Author: Dave Gomboc
Date: 21:28:00 04/10/99
Go up one level in this thread
On April 10, 1999 at 13:57:15, David Eppstein wrote: >On April 10, 1999 at 00:17:44, Richard A. Fowell (fowell@netcom.com) wrote: > >>On April 08, 1999 at 19:10:45, Charles Milton Ling wrote: >> >>>I suppose I am getting truly off-topic here, but I wasn't the first to bring up >>>Bridge :-) Does anybody know of a freeware Bridge program available on the Net? >>> Strength is not of paramount importance: I would put my "Bridge Elo" at 1200. >>>Thanks! >>>Charley >> >>See end of this message - first, let me insert some "on topic" stuff: >> >>Partition search? >> >>The page: >>http://www.cirl.uoregon.edu/research/partitionSearch.html >>discusses "partition search", and gives a paper. >> >>While applied in bridge software, it seems like it might be applicable >>to chess, as might the techniques discussed on: >>http://www.cirl.uoregon.edu/research/sysMethods.html >> >>Has anyone used these in chess? > >I was under the impression that this was just a way of identifying certain >positions (bridge hands) as equivalent even though the actual cards are >different. E.g. if you started with the Ace and King of hearts, then positions >in which you've played the Ace and still hold the King are essentially the same >as positions in which you've played the King and still hold the Ace, since all >that matters is the ordering among the unplayed cards. Is there more to it than >that? This is true, but perhaps a better example is "Spades: Ace, Queen, four ruffs". You don't care which cards out of 2 through 9 that the four low spades are, you just care that you are 6-long in spades, AQ high. So, the state space becomes much smaller as a result. IMO the potential for savings is clearer with this example. >I'm not sure that anything like this really works in chess, although I guess >PARADISE did some sort of automatic determination of moves that wouldn't make >any difference to the attack it was planning. It doesn't seem terribly applicable to chess in the general case. Of course, specific positions could benefit from it, but the overhead of finding out which do and which don't would likely kill you. Dave Gomboc
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.