Also if you don't really care about yield, many things are not that critical. I did a PhD in semi-conductor physics, and most processes for silicon are actually

Also if you don't really care about yield, many things are not that critical. I did a PhD in semi-conductor physics, and most processes for silicon are actually surprisingly robust (compared to processes of any other material). Also many things are just done the way because people know it worked before, which does not mean you can't use less fancy equipment and it would still work.