X-Message-Number: 26119
Date: Sun, 1 May 2005 10:15:45 -0400
From: Thomas Donaldson <>
Subject: CryoNet #26112 - #26117

Hello Mr Bozzonetti again:

Your answer f ails once more. You forgot that the neuron is new and 
therefore may not be on any table. Moreover you once more want to put
some number of neurons on the same chip. Ridiculous and even worse,
if you were ever to actually make such a system, disappointing.

I should explain what I just said a little further. As a new neuron,
not only won't it be on any table, but it will need an address to
start with. Even if you're guaranteeing that your system won't
work very long by solving this problem entirely with software, you 
should be more explicit. What you want is a linked list which
can be traversed in both directions.

Fundamentally, though, the mistake is to use time-sharing rather than
giving one processor to each neuron. And as I said in my last message,
such a system would work far faster than if you used time-sharing,
and wouldn't get into the problems with timing that time-sharing
creates.

             Best wishes and long long life to all,

                   Thomas Donaldson

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=26119