Page 1 of 1

Fusion progress versus computer progress ...

Posted: Wed Jan 28, 2015 12:52 am
by Tom Ligon
Dr. Park tipped me off to a presentation by Greg Hammett of PPPL.

Part of this presentation looks at a field in which stunning progress over the years is now obvious to everybody in this forum: computers. It is a pretty good bet you are presently reading this on a machine with such stunning capability that it would have changed the world a few decades back. That's true even if the computer is a smart phone or tablet.

Moore's Law predicts increases in computer speed and capacity over time, and the empirical evidence supports it.

"Moore's law" is the observation that, over the history of computing hardware, the number of transistors in a dense integrated circuit doubles approximately every two years. The observation is named after Gordon E. Moore, co-founder of the Intel Corporation, who described the trend in his 1965 paper. (Cribbed from Wikipedia)

https://en.wikipedia.org/wiki/Moore%27s_law

Hammett presents data on a plot with Moore's Law to suggest that fusion progress has been improving by a similar trend but with a much steeper slope. He also compares actual fusion funding to promised fusion funding, and shows that it is seriously underfunded. I made that observation in my last Analog article ... the funding levels for even the big programs are woefully inadequate considering the potential benefits. The result is programs so defensive of their slice of the budget that the little efforts are left scraping for crumbs.

My points above are covered in the first six slides, although slide 15 has some data on supercomputer speeds over time. Venture into the rest only if you have a deep interest in fusion modeling and algorithms.

http://w3.pppl.gov/~hammett/refs/2012/N ... tional.pdf

Re: Fusion progress versus computer progress ...

Posted: Wed Jan 28, 2015 3:53 am
by hanelyp
One difference, silicon chips have long since passed the point of being a profitable technology.

Re: Fusion progress versus computer progress ...

Posted: Wed Jan 28, 2015 4:23 am
by Tom Ligon
You telling me I should not have bought Micron?

What else could I do? EMC2 has not gone public.