Author: Eelco de Groot
Date: 18:10:25 05/16/05
Go up one level in this thread
Gentlemen, from myself, thanks very much for all the responses! It was very
interesting.
It seems that the strong point of 36.Kh1 is that the Black Queen has to move on
its own to d8 anyway, if not forced by 36.c5. But now the black d-pawn is in a
pin, and if White's next move is 37.c4-c5, Black can't respond with 37..dxc5
(would follow 38.Nxc6) Maybe a direct 37.Qg3 is strongest.
Gandalf 6.1 analysis after 36.Kh1 Qd8 37.c5:
00:00:02.1 -0,12 8 818307 Bd7 cxd6 Bh4 Qf4 gxh6 Qxh6 Ng6 Bxg6 fxg6 Qxg6+ Kh8
Qh5+ Kg8 Qf7+ Kh8 Nf5
00:00:02.6 -0,12 8 1069652 d5 Re2
00:00:04.3 0,03 8 1833393 d5 Rdd1 Bg5 Bxg5 Qxg5 h4 Qg4 Bf5 Qh5
00:00:06.1 0,12 9 2662707 d5 Rdd1 Bg5 Bxg5 Qxg5 h4 Qh5 Qg3 Qg4 Qf2
00:00:14.1 0,00 10 6420656 d5 Re2 Ng6 Rxe8+ Qxe8 Bd2 Be5 Bf5 Bb5 Nxb5 Qxb5 Qf3
00:00:25.8 0,06 11 12058030 d5 Re2 Ng6 Re3 Rxe3 Bxe3 Qc7 Bxg6 fxg6 Bf4 Qe7 Bd6
Qe4
00:00:43.3 -0,25 12 20714098 d5 Re2 Ng6 Re3 Rxe3 Bxe3 Qc7 Bxg6 fxg6 Bf4 Qe7 Bd6
Qe4
00:00:48.0 -0,57 12 22947850 d5 Re2 Ng6 Re3 Rxe3 Bxe3 Qc7 Bxg6 fxg6 Bf4 Qe7 Bd6
Qe4
00:01:13.2 -0,89 12 36132534 d5 Qg3 g6 Re2 Bb5 Nxb5 axb5 Rd1 d4 Re4 d3 Bxd3 Nxd3
Rxe8+ Qxe8 Qxd3
00:01:23.0 -0,89 12 41001604 Bd7 Nf5
00:02:07.1 -0,48 12 60402412 Bd7 Nf5 gxh6 Nxh6+ Kf8 Qxf6 Qxf6 Rxf6 Be6 Ba4 Kg7
Rf4 Kxh6 Bxe8 Kg5 Rdd4 Rxe8 cxd6
00:03:04.9 -0,48 13 88496631 Bd7 Nf5 gxh6 Nxh6+ Kf8 Bb3 dxc5 Bxf7 Bh4 Qxc5+ Kg7
Bxe8 Qxe8 Qe3 Be6 Re2 Bf6 Nf5+ Bxf5 Rxf5
00:03:58.6 -0,80 14 112586234 Bd7 Nf5 gxh6 Nxh6+ Kf8 Bb3 dxc5 Bxf7 Bh4 Qxc5+ Kg7
Bxe8 Qxe8 Qe3 Be6 Re2 Bf6 Nf5+ Bxf5 Rxf5
00:07:16.8 -1,12 14 210953588 Bd7 Nf5 gxh6 Nxh6+ Kf8 Bb3 dxc5 Bxf7 Bh4 Qxc5+ Kg7
Bxe8 Qxe8 Qe3 Be6 Re2 Bf6 Nf5+ Bxf5 Rxf5
00:08:53.1 -1,37 14 258031975 Bd7 Nf5 dxc5 Qg3 g6 Bf4 Kh7 Nd6 Bh4 Qh2 Nc6 Nxe8
Qxe8 bxc5 Rd8
00:09:42.8 -1,37 14 284268813 d5 Qg3
00:10:39.3 -0,89 14 313705413 d5 Qg3 g6 Re2 Bb5 Nxb5 axb5 Rd1 Bh4 Qf4 Qf6 g3
Qxf4 Bxf4 Bf6 Rxd5
00:16:59.3 -1,12 15 506979193 d5 Qf5 Ng6 Bxg7 Kxg7 Rdf2 Bb5 Nxb5 axb5 Qxf6+ Qxf6
Rxf6 Ne5 Bb3 d4 R6f4 d3 Bxf7 Red8
00:25:43.6 -1,05 16 780920537 d5 Qf5 Ng6 Bxg7 Kxg7 Rdf2 Bb5 Nxb5 axb5 Qxf6+ Qxf6
Rxf6 Ne5 Rd6 Nc6 Rd7 Re7 Rxe7 Nxe7 Kh2 Nc6
After 37..Bd7 38.Nf5
[D]r2qr1k1/1p1b1pp1/p2p1b1B/2P1nN2/1P6/7P/2BR1QP1/5R1K b - - 2 1
00:00:01.6 -1,79 7 618110 dxc5 Qg3+
00:00:02.1 -1,05 7 802691 dxc5 Qg3 g6 Nd6 Bh4 Qf4 g5
00:00:03.6 -0,54 8 1399720 dxc5 Qg3 g6 Nd6 Bh4 Qf4 g5 Qh2
00:00:05.7 -0,83 9 2480219 dxc5 Qg3 g6 Nd6 Bh4 Qf4 g5 Qe4 f5 Qd5+ Kh7 Nxe8 Qxe8
Bxf5+ Kxh6 Bxd7 Nxd7 Qxd7 Qxd7 Rxd7 cxb4 Rxb7
00:00:15.0 -0,83 10 7090016 dxc5 Qg3 g6 Nd6 Bh4 Qf4 g5 Qe4 f5 Qd5+ Kh7 Nxe8 Qxe8
Bxf5+ Kxh6 Bxd7 Nxd7 Qxd7 Qxd7
00:00:20.9 -0,83 10 9917261 gxh6 Nxh6
00:00:27.7 -0,48 10 13157397 gxh6 Nxh6+ Kf8 Qxf6 Qxf6 Rxf6 Be6 Ba4 Kg7 Rf4 Kxh6
Bxe8 Kg5 Rdd4 Rxe8 cxd6
00:00:52.2 -0,48 11 24780913 gxh6 Nxh6+ Kf8 Bb3 dxc5 Bxf7 Bh4 Qxc5+ Kg7 Bxe8
Qxe8 Qe3 Be6 Re2 Bf6 Nf5+ Bxf5 Rxf5
00:01:02.6 -0,80 12 29178540 gxh6 Nxh6+ Kf8 Bb3 dxc5 Bxf7 Bh4 Qxc5+ Kg7 Bxe8
Qxe8 Qe3 Be6 Re2 Bf6 Nf5+ Bxf5 Rxf5
00:01:12.3 -1,12 12 33480440 gxh6 Nxh6+ Kf8 Bb3 dxc5 Bxf7 Bh4 Qxc5+ Kg7 Bxe8
Qxe8 Qe3 Be6 Re2 Bf6 Nf5+ Bxf5 Rxf5
00:01:59.3 -1,75 12 58588052 dxc5 Qg3+
00:03:23.2 -1,37 12 98695188 dxc5 Qg3 g6 Bf4 Kh7 Nd6 Bh4 Qh2 Nc6 Nxe8 Qxe8 bxc5
Rd8
00:04:30.7 -1,37 12 132993464 Re6 Qg3
00:04:48.9 -1,37 12 141644047 dxc5 Qg3 g6 Bf4 Kh7 Nd6 Bh4 Qh2 Nc6 Nxe8 Qxe8 bxc5
Rd8
00:06:36.5 -1,50 13 197847046 dxc5 Qg3 g6 Bf4 Kh7 Nd6 Bh4 Qh2 Qf6 Nxe8 Rxe8 bxc5
Bg5 Rd6 Bxf4 Rxf4
00:07:24.4 -1,50 13 224054286 Re6 Qg3
00:08:23.2 -1,24 13 257555451 Re6 Qg3 Ng6 Nxd6 Bh4 Qf3 Ne5 Qh5 Rxh6 Qxe5 Rf6
Rfd1 Bc6
00:11:13.4 -1,24 14 342087820 Re6 Qg3 Ng6 Nxd6 Bh4 Qf3 Ne5 Qh5 Rxh6 Qxe5 Rf6
Rxf6 Qxf6 Qxf6 Bxf6 Ne4 Bc6 Nxf6+ gxf6
00:25:29.7 -0,89 15 781171926 Re6 Qg3
00:33:46.0 -0,86 15 1062959083 Re6 Qg3 Ng6 Nxd6 Qe7 Bxg6 fxg6 Bf4 Bh4 Qf3 Re1
Rd1 Rxd1 Rxd1 Rf8 Qd5+ Qe6 Qd4
Meanwhile the experiments with the Pro deo personalities have continued. The
settings for
[Pruning = MISC_58] * Recalculate passed pawns
[M58_Table = normal] * normal|midg|endg|strong
are out again for the moment. Not that I found they are very bad, but I wasn't
looking at passed pawns anyway so there was not much use in trying if they have
a beneficial effect overall.
My latest try seems a bit explosive at time, with its higher [Attacking] values.
As it has to learn that chess has a certain logic to it, and you cannot always
attack when you feel you have a superior position, I'm calling this the Vulcan
settings.. Their common characteristic is that they all have the [Pruning =
MISC_38] * New capture extension. They may not be awfully good but we'll see.
I'm writing all this down for myself too, as the personalities folder keeps
growing and growing it is easy lose track of what is what.
Expect to hear more from Pro Deo 1.1 Vulcan {9i6g}!
Eelco
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.