8 May 2003 Moore's law is killed by classical physics: can quantum information save it?
Author Affiliations +
Proceedings Volume 5115, Noise and Information in Nanoelectronics, Sensors, and Standards; (2003); doi: 10.1117/12.502062
Event: SPIE's First International Symposium on Fluctuations and Noise, 2003, Santa Fe, New Mexico, United States
In a recent study (Physics Letters A 305 (2002) 144-149) the forthcoming end of Moore's Law was predicted. The study was based on the assumption that the power dissipation was dominated by the energy dissipated during charging/discharging the CMOS gate capacitance during bit flips. In the present paper the fundamental lower limit of power dissipation during the operation of a CMOS gate is obtained. The results indicate that the energy efficiency of today’s microprocessors is extremely low. Thus, a significant improvement of the energy efficiency of microprocessors may be able to prolong the lifetime of Moore's Law. We compare the results with published data on the lower limit of power dissipation of quantum gates. Interestingly, "Classical" beats "Quantum," if we give the same chance to them. Finally, we evaluate the energy cost of Shannon-information transfer. This measure, which cannot be improved by error correcting algorithms, is the ultimate one, the real characteristic of performance versus power dissipation. In the most ideal case, the CMOS gate performs by at least an order of magnitude better than the quantum gate. It is shown that, at the same complexity of hardware, same speed and same temperature, quantum computing means more noise, less information channel capacity and greater power dissipation than the same measures in classical computers.
© (2003) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Laszlo B. Kish, "Moore's law is killed by classical physics: can quantum information save it?", Proc. SPIE 5115, Noise and Information in Nanoelectronics, Sensors, and Standards, (8 May 2003); doi: 10.1117/12.502062; https://doi.org/10.1117/12.502062


A quantum Fredkin gate (Conference Presentation)
Proceedings of SPIE (December 05 2016)
Trends in merged DRAM-logic computing
Proceedings of SPIE (November 17 2000)
VLSI digital signal processing: some arithmetic issues
Proceedings of SPIE (October 22 1996)
An extension to DNA based Fredkin gate circuits design...
Proceedings of SPIE (December 05 2005)

Back to Top