Differences between revisions 2 and 4 (spanning 2 versions)
Revision 2 as of 2010-07-17 06:36:00
Size: 1357
Revision 4 as of 2010-07-17 06:39:07
Size: 1469
Deletions are marked like this. Additions are marked like this.
Line 7: Line 7:
{{attachment:mooretran.png}} {{attachment:mooretran.png|alt Moore's Law graph for transistors|width=512}}
Line 13: Line 13:
{{attachment:mooresenergy.png}} {{attachment:mooresenergy.png|alt Hypothetical Moore's law graph for electric power|width=512}}

Moore's Law and Energy

Moore's Law describes the incredible price drop of transistors on integrated circuits over the last 4 decades. Between 1970 and 2000, transistor prices dropped 6 orders of magnitude. This is made possible by photolithographic printing, automated processing, and an incredible amount of hard work by hundreds of thousands of scientists and engineers.

Here is the price of transistors on integrated circuits, following Moore's Law, from Hans Moravec's Singularity Institute:

alt Moore's Law graph for transistors

This is the price of a transistor, on an integrated circuit die, in a package. Packaged integrated circuits cost about what they did decades ago; they just contain a million times more transistors at the same price. Packaging has not gotten a whole lot cheaper.

Meanwhile, over the last few decades, the price of generating a watt of electricity has risen. What if, starting this year, the price dropped on the same Moore's Law trajectory?

alt Hypothetical Moore's law graph for electric power

In twenty years, we could replace our entire energy infrastructure for the cost of one large power plant. Of course, a power plant are one large (and increasingly complex) package, built by hand and not printed optically. So power plant prices don't go down, but trend upwards with complexity instead.


MooresEnergy (last edited 2014-09-19 23:31:55 by KeithLofstrom)