Author: Dieter Buerssner
Date: 09:25:41 02/06/01
Go up one level in this thread
On February 05, 2001 at 19:02:26, Arturo Ochoa wrote:
>[D]r1b5/p2k1r1p/3P2pP/1ppR4/2P2p2/2P5/P1B4P/4R1K1 w - -
>
>
>Main Line: 1. Bxg6!! hxg6 2. Re7+ Rxe7 3. dxe7+ Kxe7 4. Rd8! Kxd8 5. h7 1-0
Because of the discussion about Bxg6 vs cxb5 I made a small experiment with
Yace. I suspected, that the solution found may depend on the material value used
for the pieces. Recently, I changed piece values in Yace (smaller value for
pawns, larger for minor pieces). With more convential values (P=100, N=300,
B=305) I get on AMD K6-2 475, 20 M hash, running in the backgound:
2858741 16.276 1.48 9. 1.cxb5 Bb7 2.Rd2 Re8 3.Rxe8 Kxe8 4.Bb3 Rd7
5.Be6 Rd8 {100}
3816620 22.534 1.50 10t 1.cxb5 Bb7 2.Rd2 Re8 3.Rxe8 Kxe8 4.Bb3 Rf6
5.Re2+ Kd8 6.Re7 Kc8 {100}
8092332 47.830 1.50 10. 1.cxb5 Bb7 2.Rd2 Re8 3.Rxe8 Kxe8 4.Bb3 Rf6
5.Re2+ Kd8 6.Re7 Kc8 {100}
10466626 1:03.3 1.60 11t 1.cxb5 Bb7 2.Rd2 Re8 3.Rxe8 Kxe8 4.Bb3 Rd7 5.h4
Kf8 6.Be6 Rd8 {HT} {100}
20133152 1:58.2 1.60 11. 1.cxb5 Bb7 2.Rd2 Re8 3.Rxe8 Kxe8 4.Bb3 Rd7 5.h4
Kf8 6.Be6 Rd8 {HT} {100}
28775209 2:58.4 1.65 12t 1.cxb5 Bb7 2.Rdd1 f3 3.Kf2 Re8 4.Rxe8 Kxe8
5.Bb3 Rf4 6.d7+ Kd8 7.Bg8 Ra4 {100}
31370961 3:15.3 1.66 12t+ 1.Bxg6 hxg6 2.Re7+ Rxe7 3.dxe7+ Kxe7 4.Rd8 Kxd8
5.h7 bxc4 6.h8=Q+ Kc7 7.Qe5+ Kb6 8.Qxf4 Bb7
9.Qf6+ Ka5 10.Qxg6 Rh8
33517822 3:27.9 1.69 12t 1.Bxg6 hxg6 2.Re7+ Rxe7 3.dxe7+ Kxe7 4.Rd8 Kxd8
5.h7 bxc4 6.h8=Q+ Kd7 7.Qh3+ Ke7 8.Qh4+ Ke6
9.Qxf4 Bb7 10.Qe3+ Kf6 11.Qxc5 Ba6 {196}
55338977 5:39.3 1.69 12. 1.Bxg6 hxg6 2.Re7+ Rxe7 3.dxe7+ Kxe7 4.Rd8 Kxd8
5.h7 bxc4 6.h8=Q+ Kd7 7.Qh3+ Ke7 8.Qh4+ Ke6
9.Qxf4 Bb7 10.Qe3+ Kf6 11.Qxc5 Ba6 {196}
67234675 6:49.6 1.83 13t 1.Bxg6 hxg6 2.Re7+ Rxe7 3.dxe7+ Kxe7 4.Rd8 Kxd8
5.h7 bxc4 6.h8=Q+ Kd7 7.Qh3+ Kc7 8.Qf3 Bb7
9.Qxf4+ Kb6 10.Qd6+ {HT} {96}
128058722 12:51.8 1.83 13. 1.Bxg6 hxg6 2.Re7+ Rxe7 3.dxe7+ Kxe7 4.Rd8 Kxd8
5.h7 bxc4 6.h8=Q+ Kd7 7.Qh3+ Kc7 8.Qf3 Bb7
9.Qxf4+ Kb6 10.Qd6+ {HT} {96}
So, after little bit more than 3 minutes, I see the mainline above. Of course,
I could have stopped here - position solved ...
But, at the next ply, I get a fail high for cxb5
158323250 15:54.3 1.78 14t 1.Bxg6 hxg6 2.Re7+ Rxe7 3.dxe7+ Kxe7 4.Rd8 Kxd8
5.h7 bxc4 6.h8=Q+ Kd7 7.Qh3+ Kc7 8.Qf3 Bb7
9.Qxf4+ Kb6 10.Qd6+ Bc6 11.Qxg6 Rh8 12.Qb1+ Bb5
13.a4 {196}
195159864 22:05.7 1.79 14t+ 1.cxb5 Bb7 2.Rdd1 f3 3.Kf2 Re8 4.Rxe8 Kxe8
5.Bb3 Rf4 6.Re1+ Kd7 7.Re7+ Kc8 8.Rxh7 Rf6
9.d7+ Kd8 {200}
212716679 25:36.1 2.23 14++ 1.cxb5 Bb7 2.Rdd1 f3 3.Kf2 Re8 4.Rxe8 Kxe8
5.Bb3 c4 6.Bxc4 Bd5 7.Bxd5 Kf8 {505}
262196324 32:30.9 2.60 14t 1.cxb5 Bb7 2.Rdd1 f3 3.Kf2 c4 4.Ba4 Rc8 5.b6+
Bc6 6.Re7+ Rxe7 7.dxe7+ Kxe7 8.b7 Bxb7 9.Rd7+
Ke6 10.Rxb7 Rf8 11.Rxh7 {HT} {305}
Now, with the smaller pawn value, and higher values for minor pieces:
1877059 10.887 1.28 9t 1.cxb5 Bb7 2.Rd2 Re8 3.Rxe8 Kxe8 4.Bb3 Rd7
5.Be6 Rd8 {80}
3161789 17.985 1.28 9. 1.cxb5 Bb7 2.Rd2 Re8 3.Rxe8 Kxe8 4.Bb3 Rd7
5.Be6 Rd8 {80}
4054205 23.614 1.30 10t 1.cxb5 Bb7 2.Rd2 Re8 3.Rxe8 Kxe8 4.Bb3 Rf6
5.Re2+ Kd8 6.Re7 Kc8 {80}
7986234 46.316 1.30 10. 1.cxb5 Bb7 2.Rd2 Re8 3.Rxe8 Kxe8 4.Bb3 Rf6
5.Re2+ Kd8 6.Re7 Kc8 {80}
10059038 59.433 1.37 11t 1.cxb5 Bb7 2.Rd2 Re8 3.Rxe8 Kxe8 4.Bb3 Rd7
5.Be6 Rd8 6.h4 Be4 {80}
23755034 2:22.3 1.37 11. 1.cxb5 Bb7 2.Rd2 Re8 3.Rxe8 Kxe8 4.Bb3 Rd7
5.Be6 Rd8 6.h4 Be4 {80}
35672144 3:42.0 1.45 12t 1.cxb5 Bb7 2.Rdd1 f3 3.Kf2 Re8 4.Rxe8 Kxe8
5.Bb3 Rf4 6.Bg8 Rh4 7.d7+ Kd8 {80}
66811397 6:55.3 1.45 12. 1.cxb5 Bb7 2.Rdd1 f3 3.Kf2 Re8 4.Rxe8 Kxe8
5.Bb3 Rf4 6.Bg8 Rh4 7.d7+ Kd8 {80}
82448430 8:42.6 1.82 13t 1.cxb5 Bb7 2.Rdd1 f3 3.Kf2 Rh8 4.Bb3 Rf4 5.Re7+
Kc8 6.d7+ Kd8 7.Rde1 Rff8 8.Bf7 g5 {80}
143677553 14:50.9 1.82 13. 1.cxb5 Bb7 2.Rdd1 f3 3.Kf2 Rh8 4.Bb3 Rf4 5.Re7+
Kc8 6.d7+ Kd8 7.Rde1 Rff8 8.Bf7 g5 {80}
188971890 20:06.4 2.22 14++ 1.cxb5 Bb7 2.Rdd1 f3 3.Kf2 Rh8 4.Bb3 Rf4 5.Re7+
Kc8 6.Rde1 Kd8 7.Rxb7 Rff8 {430}
265476539 28:28.1 2.42 14t 1.cxb5 Bb7 2.Rdd1 Rd8 3.Re7+ Rxe7 4.dxe7+ Kxe7
5.Rxd8 Kxd8 6.Bxg6 Bd5 7.Bxh7 Ke7 8.Bf5 Kf6
9.h7 Kg7 10.a3 Kh8 {HT} {160}
BTW. These piece values do worse in many tactical test suite positions, where
the line often starts with saccing a piece. But I have some indication, that
they will do better in "real" games.
Regards,
Dieter
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.