Author: José de Jesús García Ruvalcaba
Date: 15:47:23 06/02/00
Go up one level in this thread
On June 02, 2000 at 13:31:53, Dann Corbit wrote: >On June 02, 2000 at 04:37:15, Paulo Soares wrote: >>On June 01, 2000 at 17:01:13, José de Jesús García Ruvalcaba wrote: >>>On June 01, 2000 at 13:01:25, Vincent Lejeune wrote: >>>>On June 01, 2000 at 11:48:37, José de Jesús García Ruvalcaba wrote: >>>>>On June 01, 2000 at 05:30:31, Vincent Lejeune wrote: >>>>>>On May 31, 2000 at 13:41:45, Dann Corbit wrote: >>>>>>>CAP records (some of them do not appear to have faired well): >>>>>>>6k1/8/2P5/p6p/5bP1/7P/2B2K2/8 w - - acd 20; acn 1174673323; acs 3601; ce >>>>>>>153; pv gxh5 Kg7 Kf3 Bh2 Kg4 Kh6 Bd1 Bc7 Kf5 Bb6 Ke6 Bc7 Kd7 Bf4 Kd8 Be5; >>>>>> >>>>>>I really doubt that gxh5 is the best move because the h's pawn have the wrong >>>>>>square colour to promote, if the black bishop stay on the h2-b8 diagonal with >>>>>>can make no progress. >>>>>>1.g5 seems to be stronger >>>>> >>>>> g5 loses a pawn outright. I agree that gxh5 is a poor move. I think Kf3 is >>>>>better. >>>>>José. >>>> >>>>If 1..Bxg5?, 2.c7! wins a queen ... >>> >>> You are right! This happens to me frequently when I do not set up real board >>>and pieces to analyze. >>> After 1. g5 black must play h4 (otherwise white plays h4, then Bd1 tying >>>black's king to defent the h5 pawn, and then the white king supports the c-pawn >>>and wins). 1. Kf3 attacks the bishop, so white can play g5 the next move. >>> Interesting position, and one in which a little knowledge for opposite-coloured >>>bishop endings is extremely helpful (in this case: when trying to win make both >>>bishops "good", and when trying to draw make both bishops "bad"). >>>José. >> >>Strange, I made a long analyse from this position and it seemed >>that 1. g5 and 1. Kg3 are winner moves. >> > >On a deep search, all three of gxh5, g5, and Kf3 fare about the same. gxh5 is a mistake, which should be avoided. > Here are >the positions *after* each of those choices are made, analyzed for three hours: > > >6k1/8/2P5/p6p/5bP1/5K1P/2B5/8 b - - acd 20; acn 2045418075; acs 10002; ce >-171; pv Bc7 gxh5 Again, 2. gxh5 is a poor move. 1. Kf3 is good if followed by 2. g5. >Kg7 Kg4 Kh6 Ba4 Be5 Bd1 Bh2 Kf5 Bc7 Ke6 Bh2 Kd7 Bf4 Kd8 Bg3 >Kd7 Kg5; > >6k1/8/2P5/p6P/5b2/7P/2B2K2/8 b - - acd 21; acn 2022412756; acs 10001; ce >-161; pv Kg7 Kf3 Bc7 Kg4 Kh6 Bd1 Bd8 Kf5 Bc7 Ke6 Bh2 Kd5 Bc7 Kd4 Bb6+ Kc4 Bc7 >Bg4 Be5 Kd5 Bg3 Ke6; > >6k1/8/2P5/p5Pp/5b2/7P/2B2K2/8 b - - acd 21; acn 2095662806; acs 10001; ce >-182; pv h4 Kf3 Be5 Kg4 Bg3 Ba4 Kf7 Kh5 Kg7 Bc2 Kg8 Kh6 Bf4 Bb3+ Kh8 Kh5 Bc7 Bg3 defending the pawn is better. If white can win the h4 pawn by force it follows a rather easy win. >Kxh4 Bd8 Kg4 Bc7; > >Naturally, since we are one ply forward, we are looking for the most negative >score. 1/5 of a pawn separates the best from the worst of the three choices. > >However, the first and the third positions both had fail lows at the end of >their searchs, so these are probably the best choices. > Do not trust crafty so blindly! It is obvious crafty is missing lots of things in this position (just follow the PV's you posted). José.
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.