Re: Manhattan, Apollo and AI to the Singularity

From: Keith Henson (hkhenson@rogers.com)
Date: Fri Aug 25 2006 - 04:52:48 MDT


At 08:37 PM 8/24/2006 -0700, you wrote:
>On 8/24/06, Keith Henson <hkhenson@rogers.com> wrote:
>
>>I have scoped this out from time to time. The last time I did it I think
>>it took 150 meters of silicon on a side. It is going to take considerable
>>time before the technology is up to the task.
>
>The Japanese just built a supercomputer that runs at 10^15 ops/sec.
>The usual estimates for human-level computing power are around 10^17.
>So, two orders of magnitude or so, which works out to about 10 years
>or so in Moore's law terms, or molecular computing, whichever comes
>first. Brian Wang has also been following the quantum computing scene
>on his advanced nanotechnology blog, very interesting, the progress in
>that area - and scary.

I make the case that raw ops won't do it--which is why the monster spread
of silicon real estate to wire in the connectivity.

This says nothing about an intelligence organized on principles completely
different from natural brains. For all I know, such an AI might be able to
run on a 386 or even the chip in an Apple II.

Keith Henson



This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:57 MDT