Author: Dieter Buerssner
Date: 07:11:55 10/05/01
Go up one level in this thread
On October 05, 2001 at 05:31:08, Simon Finn wrote:
>>>>>> Taimanov-Zakharov,USSR 1969 === Black to move and win
>>>>>> [D]8/5p2/7p/5pk1/8/5KPP/8/8 b
>>>>>> Source: http://chesscafe.com/text/secpawnendings.txt
>So is 1...Kf6 (mate in 33?) better than 1...h5 in the original position?
Yace cannot find a win with Kf6. Dann has posted some lines by Yace, where Kf6
failed high. Well, I wondered about this very much. Dann did not use TBs for
that line. After a closer look, I found, that Yace totally misevaluated this,
due to a serious bug in the K+P vs. K evaluation. The bug is masked when using
TBs.
You suggested 1...Kf6 2. Kf4 Ke6 3. Ke3 Ke5 4. Kf3 which is winning for black.
But after 3. g4, Yace cannot find a win within some minutes. I get:
40761728 6:28.5 0.00 26. 3.g4 f6 4.gxf5+ Kd5 5.Kg4 Ke5 6.Kh5 Kxf5 {EGTB}
7.h4 Ke5 8.Kxh6! f5 9.h5 f4 10.Kg6 f3 11.h6! f2
12.h7! f1=Q 13.h8=Q+! Ke6 14.Qg8+ Ke7 15.Qh8
Qg1+ 16.Kf5! Qg2 17.Qb8 Qg7 18.Qc8 Qh7+ 19.Kg5
Qg7+ 20.Kh5 Qh7+ 21.Kg5 Qg7+ 22.Kh5 Qh7+ 23.Kg5
{repetition} {0}
Yace tries to give a sample line, when hitting TBs. Very often this will not
make much sense in drawn postion (especially when there are alternatives to
draw). Also the line before the TB hit may not make too much sense, because
there is no difference between an easy to achieve draw and a difficult draw.
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.