←back to thread

US Intel

(stratechery.com)
539 points maguay | 2 comments | | HN request time: 0.412s | source
Show context
cubefox ◴[] No.45027518[source]
Maybe it's worth noting that the new CEO's explosive statement, Intel would give up development of leading edge nodes (14A and beyond) unless they find a large external customer in advance, hardly got any attention on Hacker News or other websites. But that was almost certainly the reason the US administration got involved. This blog post thankfully makes this point clear.

In the past, most people (including myself) just assumed the worst case was merely Intel selling its chip manufacturing division to some other US company which would then continue to develop new advanced nodes like 14A.

But that was not at all what Lip Bu-Tan (the new Intel CEO) suggested at the earnings call. He said Intel would simply stop developing new nodes and just use the existing 18A fabs as long as there is demand. And then, presumably, closing all the fabs and fully switching to TSMC, becoming another AMD.

replies(2): >>45027746 #>>45034434 #
1. Panzer04 ◴[] No.45034434[source]
The problem with leading edge fabs is they require obscene amounts of capital to develop and are apparently (based on Intel's experience) extraordinarily risky to build.

That's a hard thing to sell in any environment.

replies(1): >>45039665 #
2. cubefox ◴[] No.45039665[source]
Well it's easy to sell if you are the one currently ahead of the competition (TSMC) and very hard to sell if you are behind (Samsung, Intel).