Author: Howard Exner
Date: 12:43:03 02/25/00
Go up one level in this thread
On February 25, 2000 at 13:59:29, Dann Corbit wrote:
>Unsolved:
>{This one by two different engines at long time control}
>1q1r3k/3P1pp1/ppBR1n1p/4Q2P/P4P2/8/5PK1/8 w - - acd 16; acn 1000; bm Rxf6; ce
>181; id "ECM.825"; pm Qd4; pv Qd4 Kg8 Rxf6 gxf6 Qxf6 Qc7 Qxh6 Rxd7 Qg5+ Kf8 Bxd7
>Qxd7 h6 Qd4 f5 b5 h7 Qg7 Qxg7+ Kxg7 axb5 axb5;
>1q1r3k/3P1pp1/ppBR1n1p/4Q2P/P4P2/8/5PK1/8 w - - acd 16; acn 425627101; acs 2501;
>bm Rxf6; ce 131; id "ECM.825"; pv Qd4 Ng8 Be4 Qc7 Bd3 Nf6 Rxf6 gxf6 Qxf6+ Kg8
>Qxh6 Rxd7 Bh7+ Kh8 Bf5+ Kg8 Bxd7 Qxd7 Kg3 Qxa4 Qxb6;
1. Rxf6 Qxe5 2. fxe5 gxf6 3. exf6 is the sequence.
It has that caged in piece theme that computers are many times blind to. In this
case the King is caged in and the rook must remain on the 8th rank. So all that
is left is for the white king to move towards the queening square.
>
>2kr3r/ppp1qpp1/2p5/2b2b2/2P1pPP1/1P2P1p1/PBQPB3/RN2K1R1 b Q - acd 13; acn
>426250253; acs 2501; bm Rh1; ce 348; id "ECM.1548"; pv Qh4 gxf5 Qh1 Rf1 g2 Nc3
>Qh4+ Kd1 gxf1=Q+ Bxf1 Qf2 Kc1 Qxf1+ Nd1 Rh2 Qc3 f6 Kc2;
This one seems to have multiple winning lines. Spassky played Rh1 but in ECM98
Amir Ban posted another quick win in Bxe3. IN ECM98 both Rh1 and Bxe3 are given
as best move.
>5r2/pp1RRrk1/4Qq1p/1PP3p1/8/4B3/1b3P1P/6K1 w - - acd 15; acn 436429868; acs
>2500; bm Qxf7+; ce 463; id "ECM.614"; pv Rxf7+ Rxf7 Qxf7+ Qxf7 c6 bxc6 bxc6 Be5
>Bd4 Bxd4 Rxf7+ Kxf7 c7 Bb6 c8=Q Kf6 Qf8+ Kg6 Qd6+ Kg7 Kg2 h5 Kf3;
Both Qxf7 and Rxf7 are best moves here, as posted. So this one is correct
according to C.A.P.
>8/5pk1/4p3/7Q/8/3q4/KP6/8 b - - acd 15; acn 413864161; acs 2500; bm Qd5+; ce
>207; id "Exner:End.12"; pv f5 b4 Qc4+ Ka3 Kf6 Qh8+ Ke7 Qe5 Qd5 Qe3 Kd6 Qf4+ Kd7
>Qf1 Ke7 b5 Qc5+ Kb3 Kd6;
Hiarcs also played the move f5 and only ended up with a draw against Fritz.
1... Qd5+ 2. Qxd5 exd5 3. b4 Kf6 4. Kb3 Ke5 and those two passed black pawns are
as protected as if they were two connected pawns. I wonder if programs know that
pawns seperated by one square protect each other from the hostile king as well
as two connected passed pawns?
>8/6Bp/6p1/2k1p3/4PPP1/1pb4P/8/2K5 b - - acd 17; acn 489219175; acs 2501; bm b2+;
>ce 28; id "exner.pp.12"; pv Kd6 g5 Ke6 Kb1 Bd4 h4 Kd6 Bf8+ Kc6 Bg7 Kd7 Bxe5 Bxe5
>fxe5 Ke6 Kb2 Kxe5 Kxb3 Kxe4;
1... b2+ 2. Kc2 exf4 3. Bxc3 f3 4. Be1 Kd4 5. Bf2+ Kxe4 6. Bc5 b1=Q+ 7.
Kxb1 Kd3 8. h4 Ke2 9. h5 g5 10. h6 f2 11. Bxf2 Kxf2 12. Kc2 Kg3
>r1b5/p2k1r1p/3P2pP/1ppR4/2P2p2/2P5/P1B4P/4R1K1 w - - acd 15; acn 418192690; acs
>2502; bm Bxg6; ce 220; id "exner.pp.04"; pv cxb5 Bb7 Rdd1 f3 Kf2 Re8 Rxe8 Kxe8
>Bb3 Rf4 Re1+ Re4 Rxe4+ Bxe4 Be6 Kd8 d7 c4 a4;
1. Bxg6 hxg6 2. Re7+ Rxe7 3. dxe7+ Kxe7 4. Rd8 Kxd8 5. h7
>r4rk1/ppq3pp/2p1Pn2/4p1Q1/8/2N5/PP4PP/2KR1R2 w - - acd 14; acn 933357056; bm
>Rxf6; ce 32; id "exner.pp.14"; pm Rfe1; pv Rfe1 Nd5 Qxe5 Qxe5 Rxe5 Nxc3 bxc3 Rf2
>Rd2 Rxd2 Kxd2 Kf8 Ke3 Ke7 Ke4 Rd8 Kf5;
1. Rxf6 Rxf6 2. Rd7 Rg6 3. Qe7 Qc8 4. Qf7+ Kh8 5. e7 Qg8 6. Qxg8+ Rxg8
7. Ne4! A nice tailender that wins material. This was one of the positions I was
thinking about when I originally posted that I would include some for future
programs/hardware to solve. The other Rxf6 is also a tough one (trapped king
one).
I'm puzzled why the Qd5+ problem is difficult for some programs?
All the effort that goes into C.A.P. is paying off. It's nice to have quick
access to winning lines of positions that reoccur here from time to time.
How many positions does C.A.P. have and how large a file is it?
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.