A time span of 60-100 years make for bad extrapolations period - let alone in complex matters with feedback responsive systems. Not to mention most papers are s

A time span of 60-100 years make for bad extrapolations period - let alone in complex matters with feedback responsive systems. Not to mention most papers are supposed to be narrow in focus as opposed to the big picture. Even if a theory in itself is right it may not pan out entirely. It is a fact that CMOS uses twice the transistors per logic gate. It is proper to say that it would fit less processing power in an area than other doping options. However when power dissipation is the defacto limit then even higher power hardware will use CMOS.