Author: Drexel,Michael
Date: 22:41:58 05/09/03
Go up one level in this thread
On May 09, 2003 at 19:01:13, Uri Blass wrote: >On May 09, 2003 at 16:53:10, Drexel,Michael wrote: > >>On May 09, 2003 at 16:11:53, Jorge Pichard wrote: >> >>>On May 09, 2003 at 15:40:24, Drexel,Michael wrote: >>> >>>>On May 09, 2003 at 13:55:55, Jorge Pichard wrote: >>>> >>>>>On May 09, 2003 at 10:12:40, Drexel,Michael wrote: >>>>> >>>>>>[D] rnbqkb1r/pp1np1pp/4p3/2pp4/3P4/2N5/PPP2PPP/R1BQKBNR w KQkq c6 0 5 >>>>>> >>>>>>What do your programs think about this well known position? >>>>>>Shredder and Fritz 8 clearly favour white here. for example : >>>>>> >>>>>>Analysis by Shredder 7.04: >>>>>> >>>>>>6.Nf3 g6 7.Ng5 Nf6 8.Bb5+ Nc6 9.0-0 Bg7 10.Re1 e5 11.Bxc6+ Qd7 >>>>>> ± (0.91) Depth: 10/10 00:00:00 6kN >>>>>>6.Nf3 g6 7.Ng5 Nf6 8.Bb5+ Nc6 9.0-0 Bg7 10.Re1 e5 11.Bxc6+ Qd7 >>>>>> ± (0.91) Depth: 11/11 00:00:00 16kN >>>>>>6.Nf3 g6 7.Ng5 cxd4 8.Bb5 Nc6 9.Nxe6 Qa5 10.Nxf8 dxc3 11.Qxd5 >>>>>> ± (0.83) Depth: 12/28 00:00:04 1435kN >>>>>>6.dxc5 >>>>>> ± (0.85) Depth: 12/28 00:00:05 1435kN >>>>>>6.dxc5 >>>>>> ± (0.85) Depth: 12/28 00:00:05 1503kN >>>>>>6.dxc5 g6 7.Qd4 Nf6 8.Nf3 Nc6 9.Nxd5 Nxd4 10.Bb5+ >>>>>> ± (0.93) Depth: 13/30 00:00:24 8276kN >>>>>>6.dxc5 g6 7.Qd4 Nf6 8.Nf3 Bg7 9.Bb5+ Bd7 10.0-0 0-0 11.Bxd7 >>>>>> ± (0.90) Depth: 14/30 00:00:51 16690kN >>>>>>6.dxc5 g6 7.Qd4 Nf6 8.Bg5 Nc6 9.Bb5 h6 10.Bf4 g5 11.Be5 Bg7 12.Bxc6+ bxc6 13.Qf4 >>>>>> ± (0.88) Depth: 15/31 00:01:44 35928kN >>>>>> >>>>>>Looks like there is a clear advantage for white. White is a pawn down and the >>>>>>evaluation is almost +1.0. >>>>>> >>>>>>However strong Chessplayers clearly seem to favour black here. Here are the >>>>>>highest rated Chessplayers who played this position in the last 6 years. All >>>>>>played with black pieces. All games ended 0-1: >>>>>> >>>>>>[Event "Santos Mario Covas 1st"] >>>>>>[Site "Santos"] >>>>>>[Date "2001.01.29"] >>>>>>[Round "3"] >>>>>>[White "Benares de sa Leitao, Ricardo"] >>>>>>[Black "Milos, Gilberto"] >>>>>>[Result "0-1"] >>>>>>[ECO "B02"] >>>>>>[WhiteElo "2276"] >>>>>>[BlackElo "2608"] >>>>>>[PlyCount "60"] >>>>>> >>>>> >>>>>Just my two cents if Milos with a 2608 had White against Benares with Black >>>>>which is a 300+ plus difference in rating and Benares wins than the case would >>>>>be open for discussion. In this case the stonger player simply outplayed the >>>>>weaker player. >>>>> >>>>>Jorge >>>> >>>> >>>>The discussion is not about this game. Its about the position in general. >>>>All this GMs wont play this opening variation if they were not convinced >>>>that the position is at least O.K. for black. 300 Elo difference is nothing in >>>>human chess. >>> >>> >>>I was not discussing this game at all, simply from from the position which you >>>mentioned. Indeed even 100 rating points difference between human opponents is >>>enough to take either side of this position and win it for the stronger player. >> >>Nonsense >> >>> >>>Jorge >> >> >>The reason why I posted this was to show that mobility is not everything >>in chess. The secret of Shredders success is: It gives very high scores >>for mobility. > >No > >I do not think that it is the secret of shredder's success. >I think that the main advantage of shredder is simply outsearching the >opponents. No way. It does outsearch weaker opponents but not the commercials. > >Too high scores for mobility may be counter productive and they may push the >program to sacrifice without compensation and to lose games. They are often NOT too high. This is a question of probability. They are sometimes too high in the opening (e.g if there are many pieces on the board) but that does not hurt, Shredder has eventually a big opening book. > > >>Therefore Shredder often seems to play with a plan (of course he doesnt) while >>other programs just shuffle around their pieces. > >I do not see how very high scores of mobilities can help you to make the >illusion of playing with a plan. I meant higher than Fritz or other commercials (except junior). What else is the reason of (0.41) at depth 1 ? Yace for example starts with (-0.51). With high scores for mobility the program theoretically always tries to improve the position of his pieces. Of course there are also apparent pawn weaknesses. However a double pawn in the center in an opening position is very often no weakness at all. In this position the double e-pawns cover many important squares. The half open f-file is also a very important factor. > >I think that there are other ways to make that illusion(of course a computer by >definition can only calculate so it is cannot be more than illusion). > >>In this position the pawn majority in the center is an important factor which is >>underestimated by many top programs. >>Both black bishops cant move and so Shredder and Fritz clearly favour white. >>They dont understand the nature of the position at all. > >The fact that the bishops cannot move in the root position may explain high >score for white at small depth but not high score at depth 15. I you look at the PV at depth 14: 6.dxc5 g6 7.Qd4 Nf6 8.Bg5 Nc6 9.Bb5 h6 10.Bf4 g5 11.Be5 Bg7 12.Bxc6+ bxc6 (13.Qf4) This position (after bxc6) is really bad for black. The mobility of the white pieces is also much higher. The variation is not important of course. Blacks moves are nonsetical. Michael > >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.