[ExI] Fermi Paradox and Transcension

[spike]

-----Original Message-----
From: extropy-chat-bounces at lists.extropy.org
[mailto:extropy-chat-bounces at lists.extropy.org] On Behalf Of Keith Henson
Sent: Wednesday, September 12, 2012 2:03 PM
To: extropy-chat at lists.extropy.org
Subject: Re: [ExI] Fermi Paradox and Transcension

On Wed, Sep 12, 2012 at 4:56 AM,  "spike" <spike66 at att.net> wrote:

On Behalf Of Keith Henson

snip

>>...If you are trying to think fast, your physical layer needs to be as
> small as you can get it... Keith
>
>>... Ja, and its temperature goes up linearly as the inverse of the radius 
> of orbit, all else being equal.  spike

>..."Orbit" is where I get the disconnect.

>...Human brain is ~20 W.  Sped up a million fold, perhaps 20 kW...I don't
want to rain on anyone's parade, but this doesn't seem like this works very
well for fundamental physics reasons....I think I first mentioned these
concerns almost two decades ago.  Keith
_______________________________________________

Hmm, we may be looking at two different models of MBrain.

I imagine the nodes being about the disk about size of a DVD only much
thinner, running a processor similar to about what we had 30 yrs ago during
my misspent youth, a Z80-ish level processor with about a million
transistors, except with today's 20 nm features and today's power-conscious
CMOS components.  Each node would have something like a Bluetooth
transmitter/receiver, able to transmit about 12 meters or so which is about
as far as a Bluetooth connection normally goes on the power levels available
to an MBrain node the size of a DVD.  This allows each node to communicate
directly with about ~7000 other nodes.

The processor's job is to simulate a neuron and the related dendrites.  I
have imagined a roughly spherical grouping of about 100 billion nodes with
an average spacing of about a meter, so that each node would occupy about a
cubic meter and the roughly spherical cluster would have a radius of about 3
km, or 10 light milliseconds.

For what I have in mind, the processors need not be particularly fast or
sophisticated.

Keith are you and I thinking about approximately the same thing?

In this model each node watches it's inputs and if everything is just right,
fires it's own simulated dendrite output.

I have no way of knowing if such a thing would ever simulate intelligence,
but I do have a way of knowing the alternative: a dead rock does not
simulate anything.

Note this isn't Rober't vision of an MBrain.  I never could convince him
that we can't really capture all the star's energy, or even most of it.

spike