Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: rook endgame test position

Author: Kurt Utzinger

Date: 13:50:39 12/15/03

Go up one level in this thread


On December 15, 2003 at 16:23:15, Dieter Buerssner wrote:

>[D] 5k2/1R4p1/5p2/4p2p/1p2P2P/r4P2/PK4P1/8 b - -
>
>This is from a game between Ruffian-Yace in the "CSS-Online-Masters" (time
>control 60 15).
>
>I think, after Ra4, white cannot make progress, and the game will end in draw.
>Yace played Re3? instead (with 0.0 score). Are there engines, that show in
>reasonable time a big score difference between Re3 and Ra4?. I believe Re3 loses
>by force. For example Re3 Rxb4 Re2 Kb3 Rxg2?! (possibly Re3 instead gives more
>resistance) a4.
>
>It is not a real good test position, because engines can show for a long time
>the same score for Ra4 and Re3, and the chosen move might be pure luck.
>Therefore also the question for scores after those moves.
>
>Regards,
>Dieter

New game
5k2/1R4p1/5p2/4p2p/1p2P2P/r4P2/PK4P1/8 b - - 0 1

Analysis by Fritz 8: (P4 1.8/32 MB hash)

1...Ra4 2.g4 g5 3.hxg5 hxg4 4.fxg4 fxg5 5.Kb3 Ra3+
  ²  (0.35)   Depth: 7/20   00:00:00  23kN
1...Rd3 2.Rxb4 Rd2+ 3.Kb3 Rxg2
  ²  (0.32)   Depth: 7/20   00:00:00  41kN
  ²  (0.32)   Depth: 8/19   00:00:00  72kN
1...Ra4 2.g3 Ra3 3.g4 Rxf3 4.gxh5 f5 5.Rxb4 Rf2+ 6.Kb3 Rf3+
  ²  (0.28)   Depth: 8/19   00:00:00  83kN
  ²  (0.32)   Depth: 11/25   00:00:01  800kN
1...Rd3!
  ²  (0.28)   Depth: 11/27   00:00:02  1399kN
  =  (0.13)   Depth: 11/27   00:00:02  1787kN
1...Re3 2.Rxb4 Re2+ 3.Kb3 Rxg2 4.a4 Rg3 5.a5 Rxf3+ 6.Ka2 g5 7.a6 Ke7 8.Ra4 Rf2+
  =  (0.10)   Depth: 11/27   00:00:03  2117kN
  =  (0.25)   Depth: 13/31   00:00:08  5115kN
1...Ra4 2.g4 g6 3.gxh5 gxh5 4.Rh7 Ra3 5.Rxh5 Rxf3 6.Rh8+ Ke7 7.Rb8 Rh3 8.Rxb4
Rxh4 9.Kb3
  =  (0.22)   Depth: 13/31   00:00:10  6267kN
  =  (0.22)   Depth: 18/57   00:04:09  162881kN, tb=88

(Utzinger, MyTown 15.12.2003)




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.