Author: Vincent Diepeveen
Date: 18:04:49 12/02/02
Go up one level in this thread
On December 02, 2002 at 20:59:01, Vincent Diepeveen wrote: >On December 02, 2002 at 19:19:15, Peter Berger wrote: > >>On December 02, 2002 at 18:27:27, Vincent Diepeveen wrote: >> >>>DIEP only considers Qf6. I myself only consider Qf6 to be best move. >>>I jumped up immediately pawn after Qe8. >>> >> >>Qe8 is a bad move agreed. My first impression was that Qf6 wins, too. But after >>trades and Ng6 the black position looks quite solid. Can you give some lines you >>think win for white? > >I remember how i felt during the game when i saw Qf6 and i was failing >low to nearly white up 3 pawns. In fact to white up 2.938. > >Basically as a human i would find it a tough position for white to >win, but if you analyze it a bit you will soon see that black cannot >to that much. basically its only hope is a knight pin to g2. > >In all lines white seemingly can make free a rook to 'hack' into >the black queen position. > >After 39.Qf6 Qxf6 40.gxf6 Ng6 then white goes for example 41.a4 with the >big problem that blacks rook cannot move from d-file because of Rd3. > >So whites seemingly weak a and c pawns are safe by means of major threats >towards the black king position. > >Even then it is not 100% clear of course like most computer positions >which are a mess this also is a mess. White has a stupid bishop and >black has a stupid knight pair. > >If black manages to eliminate 3 pawns on the queen side there is obvious >good draw chances for black. Yet in not a single analysis i managed that >when i let diep play white as it has in all lines c5 always, which also >happened in the game. > >For example: > 41... h6 > 42.a5!? Kh7 > 43.Rb2 Ngf4 > 44.c5! bxc5 > 45.Rb7 and score is already +4.4 here and if > Ne2 is not a problem for white then it is > trivial that white wins. > >So what else to try? > >Let's try the seemingly very stupid > 43... bxa5 > 44. Rb7 Ng5 > 45. c5 and one of whites passers is clearly unstoppable > in combination with h4 and Rg7 mate threats. +5.0 even > which is logical. black plays witout knights on the queen side. > >So it is clear that allowing c5 is a major problem. But can >black prevent it? I guess not. It means already you must go play >stupid computer lines with black at move 42 which go to trivially >lost bishop versus knight endings which as a FM i win blindfolded. > >I guess you can imagine that when i saw DIEP having the queen exchange >in its lines even before warp played Qf6, that i was extremely depressed, >because the whole position is extremely simple to win for white as soon >as you realize that white can so a quick c5. > >However soon i got very optimistic about winning the game when Chop >planned to give some checks and take of a stupid pawn on a7. > >Then when i was getting very happy about having still winning chances, >when Warp allowed black to not get into an endgame. > >Warp managed to still get a draw with a well timed c5 move, which is >simply saving Warps ass, but desipte that i still hoped that >all those pieces near whites king would kill it. Very lucky also the >resulting endgame after all the exchanges were a draw for white. > >I had hoped for more after the Qe8 move and seeing the pawn grabbing >mainlines at a7. > >Directly Qxa7 instead of Qf6 i don't want to analyze too well in fact. >A short line should do, because i get directly many fail lows after >after Nf4 there for white, so it is trivial that black has excellent draw >chances there 4 pieces combining against a poor white king. i don't see >how Peter can believe in that for white. He needs a chess course :) perhaps after Qxa7 a direct Qe4 is directly forcing a draw there. white can win a bunch of pawns by some sacrafices on g6 but then white has to give always perpetual check after: qxa7 qe4 qxb6 nf3 g6 hg6 rg4 qa8 rxg6 fxg6 qxe6 nxg1 qf6+ it takes diep a while to see that this is a draw. weird. looks trivial to me as long as you don't lose the d1 rook.
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.