Author: Steffen Jakob
Date: 23:22:22 03/04/99
Go up one level in this thread
On March 05, 1999 at 01:29:17, Jouni Uski wrote: Hi Jouni! >>>I added the "bm" attributes to the endgame test collection which Jouni posted. >> >>Jouni, position #1 is a draw. Kc2 doesn´t win. I also noticed that other >>positions have more than one solution. >I have not claimed, that positions are win - indicated move should be the best. >Also 34 Qe7 is draw. Oh, then it was a misunderstanding. Then I will remove #1 for testing with Hossa because I don't see why Kc2 should be better than any other move. Well, ok, against humans it might be that you get more chances... Aren't there EPD tags which tell that the position is a draw? I will check this. >In position 5 f7 is the best move leading to mate in 22 even if other moves >also win. Which positions have multiple solutions? Hmmm... when I ran #14 at home on my slower machine with less RAM Hossa claimed that Kb1 is winning, but here on my faster machine with more hash tables it would play the suggested Na4. When I force Kb1 Hossa evaluates a draw... Maybe there is something wrong with my hashtables (I always thought this). It seems as if I (well actually Hossa) was wrong here. Hossa found 29 of 47 solutions. You can see the result at http://www.jakob.at/steffen/chess/tests/endgame.html Greetings and thanx again for conrtibuting these positions, Steffen.
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.