On 3/13/2012 6:48 AM, Bruno Marchal wrote:

On 13 Mar 2012, at 01:43, Russell Standish wrote:

http://www.nature.com/news/the-unavoidable-cost-of-computation-revealed-1.10186

This about experimentally testing Landauer's principle that
computation has thermodynamic constraints.


I was worrying a bit with that title, thinking Landauer's principle was refuted, but on the contrary, it is confirmed.

The title is misleading, because it looks like computation would need energy, but the energy is needed only for erasing information, and since a paper by Hao Wang (universal, and thus all) computations can be done without ever erasing information.

This confirms also that we can transform information into energy,

That's not quite right. We can use information to make energy available for work, i.e. reduce entropy. The energy is already there, it's just not accessible for useful work.

Brent

and we can transform energy into information, and perhaps a quantum nuclear computing machine can appear on the horizon.

Note that the self-duplication, à-la W/M, produces one first person bit of information ("W or M"), which suggests that consciousness might be always associated with an energy => information flow. Energy, mass and information would be, and probably are, different aspect of the self-multiplication in arithmetic, as it should be by the UD A.

The problem is that such a matter, even a vacuum, should contain an infinite amount of energy, but similar problems occur also with quantum field theory, and thus led to renormalization theories, and we can expect important renormalization in the comp physics too.
Last, but not least, this confirms also Bennett analysis of the Maxwell Daemon. 
I think.

We are not just machine, we are steam engine.
All we need is a bit of heat, to *forget* taxes and death ...

Thanks for the link, Russell.

Bruno


http://iridia.ulb.ac.be/~marchal/




--
You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To post to this group, send email to everything-list@googlegroups.com.
To unsubscribe from this group, send email to 
everything-list+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/everything-list?hl=en.

Reply via email to