Author: Dieter Buerssner
Date: 06:23:58 01/11/02
Go up one level in this thread
On January 10, 2002 at 16:23:59, Uri Blass wrote: >On January 10, 2002 at 15:30:20, Dieter Buerssner wrote: > >>On January 10, 2002 at 10:17:25, Robert Hyatt wrote: >> >>>[...]Of course, there _will_ be those >>>positions that it thinks it is winning, but it will end up drawing due to the >>>50 move rule. But so far they have been _very_ rare. Against humans, and that >>>is the primary kind of chess I am interested in, this is not nearly such an >>>issue. Against another computer with tables, both will be wrong as one will >>>think it is winning, the other will think it is losing, when in fact it is a >>>draw... >> >>And the other computer, equipped with the same TBs and the same misfeature, will >>possibly have resigned long before it is seen, that the mate score is wrong with >>current chess rules ... >This problem is easy to fix. Yes. >tell it never to resign in comp-comp games. >It is good not only because of this reason. > >The other program may have a bug and miss mate in 1. But it also has negative consequences. In the same time, I get less test games (say for testing on ICS or in test matches). Also, when the opponent misses a mate in 1, it does not help me to judge possible engine improvements (or the contrary). It will just (in the unlikely case of bugs of the opponent) improve the rating. Also in the case (you mentioned in the other post), where the opponent does not have the right TBs, for judging my engine strength, seeing the result as a loss, when I see mating score against the engine is ok. In a tournament like CCT4, these arguments are of course void and Yace will not resign by itself. Nevertheless, many engine users like to do there private tournament, and may be happy to see, that the engine usually resigns in hopeless situations by default to save time. Even in tournaments played with the operators at the board, it seems common to resign, instead of letting the opponent show, that it really can mate. Perhaps this shows some sort of respect of the opponent programmer, like in human games. I personally would have no problem to play out a win until the last move, but different programmers may see this differently. And - how would the observers judge a draw, that "really" was a loss? Regards, Dieter
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.