Newsgroups: comp.robotics
Path: cantaloupe.srv.cs.cmu.edu!europa.chnt.gtegsc.com!news.mathworks.com!newsfeed.internetmci.com!news.sprintlink.net!howland.reston.ans.net!torn!watserv2.uwaterloo.ca!undergrad.math.uwaterloo.ca!mwtilden
From: mwtilden@math.uwaterloo.ca (Mark W. Tilden)
Subject: Re: Need more info on Tilden's Nervous Architecture
Message-ID: <DCr5Gw.1B1@undergrad.math.uwaterloo.ca>
Sender: news@undergrad.math.uwaterloo.ca (news spool owner)
Nntp-Posting-Host: math.uwaterloo.ca
Organization: University of Waterloo
References: <asnyder.807290763@grover> <3vlnq3$8ji@gap.cco.caltech.edu>
Date: Thu, 3 Aug 1995 20:41:20 GMT
Lines: 21

In article <3vlnq3$8ji@gap.cco.caltech.edu>,
Bradley Minch  <bminch@pcmp.caltech.edu> wrote:
>...It appears that he uses a
>Schmidt Trigger inverter for each "element" in the "preferred embodiment"--
>these typically have six MOS transistors each, not two...  But who's
>counting?  Hope this helps...

Well, it turns out that as long as you have a gain of six with a
hysterisis feedback, any old amplifier will do the trick.  Schmidt
triggers are just convenient because they come 6 to a pack and have
their hysterisis built in.

But as many will attest, a two transistor Nv neuron works just fine.

Is all.

-- 
Mark W. Tilden.  "Gomi no Sensei des"   _    _    ________________________
P3, LANL, Los Alamos, NM 87545, USA.   / \  / \  /________________________)
505/667-2902 <mwtilden@lanl.gov>      //\ \//\ \// ___o___________________
#include (standard.disclaimer);      //  \_/  \_/ (_______________________)
