Author: blass uri
Date: 01:16:15 06/18/98
Go up one level in this thread
On June 17, 1998 at 14:49:15, Vincent Diepeveen wrote: > >On June 17, 1998 at 12:01:56, blass uri wrote: > >> >>On June 15, 1998 at 18:56:12, Enrique Irazoqui wrote: >> >>>Those are in PGN the 13 opening lines posted by you: >>> >>>[Event "?"] >>>[Site "?"] >>>[Date "????.??.??"] >>>[Round "?"] >>>[White "R9"] >>>[Black "Diep"] >>>[Result "*"] >>>[ECO "C63"] >>>[PlyCount "9"] >>> >>>1. e4 e5 2. Nf3 Nc6 3. Bb5 f5 4. d3 d6 5. exf5 * >>> >>black can have good result in this line if white is fritz5 >>for example >>5...Bxf5 6.d4 0.50/12 Nge7 7.Bg5 0.63/11 a6 8.Bxc6+ 0.56/12 bxc6 >>9.dxe5 0.75/12 Rb8 10.Na3 0.94/11 h6 11.Bxe7 0.84/11 Qxe7 >>12.0-0 1.06/11 d5 13.Rb1 1.00/12 Bg4 14.c3 1.06/11 Rb6 > >We see that score goes clearly up for white after few moves. It's a >pawn up for white now. > >The rest of the game is even with a killerbook at own risk of course, >i'm currently seeing many time losses of diep, and positions like: > >Crafty versus Diep (Judgeturpin or DoctorWho i forgot what account) > > - = - = - = - = ... 1 ... > = - = - = - = - ... 2 ... > - = - = o = o = ... 3 ... > = - = - = o = - ... 4 ... > R o k = - O - O ... 5 ... > = - r - = - O K ... 6 ... > O = - = - = - = ... 7 ... > = - = - = - = - ... 8 ... >wow what does WWW mess this up! > >black to move > >position white Kh3,Ra4,a2,f4,g3,h4 > black Kc4,Rc3,b4,e6,f5,g6 > black to move wins easily > >Believe it or not, diep managed to lost this as it didn't play Ra3, >it played Re3. >This is so simple, but i haven't worked at endgame yet. Fritz5 is not better in this position it needed more than 80 seconds to find Ra3 is winning before that(after 1 minute on my pentium 200MMX) it prefered Re3 It intended to play Ra3 without seeing it is winning after some seconds but change its mind > >Diep needs about a minute to see that Ra3 wins! > >So even a simple won position doesn't garantuee anything at blitz especially, >but at 2 minutes a move, or in Sweden at slower hardware >3 minutes a move, and commercial programs which have worked >at their weak points, it gets harder to mess up. > >Then +1.xx out of book and having positions which your prog plays >well, which you know in advance, is a huge advantage. fritz5 cannot find better moves from moves 15 to 24 if I give it 20 minutes per move after 29 minutes and 30 seconds it found 20.Ne3 instead of 20.h3 Uri > > >>15.Re1 1.06/12 Qf7 16.e6 1.00/12 Qf4 17.Nc2 0.94/12 Be7 >>18.Qd4 1.03/12 Qxd4 19.Nfxd4 1.06/13 c5 20.h3 0.66/13 Bh5 >>21.Nf5 0.50/12 Bg6 22.g4 0.50/12 Bxf5 23.gxf5 0.59/12 Rf8 >>24.Ne3 0.34/12 d4 25.Nc4 0.25/12 Rb8 26.cxd4 0.22/12 cxd4 >>27.Re4 0.22/12 Rb5 28.Rxd4 0.03/12 Rbxf5 29.Rf1 0.06/12 R8f6 >>30.Re4 0.03/12 Bc5 31.Ne3 0.00/11 Rg5+ 32.Kh2 -0.13/13 Rgg6 >>33.f3 -0.16/12 Bd6+ 34.Kh1 -0.25/13 Rxe6 35.Nc4 -0.31/11 Rxe4 >>36.fxe4 -0.56/12 Rg3 37.e5 -0.66/13 Rxh3+ 38.Kg2 -0.84/14 Rh4 >>39.b3 -0.97/13 Be7 40.Rd1 -0.94/12 h5 >>this game was played by fritz5 against itself 160 minutes per 40 moves >>on my pentium 200MMX. > >>the opening can be quilty only if fritz5 had some moves in its power book >>after move 5 that are different from this game. >> >>Uri >> >---------------------------------- >>> >>>All moves are played by Rebel 9 and Diep while still in book. I checked these >>>positions on a PII/300 at one minute per move and from the point of view of Diep >>>in the first move out of book: >>> >>> Fritz 5 Crafty 14.12 Diep Hiarcs 6 >>>1- -0.50 -0.53 -1.00 >>>2- -0.16 -0.20 -0.84 >>>3- +0.06 -0.20 -1.01 >>>4- 0.00 -0.16 -0.24 >>>5- -1.25 -1.13 -1.16 -1.46 >>>6- -1.06 -1.87 -1.24 -1.48 >>>8- -0.56 -0.45 -0.58 >>>11- -0.63 -0.88 -0.78 >>>12- -0.13 -0.59 -0.44 >>> >>>Lines 6, 7, 9 and 10 are identical, and so are lines 5 and 13. >>> >>>Lines 5 and 6 are the most favorable to Rebel and the only ones that could be >>>suspected. Line 5 was played in a game Karklins-Leverett in the Chicago US Open >>>of 1989. This game ended in a draw. Line 6 was played in the game Hazai-Moehring >>>of Halle2 1981 and it also ended in a draw. No special computer killer lines >>>here, although no one would be very happy playing black's side. >>> >>>Nothing strange about the other lines. Black is slightly down when it leaves >>>book, but this is not uncommon playing black. >>> >>>What happened in the games Fritz 5 - Rebel 9 played by the SSDF has nothing to >>>do with cooked lines. Rebel was unable to save its games and therefore to >>>"learn", and it fell over and over on the same (not cooked) lines. >>> >>>Enrique
This page took 0.01 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.