Author: KarinsDad
Date: 16:06:29 06/17/99
Go up one level in this thread
On June 17, 1999 at 18:26:40, Dave Gomboc wrote: >On June 17, 1999 at 14:42:44, KarinsDad wrote: > >>On June 17, 1999 at 13:32:19, Dave Gomboc wrote: >> >>>Imagine: >>> >>>White Ra1, Pa4, Pb4, Pc3 >>>Black Ra8, Pa6, Pb5, Pc6 >>> >>>The a6 pawn should be evaluated as either isolated or backward (depending on who >>>is to move), assuming other pieces do not interact with this configuration. >>> >>>Dave >> >>r.. >>... >>p.p >>.p. >>PP. >>..P >>... >>R.. >> >>The a6 pawn cannot be considered isolated unless black plays ba (which is not >>forced). >> >>The reason that it appears that the a6 pawn is backwards is due to the white >>rook having control of the a5 square indirectly (since white can play ab at any >>time). If black had another rook protecting his a8 rook, then the a6 pawn would >>not be considered backwards due to the a6 pawn not being pinned (i.e. the a6 >>pawn could be pushed without the penalty of losing a pawn). >> >>I think that examples like this are excellent since they illustrate that some >>concepts are MUCH more complex than is readily apparent (and hence the need for >>a search engine). >> >>KarinsDad :) > >What I said wasn't exactly what I meant, but the main point is yeah, there's a >lot more going on than the obvious stuff. More specifically regarding >backwardness, if it is Black's move, and allowing White to play ab would be bad, Why? 1. ab cb, what is white's follow up? If 2. c4 bc 3. Ra5 (3. b5 a5) c3, 4. Rc5 Rb8 and black has won a pawn and has 2 pawns to white's 1. ab is only good for white if black pushes a5, so black's a pawn is definitely backwards. Even if it is black's move, he can play R practically anywhere to not push a pawn. The a6 pawns is backwards, but there is no reason to isolate any of the pawns (at least not yet). >then Black can play ba and accept the isolated pawn penalty instead of the >backward pawn penalty. But, if ba is unplayable, then the pawn is backward. >You could for sure apply the minimum penalty of the two in such a case. > >Realistically speaking, there are often tons of other interactions on the board, >so it might be too time-consuming to resolve it all. Maybe it could be done >incrementally? I haven't thought about it enough. Because of this (lack of >thought :-), what I am saying is a probably bit muddy, but I think you >appreciate what I am trying to get at. Definitely. Due to the interaction of pieces, it is extremely difficult to come up with a set formula to figure out things such as backwards pawns, etc. However, there are probably some rules of thumb that can be used such as who controls the square in front of the pawn. However, these types of rules will not always be totally accurate as you can see in your example. It is hard to say that the c3 or the c6 pawn is backwards since either of them can be pushed and it is not guaranteed that a pawn will be lost (ps. I am making this claim as a lousy chess player, I have not analyzed this fraction of a position with a program). KarinsDad :) For instance, if the a8-rook is defended >by a rook on f8, great, but now if we put another rook on a2 we are back to >where we were. (Shades of DB-Kasparov. :-) > >Dave
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.