My understanding is that about 80% of equipment of 22nm node is re-usable for 14nm. So I would consider any fabs with 22nm to be good candidates for 14nm. To upgrade Ireland from 65nm is a huge investment. Maybe it is better to build a new fab there instead. Israel is at 28nm but the fab is about 5 years old. How many tools can they use for 14nm or even 10nm? This involves also retrofit of cleanroom.
For the introduction of P1272 14-nm at Leixlip, building a new fab module is effectively what Intel did.
The stripped the old fab back to a bare dirty shell of a building taking out the waffle flooring, etc.
Blog Doing Math in FPGAs Tom Burke 18 comments For a recent project, I explored doing "real" (that is, non-integer) math on a Spartan 3 FPGA. FPGAs, by their nature, do integer math. That is, there's no floating-point ...