Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Shortest mate that programs cannot find or take long time

Author: Alessandro Scotti

Date: 15:49:25 12/17/05

Go up one level in this thread


On December 17, 2005 at 16:37:35, David Dahlem wrote:

>Here is a mate in 1 (probably not legal). ChestUCI finds the solution, but other
>engines can't seem to find it. :-)

This position causes an explotion of quiesce nodes, here's Kiwi with "unlimited"
qsearch (some lines omitted):

 3/ 3  +0.00  00:22       111  Kh7 Nxd5 dxe4
 3/ 3  +1.84  00:50      2303  Kh7 fxg5 dxe4
 3/ 3  +2.68  00:55      3790  Nxc6 fxg5 dxe4
 3/ 3  +2.92  01:10      4118  Nxc6 fxg5 dxe4
 3/ 3  +3.52  01:23      5861  Qd1 exd5 fxe4
 3/ 3  +4.60  01:54      7604  Qd1 exd5 fxe4
 3/ 3  +4.76  02:04      8300  Nxc4 cxd5 dxe4
 3/ 4  +5.84  02:14      9903  Qf1+ Kh2 fxe4 exd5 cxd4
 3/ 4  +9.20  03:04     15235  Qf1+ Kh2 Rxd4 fxg5 Qf2+
 3/ 5  +9.28  04:33     22833  dxe4 bxa5 Bxd4
 3/ 5 +199.99  04:40     23268  Qh1+
 4/ 0 +199.99  04:40     23313  Qh1+
 5/ 0 +199.99  04:41     30181  Qh1+

So almost 5 minutes to see the mate! But with limited (3*nominal search depth)
qsearch:

 3/ 3  +0.00  00:00        99  Kh7 Qxd5 Qf1+
 3/ 3  +3.80  00:01      5660  Nxc6 exd5 dxe4
 3/ 3  +4.52  00:01      7739  Qd1 exd5 fxe4
 3/ 3 +199.99  00:02      9659  Qh1+
 4/ 0 +199.99  00:02     10684  Qh1+
 5/ 0 +199.99  00:02     15983  Qh1+
 6/ 0 +199.99  00:03     68534  Qh1+



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.