jms@antares.UUCP (joe smith) (08/02/88)
[This is a repost - it didn't get out on 19-Jul] In the August 1988 issue of Scientific American, A. K. Dewdney describes a cellular automaton program in the COMPUTER RECREATIONS section. The program is called hodgepodge by its creators, Martin Gerhardt and Heike Schuster of the University of Bielefeld in West Germany. The description is incomplete in the article (due to space restrictions). Dewdney asks that readers who would like a more complete algorithmic description of the hodgepodge machine should write to him in care of Scientific American, and include check or money order for $2 to cover postage (worldwide), copying, and other costs. The details sound interesting; could someone post them to the net? I wrote a little program in AmigaBASIC just to get a feel of its operation. In doing so, I found some ambiguities due to the abbreviated description. The definition of S includes self, but does the count A include self? If not, divide by zero occurs with a solitary infected cell. The definition of S appears to include ill cells as well as infected cells, but it is divided by A instead of (A+B). One comment about von Newmann neighborhoods: if all cells outside a given rectangle are healthy, then the infection will never spread outside that rectangle as long as K1 and K2 are both greater than 1. This caused all my simulations to shrink until all cells were healthy. I must say, it is fascinating to watch. | TYMNET: JMS@F29 UUCP: {ames|pyramid}oliveb!tymix!antares!jms | | INTERNET: JMS%F29.Tymnet@Office-1.ARPA PHONE: Joe Smith @ (408)922-6220 | -- +-----------------------------------------------------------------------------+ | TYMNET: JMS@F29 UUCP: {ames|pyramid}oliveb!tymix!antares!jms | | INTERNET: JMS%F29.Tymnet@Office-1.ARPA PHONE: Joe Smith @ (408)922-6220 | +-----------------------------------------------------------------------------+