1) The article is primarily concerned with Dirac's relativistic formulation of quantum mechanics. This is why they make the claim
'Without understanding the origin of spin, and the Dirac statistics, you wouldn't have mobile phones, computers or anything else that runs on electronics.'
However, Fermi-Dirac statistics can be derived directly from the properties of fermions. True, you can derive the fact that electrons will be fermions because of their spin from the spins-statistics theorem, but you only need to know the Pauli Exclusion principle to get the F-D statistics.
2) My understanding, and speaking with engineers in the field, is that most of the applied industrial research uses empirically derived band structures (which definitely does require Rel. QM to derive from first principles) and regular QM to design devices. This is also the approach I've seen in most semiconductor simulators. However, it is entirely possible that Intel or AMD take a different approach. Not having access to the tools they use, I can't speak to their tech.
However, you seem to be a knowledgeable person, so I'd be more than happy to be informed that I'm full of shit here. :)
1) The article is primarily concerned with Dirac's relativistic formulation of quantum mechanics. This is why they make the claim
'Without understanding the origin of spin, and the Dirac statistics, you wouldn't have mobile phones, computers or anything else that runs on electronics.'
However, Fermi-Dirac statistics can be derived directly from the properties of fermions. True, you can derive the fact that electrons will be fermions because of their spin from the spins-statistics theorem, but you only need to know the Pauli Exclusion principle to get the F-D statistics.
2) My understanding, and speaking with engineers in the field, is that most of the applied industrial research uses empirically derived band structures (which definitely does require Rel. QM to derive from first principles) and regular QM to design devices. This is also the approach I've seen in most semiconductor simulators. However, it is entirely possible that Intel or AMD take a different approach. Not having access to the tools they use, I can't speak to their tech.
However, you seem to be a knowledgeable person, so I'd be more than happy to be informed that I'm full of shit here. :)