Author: Dieter Buerssner
Date: 14:18:47 01/08/06
Go up one level in this thread
On January 07, 2006 at 19:16:19, Vincent Lejeune wrote:
>On January 07, 2006 at 14:39:32, Dieter Buerssner wrote:
>...
>>> 3b4/3B4/4n3/5kp1/8/6P1/6K1/8 b - - 0 1 AM g4??
>>
>>Not avoided in 10 minutes. Can you please show a sample winning line (I did not
>>search for one myself yet).
>
>It's a position you posted 2 years ago ;o)
Oops
>http://chessprogramming.org/cccsearch/ccc.php?art_id=361560
>
>After g4 the position is DRAW but I'm not sure not pushing g4 is winning.
I looked at it again, and I am sure it is winning. After 1...Ke5 and a bit of
backwards analysis, I get:
22/41 27:24 -4.95 2.Kf3 Sd4+ 3.Ke3 La5 4.Lc8 Le1 5.g4 La5
6.Lb7 Lb6 7.La8 Sb3+ 8.Kd3 Kf4 9.Ld5 Sd4
10.Lf7 (3.055.536.632) 1857
TB:10.297
This score means, that a KBNP vs. KB endgame can be forced. This always should
be an easy win. I actually followed quite a few lines further, and due to
Shredder Online, the always ended in TB mate. I started the backwards analysis
with a "goal score" of -9; it was no big problem, but then I lost patience, and
thought -5 would be enough of a proof. I guess with 6-men TBs a mate score can
be proved with some effort.
The position itself is not an excellent test position for non interactive "batch
testing" (although it is interesting to analyse, and to look how engines
behave), because g4 can be avoided with only a slight difference in score.
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.