Author: Dieter Buerssner
Date: 13:38:35 09/26/03
Go up one level in this thread
On September 26, 2003 at 16:22:27, Peter Fendrich wrote:
>On September 26, 2003 at 14:38:46, Dieter Buerssner wrote:
>> 201785128 9:44.5 Mat34 22t 1.Be8 a3 2.Bxg6 hxg6 3.h7 a2 4.h8=B Ka4 5.Kc2
>> Ka3 6.Ba1 Ka4 7.Kb2 b3H 8.Kc3H Ka3H 9.Bb2+H
>> Ka4H 10.Kc4H Ka5H 11.Kxb3H Kb5H 12.Kxa2H {EGTB}
>> 12...Kb4 13.Kb1! Kb5 14.Kc1 Kc5 15.Kd2! Kc6
>> 16.Kd3 Kd5! 17.Bc3 Kc5 18.Ke4! Kd6! 19.Bb4+
>> Ke6! 20.Bc5 Kd7 21.Ke5! Kd8 22.Kf6! Kd7! 23.Bg1
>> Kd6! 24.Kxg6! Ke6 25.Kh6 Ke7 26.g6! Ke8 27.g7
>> Kf7! 28.Kh7! Ke7 29.g8=Q! Kd7 30.Qd5+ Kc7!
>> 31.Bf2 Kc8! 32.Qc6+ Kb8! 33.Qd7! Ka8 34.Qa7#
>> {1351}
>>
>Wow, this is extrememly fast!
>Singular Extensions?
No. If you refer to the time, where the mate was found: Well, here the TBs help.
Above was with a default setup. When I use more aggressive probing (tb_u 4 for
Winboard/console users, for UCI it is in the engine parameters), the nodes/s go
down significantly, but a mate is found already at depth 18:
14669796 1:37.4 8.70 18++ 1.Be8 a3 2.Bxg6 hxg6 3.h7 Kb2 4.h8=Q+ Kb1
14968579 1:40.2 9.70 18++ 1.Be8 a3 2.Bxg6 hxg6 3.h7 Kb2 4.h8=Q+ Ka2
20105705 3:07.4 14.70 18++ 1.Be8 a3 2.Bxg6 a2 3.Bf7+ Ka3 4.Bxa2 Kb2 5.g6
Ka3 6.gxh7 b3 7.h8=Q Ka4
36183300 6:54.9 Mat39 18t 1.Be8 a3 2.Bxg6 hxg6 3.h7 a2 4.h8=B Ka3 5.Kc2
Ka4 6.Bb2 Kb5H 7.Kc1H b3H 8.Bc3H a1=Q+H 9.Bxa1H
Ka4H 10.Bf6H Ka3H 11.Bc3H Ka2H 12.Bg7H Ka3H
13.Kb1H Kb4H {270}
36978014 7:03.9 Mat39 18. 1.Be8 a3 2.Bxg6 hxg6 3.h7 a2 4.h8=B Ka3 5.Kc2
Ka4 6.Bb2 Kb5 7.Kc1 b3 8.Bc3 a1=Q+ 9.Bxa1 Ka4
10.Bf6 Ka3 11.Bc3 Ka2 12.Bg7 Ka3 13.Kb1 Kb4
{270}
38319901 7:12.4 Mat34 19t 1.Be8 a3 2.Bxg6 hxg6 3.h7 a2 4.h8=B Ka3 5.Kc2
Ka4 6.Bb2 Kb5 7.Kb3 Ka5 8.Kxa2 Ka4 9.Bg7H b3+H
10.Kb2H Kb4H 11.Bh8H Kc4H 12.Bc3H Kd5H 13.Kxb3H
{EGTB} 13...Kc5! 14.Be5 Kd5 15.Bf6 Kd6 16.Kc4!
Kc6! 17.Be5! Kd7! 18.Kc5 Ke7 19.Kd5! Kd7!
20.Bd6 Kd8 21.Ke6! Ke8 22.Bc7 Kf8 23.Kd7! Kf7
24.Bd6! Kg8! 25.Ke7 Kg7! 26.Bc5 Kg8 27.Kf6!
Kh7! 28.Bd6 Kh8! 29.Kxg6! Kg8 30.Be7 Kh8
31.Kf7! Kh7 32.Bf8! Kh8 33.Bg7+! Kh7 34.g6#!
{430}
I guess, other engines will be faster. All was on P4, 2.53 GHz, although running
in the background with many other applications open. This might reduce the speed
significantly, because there is less space for the HD-cache, which could help to
speed up the TB-accesses.
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.