Carbon nanotube chips could eventually bring us pocket-sized super computers
By Andrew Grush July 16, 2014
Think today’s quad and octa-core mobile devices are impressive? You haven’t seen anything yet. Speaking at a talk at Semicon West, H.S. Philip Wong, Stanford University’s Professor of Electrical Engineering, talked a bit about the future of computing power, focusing on a theoretical 3D chip stack that interleaves next-gen memory and logic technologies using carbon nanotubes.
Of course, just because carbon nanotubes eventually have the potential to deliver amazing results, doesn’t mean we can expect the Samsung Galaxy S6, S7 or even the S8 to utilize this kind of amazing power. There are still challenges to be met before such a chip stack can move beyond theory. In particular, Eetasia says Wong notes three challenges ahead:
Carbon nanotube chips could bring us pocket-sized super computers
By Andrew Grush July 16, 2014
Think today’s quad and octa-core mobile devices are impressive? You haven’t seen anything yet. Speaking at a talk at Semicon West, H.S. Philip Wong, Stanford University’s Professor of Electrical Engineering, talked a bit about the future of computing power, focusing on a theoretical 3D chip stack that interleaves next-gen memory and logic technologies using carbon nanotubes.
The theoretical chip stack could eventually result in a computing system as powerful as the IBM Watson super computer
Though Wong admits the material faces huge technical challenges, the idea is that the interleaving of layers of resistive and magnetic RAM with logical layers made from 1D and 2D field effect transistors could result in a computing system that would theoretically be as powerful as the IBM Watson super computer, but in pocket-sized form. This is more than impressive, considering Watson packs 2,880 3.5GHz IBM Power 7 cores and delivers 80TFLOPS.
Of course, just because carbon nanotubes eventually have the potential to deliver amazing results, doesn’t mean we can expect the Samsung Galaxy S6, S7 or even the S8 to utilize this kind of amazing power. There are still challenges to be met before such a chip stack can move beyond theory. In particular, Eetasia says Wong notes three challenges ahead:
The material is not suitable for the high-temperature doping processes used in today’s chip fabs. Researchers still need to improve the purity of the material they grow. And, like all transistor materials, it faces challenges when contacts scale to increasingly small sizes.
Bottom-line, there will come a time when carbon nanotubes provide computing power than is significantly more impressive than what we see today — even if not necessarily as aggressive of a jump as Wong’s stack claims to provide. Outside of Wong’s own theoretical stack, there are other companies exploring ways to utilize carbon nanotubes in order to advance computing technology. The most notable example is IBM, who announced earlier this month that they have plans to produce the first commercial carbon nanotube chips by early 2020.
Carbon nanotube chips could bring us pocket-sized super computers