Everyone tends to forget that Samsung is a provider of leading edge semis manufacturing. It is important to remember them because there is a non-zero chance that they may not remain an alternative.
China’s semis industry has made incredible progress over the past ten years, but still has a long way to go.
Semis are changing, so it makes sense that some foundries are going to adopt new customer acquisition models. So just like we saw “Stealth IT” encourage cloud adoption, maybe we are going to see “Stealth Semis”.
Purchasing a chip based solely on public benchmarks is like picking your Viking raiding band based on their performance at an Olympic Fencing duel.
AWS cannot control the software it runs, but it can still benefit from having its own CPU by reducing power consumption and thus increasing capacity of its data centers. And maybe that also introduces a new form of customer lock-in.
The optical industry is plagued by manufacturing and packaging obstacles, but may finally be closing in the Holy Grail of Silicon Photonics. This time is totally different….
Much of modern technology is built on the idea of abstracting different layers of systems so that they can operate independently. For semis, the pendulum is now swinging back the other direction towards less abstract, custom chips.
There probably will not be thousands of companies designing their own chips – the upfront costs are just too prohibitive – but there will be enough to mark a major shift for the industry.
By our count 28 non-chip companies are developing their own chips. Of these, at least 16 have delivered multiple generations of their chips, signaling a long-term commitment.
Google is promoting the growth of open source tools for designing semis. The science fiction version of this story leads to everyone designing chips, the reality is going to be much narrower, but still means many more companies designing their own semis.