Author: Randy Lovelace
Date: 20:48:00 03/04/99
Go up one level in this thread
On March 04, 1999 at 18:29:05, blass uri wrote: > >On March 04, 1999 at 11:48:36, Robert Hyatt wrote: > >>On March 04, 1999 at 11:00:46, Randy Lovelace wrote: >> >>>In this position La Dame Blanche played h4!! in only 5 sec >>>Rebel 10c cant find this move after 10min on my machine >>>This was a game played by my program RLchess once again a passed >>>pawn kills me >>>4bR2/4P3/1np3pk/1p2B3/p3R3/8/2q3PP/7K w - - bm h4!!;id Ldb-RLchess; >>>Randy Lovelace >> >> >>You must have set the position up wrong. I ran this on crafty and got >>the following: >> >> 6 0.27 0.46 1. Ref4 Qc1+ 2. Rf1 Qxf1+ 3. Rxf1 Nc4 >> 4. Bh8 Nd6 5. Be5 Ne4 >> 6 0.66 ++ 1. h4!! >> 6-> 0.70 0.85 1. h4 >> 7 0.98 ++ 1. h4!! >> 7 2.11 3.45 1. h4 Qb1+ 2. Kh2 g5 3. Rxe8 Qxe4 4. >> Rh8+ Kg6 5. e8=Q+ Kf5 6. Qf7+ Kxe5 >> 7. Re8+ Kd4 8. Rxe4+ Kxe4 9. hxg5 >> 7-> 2.43 3.45 1. h4 Qb1+ 2. Kh2 g5 3. Rxe8 Qxe4 4. >> Rh8+ Kg6 5. e8=Q+ Kf5 6. Qf7+ Kxe5 >> 7. Re8+ Kd4 8. Rxe4+ Kxe4 9. hxg5 >> 8 3.25 ++ 1. h4!! >> 8 6.18 7.19 1. h4 g5 2. Rxe8 Kg6 3. Rf8 Qd1+ 4. >> Kh2 Qd7 5. e8=Q+ Qxe8 6. Rxe8 >> 8-> 6.87 7.19 1. h4 g5 2. Rxe8 Kg6 3. Rf8 Qd1+ 4. >> Kh2 Qd7 5. e8=Q+ Qxe8 6. Rxe8 >> 9 8.78 7.19 1. h4 g5 2. Rxe8 Kg6 3. Rf8 Qd1+ 4. >> Kh2 Qd7 5. e8=Q+ Qxe8 6. Rxe8 >> 9-> 10.34 7.19 1. h4 g5 2. Rxe8 Kg6 3. Rf8 Qd1+ 4. >> Kh2 Qd7 5. e8=Q+ Qxe8 6. Rxe8 >> 10 16.98 7.14 1. h4 g5 2. Rxe8 Qd1+ 3. Kh2 Qd7 4. >> Rh8+ Kg6 5. Bg3 gxh4 6. Bxh4 Nc4 7. >> Rg8+ Kf5 8. e8=Q Qxe8 9. Rgxe8 >> >>I can't imagine a program like Rebel not finding that. > >I can imagine that there is a bug that Rebel10C can never find some simple moves >in some rare cases because I found similiar problems with other programs. > >I know that there are rare cases when Junior5 can never find simple moves(other >computers have no problem to find and I am not talking about underpromotions). > >Fortunately this cases are rare enough so Junior is a good program. > >I know from analyzing studies that there is a bug in Genius3(this program can go >to an infinite loop and lose on time). > >I also found that I cannot be sure that the program is correct when it say mate. >These cases are rare and almost never happen in games. > >Uri this position rebel 8 coundn't find 7r/1Q3p1p/3p1k1b/4p3/3n4/N2q3P/PP3PPR/R5K1 b bm Rb8! this is a deflection sacrifice I believe Ed Schroeder is dropping duffer moves in the early iterations Maybe He also fails in the famous Sanz rook sacrifice in the endgame Randy Lovelace
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.