Path: utzoo!censor!geac!torsqnt!news-server.csri.toronto.edu!cs.utexas.edu!ut-emx!dking From: dking@ut-emx.uucp (David L. King) Newsgroups: comp.theory Subject: Re: Non recursive "Towers of Hanoi" Summary: P.s. to tower of hanoi Keywords: Towers of Hanoi, Recursivity, Storage Message-ID: <40607@ut-emx.uucp> Date: 3 Dec 90 20:33:41 GMT References: <1197@disuns2.epfl.ch> <1990Nov27.112713.26729@actrix.gen.nz> Organization: The University of Texas at Austin; Austin, Texas Lines: 17 Postscript to the last post. I take it back: finding the fastest solution is not as easy as I said, in general cases. The algorithm IS fastest where all are initially on one post. But positions can arise in which it would not be the smallest one's 'turn', yet only the smallest one can move (all have been stacked on one post). In this case move the smallest again (in the same old indicated direction) and you will be on the quickest road from where you are currently. I.e, the algorithm gets you there in any case. However, the fact that you moved one disk twice to where it could be moved in one move shows that it is not the fastest solution in these cases. I'll let others use that archetypal case to arrive at the ultimate solution. It's still not very hard, but there's a little fun left.... David L. King University of Texas System Center for High Performance Computing Brought to you by Super Global Mega Corp .com