I have been dealing with a fair amount of stress in my life, and have
much more important things to do so therefore my brain suddenly came up
with a very plausible method of constructing an entry-level nanocomputer
with mostly available materials.

This posting is intended as food for thought because I am a hundred
miles from actually being competent in any of the fields I'm going to be
touching on. =P

Molecular Manufacturing is either still a decade off or a closely
guarded military secret. Who knows which... However, a computational
architecture does seem feasible using existing techniques.

A computer is a machine defined not by it's structure, but by the
information it contains. Therefore the physical structure of a computer
can be extremely simple.

http://www.flickriver.com/photos/alteracorp/3637042466/

Basically, a simple repeating pattern of nodes should be completely
sufficient. A functional machine would effectively operate on FPGA or
CAM principles.

http://en.wikipedia.org/wiki/Protein_crystallization

So basically what you need is an amino acid sequence, or a sequence of
any type of polymer actually, that will

--> fold into a cubic or nearly cubic shape.
--> Exhibit exploitable, programmable computational properties at some
reasonable temperature.

The hypothesis is that such a chain of molecules exists.

There are several mechanical basises which it might operate, in
approximate order of increasing desirability:

--> rod/gear logic.
--> electronics
--> "spintronics"
--> photonics.
--> quantum entanglement.

To start with, you would only want maybe ten microns cubed to prove your
concept so bulk fabrication shouldn't be a prohibitive roadblock, at
first at least.

http://en.wikipedia.org/wiki/Gene_synthesis

Eventually you'd want to scale up to a brick the size of a good old
full-height drive bay. (a typical desktop CD-rom drive is a half-height
device).

Obviously, nailing down a computationally useful crystal shape is going
to be challenging.

To use it, you're going to need to power it, communicate with it, and
cool it. For a micro-volume, the latter probably won't come up but at
brick size, hell yeah, it'll matter.

The problem with communicating with the thing is that it will be
extremely tiny, you'd have to have exceptionally tiny wires. Also, the
crystal is likely to be extremely fragile to both thermal and
electrical, though probably not mechanical, stresses. So yeah, even if
you had a nanocomputer crystal, actually accessing it would be a
challenge in its own right... =\

Nanotubes might work. Extreme UV lasers might be an option... You'd need
a STM

http://en.wikipedia.org/wiki/Scanning_tunneling_microscope

To even see what you're doing when you're wiring the thing up...

Still, the approach seems plausible... If it were to work, It'd reduce
the singularity to a software problem... =P

-- 
E T F
N H E
D E D

Powers are not rights.



-------------------------------------------
AGI
Archives: https://www.listbox.com/member/archive/303/=now
RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-c97d2393
Modify Your Subscription: 
https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-2484a968
Powered by Listbox: http://www.listbox.com

Reply via email to