Author: Jeremiah Penery
Date: 00:52:29 03/11/00
Go up one level in this thread
On March 10, 2000 at 13:42:27, Eric Tom wrote:
>Hi!
>
>I'm very sorry about not posting the test positions correctly. Thank you again
>to Laurence Chen for seeing that. As I said before, I'm extremely curious if
>any chess programs out there can solve any of the following test positions. For
>those who find these positions familiar, they are indeed from famous games.
>
>Position Number One, White to move: b3
>
>Position Number One, Black to move: Rb4
>
>Position Number One, Black to move: Re6
>
>Position Number One, Black to move: Rc4
>
>1rbq1rk1/p3ppbp/3p2p1/n2P4/1p3B2/6P1/PPPQNPBP/R4RK1 w - - 0 1
>
>1r3rk1/2qnppb1/b2p2pp/p1pP4/P1P5/3B1NP1/2QBPP1P/1R3RK1 b - - 0 1
>
>3rq1k1/4rppp/2n3b1/pp2P3/2pP1QB1/P1P1R3/1B4PP/4R1K1 b - - 0 1
>
>2q2rk1/p1rn1ppp/1p2pn2/8/Q2P4/2P2N2/P2B1PPP/2RR2K1 b - - 0 1
>
>Thank you,
>Eric Tom
Here's what my modified Crafty on PII-300 finds. Each position was searched for
an hour.
1rbq1rk1/p3ppbp/3p2p1/n2P4/1p3B2/6P1/PPPQNPBP/R4RK1 w - - acd 14; acn 415965948;
acs 3621; bm b3; ce -8; id "b3"; pv c3 Nc4 Qc1 bxc3 bxc3 Bb7 Bh6 e6 dxe6 fxe6
Bxg7 Bxg2 Bxf8 Bxf1 Be7 Qxe7 Qxf1;
But when I search b3 for the same depth, this is what I get:
14-> 18:57 0.15 1. b3 Bxa1 2. Rxa1 Ba6 3. a3 Bxe2 4.
Qxe2 bxa3 5. Rxa3 Qc7 6. Be3 Ra8 7.
Qe1 Nb7 8. Bxa7 Qxc2 9. Qxe7
The score is better than for the 'best' move it found before. So for some
reason this move is getting pruned from the tree, but when searched separately
it should be found as the 'best' move.
1r3rk1/2qnppb1/b2p2pp/p1pP4/P1P5/3B1NP1/2QBPP1P/1R3RK1 b - - acd 15; acn
411960695; acs 3602; bm Rb4; ce -18; id "Rb4"; pv Rb4 Rfc1 Nb6 Bxb4 cxb4 Nd2 Rc8
Qd1 Qd7;
So this one is solved.
3rq1k1/4rppp/2n3b1/pp2P3/2pP1QB1/P1P1R3/1B4PP/4R1K1 b - - acd 14; acn 453284604;
acs 3601; bm Re6; ce -126; id "Re6"; pv Nb8 e6 f6 Bf3 Nc6 Bc1 Bh5 Be4 Bg6 d5
Bxe4 Qxe4 Ne5 Rf1;
The evaluation for Re6, at least up to 12 ply or so, is almost a pawn worse
still.
2q2rk1/p1rn1ppp/1p2pn2/8/Q2P4/2P2N2/P2B1PPP/2RR2K1 b - - acd 16; acn 510566371;
acs 3601; bm Rc4; ce 8; id "Rc4"; pv Ne4 Qc2 Nxd2 Rxd2 Nf6 Re2 Qa6 c4 Rfc8 Nd2
Qb5 f4 Qd7 Qd3 Rd8 Nf3;
16-> 11:30 -0.19 1. ... Rc4 2. Qxa7 Qc6 3. Qa3 Rc8 4.
h3 h6 5. Rc2 Nd5 6. Qb3 Nxc3 7. Bxc3
Rxc3 8. d5 exd5 9. Rxc3 Qxc3 10. Qxd5
That's the output for searching the 'correct' move to the same depth. The score
is slightly less than 1/3 of a pawn worse than Ne4. So this one is close...
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.