From: Peter C. McCluskey (pcm@rahul.net)
Date: Wed Dec 04 2002 - 10:02:37 MST
bradbury@aeiveos.com (Robert J. Bradbury) writes:
>I think what many fail to consider is that as the size scale decreases
>the costs fall, the parallelism rises and the experimental approaches
>that can be used become much different (and more productive).
>
>Now perhaps simulations may be the way to go -- but you are assuming
>that the costs of developing real nanotech parts and running experiments
>on them does not decrease below the cost of the simulations (which I
>presume are running on non-nano electricity). It isn't clear that is
>how it may play out.
I assume that there are wide variations from problem to problem in the
costs of experiments and in the value of simulation, so that there won't
be any point where "the" cost of experiments drops below "the" cost of
simulation.
You seem to be extrapolating from drug development. There may be some
steps in the development of an assembler for which that is a good analogy,
but I see two reasons for suspecting it underestimates the value of
simulation:
Drug developers have already tested most well-known molecules (i.e. picked
the low-lying fruit). Assembler designers won't have a similar reason for
avoiding well-known molecules for a while.
Drug developers do their hardest experiments on clinical trials, because
they need to deal with a system (the human body) that is too complex to
simulate well. Assembler designers have some ability to choose their
system so that it can be mostly analyzed by simulating subsystems.
-- ------------------------------------------------------------------------------ Peter McCluskey | Free Jon Johansen! http://www.rahul.net/pcm |
This archive was generated by hypermail 2.1.5 : Wed Jan 15 2003 - 17:58:34 MST