Author: Dieter Buerssner
Date: 13:50:04 08/11/03
Go up one level in this thread
On August 11, 2003 at 00:10:13, Mike Byrne wrote:
>new beta version of Crafty personality "Staunton"
>
>beautiful moves btw ....
>
>New game
>8/6Bp/6p1/2k1p3/4PPP1/1pb4P/8/2K5 b - - 0 1
It was interesting, to test this position. Manfred already showed some analysis
by Yace. I made some interesting observations - TBs and bitbases seem to help
here. I have one version, that is almost identical to the version Manfred used,
with the main difference, that kppkp TB is available from RAM (and will other
than TBs be probed until the leafs). This seemed to help pretty much to find a
good score much faster. Also, by default Yace uses rather conservative settings
for null move (R=2 and R=1 in later endgames with only one minor piece on each
side). When I change this to R=3/2 (Heinz adaptive nullmove) and R=2 in the
endgames, I get a pretty convincing score fast.
P4, 2.53 GHz, 300 Mb hash:
[ nodes time score depth]
16905465 17.630 -0.02 16. 1...Kd6 2.f5 Ke7 3.Bh6 Kf7 4.h4 Bd4 5.Bg5 gxf5
6.gxf5 h5 7.Kd2 Ke8 8.Kd3 b2 9.Kc2 Kd7 10.Kb1
Kd6
19980679 21.121 -0.04 17t 1...Kd6 2.f5 Ke7 3.Bh6 Kf7 4.h4 Bd4 5.h5 gxh5
6.gxh5 Kf6 7.Bd2 Ba1H 8.Be3H Kf7H {HT}
23762267 25.075 -0.03 17t+ 1...b2+ 2.Kb1 exf4 3.Bxc3 f3 4.Be1 Kd4 5.Kxb2
Kxe4 6.Kb3 Ke3 7.Bf2+ Kxf2 8.g5 h5 9.gxh6 Kg3
10.h7 f2 11.h8=R f1=Q 12.h4 {501}
26879476 28.387 0.37 17t 1...b2+ 2.Kc2 exf4 3.Bxc3 f3H 4.Be1H Kd4H 5.h4H
b1=Q+H 6.Kxb1H Kxe4H 7.g5H Ke3H 8.Bd2+H Kxd2H
9.Ka2H h5H 10.gxh6H f2H 11.h7H f1=QH 12.h5H {HT}
31712335 33.972 0.37 17. 1...b2+ 2.Kc2 exf4 3.Bxc3 f3 4.Be1 Kd4 5.h4
b1=Q+ 6.Kxb1 Kxe4 7.g5 Ke3 8.Bd2+ Kxd2 9.Ka2 h5
10.gxh6 f2 11.h7 f1=Q 12.h5
33062660 35.366 0.77 18++ 1...b2+ 2.Kc2 exf4 3.Bxc3 f3 4.Be1 Kd4 5.Kxb2
Kxe4 6.h4 Ke3 7.h5 f2 8.Bd2+ Kxd2 9.hxg6 f1=Q
10.g7 Qc4 11.g8=R
35047121 37.495 1.77 18++ 1...b2+ 2.Kc2 exf4 3.Bxc3 f3 4.Be1 Kd4 5.Kxb2
Kxe4 6.h4 Ke3 7.h5 f2 8.Bd2+ Kxd2 9.hxg6 f1=Q
10.g7 Qc4 11.g8=R
52653911 58.472 6.00 18t 1...b2+ 2.Kc2 exf4 3.Bxc3 f3 4.Be1 Kd4 5.h4
Kxe4 6.Bf2 b1=Q+ 7.Kxb1 Kd3H 8.Be1H Ke2H 9.Kc2H
Kxe1H 10.Kd3H f2H 11.h5H f1=Q+H 12.Kc3H {HT}
66852953 1:16.0 6.00 18. 1...b2+ 2.Kc2 exf4 3.Bxc3 f3 4.Be1 Kd4 5.h4
Kxe4 6.Bf2 b1=Q+ 7.Kxb1 Kd3 8.Be1 Ke2 9.Kc2
Kxe1 10.Kd3 f2 11.h5 f1=Q+ 12.Kc3
98864872 1:59.8 6.32 19t 1...b2+ 2.Kc2 exf4 3.Bxc3 f3 4.Be1 Kd4 5.e5
Kxe5 6.Bf2 Ke4 7.h4 b1=Q+ 8.Kxb1H h5H {HT}
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.