Author: Dieter Buerssner
Date: 18:27:18 04/18/01
Go up one level in this thread
On April 18, 2001 at 19:05:18, Landon Rabern wrote:
>Would someone solve these and post the results here? It is hard to test without
>knowing what the results should be.
Yace, all 4-men and very few 5-men TBs
>K7/8/8/7p/N2k4/8/8/8 w - -
nodes time score depth
196 3.141 0.00 62. 1.Nb2 {EGTB} 1...h4 2.Nd1 h3 3.Nf2 h2 4.Ka7
h1=Q 5.Nxh1! {No mating material} {340}
>8/8/8/2k5/p1b5/8/1P1K4/8 w - -
I don't know the solution. Here Yace shows very strange jumping of scores. I
just give you the PV I get after some time:
43441723 6:41.2 -5.47 22t 1.Kc2 Kd4 2.Kb1 Bb3 3.Ka1 Ke3 4.Kb1 Bd5 5.Kc2
Kd4 6.Kb1 Bc4 7.Ka1H Kc5H 8.Kb1H Kb4H 9.Ka1H
Kb3H 10.Kb1H Be2H 11.Ka1H Bf3H 12.Kb1H Bc6H
13.Ka1H Kc2H 14.Ka2H Bd5+H 15.Ka1H Kd2H 16.Kb1H
Bc4H 17.Ka1H Bd5H 18.Kb1H Bc4H 19.Ka1H Bd5HR
{-350}
The "H" at the end of a move means, that this move was picked from the hash
tables at display time (and not at search time). So, these moves may not be
reliable. The "HR" means, that this is a repetition of position in the hash
tables. So, this will not fit the score at all. Others, with better programs,
mor patience or more TBs will be able to give a better result.
>6k1/3p4/8/8/8/B7/P7/7K w - -
33045691 4:06.4 17.59 20t 1.Bb4 Kf7 2.a4 Ke6 3.a5 Kd5 4.a6 Kc6 5.Ba5 d5
6.Kg2 d4 7.Kf3 d3 8.Ke3 Kb5 9.a7 Kc6 10.a8=Q+
Kd6 11.Qa6+ Kd5H {HT} {1271}
>8/8/2B5/7b/7k/2K5/1P6/8 w - -
45414533 7:12.8 Mat22 22t 1.Bb4 Kf7 2.a4 Ke6 3.a5 Kd5 4.a6 Kc6 5.Ba5 d6
6.Kg2 d5 7.Kf3 Kb5H 8.a7H Kc4H 9.a8=QH Kc5H
10.Qe8H Kd6H {HT} {1271}
>8/4K3/8/3k4/8/1N1B4/6p1/8 w - -
35125 1.768 0.00 8t 1.Bc4+ Kc6 2.Nd4+ Kc5 3.Nf3 Kxc4 {EGTB} 4.Nd2+
Kd4 5.Nf3+! Ke4 6.Ng1 Kf4 7.Ne2+ Kg4 8.Ng1 Kh4
9.Ne2 g1=Q 10.Nxg1! {No mating material} {340}
>7K/6p1/8/7p/8/8/R7/6k1 w - -
866 3.926 Mat29 62t 1.Kh7 {EGTB} 1...h4! 2.Kg6! h3! 3.Kg5 h2! 4.Kg4
{Not all positions found in TB} {340}
>8/R7/3P4/6r1/8/8/8/1k5K w - -
741 3.060 Mat33 62t 1.d7 {EGTB} 1...Rd5! {Not all positions found
in TB} {80}
The {Not all ...} means, that Yace is not able to display the whole mating line,
because some TBs are missing.
To your former question, about endgame test position. Have a look at
www.jakob.at. There you will find a very interesting endgame testsuite. Some of
the positions can be solved by TBs. In other positions, TBs will help.
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.