Author: blass uri
Date: 08:29:30 01/22/99
Go up one level in this thread
On January 22, 1999 at 09:36:37, Enrique Irazoqui wrote: >On January 22, 1999 at 08:37:43, blass uri wrote: > >>I am not sure if the results in games >>10,11,15,16,17,19,32,43,44,45,46,54,64,85,138,140,143,144,146 is the right >>result. (games were posted in 31.12) >> >>These games were adjudicated. >>I believe the results are the right result in more than 80% of them but I think >>they were adjudicated too early. >> >>These are not the only games(I believe adjudicated too early) but only games >>from games 1-150 >> >>Uri > >10 >11 No progress There is a chance that after many moves the programs will start to do some progress. I can imagine in number 10 black sacrificing a knight for a pawn, after it go with the king forward and force white to sacrifice a bishop for a pawn and win or lose the pawn ending. In 11 white can try Kg2 and h4. Junior5 may try to do some progress after many moves if it does not want a draw by the 50 move rule. >15 Evaluation 0.00, dead draw I agree the position is drawn but I am not sure if the computers will do a draw I can imagine that black is going to put the king far from the pawn and go to a position that it suddenly discover that it cannot defend the pawn. I saw cases when programs managed to lose a dead draw position. For example Junior5 lost a dead draw position with no progress for many moves against crafty in game 169. >16 No progress, different color bishops, dead draw I am not sure if it is a dead draw between programs. Program can trade pawns and get passed pawn for both sides. >17 >19 No progress, different color bishops, dead draw I am not sure if 19 is a dead draw. Maybe white can sacrifice something after 50 boring moves. White has practical chances to win by sacrificing f2 and going with the king to support the passed pawns. Kf1,Ke1,Qd2 is a practical plan against a computer because white has good chance that black will play Qxd2+ Kxd2 Bxf2. white may try it after many boring moves to avoid the 50 move rule. >32 KQ vs KR, queen wins The question is if Hiarcs7 knows how to win it. The evaluation of Hiarc did not improve. 432 at move 63, 391 at move 65 >43 No progress for over 30 moves, dead draw I am not sure if it is a dead draw. I do not see a way to win but I am not close to be a grandmaster. If I were white then I would not agree to a draw and try to do something(for example to push a5) >44 Eval 0.01, dead draw I am not sure about the ability of programs in endgames. It is possible that the sides are going to trade knights and go to a lost endgame for one side. I believe there is a chance of 90% for a draw but I am not close to be sure about it. >45 Different color bishops, dead draw I am not sure about the result between computers. Both sides can do passed pawns. Maybe after it Junior5 is going to sacrifice the bishop for a pawn. KRB vs KRP is a dead draw but Junior5 lost it against crafty so I am not totally sure that it is a draw. >46 Perpetual check, draw I do not see the perpetual. The programs do not see it. >54 >64 No progress in 16 moves, dead draw You are probably right but I am not sure about it Hiarcs evaluate the position as 149 advantage for itself. It may try something dangerous and lose. >85 Even material, obe pawn left, dead draw I am not sure about programs ability to play endgames maybe white can put the king in a bad position and lose. >138 Eval 0.00, dead draw Junior5 wants to play Nb6 in the final position with evaluation of 0.06 advantage for itself at depthes 17,18 and 0.15 for itself at depth 19 with the line 45...Nb6 46.Ra6 Nc4 47.Be7 Rh2 48.Bb4 Rb2 49.Be1 Rb3 50.Kf2 Rxe3 51.Bb4 Rxe4 52.Be7 Rd4 53.Rxf6+ This is probably a draw but I am not sure that it is a draw because programs may trade pawns and have passed pawns. >140 >143 Perpetual check. draw I am not sure about it. I did not check it but even if a computer evaluates a position as a perpetual check it may be because of the fact that the best thing it can do is to repeat to a previous position of the game. It may repeat the previous position and find a win from it. I do not say that it is the case here but only that I am not sure. Uri
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.