Author: Stuart Cracraft
Date: 14:00:33 08/26/04
Go up one level in this thread
On August 26, 2004 at 08:07:36, Brian Richardson wrote: >On August 25, 2004 at 23:51:50, Stuart Cracraft wrote: > >>On August 25, 2004 at 12:58:23, Brian Richardson wrote: >> >>>On August 25, 2004 at 12:42:34, Volker Böhm wrote: >>> >>>> >>>>Hi, >>>> >>>>the following position of the iq - test (position 166) cannot be solved from >>>>spike in reasonalbe time (> 2h): >>>> >>>>r4r1k/6pp/3p3b/1p1Npb2/3nB2q/2N3P1/PP3P1P/R2Q1RK1 b - - 0 1 >>>> >>>>(The Move is Qxe4) >>>> >>>>The reason is a nullmove-problem. After the queen-sac there are many moves that >>>>leads to a mate-thread. But one queen back spike tends to prune because of >>>>nullmove result (one move is not enough to show the mate thread thus going back >>>>to window). >>>> >>>>I tested the position with fritz 8, he solves it in 8 ply in nearly no time. >>>>Are there non commercial engines that solves this position too in a short amount >>>>of time? >>>>Maybe the author of this engine would be so kind to explain the rule that solves >>>>the nullmove-problem. >>>> >>>>Thanks Volker >>> >>>I have two versions of Tinker that I have been experimenting with for some time. >>> One is "regular" Tinker with full evaluation, and the other uses only pawn >>>evaluation and some simple PCSQ terms (e.g., no king safety, etc). The search >>>is the same in both versions. Regular Tinker solves this in a few seconds at 8 >>>ply, and the "simple" Tinker does not after a long time. I suspect the >>>difference is the king safety and not so much null move or pruning, at least for >>>Tinker (which would normally lose to Spike anyway). >>>Brian >> >>Can you be more specific about regular Tinker's king safety? What specifically >>does it have? >> >>Stuart > >Like most programs, Tinker has quite a few king related evaluation terms, some >of which depend on the game phase and where the king is, and are scaled by >opponent material, and some are asymmetric. In addition to piece square tables >there are penalties for controlling squares around the king. This is one area >that I am constantly tinkering with. The following should give you an idea: > >#define KING_EARLY_PIECES >#define KING_MOVED_EARLY_PENALTY >#define KING_MOVED_PENALTY >#define KING_FILE_OPEN_PENALTY >#define KING_XRAY_BONUS >#define KING_PIN_BONUS >#define KING_PAWN_MOVED_PENALTY >#define KING_AH_NOPAWN_PENALTY >#define KING_AH_ATTACK_PENALTY >#define KING_AH_OPEN_PENALTY >#define KING_PAWN_COVER_BONUS >#define KING_QUAD_PENALTY >#define KING_ZONE_PENALTY >#define KING_ATTACK_PENALTY >#define KING_FIANCHETTO_BONUS >#define KING_KINGSIDE_PAWN_BONUS >#define KING_QUEENSIDE_PAWN_BONUS >#define KING_CASTLED_PAWN_BONUS >#define KING_QUEEN_FLANK_PENALTY >#define KING_CENTER_PENALTY >#define CASTLE_SCORE > >Brian Interesting -- thanks.
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.