Author: Robert Hyatt
Date: 09:36:48 08/24/05
Go up one level in this thread
On August 24, 2005 at 11:59:58, Uri Blass wrote: >On August 24, 2005 at 11:14:48, Robert Hyatt wrote: > >>On August 24, 2005 at 03:05:25, Uri Blass wrote: >> >>>On August 24, 2005 at 01:23:26, Majd Al-Ansari wrote: >>> >>>>A human will immediately assess that black has a passed pawn on kingside so his >>>>own passed pawn on queenside cannot be supported by the king .... therefore an >>>>obvious draw without any calculation needed. >>> >>>I think that you overestimate humans. >>>I was not talking about the top humans but about the big majority of humans who >>>play in tournaments. >>> >>>Of course humans can see that black cannot support the passed pawns by the king >>>after some thinking but not immediately. >> >>Sorry, but I disagree here. But then I have worked on this exact code in Crafty >>many times. I don't even have to think about the king-side majority that >>becomes a passed pawn. So when I saw the position, my only question was "can >>black do anything on the queenside without the king, because the king has to >>restrain white's passed pawn on the kingside. >> >>After looking for a few seconds, I thought "no". So now it does become some >>calculation to see if either can win when both have passers on opposite wings. >>Since white's king is positioned OK, didn't take me much to come up with >>"draw"... > >I know that basically 2 pawns cannot win without the king but I also know >basically that with 2 pawn advantage in pawn endgame you can win and when you >have 2 passed pawns and the oponnent only has pawn majority it seems to me >thatit probably a win for the side with 2 passed pawns. > >Only after thinking about it I see that the king cannot support the pawns. > >pawn majority of 2 against 1 is not something that I am afraid of based on >experience and here is an example > >[D]8/8/pk4pp/8/PP5K/8/8/8 w - - 0 1 > >pawn majority does not help white and note that black has only one pawn >advantage and not 2 pawn advantage. > >Uri that isn't a pawn majority in my calculations. The pawn can't advance to make a passed pawn, it is just a pawn liquidation that loses a pawn. Also, two connected passers against a single passer loses. The king can't stop two connected passers. The only calculation I have to do here is to answer the question "can one side produce two connected passers against the other side's single passer?" If so, the two connected passers have to be stopped, losing the single passer and the game... I actually have code in Crafty to handle this specific case (a protected passer wins against an isolated passer if the king can catch the isolated passer)...
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.