Author: KarinsDad
Date: 16:32:03 06/15/99
Go up one level in this thread
On June 15, 1999 at 18:51:21, Pete Galati wrote: >On June 15, 1999 at 18:29:26, m.d.hurd wrote: > >>On June 15, 1999 at 18:20:24, Tom King wrote: >> >>>On June 15, 1999 at 18:17:19, Dann Corbit wrote: >>> >>>>On June 15, 1999 at 18:14:21, m.d.hurd wrote: >>>>[snip] >>>>>Why do you assume junior will beat hiarcs ? >>>>I see it as a toss up, but Junior is the #1 ranked program currently. >>>>The thing that amazes me is that P.ConNerS and Zugzwang are dead in the middle >>>>of the pack with a hundred times the horsepower of the other entries. >>>>!! >>> >>>but anything can happen in three rounds..let's see who is near the top after 7 >>>rounds. Shame it can't be like the WMCCC, with 11 rounds. >>> >>>Rgds, >>>Tom >> >>I agree, the more rounds the more likely the best program will triamph. As a >>side issue does any one know which opening books the programs are using, >>commercial or special ones to prevent the oponents from prepairing against them >>? > >Are they allowed to change their opening book from round to round? Posibly >having one more or less taylored for the playing style of each opponent? > >Pete Galati The rules do not explicitly prohibit changing opening books and I know that people make code changes between rounds. So, it would appear to be legal. However, if ICCA ever decided to make it illegal, you could get around it by having a superset of openings with subsets inside of it for each opponent. When you enter the name of your opponent, the program could pick the appropriate opening "book". So, nobody would know the difference since it would be part of the program. I am surprised that the ICCA does not insist that the program (parameters, computing system, opening book, etc.) remain the same throughout the tournament. It is a strange thing that bug fixes and other code changes can be made in mid-tournament. The reason I think that this is strange is that one of the tie breaks is based off which programs were played. Since the games are played at different times in the tournament (with potentially weaker or stronger programs of the same name), a tie break could be won when program A beat buggy program C early in the rounds whereas program B drew to not so buggy program C later in the rounds, but program C counted the same towards both programs A and B (and B lost tie break points over it). KarinsDad :)
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.