[ExI] Slow thinking

Keith Henson hkeithhenson at gmail.com
Sat Oct 1 16:50:10 UTC 2011


On Sat, Oct 1, 2011 at 5:00 AM,  "spike" <spike66 at att.net> wrote:

> On Behalf Of john clark

>>.On Fri, 9/30/11, Keith Henson <hkeithhenson at gmail.com> wrote:
>
> "And the point of this 10^15 larger brain that thinks no faster than a human
> would be?"  Keith

>>.It would have 10^15 times more memory than a human, and be able to think
> about 10^15 times more things at the same time than a human. John K Clark

Not impressed.  You are talking about something that isn't that far
off of the present where we have ~10^10 people and with access to
information on that kind of multiple through the internet.
>
> I have stayed out of this discussion of latency in an MBrain, but do keep in
> mind there are a bunch of variations on the MBrain theme which complicate
> the question of latency vs node size.  For instance, we can have an MBrain
> with SBrains orbiting within: an SBrain is a cluster of a few thousand, a
> few million or a few billion nodes which co-orbit a common center of mass,
> while that common center of mass orbits the star.  This creates something
> analogous to a human city or population center, where local latency is very
> low.
>
Human clock rate is perhaps 200 Hz.  I don't know where computers will
limit, but we are already GHz rates, if we get up to 200 GHz, latency
becomes a big problem with maximum dimensions for a brain being in the
mm range and the subjective time to get a message back from the far
side of an MBrain is 40,000 years.  All of a sudden, galactic
communication problems have come to the solar system.

Power, cooling and speed of light delays all conspire against brains
or even civilizations being physically large if they are running at
high clock rates.  Round trip to the moon is about 2.5 seconds.  At an
easy million to one subjective speed up, it would take about 29 days
to get a message back.  Even the 60 ms low latency optical fibers from
New York to London would seem like 16 hours.

It occurred to me just now that this is an alternative answer to the
Fermi question.  Once intelligent life forms start moving into faster
hardware, it may be that they can't leave their home planet without
getting hopelessly out of touch with whatever is going on.

Keith




More information about the extropy-chat mailing list