Position taken from a blitz game I played earlier at FICS.
The best move, which I played is 18..f5 but 18..c4 might be better?
Let your engines run!
[d]2r1r1k1/pb1q1ppp/3b4/1pp5/3pP3/1P1P3P/P1Q1BPPN/R4RK1 b - - 0 18
Uncertain Regards,
Terry
Test for Dedicated Chess Computers and Chess Engines Too!
Moderator: Ras
-
pedrox
- Posts: 1056
- Joined: Fri Mar 10, 2006 6:07 am
- Location: Basque Country (Spain)
Re: Test for Dedicated Chess Computers and Chess Engines Too
In this position, darks seem to have advantage, they have the pair of bishops and his rooks are more active, lights also have a delayed central pawn and the knight in h2.
But it is complicated to decide if c4 or f5 is better (at least for my program) since the evaluation not difere much.
FEN: 2r1r1k1/pb1q1ppp/3b4/1pp5/3pP3/1P1P3P/P1Q1BPPN/R4RK1 b - - 0 18
DanaSah v.3.13:
2 00:00 505 505 +1,45 Tc6 Cg4
2 00:00 701 701 +1,50 Te6 Cg4
2 00:00 1.099 1.099 +1,53 Ae5 Cg4
2 00:00 1.576 1.576 +1,66 Af4 Cg4
3 00:00 3.082 3.082 +0,86 Af4 Ag4 Axh2+ Rxh2
3 00:00 3.924 3.924 +1,20 Axh2+ Rxh2 Dd6+ Rg1 Df4
3 00:00 6.277 627.700 +1,55 Tc6 Tfc1 Dc7
4 00:00 8.604 860.400 +1,46 Dc7 Cg4 Af4 a4
4 00:00 19.863 662.100 +1,54 f5 Af3 Dc7 g3 fxe4 Axe4
5 00:00 27.251 908.366 +1,57 f5 Af3 fxe4 dxe4 Te6 Cg4
6 00:00 100.808 840.066 +1,50 c4 Ag4 Axh2+ Rxh2 Dc7+ f4 cxd3 Dxd3 Axe4 Dxd4
7 00:00 144.197 801.094 +1,48 c4 bxc4 bxc4 Ag4 Axh2+ Rxh2 Dc7+ f4 cxd3 Dxd3 Axe4 Dxd4
8 00:00 262.271 846.035 +1,48 c4 bxc4 bxc4 Ag4 Axh2+ Rxh2 Dc7+ f4 cxd3 Dxd3 Axe4 Dxd4
9 00:00 991.914 927.022 +1,51 c4 Ag4 Axh2+ Rxh2 Dd6+ f4 cxd3 Dxd3 Axe4 Dxb5 Tb8 Dc4
10 00:01 1.479.980 942.662 +1,50 c4 Ag4 Axh2+ Rxh2 Dd6+ f4 cxd3 Dxd3 Axe4 Dxb5 Tb8 Dg5 Ad5
11 00:02 2.333.637 933.454 +1,50 c4 Ag4 Axh2+ Rxh2 Dd6+ f4 cxd3 Dxd3 Axe4 Dxb5 Tb8 Dg5 h6 Da5
12 00:04 4.585.196 975.573 +1,55 c4 bxc4 bxc4 Db1 cxd3 Axd3 Ac6 Cg4 Tb8 Dc2 h5 Ch2 Af4 Cf3
13 00:09 8.788.121 929.959 +1,55 c4 bxc4 bxc4 Db1 cxd3 Axd3 Ac6 Cg4 Tb8 Dc2 h5 Ch2 Af4 Cf3 Dd6
14 00:14 13.900.051 929.769 +1,55 c4 bxc4 bxc4 Db1 cxd3 Axd3 Ac6 Cg4 Tb8 Dc2 h5 Ch2 Aa4 De2 Dc7 Cf3
15 00:27 24.883.436 921.608 +1,55 c4 bxc4 bxc4 Db1 cxd3 Axd3 Ac6 Cf3 Af4 Ac2 Tc7 Ad3 Tb7 Dc2 Aa4 Dc4 Ab5
16 00:57 52.828.734 926.819 +1,55 c4 bxc4 bxc4 Db1 cxd3 Axd3 Af4 Cf3 Ac6 Ac2 Tc7 Ad3 h6 Ch2 Tb7 De1 Ae5 Cf3
17 18:55 1.038.144.734 914.664 +1,50 f5 Cf3 Df7 Cd2 Af4 Tae1 fxe4 Cxe4 Axe4 dxe4 c4 bxc4 bxc4 Ag4 Tc5 Dd1 d3 Da4 Tee5 Dd7
18 31:00 1.674.116.104 900.062 +1,50 f5 Cf3 Df7 Cd2 Af4 Tae1 fxe4 Cxe4 Axe4 dxe4 c4 bxc4 bxc4 Ag4 Tc5 De2 Tf8 a3 d3 Db2 Ag5
But it is complicated to decide if c4 or f5 is better (at least for my program) since the evaluation not difere much.
FEN: 2r1r1k1/pb1q1ppp/3b4/1pp5/3pP3/1P1P3P/P1Q1BPPN/R4RK1 b - - 0 18
DanaSah v.3.13:
2 00:00 505 505 +1,45 Tc6 Cg4
2 00:00 701 701 +1,50 Te6 Cg4
2 00:00 1.099 1.099 +1,53 Ae5 Cg4
2 00:00 1.576 1.576 +1,66 Af4 Cg4
3 00:00 3.082 3.082 +0,86 Af4 Ag4 Axh2+ Rxh2
3 00:00 3.924 3.924 +1,20 Axh2+ Rxh2 Dd6+ Rg1 Df4
3 00:00 6.277 627.700 +1,55 Tc6 Tfc1 Dc7
4 00:00 8.604 860.400 +1,46 Dc7 Cg4 Af4 a4
4 00:00 19.863 662.100 +1,54 f5 Af3 Dc7 g3 fxe4 Axe4
5 00:00 27.251 908.366 +1,57 f5 Af3 fxe4 dxe4 Te6 Cg4
6 00:00 100.808 840.066 +1,50 c4 Ag4 Axh2+ Rxh2 Dc7+ f4 cxd3 Dxd3 Axe4 Dxd4
7 00:00 144.197 801.094 +1,48 c4 bxc4 bxc4 Ag4 Axh2+ Rxh2 Dc7+ f4 cxd3 Dxd3 Axe4 Dxd4
8 00:00 262.271 846.035 +1,48 c4 bxc4 bxc4 Ag4 Axh2+ Rxh2 Dc7+ f4 cxd3 Dxd3 Axe4 Dxd4
9 00:00 991.914 927.022 +1,51 c4 Ag4 Axh2+ Rxh2 Dd6+ f4 cxd3 Dxd3 Axe4 Dxb5 Tb8 Dc4
10 00:01 1.479.980 942.662 +1,50 c4 Ag4 Axh2+ Rxh2 Dd6+ f4 cxd3 Dxd3 Axe4 Dxb5 Tb8 Dg5 Ad5
11 00:02 2.333.637 933.454 +1,50 c4 Ag4 Axh2+ Rxh2 Dd6+ f4 cxd3 Dxd3 Axe4 Dxb5 Tb8 Dg5 h6 Da5
12 00:04 4.585.196 975.573 +1,55 c4 bxc4 bxc4 Db1 cxd3 Axd3 Ac6 Cg4 Tb8 Dc2 h5 Ch2 Af4 Cf3
13 00:09 8.788.121 929.959 +1,55 c4 bxc4 bxc4 Db1 cxd3 Axd3 Ac6 Cg4 Tb8 Dc2 h5 Ch2 Af4 Cf3 Dd6
14 00:14 13.900.051 929.769 +1,55 c4 bxc4 bxc4 Db1 cxd3 Axd3 Ac6 Cg4 Tb8 Dc2 h5 Ch2 Aa4 De2 Dc7 Cf3
15 00:27 24.883.436 921.608 +1,55 c4 bxc4 bxc4 Db1 cxd3 Axd3 Ac6 Cf3 Af4 Ac2 Tc7 Ad3 Tb7 Dc2 Aa4 Dc4 Ab5
16 00:57 52.828.734 926.819 +1,55 c4 bxc4 bxc4 Db1 cxd3 Axd3 Af4 Cf3 Ac6 Ac2 Tc7 Ad3 h6 Ch2 Tb7 De1 Ae5 Cf3
17 18:55 1.038.144.734 914.664 +1,50 f5 Cf3 Df7 Cd2 Af4 Tae1 fxe4 Cxe4 Axe4 dxe4 c4 bxc4 bxc4 Ag4 Tc5 Dd1 d3 Da4 Tee5 Dd7
18 31:00 1.674.116.104 900.062 +1,50 f5 Cf3 Df7 Cd2 Af4 Tae1 fxe4 Cxe4 Axe4 dxe4 c4 bxc4 bxc4 Ag4 Tc5 De2 Tf8 a3 d3 Db2 Ag5
Last edited by pedrox on Tue Jul 17, 2007 12:23 am, edited 1 time in total.
-
JVMerlino
- Posts: 1398
- Joined: Wed Mar 08, 2006 10:15 pm
- Location: San Francisco, California
Re: Test for Dedicated Chess Computers and Chess Engines Too
After forcing each move and letting CM9_R1 search for a few minutes, here's the output:
If 18...c4:
If 18...f5:
So The King seems to think that 18...c4 is slightly better.
jm
If 18...c4:
Code: Select all
Time Depth Score Positions Moves
0:00 1/3 -1.74 1202 19.bxc4 bxc4 20.Nf3 cxd3 21.Qxd3
0:00 1/4 -1.83 4531 19.bxc4 bxc4 20.Bg4 Qc7 21.Nf3
cxd3 22.Qxd3
0:00 1/5 -2.08 11731 19.bxc4 bxc4 20.Nf3 cxd3 21.Qxd3
Bxe4 22.Qxd4
0:00 1/5 -1.88 17156 19.Nf3 cxd3 20.Qxd3 Bxe4 21.Qxb5
0:00 1/6 -1.79 46484 19.Nf3 cxd3 20.Qxd3 Bxe4 21.Qxb5
Bc6
0:00 1/7 -2.06 130222 19.Nf3 Bf4 20.a4 cxd3 21.Qxd3 Bxe4
22.Qxb5 Qd6
0:01 1/7 -1.59 167309 19.Bg4 Qc7 20.Nf3 cxd3 21.Qxd3
Bxe4 22.Qxb5 Rb8
0:01 1/8 -1.95 267736 19.Bg4 Qc7 20.Nf3 cxd3 21.Qxd3
Bxe4 22.Qxb5 Rb8 23.Qc4
0:04 1/8 -1.57 739569 19.Nf3 cxd3 20.Qxd3 Rxe4 21.Rad1
Rc3 22.Qxb5 Bc6
0:10 1/9 -1.77 1710836 19.Nf3 cxd3 20.Qxd3 Rc3 21.Qxb5
Bc6 22.Qa5 Bc7 23.Qb4 Rxe4
0:11 1/9 -1.45 2000849 19.Bg4 Qc7 20.Nf3 cxd3 21.Qxd3
Bxe4 22.Qxb5 Rb8 23.Qa4 f5 24.Rac1
0:16 1/10 -1.65 2864099 19.Bg4 Qc7 20.Nf3 cxd3 21.Qd2 Rcd8
22.Rac1 Qb6 23.Bf5 Ba3
0:56 1/11 -1.51 10496424 19.Bg4 Qc7 20.Nf3 cxd3 21.Qxd3
Bxe4 22.Qxb5 Rb8 23.Qa4 f5 24.Rac1
Qf7 25.Ng5
2:05 1/12 -1.64 23895495 19.Bg4 Qc7 20.Nf3 cxd3 21.Qxd3
Bxe4 22.Qxb5 a6 23.Qd7 Rcd8 24.Qxc7
Bxc7 25.Rfe1 d3 26.Rac1
Code: Select all
Time Depth Score Positions Moves
0:00 1/3 -2.04 2900 19.a4 fxe4 20.axb5
0:00 1/3 -2.01 4673 19.exf5 Qxf5 20.Nf3 Qf6 21.Qd2
0:00 1/4 -2.19 10555 19.exf5 Qxf5 20.Bg4 Qf4 21.g3
0:00 1/4 -1.93 13198 19.a4 Qc7 20.g3 fxe4 21.axb5
0:00 1/4 -1.72 21273 19.Bf3 Qf7 20.Rfe1 fxe4 21.Bxe4
0:00 1/5 -1.81 34318 19.Bf3 Qf7 20.Rfe1 fxe4 21.Bxe4 Qf4
0:00 1/6 -2.60 74419 19.Bf3 fxe4 20.Bxe4 Bxe4 21.dxe4
d3 22.Qd2 Bxh2+ 23.Kxh2 Rxe4
0:00 1/6 -2.47 86752 19.exf5 Bxh2+ 20.Kxh2 Qd5 21.Bf3
Qe5+ 22.Kh1 Bxf3 23.gxf3 Qxf5
0:00 1/6 -1.73 125765 19.Rfe1 fxe4 20.Bg4 Qc7 21.Nf1
exd3 22.Qxd3
0:02 1/7 -1.87 316460 19.Rfe1 c4 20.bxc4 bxc4 21.exf5
c3 22.Nf3 Bf4
0:04 1/7 -1.72 638141 19.Nf3 fxe4 20.Nd2 exd3 21.Bxd3
g6 22.Ne4 c4 23.Nxd6 cxd3
0:09 1/8 -1.58 1295669 19.Nf3 fxe4 20.Nd2 c4 21.bxc4 Qc7
22.Nxe4 Bxe4 23.dxe4 bxc4
0:21 1/9 -1.60 3232115 19.Nf3 Qf7 20.Nd2 Bf4 21.a4 b4
22.Rfe1 Bxd2 23.Qxd2 fxe4 24.Bg4
0:51 1/10 -1.56 8372493 19.Nf3 fxe4 20.Nd2 c4 21.bxc4 bxc4
22.Nxc4 Qc7 23.Rac1 Bh2+ 24.Kh1 Bf4
3:17 1/11 -1.39 32823448 19.Nf3 fxe4 20.Ng5 Qf5 21.Nxe4
Rxe4 22.dxe4 Bxe4 23.Qc1 Qe5 24.f4
Qd5 25.Bxb5 Bxg2
5:42 1/12 -1.51 59008892 19.Nf3 fxe4 20.Ng5 Qf5 21.Nxe4
Rxe4 22.dxe4 Bxe4 23.Qc1 d3 24.Bg4
Qe5 25.f4 Qd4+ 26.Kh2
jm
-
Eelco de Groot
- Posts: 4662
- Joined: Sun Mar 12, 2006 2:40 am
- Full name: Eelco de Groot
Re: Test for Dedicated Chess Computers and Chess Engines Too
Alaric eventually chooses Terry's move:
2r1r1k1/pb1q1ppp/3b4/1pp5/3pP3/1P1P3P/P1Q1BPPN/R4RK1 b - -
Engine: Alaric 707 (128 MB)
gemaakt door Peter Fendrich
9.00 0:00 +1.10 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Db1 Tcd8
5.Dxd3 Lxe4 6.Dxb5 (301.317)
10.00 0:00 +1.10 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Db1 Tcd8
5.Dxd3 Lxe4 6.Dxb5 (568.551) 613
11.01 0:02 +1.17 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dd2 Lf4
5.Dxd3 Lxe4 6.Dxb5 Tcd8 (1.348.250) 628
12.01 0:03 +1.05 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 Tb8 6.Da4 d3 7.Tac1 (2.470.263) 627
13.01 0:09 +1.06 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 Tb8 6.Da6 d3 7.Tac1 Lc5 (5.831.995) 639
14.01 0:18 +0.95 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 Tb8 6.Da4 f5 7.Lh5 g6 8.Tac1 (12.310.852) 650
15.01 0:52 +0.85 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 a6 6.Dxa6 f5 7.Lh5 g6 8.Lxg6 hxg6
9.Pxd4 (34.729.281) 656
16.01 2:06 +0.85 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 a6 6.Dxa6 f5 7.Lh5 g6 8.Lxg6 hxg6
9.Pxd4 (83.322.010) 656
17.01 5:59 +0.99 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 a6 6.Dxa6 f5 7.Lxf5 Lxf5
8.Pxd4 Ta8 9.Db5 Le4 (235.429.816) 655
18.01 8:53 +0.91 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 a6 6.Dxa6 f5 7.Lxf5 Lxf5
8.Pxd4 Le4 9.Tad1 Dd7 10.Tfe1 (350.044.605) 656
18.19 19:10 +1.01 1...f5 2.Pf3 Df7 3.Pd2 Lf4 4.Tae1 fxe4
5.Pxe4 Lxe4 6.dxe4 c4 7.g3 d3 8.Lxd3 cxd3
9.Dxd3 Ted8 10.Dxb5 (752.820.939) 654
19.01 28:53 +1.02 1...f5 2.Pf3 Df7 3.Pd2 Lf4 4.Tae1 fxe4
5.Pxe4 Lxe4 6.dxe4 c4 7.Lg4 d3 8.Dc3 d2
9.Tb1 Tc6 10.g3 Lg5 11.e5 (1.134.324.948) 654
Alaric Deep IIB this time is faster:
2r1r1k1/pb1q1ppp/3b4/1pp5/3pP3/1P1P3P/P1Q1BPPN/R4RK1 b - -
Engine: Alaric 707 Deep IIb (128 MB)
gemaakt door Peter Fendrich
9.00 0:00 +1.10 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Db1 Tcd8
5.Dxd3 Lxe4 6.Dxb5 (336.104)
10.00 0:00 +1.10 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Db1 Tcd8
5.Dxd3 Lxe4 6.Dxb5 (639.141) 702
11.01 0:02 +1.18 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxd4 Lc5 6.Dd7 a6 (1.372.338) 682
12.01 0:03 +1.05 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 Tb8 6.Da4 d3 7.Tac1 (2.223.620) 681
13.01 0:07 +1.06 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 Tb8 6.Da6 d3 7.Tac1 Lc5 (4.895.347) 689
14.01 0:12 +0.95 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 Tb8 6.Da4 f5 7.Lh5 g6 8.Tac1 (8.631.718) 692
15.01 0:25 +0.85 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 a6 6.Dxa6 f5 7.Lh5 g6 8.Lxg6 hxg6
9.Pxd4 (17.768.430) 696
16.01 1:23 +0.85 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 a6 6.Dxa6 f5 7.Lh5 g6 8.Lxg6 hxg6
9.Pxd4 (57.135.696) 682
17.01 2:26 +1.07 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 a6 6.Dxa6 f5 7.Lh5 g6 8.Lxg6 hxg6
9.Pxd4 Dd7 (100.764.903) 686
18.01 3:21 +0.81 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 a6 6.Dxa6 f5 7.Lh5 g6 8.Lxg6 hxg6
9.Pxd4 Ta8 10.Db5 (140.125.689) 694
18.19 6:22 +0.91 1...f5 2.Pf3 fxe4 3.Pg5 c4 4.Lg4 exd3
5.Dxd3 cxd3 6.Lxd7 Ld5 7.f3 d2 8.Pe4 Lf4
9.Lxe8 Txe8 10.Kf2 Lxe4 11.fxe4 Txe4 (268.358.150) 701
With these settings Alaric is about as fast in reaching plydepths as Rybka Beta. And then I mean the real Rybka plydepths, reported plydepths plus 3. For instance here in another position:
Alaric Deep IIb
22.01 99:37 +0.35 17.Kb1 Dxf2 18.Lf3 Dxd2 19.Txd2 Tae8
20.g3 Te5 21.Thd1 h5 22.c3 Kh7
23.Ka2 Tfe8 24.Kb3 b6 25.Pc7 T8e7
26.Td7 Lf8 27.Td8 Txc7 28.Txf8 (2.578.970.046) 431
and in parallel
Rybka 1.0 Beta
19.01 100:14 +0.46 17.Kb1 Dxf2 18.Ld3 (224.898.269) 38
Rybka's nodenumbers here probably have to be multiplied by a factor of ten at least to get the real NPS.
Unfortunately sometimes there can be some drawbacks to skipping, or pruning, so many nodes per ply compared to the default settings in Alaric..
I also noticed that when analyzing in parallel like here, Alaric only gets 40% of the CPU and the second engine gets 60%. At least that is what the taskmanager reports. If this actually helps against freezing up other resources on a single CPU computer, I don't know whether it does, this is fine with me.
Eelco
2r1r1k1/pb1q1ppp/3b4/1pp5/3pP3/1P1P3P/P1Q1BPPN/R4RK1 b - -
Engine: Alaric 707 (128 MB)
gemaakt door Peter Fendrich
9.00 0:00 +1.10 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Db1 Tcd8
5.Dxd3 Lxe4 6.Dxb5 (301.317)
10.00 0:00 +1.10 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Db1 Tcd8
5.Dxd3 Lxe4 6.Dxb5 (568.551) 613
11.01 0:02 +1.17 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dd2 Lf4
5.Dxd3 Lxe4 6.Dxb5 Tcd8 (1.348.250) 628
12.01 0:03 +1.05 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 Tb8 6.Da4 d3 7.Tac1 (2.470.263) 627
13.01 0:09 +1.06 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 Tb8 6.Da6 d3 7.Tac1 Lc5 (5.831.995) 639
14.01 0:18 +0.95 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 Tb8 6.Da4 f5 7.Lh5 g6 8.Tac1 (12.310.852) 650
15.01 0:52 +0.85 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 a6 6.Dxa6 f5 7.Lh5 g6 8.Lxg6 hxg6
9.Pxd4 (34.729.281) 656
16.01 2:06 +0.85 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 a6 6.Dxa6 f5 7.Lh5 g6 8.Lxg6 hxg6
9.Pxd4 (83.322.010) 656
17.01 5:59 +0.99 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 a6 6.Dxa6 f5 7.Lxf5 Lxf5
8.Pxd4 Ta8 9.Db5 Le4 (235.429.816) 655
18.01 8:53 +0.91 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 a6 6.Dxa6 f5 7.Lxf5 Lxf5
8.Pxd4 Le4 9.Tad1 Dd7 10.Tfe1 (350.044.605) 656
18.19 19:10 +1.01 1...f5 2.Pf3 Df7 3.Pd2 Lf4 4.Tae1 fxe4
5.Pxe4 Lxe4 6.dxe4 c4 7.g3 d3 8.Lxd3 cxd3
9.Dxd3 Ted8 10.Dxb5 (752.820.939) 654
19.01 28:53 +1.02 1...f5 2.Pf3 Df7 3.Pd2 Lf4 4.Tae1 fxe4
5.Pxe4 Lxe4 6.dxe4 c4 7.Lg4 d3 8.Dc3 d2
9.Tb1 Tc6 10.g3 Lg5 11.e5 (1.134.324.948) 654
Alaric Deep IIB this time is faster:
2r1r1k1/pb1q1ppp/3b4/1pp5/3pP3/1P1P3P/P1Q1BPPN/R4RK1 b - -
Engine: Alaric 707 Deep IIb (128 MB)
gemaakt door Peter Fendrich
9.00 0:00 +1.10 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Db1 Tcd8
5.Dxd3 Lxe4 6.Dxb5 (336.104)
10.00 0:00 +1.10 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Db1 Tcd8
5.Dxd3 Lxe4 6.Dxb5 (639.141) 702
11.01 0:02 +1.18 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxd4 Lc5 6.Dd7 a6 (1.372.338) 682
12.01 0:03 +1.05 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 Tb8 6.Da4 d3 7.Tac1 (2.223.620) 681
13.01 0:07 +1.06 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 Tb8 6.Da6 d3 7.Tac1 Lc5 (4.895.347) 689
14.01 0:12 +0.95 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 Tb8 6.Da4 f5 7.Lh5 g6 8.Tac1 (8.631.718) 692
15.01 0:25 +0.85 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 a6 6.Dxa6 f5 7.Lh5 g6 8.Lxg6 hxg6
9.Pxd4 (17.768.430) 696
16.01 1:23 +0.85 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 a6 6.Dxa6 f5 7.Lh5 g6 8.Lxg6 hxg6
9.Pxd4 (57.135.696) 682
17.01 2:26 +1.07 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 a6 6.Dxa6 f5 7.Lh5 g6 8.Lxg6 hxg6
9.Pxd4 Dd7 (100.764.903) 686
18.01 3:21 +0.81 1...c4 2.Lg4 Dc7 3.Pf3 cxd3 4.Dxd3 Lxe4
5.Dxb5 a6 6.Dxa6 f5 7.Lh5 g6 8.Lxg6 hxg6
9.Pxd4 Ta8 10.Db5 (140.125.689) 694
18.19 6:22 +0.91 1...f5 2.Pf3 fxe4 3.Pg5 c4 4.Lg4 exd3
5.Dxd3 cxd3 6.Lxd7 Ld5 7.f3 d2 8.Pe4 Lf4
9.Lxe8 Txe8 10.Kf2 Lxe4 11.fxe4 Txe4 (268.358.150) 701
With these settings Alaric is about as fast in reaching plydepths as Rybka Beta. And then I mean the real Rybka plydepths, reported plydepths plus 3. For instance here in another position:
Alaric Deep IIb
22.01 99:37 +0.35 17.Kb1 Dxf2 18.Lf3 Dxd2 19.Txd2 Tae8
20.g3 Te5 21.Thd1 h5 22.c3 Kh7
23.Ka2 Tfe8 24.Kb3 b6 25.Pc7 T8e7
26.Td7 Lf8 27.Td8 Txc7 28.Txf8 (2.578.970.046) 431
and in parallel
Rybka 1.0 Beta
19.01 100:14 +0.46 17.Kb1 Dxf2 18.Ld3 (224.898.269) 38
Rybka's nodenumbers here probably have to be multiplied by a factor of ten at least to get the real NPS.
Unfortunately sometimes there can be some drawbacks to skipping, or pruning, so many nodes per ply compared to the default settings in Alaric..
I also noticed that when analyzing in parallel like here, Alaric only gets 40% of the CPU and the second engine gets 60%. At least that is what the taskmanager reports. If this actually helps against freezing up other resources on a single CPU computer, I don't know whether it does, this is fine with me.
Eelco