This one seems to be pretty tough for all of the engines I have, the best move here is what Morozevich played 24.Ba6! perhaps someone else has an engine that can find it.
[d] 2k2rr1/ppqb4/2n3np/2PpP1p1/6R1/P2B1N2/2PBQPP1/1R2K3 w - - 0 1
Test Position: Morozevich vs Pelletier, Biel 2003
Moderators: hgm, Rebel, chrisw
-
acase
- Posts: 981
- Joined: Wed Mar 15, 2006 1:14 am
- Location: Columbus, Ohio USA
- Full name: Andrew R Case
-
rightrook
- Posts: 1452
- Joined: Wed Mar 08, 2006 8:45 pm
Re: Test Position: Morozevich vs Pelletier, Biel 2003
Hi....That looks like another Paul Morphy move....does not have a forced win...so probably not a good test move.
hiarcs 11.2 plays....e6..
regards
Robert
hiarcs 11.2 plays....e6..
regards
Robert
-
rlsuth
- Posts: 322
- Joined: Wed Mar 08, 2006 9:37 pm
Re: Test Position: Morozevich vs Pelletier, Biel 2003
Rybka 3 plays Ba6 after about a minute on my PC.
-
acase
- Posts: 981
- Joined: Wed Mar 15, 2006 1:14 am
- Location: Columbus, Ohio USA
- Full name: Andrew R Case
Re: Test Position: Morozevich vs Pelletier, Biel 2003
rlsuth wrote:Rybka 3 plays Ba6 after about a minute on my PC.
Rybka 3 does find this one, ( I thought I had tried Rybka 3...it appears I'm mistaken ) however it takes MUCH longer than one minute on my quad. Perhaps you are using some "special settings"?
New game
2k2rr1/ppqb4/2n3np/2PpP1p1/6R1/P2B1N2/2PBQPP1/1R2K3 w - - 0 1
Analysis by Rybka 3:
1.e6 Be8 2.Kf1
+- (1.69) Depth: 6 00:00:00 6kN
1.e6 Re8 2.Bf5 Kb8 3.Kf1 Nd8 4.Rgb4 Bxe6 5.Bxe6 Rxe6 6.Rxb7+ Qxb7 7.Rxb7+ Kxb7 8.Qd3
+/- (1.38) Depth: 7 00:00:00 158kN
1.e6 Re8 2.Bf5 Rxe6 3.Qxe6 Bxe6 4.Bxe6+ Kb8 5.Bxg8 Qc8 6.Nd4 Nxd4
+/- (0.73) Depth: 8 00:00:01 274kN
1.e6 Re8 2.Bf5 Rxe6 3.Qxe6 Bxe6 4.Bxe6+ Kb8 5.Bxg8 Qc8 6.Nd4 Nxd4 7.Rxd4 Qxg8 8.Rdb4 Qe6+ 9.Be3 Qa6 10.f3
+/- (0.71) Depth: 9 00:00:01 322kN
1.e6 Re8 2.Bf5 Rxe6 3.Qxe6 Bxe6 4.Bxe6+ Kb8 5.Bxg8 Qc8 6.Nd4 Nxd4 7.Rxd4 Qxg8 8.Rdb4 Qe6+ 9.Be3 Qa6 10.Kd2 Ne5
+/- (0.71) Depth: 10 00:00:01 398kN
1.e6 Re8 2.Bf5 Rxe6 3.Qxe6 Bxe6 4.Bxe6+ Kb8 5.Bxg8 Qc8 6.Nd4 Nxd4 7.Rxd4 Qxg8 8.Rdb4 Qe6+ 9.Be3 Qa6 10.Kd2 Ne5
+/= (0.70) Depth: 11 00:00:02 543kN
1.e6 Re8 2.Bf5 Rxe6 3.Qxe6 Bxe6 4.Bxe6+ Kb8 5.Bxg8 Qc8 6.Nd4 Nxd4 7.Rxd4 Qxg8 8.Rdb4 Qe6+ 9.Be3 Qa6 10.Kd2 Ne5 11.Rxb7+ Kc8 12.Rb8+ Kc7
+/= (0.70) Depth: 12 00:00:03 897kN
1.e6 Re8 2.Bf5 Rxe6 3.Qxe6 Bxe6 4.Bxe6+ Kb8[] 5.Bxg8 Qc8 6.Nd4 Nxd4 7.Rxd4 Qxg8 8.Rdb4 Qe6+ 9.Be3 Qa6 10.Kd2 Kc8 11.Rxb7 Ne5 12.Rb8+
+/= (0.70) Depth: 13 00:00:07 2147kN
1.e6 Re8 2.Bf5 Rxe6 3.Qxe6 Bxe6 4.Bxe6+ Kb8[] 5.Bxg8 Qc8 6.Nd4 Nxd4 7.Rxd4 Qxg8 8.Rdb4 Qe6+ 9.Be3 Qe5 10.Kd2 Kc8 11.g3 a5 12.f4 gxf4 13.gxf4 Qh5 14.Rxb7 Qh2+
+/= (0.68) Depth: 14 00:00:11 3190kN
1.e6 Re8 2.Bf5 Rxe6 3.Qxe6[] Bxe6[] 4.Bxe6+ Kb8[] 5.Bxg8[] Qc8 6.Nd4 Nxd4 7.Rxd4 Qxg8 8.Rdb4 Qe6+ 9.Be3 Qe5 10.Kd2 Kc8 11.g3 Qf6 12.Rxb7 d4 13.Rb8+ Kd7 14.R1b7+ Ke6[] 15.Re8+ Kf5[] 16.g4+
+/= (0.57) Depth: 15 00:00:20 5755kN
1.e6 Re8 2.Bf5 Rxe6 3.Qxe6 Bxe6 4.Bxe6+ Kb8[] 5.Bxg8[] Qc8 6.Nd4 Nxd4 7.Rxd4 Qxg8 8.Rdb4 Qe6+ 9.Be3 Qe5 10.Kd2 Kc8 11.g3 Qf6 12.Rxb7 d4 13.Rb8+ Kd7 14.R1b7+ Ke6[] 15.Re8+ Kf5[] 16.g4+
+/= (0.57) Depth: 16 00:00:34 9616kN
1.e6 Re8 2.Bf5 Rxe6 3.Qxe6 Bxe6 4.Bxe6+ Kb8[] 5.Bxg8[] Qc8 6.Nd4 Nxd4 7.Rxd4 Qxg8 8.Rdb4 Qe6+ 9.Kd1 Qa6 10.Be3 Nh4 11.Rxb7+ Qxb7 12.Rxb7+[] Kxb7 13.g4 Kc6[] 14.Ke2 Ng6 15.Bd4 Nf4+
+/= (0.50) Depth: 17 00:01:06 19666kN
1.e6 Re8 2.Bf5 Rxe6[] 3.Qxe6 Bxe6 4.Bxe6+ Kb8[] 5.Bxg8[] Qc8 6.Nd4 Nxd4 7.Rxd4 Qxg8 8.Rdb4 Qe6+ 9.Kd1 Qa6 10.Be3 Nh4 11.Rxb7+ Qxb7 12.Rxb7+ Kxb7 13.g4 Kc6[] 14.Ke2 Ng6 15.Bd4 Nf4+ 16.Ke3
+/= (0.52) Depth: 18 00:02:13 39759kN
1.e6 Re8 2.Bf5 Rxe6 3.Qxe6[] Bxe6[] 4.Bxe6+ Kb8[] 5.Bxg8[] Qc8 6.Nd4 Nxd4 7.Rxd4 Qxg8 8.Rdb4 Qe6+ 9.Kd1 Qa6 10.Be3 Nh4 11.Rxb7+ Qxb7 12.Rxb7+ Kxb7 13.g4 Kc6[] 14.Ke2 Ng6 15.Bd4 Nf4+ 16.Ke3
+/= (0.46) Depth: 19 00:04:54 89961kN
1.Ba6
+/- (0.72 !) Depth: 19 00:14:07 219mN
1.Ba6 Bxg4 2.Rxb7[] Bxf3 3.Rxc7+[] Kxc7 4.gxf3[] Ngxe5 5.Ba5+ Kd7 6.Kf1 Rg6 7.Bc3 Re6 8.Qd1 Ke7 9.Bxe5 Rxe5 10.Bb7 Rf6 11.Qd3 Kf8 12.Qh7 Nd4 13.Qh8+ Kf7 14.f4 gxf4
+/- (0.86) Depth: 19 00:22:04 332mN
1.Ba6 Bxg4 2.Rxb7[] Bxf3 3.Rxc7+[] Kxc7 4.gxf3[] Ngxe5 5.Ba5+ Kd7 6.Kf1 Rg6 7.Bc3 Re6 8.Qd1 Ke7 9.Bxe5 Rxe5 10.Bb7 Rf6 11.Qd3 Kf8 12.Qh7 Ree6 13.Bxc6 Rxc6 14.Qxa7 Ra6 15.Qb8+ Kg7 16.Qb7+
+/- (0.94) Depth: 20 00:36:07 532mN
-
rlsuth
- Posts: 322
- Joined: Wed Mar 08, 2006 9:37 pm
Re: Test Position: Morozevich vs Pelletier, Biel 2003
I'm using an 8 core PC if that helps?
-
Eelco de Groot
- Posts: 4561
- Joined: Sun Mar 12, 2006 2:40 am
- Full name:
Re: Test Position: Morozevich vs Pelletier, Biel 2003
The later builds that have the amputated qsearches can't find it, but build 65 is a bit better, it can find this. But not as fast as Rybka 3 I think on a quad. I'm not totally sure I could rebuilt it, I do hope so but anyway it's not finished yet. If I can find back the right search_bld065.cpp, I will try to combine that with the evaluation_bld074.cpp that computes King mobility again in King Safety and see if I can beat the 14 minutes from your quad Richard 
Only need to shave off three minutes of the time, but probably it will have too slow nodes per second to be competitive.
[D]2k2rr1/ppqb4/2n3np/2PpP1p1/6R1/P2B1N2/2PBQPP1/1R2K3 w - -
Engine: Rainbow Serpent 1.6.3s(dc) Build 65 (Athlon 2009 MHz, 128 MB)
by Tord Romstad, Marco Costalba, Joona Kiiski Modifications: Dann Corbit
1.00 0:00 +2.26 1.e6 (1.009) 9
2.00 0:00 +2.62 1.e6 Be8 (1.169) 10
3.00 0:00 +3.03 1.e6 Be8 2.Kf1 (1.628) 14
4.00 0:00 +2.22 1.e6 Be8 2.Kf1 Nf4 (2.714) 21
5.00 0:00 +2.02 1.e6 Be8 2.Bc3 Nf4 3.Qe3 (9.300) 66
6.00 0:00 +2.06 1.e6 Be8 2.Kf1 Nge5 3.Nxe5 Qxe5
4.Qxe5 Nxe5 5.Rd4 Rf6 6.Rdb4 Nxd3
7.cxd3 (15.825) 113
7.00 0:00 +2.02 1.e6 Be8 2.Kf1 Nge5 3.Nxe5 Qxe5
4.Qxe5 Nxe5 5.Rd4 Rf6 6.Rdb4 Nxd3
7.cxd3 Bg6 8.R1b3 Rgf8 (45.524) 207
8.00 0:00 +1.93 1.e6 Be8 2.Kf1 Nge5 3.Nxe5 Qxe5
4.Qxe5 Nxe5 5.Rd4 Rf6 6.Re1 Nxd3
7.cxd3 Rgg6 8.e7 Rf5 9.f3 (76.979) 259
9.00 0:00 +2.14++ 1.e6 Be8 2.Kf1 Nge5 3.Nxe5 Qxe5
4.Qxe5 Nxe5 5.Rd4 Rf6 6.Re1 Nxd3
7.cxd3 Rgg6 8.e7 Rf5 9.g4 (109.528) 305
10.00 0:01 +1.97 1.e6 Re8 2.Bf5 Kb8 3.Nd4 Nxd4 4.Rxd4 Qh2
5.Rdb4 Bc6 6.Bc3 d4 7.Bxg6 Rxg6 (443.298) 411
11.01 0:12 +1.25 1.e6 Re8 2.Bf5 Kb8 3.Kf1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Qxg6 Bxg4 7.Qxh6 Rg8
8.Bxg5 Bxf3 9.gxf3 Ne5 10.f4 Nc4
11.Qe6 (5.877.182) 473
12.01 0:18 +1.17 1.e6 Re8 2.Bf5 Kb8 3.Kf1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Qxg6 Bxg4 7.Qxh6 Rg8
8.Bxg5 Bxf3 9.gxf3 Ka8 10.f4 Qc8
11.Qh5 Rf8 12.Rd1 Qf5 13.c3 (8.719.928) 483
13.01 0:44 +0.88 1.e6 Re8 2.Bf5 Kb8 3.Kf1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Qxg6 Bxg4 7.Qxh6 Rg8
8.Bxg5 Bxf3 9.gxf3 Ka8 10.Rd1 Qf7
11.f4 Qf5 12.Rd2 d4 13.Re2 Qxc5
14.Qh5 (22.460.984) 499
14.01 2:11 +0.40 1.e6 Re8 2.Bf5 Kb8 3.Kd1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Rgb4 Nxb4 7.Qxg6 Bf5
8.Qxh6 Qxc5 9.Rxb4 Qxc2+ 10.Ke1 Re8+
11.Kf1 Qd1+ 12.Be1 Bd3+ 13.Kg1 Rxe1+
14.Nxe1 (65.620.347) 499
15.01 3:31 +1.01 1.e6 Re8 2.Bf5 Kb8 3.Kd1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Rgb4 Nxb4 7.Qxg6 Nxc2
8.c6 Qxc6 9.Qxc6 Bxc6 10.Kxc2 d4
11.Kd3 Bxf3 12.gxf3 Rxf3+ 13.Kxd4 Rxf2
14.Bb4 (105.268.041) 498
16.01 7:49 +0.64 1.e6 Re8 2.Bf5 Kb8 3.Kd1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Bf4 gxf4 7.Rxg6 Bf5
8.Rxc6 Qxc6 9.Nd4 Bxd3 10.Nxc6+ Kc7
11.Nxa7 Bc4 12.Nb5+ Bxb5 13.Rxb5 Rg8
14.Ke2 (230.691.830) 491
17.01 13:15 +0.64 1.e6 Re8 2.Bf5 Kb8 3.Kd1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Bf4 gxf4 7.Rxg6 Bf5
8.Rxc6 Qxc6 9.Nd4 Bxd3 10.Nxc6+ Kc7
11.Nxa7 Bc4 12.Nb5+ Bxb5 13.Rxb5 Rg8
14.Ke2 (388.919.637) 488
17.06 17:29 +0.84 1.Ba6 Bxg4 2.Rxb7 Bxf3 3.Rxc7+ Kxc7
4.gxf3 Ngxe5 5.Kf1 Rb8 6.Ba5+ Kd7
7.Be1 Rg6 8.Qd1 Ke6 9.Bc3 d4 10.Bd2 Kf7
11.f4 Ne7 12.Be2 (514.064.568) 489
best move: Bd3-a6 time: 18:47.672 min n/s: 491.533 nodes: 554.280.148
Only need to shave off three minutes of the time, but probably it will have too slow nodes per second to be competitive.[D]2k2rr1/ppqb4/2n3np/2PpP1p1/6R1/P2B1N2/2PBQPP1/1R2K3 w - -
Engine: Rainbow Serpent 1.6.3s(dc) Build 65 (Athlon 2009 MHz, 128 MB)
by Tord Romstad, Marco Costalba, Joona Kiiski Modifications: Dann Corbit
1.00 0:00 +2.26 1.e6 (1.009) 9
2.00 0:00 +2.62 1.e6 Be8 (1.169) 10
3.00 0:00 +3.03 1.e6 Be8 2.Kf1 (1.628) 14
4.00 0:00 +2.22 1.e6 Be8 2.Kf1 Nf4 (2.714) 21
5.00 0:00 +2.02 1.e6 Be8 2.Bc3 Nf4 3.Qe3 (9.300) 66
6.00 0:00 +2.06 1.e6 Be8 2.Kf1 Nge5 3.Nxe5 Qxe5
4.Qxe5 Nxe5 5.Rd4 Rf6 6.Rdb4 Nxd3
7.cxd3 (15.825) 113
7.00 0:00 +2.02 1.e6 Be8 2.Kf1 Nge5 3.Nxe5 Qxe5
4.Qxe5 Nxe5 5.Rd4 Rf6 6.Rdb4 Nxd3
7.cxd3 Bg6 8.R1b3 Rgf8 (45.524) 207
8.00 0:00 +1.93 1.e6 Be8 2.Kf1 Nge5 3.Nxe5 Qxe5
4.Qxe5 Nxe5 5.Rd4 Rf6 6.Re1 Nxd3
7.cxd3 Rgg6 8.e7 Rf5 9.f3 (76.979) 259
9.00 0:00 +2.14++ 1.e6 Be8 2.Kf1 Nge5 3.Nxe5 Qxe5
4.Qxe5 Nxe5 5.Rd4 Rf6 6.Re1 Nxd3
7.cxd3 Rgg6 8.e7 Rf5 9.g4 (109.528) 305
10.00 0:01 +1.97 1.e6 Re8 2.Bf5 Kb8 3.Nd4 Nxd4 4.Rxd4 Qh2
5.Rdb4 Bc6 6.Bc3 d4 7.Bxg6 Rxg6 (443.298) 411
11.01 0:12 +1.25 1.e6 Re8 2.Bf5 Kb8 3.Kf1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Qxg6 Bxg4 7.Qxh6 Rg8
8.Bxg5 Bxf3 9.gxf3 Ne5 10.f4 Nc4
11.Qe6 (5.877.182) 473
12.01 0:18 +1.17 1.e6 Re8 2.Bf5 Kb8 3.Kf1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Qxg6 Bxg4 7.Qxh6 Rg8
8.Bxg5 Bxf3 9.gxf3 Ka8 10.f4 Qc8
11.Qh5 Rf8 12.Rd1 Qf5 13.c3 (8.719.928) 483
13.01 0:44 +0.88 1.e6 Re8 2.Bf5 Kb8 3.Kf1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Qxg6 Bxg4 7.Qxh6 Rg8
8.Bxg5 Bxf3 9.gxf3 Ka8 10.Rd1 Qf7
11.f4 Qf5 12.Rd2 d4 13.Re2 Qxc5
14.Qh5 (22.460.984) 499
14.01 2:11 +0.40 1.e6 Re8 2.Bf5 Kb8 3.Kd1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Rgb4 Nxb4 7.Qxg6 Bf5
8.Qxh6 Qxc5 9.Rxb4 Qxc2+ 10.Ke1 Re8+
11.Kf1 Qd1+ 12.Be1 Bd3+ 13.Kg1 Rxe1+
14.Nxe1 (65.620.347) 499
15.01 3:31 +1.01 1.e6 Re8 2.Bf5 Kb8 3.Kd1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Rgb4 Nxb4 7.Qxg6 Nxc2
8.c6 Qxc6 9.Qxc6 Bxc6 10.Kxc2 d4
11.Kd3 Bxf3 12.gxf3 Rxf3+ 13.Kxd4 Rxf2
14.Bb4 (105.268.041) 498
16.01 7:49 +0.64 1.e6 Re8 2.Bf5 Kb8 3.Kd1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Bf4 gxf4 7.Rxg6 Bf5
8.Rxc6 Qxc6 9.Nd4 Bxd3 10.Nxc6+ Kc7
11.Nxa7 Bc4 12.Nb5+ Bxb5 13.Rxb5 Rg8
14.Ke2 (230.691.830) 491
17.01 13:15 +0.64 1.e6 Re8 2.Bf5 Kb8 3.Kd1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Bf4 gxf4 7.Rxg6 Bf5
8.Rxc6 Qxc6 9.Nd4 Bxd3 10.Nxc6+ Kc7
11.Nxa7 Bc4 12.Nb5+ Bxb5 13.Rxb5 Rg8
14.Ke2 (388.919.637) 488
17.06 17:29 +0.84 1.Ba6 Bxg4 2.Rxb7 Bxf3 3.Rxc7+ Kxc7
4.gxf3 Ngxe5 5.Kf1 Rb8 6.Ba5+ Kd7
7.Be1 Rg6 8.Qd1 Ke6 9.Bc3 d4 10.Bd2 Kf7
11.f4 Ne7 12.Be2 (514.064.568) 489
best move: Bd3-a6 time: 18:47.672 min n/s: 491.533 nodes: 554.280.148
Debugging is twice as hard as writing the code in the first
place. Therefore, if you write the code as cleverly as possible, you
are, by definition, not smart enough to debug it.
-- Brian W. Kernighan
place. Therefore, if you write the code as cleverly as possible, you
are, by definition, not smart enough to debug it.
-- Brian W. Kernighan
-
Eelco de Groot
- Posts: 4561
- Joined: Sun Mar 12, 2006 2:40 am
- Full name:
Re: Test Position: Morozevich vs Pelletier, Biel 2003
It seemed possible to do this, but the evaluation path was not so successful. To do this right I think you have to have a very good material imbalance evaluator and those tables/polynomials are I think beyond any normal human tuning, only the Stockfish Server ® or Diepeveen's neural networks can improve that
I could recover build 65, at least I think so, and after several tries finally a new search that could beat Rybka's time on Andrew's quad
Some luck; it is easy to give too high a score to e6 so that you never find Ba6, because for that move there is just a very narrow move-path to take, to get a little more edge out of it than for the passed pawn on e6 and because lots of material is exchanged it is possible that the endgames do not have much edge left anymore, usually you end up with a pair of Bishops fighting two Knights in some combination with other pieces. Very interesting for Tord's material imbalance tables, how to compare the Bishop pair with a Knight pair on the other side, add to that the imbalance of Queen for two Rooks and some doubled pawns, it gets very complex.
(I did not wait what exactly would happen at plydepth 20, the difference between Ba6 and e6 is still small there and it is possible that the engine still might switch moves.)
[D]2k2rr1/ppqb4/2n3np/2PpP1p1/6R1/P2B1N2/2PBQPP1/1R2K3 w - -
Engine: Rainbow Serpent 1.6.3s(dc) Build 82 (Athlon 2009 MHz, 128 MB)
by Tord Romstad, Marco Costalba, Joona Kiiski Modifications: Dann Corbit
1.00 0:00 +2.26 1.e6 (1.009) 9
2.00 0:00 +2.62 1.e6 Be8 (1.169) 10
3.00 0:00 +3.03 1.e6 Be8 2.Kf1 (1.628) 14
4.00 0:00 +2.22 1.e6 Be8 2.Kf1 Nf4 (2.714) 24
5.00 0:00 +2.02 1.e6 Be8 2.Bc3 Nf4 3.Qe3 (9.300) 74
6.00 0:00 +2.06 1.e6 Be8 2.Kf1 Nge5 3.Nxe5 Qxe5
4.Qxe5 Nxe5 5.Rd4 Rf6 6.Rdb4 Nxd3
7.cxd3 (15.825) 112
7.00 0:00 +2.02 1.e6 Be8 2.Kf1 Nge5 3.Nxe5 Qxe5
4.Qxe5 Nxe5 5.Rd4 Rf6 6.Rdb4 Nxd3
7.cxd3 Bg6 8.R1b3 Rgf8 (45.524) 207
8.00 0:00 +1.93 1.e6 Be8 2.Kf1 Nge5 3.Nxe5 Qxe5
4.Qxe5 Nxe5 5.Rd4 Rf6 6.Re1 Nxd3
7.cxd3 Rgg6 8.e7 Rf5 9.f3 (76.979) 273
9.00 0:00 +2.14++ 1.e6 Be8 2.Kf1 Nge5 3.Nxe5 Qxe5
4.Qxe5 Nxe5 5.Rd4 Rf6 6.Re1 Nxd3
7.cxd3 Rgg6 8.e7 Rf5 9.g4 (110.862) 322
10.00 0:01 +1.25-- 1.e6 Re8 2.Bf5 Kb8 3.Kf1 Rgf8 (565.889) 446
11.01 0:06 +1.57 1.e6 Re8 2.Bf5 Kb8 3.Nd4 Nxd4 4.Rxd4 Rgf8
5.g4 Nh4 6.Rdb4 Bc6 7.Qa6 d4 8.f4 Nf3+
9.Kd1 Nxd2 10.Kxd2 Qxf4+ 11.Kd1 (3.215.109) 489
12.01 0:20 +0.92 1.e6 Re8 2.Bf5 Kb8 3.Kf1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Qxg6 Bxg4 7.Qxh6 Re8
8.Qxg5 Bxf3 9.gxf3 Qh7 10.Qh6 Qxc2
11.Qd6+ Ka8 (10.384.593) 513
13.01 0:51 +0.72 1.e6 Re8 2.Bf5 Kb8 3.Kf1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Qxg6 Bxg4 7.Qxh6 Qc8
8.Qxg5 Bxf3 9.gxf3 Qh3+ 10.Qg2 Qh7
11.Rb3 Qxc2 12.Qg7 Qxb3 13.Qxf8+ Kc7
14.Qf7+ (27.101.962) 521
14.01 2:08 +0.48 1.e6 Re8 2.Bf5 Kb8 3.Kf1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Qxg6 Bxg4 7.Qxh6 Qc8
8.Ng1 Ka8 9.Qg7 Re8 10.Re1 Rxe1+
11.Bxe1 Bf5 12.Nf3 Bxc2 13.Nxg5 Bb3
14.Bb4 (66.959.177) 522
15.01 2:36 +0.60 1.e6 Re8 2.Bf5 Kb8 3.Kf1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Qxg6 Bxg4 7.Qxh6 Qc8
8.Ng1 Ka8 9.Qg7 Re8 10.Re1 Rxe1+
11.Bxe1 Bf5 12.Nf3 Bxc2 13.Nxg5 Bb3
14.Kg1 (81.956.651) 523
[col0r=#FF0000]15.07 4:38 +1.17++ 1.Ba6 Bxg4 2.Rxb7 Qxb7 3.Qb5 Qxa6
4.Qxa6+ Kc7 5.Ba5+ Nxa5 6.Qxa5+ Kb8
7.Qb4+ (147.899.987) 530
16.01 6:09 +1.17 1.Ba6 Bxg4 2.Rxb7 Qxb7 3.Qb5 Qxa6
4.Qxa6+ Kc7 5.Ba5+ Nxa5 6.Qxa5+ Kc8
7.Qxa7 Bd7 8.e6 Be8 9.Qa8+ Kc7
10.Nd4 Rh8 11.Qxd5 Rh7 12.f3 Rfh8
13.c4 Nf4 14.Qd6+ (197.917.164) 534
17.01 13:07 +0.96 1.Ba6 Bxg4 2.Rxb7 Bxf3 3.Rxc7+ Kxc7
4.gxf3 Ngxe5 5.Kf1 Rg6 6.Bb5 Rf5
7.Bc3 d4 8.Bd2 Rgf6 9.f4 Nd7 10.Qe8 Rxc5
11.fxg5 Rf8 12.Qe2 hxg5 (427.961.336) 543
18.01 28:53 +0.76 1.Ba6 Bxg4 2.Rxb7 Bxf3 3.Rxc7+ Kxc7
4.gxf3 Ngxe5 5.Kf1 Rg6 6.Bb5 Rgf6
7.f4 Nd7 8.fxg5 Rxf2+ 9.Qxf2 Rxf2+
10.Kxf2 hxg5 11.Bxg5 Nxc5 12.Ke3 d4+
13.Kf3 Kd6 14.Kg4 (967.913.887) 558
19.01 47:42 +1.37++ 1.Ba6 Bxg4 2.Rxb7 Bxf3 3.Rxc7+ Kxc7
4.gxf3 Ngxe5 5.Kf1 Rg6 6.Bb5 Rgf6
7.f4 Nd7 8.c4 d4 9.Bxc6 Rxc6 10.Qe4 Rxc5
11.fxg5 hxg5 12.Bb4 Rf4 13.Qe7 Kc6
14.Bxc5 (1.553.277.315) 542
20.01 59:50 +1.01 1.Ba6 Bxg4 2.Rxb7 Bxf3 3.Rxc7+ Kxc7
4.gxf3 Ngxe5 5.Kf1 Rg6 6.Bb5 Rgf6
7.f4 gxf4 8.Ba5+ Kc8 9.Bc3 f3 10.Qe3 d4
11.Ba6+ Kc7 12.Ba5+ Nxa5 13.Qxe5+ Kd7
14.Qxd4+ (1.927.532.884) 536 [/color]
best move: Bd3-a6 time: 66:47.672 min n/s: 537.703 nodes: 2.154.930.285
I could recover build 65, at least I think so, and after several tries finally a new search that could beat Rybka's time on Andrew's quad
Some luck; it is easy to give too high a score to e6 so that you never find Ba6, because for that move there is just a very narrow move-path to take, to get a little more edge out of it than for the passed pawn on e6 and because lots of material is exchanged it is possible that the endgames do not have much edge left anymore, usually you end up with a pair of Bishops fighting two Knights in some combination with other pieces. Very interesting for Tord's material imbalance tables, how to compare the Bishop pair with a Knight pair on the other side, add to that the imbalance of Queen for two Rooks and some doubled pawns, it gets very complex. (I did not wait what exactly would happen at plydepth 20, the difference between Ba6 and e6 is still small there and it is possible that the engine still might switch moves.)
[D]2k2rr1/ppqb4/2n3np/2PpP1p1/6R1/P2B1N2/2PBQPP1/1R2K3 w - -
Engine: Rainbow Serpent 1.6.3s(dc) Build 82 (Athlon 2009 MHz, 128 MB)
by Tord Romstad, Marco Costalba, Joona Kiiski Modifications: Dann Corbit
1.00 0:00 +2.26 1.e6 (1.009) 9
2.00 0:00 +2.62 1.e6 Be8 (1.169) 10
3.00 0:00 +3.03 1.e6 Be8 2.Kf1 (1.628) 14
4.00 0:00 +2.22 1.e6 Be8 2.Kf1 Nf4 (2.714) 24
5.00 0:00 +2.02 1.e6 Be8 2.Bc3 Nf4 3.Qe3 (9.300) 74
6.00 0:00 +2.06 1.e6 Be8 2.Kf1 Nge5 3.Nxe5 Qxe5
4.Qxe5 Nxe5 5.Rd4 Rf6 6.Rdb4 Nxd3
7.cxd3 (15.825) 112
7.00 0:00 +2.02 1.e6 Be8 2.Kf1 Nge5 3.Nxe5 Qxe5
4.Qxe5 Nxe5 5.Rd4 Rf6 6.Rdb4 Nxd3
7.cxd3 Bg6 8.R1b3 Rgf8 (45.524) 207
8.00 0:00 +1.93 1.e6 Be8 2.Kf1 Nge5 3.Nxe5 Qxe5
4.Qxe5 Nxe5 5.Rd4 Rf6 6.Re1 Nxd3
7.cxd3 Rgg6 8.e7 Rf5 9.f3 (76.979) 273
9.00 0:00 +2.14++ 1.e6 Be8 2.Kf1 Nge5 3.Nxe5 Qxe5
4.Qxe5 Nxe5 5.Rd4 Rf6 6.Re1 Nxd3
7.cxd3 Rgg6 8.e7 Rf5 9.g4 (110.862) 322
10.00 0:01 +1.25-- 1.e6 Re8 2.Bf5 Kb8 3.Kf1 Rgf8 (565.889) 446
11.01 0:06 +1.57 1.e6 Re8 2.Bf5 Kb8 3.Nd4 Nxd4 4.Rxd4 Rgf8
5.g4 Nh4 6.Rdb4 Bc6 7.Qa6 d4 8.f4 Nf3+
9.Kd1 Nxd2 10.Kxd2 Qxf4+ 11.Kd1 (3.215.109) 489
12.01 0:20 +0.92 1.e6 Re8 2.Bf5 Kb8 3.Kf1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Qxg6 Bxg4 7.Qxh6 Re8
8.Qxg5 Bxf3 9.gxf3 Qh7 10.Qh6 Qxc2
11.Qd6+ Ka8 (10.384.593) 513
13.01 0:51 +0.72 1.e6 Re8 2.Bf5 Kb8 3.Kf1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Qxg6 Bxg4 7.Qxh6 Qc8
8.Qxg5 Bxf3 9.gxf3 Qh3+ 10.Qg2 Qh7
11.Rb3 Qxc2 12.Qg7 Qxb3 13.Qxf8+ Kc7
14.Qf7+ (27.101.962) 521
14.01 2:08 +0.48 1.e6 Re8 2.Bf5 Kb8 3.Kf1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Qxg6 Bxg4 7.Qxh6 Qc8
8.Ng1 Ka8 9.Qg7 Re8 10.Re1 Rxe1+
11.Bxe1 Bf5 12.Nf3 Bxc2 13.Nxg5 Bb3
14.Bb4 (66.959.177) 522
15.01 2:36 +0.60 1.e6 Re8 2.Bf5 Kb8 3.Kf1 Rgf8 4.Bxg6 Rxe6
5.Qd3 Rxg6 6.Qxg6 Bxg4 7.Qxh6 Qc8
8.Ng1 Ka8 9.Qg7 Re8 10.Re1 Rxe1+
11.Bxe1 Bf5 12.Nf3 Bxc2 13.Nxg5 Bb3
14.Kg1 (81.956.651) 523
[col0r=#FF0000]15.07 4:38 +1.17++ 1.Ba6 Bxg4 2.Rxb7 Qxb7 3.Qb5 Qxa6
4.Qxa6+ Kc7 5.Ba5+ Nxa5 6.Qxa5+ Kb8
7.Qb4+ (147.899.987) 530
16.01 6:09 +1.17 1.Ba6 Bxg4 2.Rxb7 Qxb7 3.Qb5 Qxa6
4.Qxa6+ Kc7 5.Ba5+ Nxa5 6.Qxa5+ Kc8
7.Qxa7 Bd7 8.e6 Be8 9.Qa8+ Kc7
10.Nd4 Rh8 11.Qxd5 Rh7 12.f3 Rfh8
13.c4 Nf4 14.Qd6+ (197.917.164) 534
17.01 13:07 +0.96 1.Ba6 Bxg4 2.Rxb7 Bxf3 3.Rxc7+ Kxc7
4.gxf3 Ngxe5 5.Kf1 Rg6 6.Bb5 Rf5
7.Bc3 d4 8.Bd2 Rgf6 9.f4 Nd7 10.Qe8 Rxc5
11.fxg5 Rf8 12.Qe2 hxg5 (427.961.336) 543
18.01 28:53 +0.76 1.Ba6 Bxg4 2.Rxb7 Bxf3 3.Rxc7+ Kxc7
4.gxf3 Ngxe5 5.Kf1 Rg6 6.Bb5 Rgf6
7.f4 Nd7 8.fxg5 Rxf2+ 9.Qxf2 Rxf2+
10.Kxf2 hxg5 11.Bxg5 Nxc5 12.Ke3 d4+
13.Kf3 Kd6 14.Kg4 (967.913.887) 558
19.01 47:42 +1.37++ 1.Ba6 Bxg4 2.Rxb7 Bxf3 3.Rxc7+ Kxc7
4.gxf3 Ngxe5 5.Kf1 Rg6 6.Bb5 Rgf6
7.f4 Nd7 8.c4 d4 9.Bxc6 Rxc6 10.Qe4 Rxc5
11.fxg5 hxg5 12.Bb4 Rf4 13.Qe7 Kc6
14.Bxc5 (1.553.277.315) 542
20.01 59:50 +1.01 1.Ba6 Bxg4 2.Rxb7 Bxf3 3.Rxc7+ Kxc7
4.gxf3 Ngxe5 5.Kf1 Rg6 6.Bb5 Rgf6
7.f4 gxf4 8.Ba5+ Kc8 9.Bc3 f3 10.Qe3 d4
11.Ba6+ Kc7 12.Ba5+ Nxa5 13.Qxe5+ Kd7
14.Qxd4+ (1.927.532.884) 536 [/color]
best move: Bd3-a6 time: 66:47.672 min n/s: 537.703 nodes: 2.154.930.285
Debugging is twice as hard as writing the code in the first
place. Therefore, if you write the code as cleverly as possible, you
are, by definition, not smart enough to debug it.
-- Brian W. Kernighan
place. Therefore, if you write the code as cleverly as possible, you
are, by definition, not smart enough to debug it.
-- Brian W. Kernighan
-
rightrook
- Posts: 1452
- Joined: Wed Mar 08, 2006 8:45 pm
Re: Test Position: Morozevich vs Pelletier, Biel 2003
>>>the difference between Ba6 and e6 is still small there and it is possible that the engine still might switch moves.) <<<
Yes, that's why I said it probably is not a good test move...!
regards
Robert
A true sacrifice is where it "must be accepted"...
(Just like the sacrifice of Christ on the cross *must be accepted* if a person is to become saved.)
Yes, that's why I said it probably is not a good test move...!
regards
Robert
A true sacrifice is where it "must be accepted"...
(Just like the sacrifice of Christ on the cross *must be accepted* if a person is to become saved.)
-
yanquis1972
- Posts: 1766
- Joined: Wed Jun 03, 2009 12:14 am
Re: Test Position: Morozevich vs Pelletier, Biel 2003
firebird seems to agree -- it found the move in a few mins at d23 or so, but by d25 had dropped its evaluation about .15 pawns. had to abort at that point & unfortunately already pasted over the analysis. it may have even switched back to e6 had i let it run longer.