get the impression that people think that Intel's stagnation after 14nm was due to complacency and the desire to avoid taking risks. That is literally the exact opposite of the truth because almost all of their problems with the 10nm node stemmed from it being far too ambitious. It incorporated a large number of extremely cutting edge technologies which were leaps ahead of anything else being attempted at the time. This over-ambition lead to nearly half a decade being wasted because, as it turns out, they were a bridge too far. If Intel had been able to pull off what they originally intended with 10nm when they intended it they would have been absurdly far ahead of all of their competition but unfortunately that choice actually resulted in them falling behind.

The first half of their post-14nm stagnation was because they were overly ambitious for 10nm. But the second half of that era was because of their hubris, denial that things weren't going to plan, and inability to adjust plans. They eventually admitted defeat (to some extent) and back-ported one of their 10nm designs to 14nm as a stopgap, but they started that effort at least two years later than they should have.

I’m not denying most of that, but it feels extremely one-sided. I feel like I’m reading Intel investor relations right now. I am not suggesting Intel was not working on cutting edge technology, but so were other companies, and some of them succeeded where Intel failed. Meanwhile, Intel had multiple business units with similar issues, especially their modem business, and I doubt the postmortem said “Oops, we were too ambitious!” Bullshit. They had organizational issues and complacency. I’m sure 10nm was plagued with many problems that weren’t complacency too. That’s part of the table stakes when you are in the business of defying physics.

The financials tell a different story.

Intel’s process technology lead was the core of their strategy and explains 95% of what’s gone wrong. The 5G modem design was an irrelevant sideshow that had no impact on 10nm.

I’m not suggesting the same exact thing happened, but it’s hard to believe there weren’t some overarching management issues going on, and it seems likely they contributed to problems further down the line. A couple blunders could’ve been a fluke, but I think we’re past the “it could’ve been a fluke” stage.

Of course, this is all pure conjecture, but nobody can really be all that sure. I’m sure even people at Intel have differing opinions about what was wrong.

Word on the street is also that the 14nm death march resulted in a lot of engineers taking early retirement which left them in a worse position to pursue 10nm.

From what I've heard, that was Intel's doing. They intentionally pushed older engineers into early retirement to save money.

"These geese lay nice, gold eggs but cost too much to feed. Let's get rid of them."

Well the old guard were the ones who screwed up 14nm, so were they really the geese?

People forget that around the 14nm timeline Intel was completely DOMINATING in terms of performance. And this was not THAT long ago. Will be interesting to see where everything is in 2 years

In this case, we'd expect them to have gotten some R&D out of those overambitious failures, right? So we should hope to see some really snappy progress in post-10nm nodes?