Xref: utzoo comp.theory.cell-automata:238 sci.electronics:15335
Path: utzoo!attcan!uunet!ogicse!dali.cs.montana.edu!rpi!rpi.edu!hiebeler
From: hiebeler@turing.cs.rpi.edu (Dave Hiebeler)
Newsgroups: comp.theory.cell-automata,sci.electronics
Subject: Re: John Conway's "Life"/Pseudo-Neural hardware implementation
Message-ID: <~59%%F^@rpi.edu>
Date: 2 Nov 90 00:25:23 GMT
References: <1990Nov1.185409.25802@bradley2.bradley.edu>
Organization: Rensselaer Polytechnic Institute Computer Science, Troy NY
Lines: 27
In-Reply-To: pwh@bradley2.bradley.edu's message of Thu, 1 Nov 90 18:54:09 GMT


  The people in the Information Mechanics Group at MIT designed and
built a series of cellular automata machines (CAMs).  

  I have one in my PC right now; it's a plug-in board that runs
cellular automata at about the speed of the earlier cray's: 256x256
array, updated 60 times/second.  Much cheaper than a Cray, too. :-)
It's not a truly parallel machine; it's basically a tight pipeline
that loops over the array to update the cells.

  If anyone's interested, I can send more information.  I don't want
to sound too much like an advertisement.

  There have been other CA hardware implementations, such as LGM-1,
RAP2, but I am not as familiar with them.  (Although I can supply a
couple of references if there's interest).

  There is also, of course, the Connection Machine, as someone
mentioned.  It can run CA pretty darn quickly, although it's not a
dedicated CA machine -- people have written all sorts of applications
on the CM.

--
Dave Hiebeler                           | Internet: hiebeler@turing.cs.rpi.edu
Computer Science Dept., Amos Eaton Bldg.|           hiebeler@heretic.lanl.gov
RPI                                     | Bitnet: userF3JL@rpitsmts
Troy, NY 12180-3590                     |