Not logged inRybka Chess Community Forum
Up Topic The Rybka Lounge / Computer Chess / For the engines: An unbalanced endgame
- - By Dadi Jonsson (Silver) Date 2009-02-26 13:13
How do the engines evaluate this position? Can White win? He can immediately pick up the a4-pawn with 1.Qe8+ if he wants to.

7Q/1p6/6k1/3p1n2/p2n4/2K5/8/8 w - -
Parent - - By Vempele (Silver) Date 2009-02-26 13:18

> Can White win?


It's a draw if you remove b7 and a4 so I guess not.
Parent - By Kappatoo (*****) Date 2009-02-26 13:20
Seems like you beat me ...
Parent - - By Dadi Jonsson (Silver) Date 2009-02-26 13:23
Fast :) That's what I think too, but like Kappatoo I was wondering if "I'm missing some subtlety" that one of the engines would find.
Parent - By Vempele (Silver) Date 2009-02-26 13:26
It's also a draw with just the b7 pawn, only a4 loses - the winning line is very specific and requires the use of a8 as a checking square.
Parent - - By CSullivan (**) Date 2009-02-26 16:02
This looks like more analysis from the Botvinnik-Bronstein game.  I noticed that you had previously made the comment that you had run into Q vs. N+N endings.  I then looked at some of my analysis and realized that I had evaluated (incorrectly) these as wins for the side with the queen. 

I wonder if you have found a possible drawing variation for Black in one of the main lines????
Parent - By Dadi Jonsson (Silver) Date 2009-02-27 12:41

> This looks like more analysis from the Botvinnik-Bronstein game


Yes, that's correct :) I decided to do some additional analysis of 55...N8e7 56.Bg5 Ke6 and then I ran into this endgame again.

> I wonder if you have found a possible drawing variation for Black in one of the main lines????


I'm not sure what you mean by the main lines, but I will probably have a closer look at the above variation and perhaps 55...Kf5 too, although 56.Bh2 seems strong.
Parent - By Kappatoo (*****) Date 2009-02-26 13:20
Sorry, no engines here ...
Maybe I'm missing some subtlety, but this

7Q/8/6k1/3p1n2/3n4/2K5/8/8 w - - 0 1


is a tablebase draw - so I find it very hard to believe that your position is not drawn, too.
Parent - - By Debaser (***) Date 2009-02-27 13:38
According to theory, the knights pair usually can hold a queen, in this case I should say it is sure because they are connected and centered and the king near them

Monte Carlo gives a result of 85% draws in default settings and after 1000 games

The contrary happens with the bishop pair, only Lolli's fortress draws
Parent - - By Vempele (Silver) Date 2009-02-27 13:52

> The contrary happens with the bishop pair, only Lolli's fortress draws


There are similar fortresses that can push the win past 50 moves in some cases.

Bishop + knight also loses with the exception of Bb7+Nd5 and mirrors. I don't know if there's another fortress that even comes close.
Parent - By Debaser (***) Date 2009-02-27 14:00
You are right, more positions can be drawn with 50 moves rule ;)

Yes too according to theory, Bb7+Nd5 and mirrors are the exceptions, B+N are lost in the rest of the cases
Up Topic The Rybka Lounge / Computer Chess / For the engines: An unbalanced endgame

Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill