Economics, not physics, will make Moore's Law no more

Former Intel chief chip architect says law governing transistor shrinkage could be dead and buried by 2020

A leading silicon scientist has claimed Moore's Law will be moribund within a decade, with economics – rather than physics – the likely cause of death.

Named after Intel founder Gordon Moore, Moore's Law dictates that the number of transistors on integrated circuits doubles roughly every two years. Over 30 years, it has been responsible for a 3,500-fold increase in microchip speed, enabling manufacturers to assemble ever-more powerful notebooks, tablets and smartphones.

Although proclamations of its demise are nothing new, former Intel chief chip architect Robert Colwell claimed on Monday it could be dead and buried within seven years, at the 7nm node.

"I pick 2020 as the earliest date we could call it dead," he said in his keynote speech at the Hot Chips Conference at Stanford University, according to EE Times.

Colwell said he could be talked into 2022 and 5nm, but rejected suggestions it could go much further, adding that the law's demise could be as much about economics as physics.

Intel is only willing to plough billions into the fab plants and design teams to design new chips as long as it thinks that investment will repay itself, he said.

"If there is doubt that those profits will arrive, and possibly if they just doubt they can come up with the necessary silicon improvements, they may not want to make the investment at all," he was reported to have said by CNET. "Should a major player like Intel make such a call, that would effectively end Moore's Law all by itself, because then the various companies that make the super expensive tools for chip production will themselves not make the investments needed to keep Moore's Law alive."