Author: Uri Blass
Date: 02:52:51 07/12/01
Go up one level in this thread
On July 12, 2001 at 04:38:17, Vincent Diepeveen wrote: >On July 11, 2001 at 20:18:05, James T. Walker wrote: > >>On July 11, 2001 at 18:59:27, Vincent Diepeveen wrote: >> >>>On July 11, 2001 at 17:25:34, James T. Walker wrote: >>> >>>>Junior 7 likes Nxd5 after 5 seconds on my Athlon 900. I holds it for over 3 >>>>minutes with a score of less than 1. Junior 7 likes Nxd5 cxd5 Rxd5 but then >>>>likes Nc4 Bc5 as a followup. If I force feed it Rd8 instead of Nc4 the score >>>>goes slightly over +1. So anyway Junior 7 would play this (Nxd5) move probably >>>>in blitz. >>>>Jim >>> >>>we need a mainline. >>> >>>Nxd5 can be played because of 2 reasons >>> a) patzerscores >>> b) it sees the whole trick >>> >>>in case of a it didn't solve it. In case of b it must show >>>at least next moves in mainline: >>> nxd5 cd rd5 nc4 bf1! >>> >>>if other moves as bf1 shown then forget it. >> >> >>Hello Vincent, >>You may be right since you are a good chessplayer but my question is why do you >>insist on Bf1? Since Bf1 was not played in the game and white won anyway why do > >"bf1 was not played in the game" is not a valid argument. Opponent made >huge blunder as he played Rd8. > >>you believe Bf1 is needed? Also if I play the black moves from the Nxd5 move > >Junior is PSQ please use a non preprocessor you will see score drop >to near zero at bigger depths when playing the line > Nd5 cd Rd5 Nc4 Rf5 Junior7 is not a preprocessor. Maybe you got the impression that Junior7 is a preprocessor from another post of mine. I said that something surprised me because the evaluation after the move was clearly different and I remember that I said that I believed that Junior is not a preprocessor. I said that I believed that Junior is not a preprocessor in order to explain the reason that I was surprised to see something and not in order to say that Junior is a preprocessor. I still believe Amir Ban that junior is not a preprocessor and the reason for the difference in the evaluation was different. The reason is probably simply the fact that junior does not go linear in depthes so if I give it to analyze without learning it can jump from depth x to x+2 and never be in depth x+1 when it has different evaluation. Junior can see a drop in the evaluation for Nd5 in the root position and here is the analysis (now on p800(64 Mbytes hash) with time control of 6 hours/40 moves) It can also see Bf1 at depth 17 Junior 7 - Blass,U r5nr/pp3kpp/2pb4/3pn1P1/8/2N1B2B/PPP2P1P/2KRR3 w - - 0 1 Analysis by Junior 7: 1.g6+ hxg6 2.Bf1 Rxh2 = (-0.23) Depth: 3 00:00:00 1.Ne4 Bb4 2.Bd2 Be7 3.Ng3 Nf3 4.Be6+ Ke8 5.Bxg8 Rxg8 ± (0.77) Depth: 6 00:00:00 16kN 1.Ne4 Bb4 2.Bd2 Be7 3.Ng3 Nf3 4.Be6+ Ke8 5.Bxg8 Rxg8 ± (0.77) Depth: 6 00:00:00 16kN 1.Ne4 Bb4 2.Bd2 Be7 3.Ng3 Nf3 4.Be6+ Ke8 5.Bxg8 Rxg8 ± (0.77) Depth: 6 00:00:00 16kN 1.Ne4 Bb4 2.Bd2 Be7 3.Ng3 Nf3 4.Be6+ Ke8 5.Bxg8 Rxg8 ± (0.77) Depth: 6 00:00:00 16kN 1.Ne4 Bb4 2.Bd2 Be7 3.Ng3 Nf3 4.Be6+ Ke8 5.Bxg8 Rxg8 ± (0.77) Depth: 6 00:00:00 16kN 1.Ne4 Bb4 2.Bd2 Be7 3.Ng3 Nf3 4.Be6+ Ke8 5.Re3 Nxd2 6.Kxd2 Bxg5 ² (0.54) Depth: 9 00:00:00 188kN 1.Bg2 Ne7 2.Bd4 N5g6 3.h3 Bf4+ 4.Be3 Bxe3+ 5.Rxe3 ² (0.59) Depth: 9 00:00:00 200kN 1.f4 Nc4 2.Bd2 Nxd2 3.Rxd2 b5 4.Be6+ Kg6 5.Ne2 ² (0.68) Depth: 9 00:00:00 323kN 1.f4 Nc4 ² (0.38) Depth: 12 00:00:02 1401kN 1.Bg2 Ne7 2.Ne4 Bb4 3.Nd2 Ng4 4.c3 Nxe3 5.Rxe3 Bd6 ² (0.59) Depth: 12 00:00:02 1629kN 1.Nxd5 cxd5 2.Rxd5 Nc4 3.Rf5+ Kg6 4.Bd4 Ne7 5.Re6+ Kh5 6.Rf7 Rac8 7.Bxg7 Rhe8 ± (0.81) Depth: 12 00:00:06 3685kN 1.Nxd5 cxd5 2.Rxd5 Nc4 3.Rf5+ Kg6 4.Bd4 Ne7 5.Re6+ Kh5 6.Rf7 Ng6 7.Bxg7 Bf4+ 8.Kb1 Rhd8 9.Re4 ± (0.77) Depth: 15 00:00:45 29351kN 1.Nxd5 cxd5 2.Rxd5 Nc4 3.Bf1 Nxe3 4.Rxe3 Bxh2 5.Bc4 Kg6 6.Bd3+ Kf7 7.Rf5+ Nf6 8.gxf6 gxf6 9.Bc4+ Kg6 10.Rd5 ² (0.57) Depth: 17 00:06:03 225373kN (Blass, Tel-aviv 12.07.2001) > >This line is very unclear and needs deep study to reveal whether >it's: > a) won for black > b) draw > c) won for white > >This where Nd5 cd Rd5 Nc4 Rf5 Bf1 is a clear win for white. The fact that Junior does not see Bf1 for the tactical reason does not prove that the evaluation of junior is wrong. Junior did not claim that Nd5 is a winning move. 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.