Has this been posted here before (its from last year - I apologize in advance if it has):

http://www.pcmag.com/article2/0,2817,2409635,00.asp

In this article these statements stood out to me which contradicts what Peter is saying:

"Of course, those tiny transistors may not actually be made of silicon at that point, according to Bohr. The Intel fellow was hesitant to spill too many details about the "exotic" approaches he said Intel's most far-reaching researchers are now experimenting with to achieve process technologies beyond 10nm".

" It looks like we have a a solution for 10nm and I'm confident we will have solutions for 7nm and 5nm"