Author: Dann Corbit
Date: 20:55:47 11/10/98
Go up one level in this thread
On November 10, 1998 at 23:35:38, Mark Young wrote:
>I was knocking out some positions with Rebel 10, when I tried what I thought
>would be a simple stock sac that Rebel 10 would find easy.
>
>[Site "?"]
>[Date "????.??.??"]
>[Round "?"]
>[White "Fischer"]
>[Black "Larsen"]
>[Result "*"]
>[Annotator "Young,M"]
>[SetUp "1"]
>[FEN "2r3k1/2r1ppb1/3p2p1/pq4Pn/1p1BP3/1B3P2/PPPQ4/1K1R3R w - - 0 1"]
>[PlyCount "0"]
>
>The Key move is RxN !, Rebel instead finds even after 10 min. BxB.
>
>All the programs I tried Junior 5, Fritz 5, Crafty 15.18, ect. ect , find RxN!
>in a matter of a few seconds.
>
>My question is
>(a) Is BxB just an alternative winning line in this position that Rebel 10
>prefers?
>
>(b) Is BxB just a misevaluation that all programs do in certain types of
>position?
Crafty likes BxB for 36 seconds, then chooses the rook move and sticks with it.
I would be interested to see the eval from Rebel. It may have discovered some
interesting line. Crafty thinks the eval went negative at ply 8 with the bishop
capture. Rebel 10 has two different modes for positional solving. Did you try
them both?
Here is a screen dump of Crafty 16.00:
play best book line after search.
EGTB access enabled
using tbpath=e:/crafty/release/tb
5 piece tablebase files found
Crafty v16.0
White(1): setboard 2r3k1/2r1ppb1/3p2p1/pq4Pn/1p1BP3/1B3P2/PPPQ4/1K1R3R w - - 0 1
White(1): go
clearing hash tables
time surplus 0.00 time limit 12:00 (12:00)
depth time score variation (1)
4-> 0.19 0.65 1. Bxg7 Kxg7 2. Rc1 a4 3. Bd5
5 0.30 0.55 1. Bxg7 Kxg7 2. Qe3 a4 3. Rd5 Rc5
5-> 0.53 0.55 1. Bxg7 Kxg7 2. Qe3 a4 3. Rd5 Rc5
6 0.89 0.50 1. Bxg7 Kxg7 2. Qe3 a4 3. Rd5 Rc5 4.
Qd4+ Kg8
6-> 1.45 0.50 1. Bxg7 Kxg7 2. Qe3 a4 3. Rd5 Rc5 4.
Qd4+ Kg8
7 2.44 0.20 1. Bxg7 Kxg7 2. Qd4+ Kg8 3. Rh2 a4
4. Bd5 a3 5. Rdh1 axb2 6. Kxb2
7-> 4.34 0.20 1. Bxg7 Kxg7 2. Qd4+ Kg8 3. Rh2 a4
4. Bd5 a3 5. Rdh1 axb2 6. Kxb2
8 6.71 -0.04 1. Bxg7 Kxg7 2. Qd4+ Kg8 3. Qd5 Qxd5
4. Rxd5 Ra7 5. Rd4 Nf4 6. Rh4 Ne2
8-> 12.46 -0.04 1. Bxg7 Kxg7 2. Qd4+ Kg8 3. Qd5 Qxd5
4. Rxd5 Ra7 5. Rd4 Nf4 6. Rh4 Ne2
9 18.32 -0.13 1. Bxg7 Kxg7 2. Qd4+ Kg8 3. Qd5 Qxd5
4. Rxd5 Ra7 5. Rb5 Nf4 6. Bd5 Rac7
9 36.45 -0.07 1. Rxh5 gxh5 2. g6 Bxd4 3. gxf7+ Kf8
4. Qxd4 Qe5 5. Qe3 a4
9-> 38.25 -0.07 1. Rxh5 gxh5 2. g6 Bxd4 3. gxf7+ Kf8
4. Qxd4 Qe5 5. Qe3 a4
10 1:04 0.23 1. Rxh5 Bxd4 2. Rhh1 Be5 3. f4 Bg7
4. Qh2 a4 5. Qh7+ Kf8 6. Rd5 Rc5
10-> 1:36 0.23 1. Rxh5 Bxd4 2. Rhh1 Be5 3. f4 Bg7
4. Qh2 a4 5. Qh7+ Kf8 6. Rd5 Rc5
11 2:18 0.09 1. Rxh5 Bxd4 2. Rh4 Be5 3. f4 Bg7 4.
Rdh1 a4 5. Qh2 Bxb2 6. Bxf7+ Kxf7 7.
Kxb2
11-> 4:05 0.09 1. Rxh5 Bxd4 2. Rh4 Be5 3. f4 Bg7 4.
Rdh1 a4 5. Qh2 Bxb2 6. Bxf7+ Kxf7 7.
Kxb2
12 5:49 0.25 1. Rxh5 Bxd4 2. Rh4 Bg7 3. Rdh1 Qe5
4. Qc1 e6 5. R4h2 Rc5 6. Rg2 <HT>
12-> 9:22 0.25 1. Rxh5 Bxd4 2. Rh4 Bg7 3. Rdh1 Qe5
4. Qc1 e6 5. R4h2 Rc5 6. Rg2 <HT>
13 9:30 1/45 1. Rxh5
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.