Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Revised ECM Test Suite (some additional analysis needed)

Author: Scott Gasch

Date: 11:30:46 07/19/04

Go up one level in this thread


I run ecm/20 as my main test suite (and am amazed at how well the pros do in 10
sec / move!).  My engine does get one of the below positions:

>r1b2r1k/ppppq1pp/2n1n3/6N1/2B2P2/4B3/PPP3PP/R2Q1RK1 w - - bm Nxh7; id "ECM.674";

 1.  +1.87    0.01     81               Qd3
 2.  +1.83    0.01     763              Qd3 g6
 3.  +2.04    0.03     2343             Qd3 g6 c3
 4.  +1.28    0.11     21370            Qd3 Nxg5 fxg5 d6
 5.  +1.64    0.16     35761            Qd3 Nxg5 fxg5 Ne5 Qe4
 6.  +1.04    0.66     187144           Qd3 Nxg5 fxg5 Ne5 Qe4 d6
 7.  +1.20    1.42     441866           Nxe6 dxe6 Qh5 Rf5 Qg4 Bd7 Rad1
 8.  +1.21    4.17     1349493          Bxe6 dxe6 Qd3 g6 Qc3+ Qf6 Qxf6+ Rxf6 Kh1
b6 Rad1
 9.  +1.57    9.57     3289301          Qd3 Nxg5 fxg5 Ne5 Qe4 Re8 Bd5 Ng6 Qxe7
Rxe7
10.  +1.74   19.48     7203198          Nxh7 Kxh7 Qh5+ Kg8 f5 Ne5 Bb3 g6 fxg6
Rxf1+ Rxf1 Qg7 Bxe6+ dxe6

I'll run this longer when I have time tonight.  Though I think this is just a
"lucky guess" by the engine.  For the record, my program fails to solve every
other position in your list.

Scott



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.