Author: Uri Blass
Date: 01:37:15 09/12/02
Go up one level in this thread
On September 12, 2002 at 02:58:52, Uri Blass wrote: >Fritz 7.0.0.8 - XP Machëide x2cou_51 >[D]6r1/r6k/3p1Qp1/2bPp2p/B1P1K2P/1P4P1/8/8 w - - 0 52 bm g4 am Qf1 > > > >I analyazed the first game and I think that the losing mistake of Fritz was >52.Qf1. > >Some analysis suggest that white has a draw by >52.g4 hxg4 53.h5 gxh5 54.Qf5+ Kh6 55.Qf6+ Rg6 56.Qf8+ Kh7 57.Qf5 Re7 58.Be8 Rxe8 >59.Qd7+ Kh6 60.Qxe8 Rf6 61.Qh8+ Kg6 62.Qg8+ > >This was the main line of yace after hours. > >The problem with that is that Fritz probably wanted to win the game >and the default contempt factor told it to prefer -0.12 for Qf1 and not a draw >for g4. > >I say probably because I do not have Fritz7 to analyze and I found the line that >was posted after giving yace to analyze all night when it went backward in the >game. > >Movei is analyzing the position and it likes to play Kf3 with 0.00 score but it >seems that Kf3 is also bad > >Here is some analysis of yace(note that yace learned that Qf1 is bad). > >52.Kf3 Rgg7 53.Be8 e4+ 54.Kf4 e3 55.Kg5 Rae7 56.Bxg6+ Kg8 57.Qa1 Re5+ 58.Kf6 Ra7 >59.Qe1 e2 60.Bxh5 Bf2 61.Qxf2 > = (-0.23) Depth: 15 00:05:40 125927kN >52.Kf3 Rgg7 53.Kg2 Rgf7 54.Qg5 e4 > ³ (-0.55) Depth: 16 00:17:40 385158kN >52.g4 hxg4 > ³ (-0.54) Depth: 16 00:35:12 758607kN >52.g4 hxg4 53.h5 gxh5 54.Qf5+ Kh6 55.Qf6+ Rg6 56.Qf8+ Kh7 57.Be8 Rg5 58.Qf6 Rag7 >59.Kd3 > ³ (-0.36) Depth: 16 00:40:47 875637kN >52.g4 hxg4 53.h5 gxh5 54.Qf5+ Kh6 55.Qf6+ Rg6 56.Qf8+ Kh7 57.Qf5 Re7 58.Be8 Rxe8 >59.Qd7+ Kh6 60.Qxe8 > = (0.00) Depth: 17 01:04:41 1373474kN >52.g4 hxg4 53.h5 gxh5 54.Qf5+ Kh6 55.Qf6+ Rg6 56.Qf8+ Kh7 57.Qf5 Re7 58.Be8 Rxe8 >59.Qd7+ Kh6 60.Qxe8 Bb4 61.Qf8+ Kg5 62.Qf5+ Kh6 > = (0.00) Depth: 18 02:03:55 2561567kN >52.g4 hxg4 53.h5 gxh5 54.Qf5+ Kh6 55.Qf6+ Rg6 56.Qf8+ Kh7 57.Qf5 Re7 58.Be8 Rxe8 >59.Qd7+ Kh6 60.Qxe8 Rf6 61.Qh8+ Kg6 62.Qg8+ > = (0.00) Depth: 19 03:38:44 263775kN > >(blass, tel-aviv 12.09.2002) > >Uri Movei needs only depth 15 but more than an hour to find g4 Here is the analysis of movei(p850) under winboard only at depth 15 it get a sligthly negative score for Kf3 (losing 0.01 pawn) so it can find g4. depth 14 (0.00) 670.31 seconds 104955896 nodes Kf3 Rgg7 Be8 Rg8 Bf7 Rg7 Be8 depth 15 (-0.01) 1218.19 seconds 191246051 nodes Kf3 Rgg7 Be8 e4+ Kf4...the rest is not important depth 15 (0.00) 4585.45 seconds 695460907 nodes g4 depth 15 (0.00) 4832.23 seconds 733997619 nodes g4 hg4 h5 g3 Qh4?(Be8 is the right move) g2?(g5 is better) hxg6+ Kxg6 Qg4+ Kf6 Qf5+ Kg7 Qg5+ Kf7 Qf5+ Ke7 Qe6+ Kf8 Qf6+ Rf7 Qd8+ Kg7 Qg5+ Kh8 Qh6+ Rh7 Qf6+ Rgg7 Qf8+ Rg8 Qf6+ Uri
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.