Discussion of anything and everything relating to chess playing software and machines.
Moderator: Ras
Henk
Posts: 7261 Joined: Mon May 27, 2013 10:31 am
Post
by Henk » Tue May 10, 2016 2:25 pm
My chess program still thinks this position is equal. Right or wrong ?
[d] r2q1rk1/5pb1/1pn1b1np/p1p1p1p1/2PpP3/P2P1PP1/1PQN2BP/1RB1NRK1 b - - 0 1
Code: Select all
Depth Value Time(seconds) Nodes
24 0.1280 57.89 27972829 Be6-d7 b2-b3 Qd8-e7 Kg1-f2 Nc6-d8 h2-h3
23 0.1280 25.40 12408755 Be6-d7 b2-b3 Qd8-e7
22 0.1280 16.67 7963200 Be6-d7 b2-b3 Qd8-e7
21 0.1280 10.00 4787204 Be6-d7 b2-b3 Qd8-e7 Kg1-f2 Nc6-d8 h2-h3 Nd8-e6 a3-a4 Rf8-c8 Rf1-h1 Ra8-a7
20 0.1152 6.65 3084954 Be6-d7 b2-b3
19 0.1152 3.78 1767336 Be6-d7 b2-b3
18 0.1152 1.77 839519 Be6-d7 b2-b3 Kg8-h8 Kg1-f2 Rf8-e8 Rf1-h1 f7-f5 h2-h3
17 0.1280 0.59 323973 Be6-d7 Kg1-f2
16 0.1280 0.39 164949 Be6-d7 Kg1-f2
15 0.1280 0.64 253868 Be6-d7 Kg1-f2 Kg8-h8 h2-h3 Bg7-f6 Rf1-h1 Bf6-e7 b2-b3 f7-f5
14 0.1024 0.14 67425 Be6-d7 b2-b3 Bg7-f6 Kg1-f2 Bf6-e7 h2-h3
13 0.0896 0.14 46963 Be6-d7 b2-b3 Rf8-e8 Rb1-b2
12 0.0896 0.14 46074 Be6-d7 b2-b3 Rf8-e8 Rb1-b2 Re8-e6 a3-a4
11 0.1024 0.06 20645 Be6-d7 b2-b3 Rf8-e8
10 0.1024 0.02 12829 Be6-d7 b2-b3 Rf8-e8
9 0.1024 0.02 5268 Be6-d7 b2-b3 Rf8-e8 Rb1-b2 Re8-e6
8 0.0896 0.01 6516 Be6-d7 b2-b3
7 0.0896 0.00 1144 Be6-d7 b2-b3
6 0.0896 0.00 1146 Be6-d7 b2-b3 Rf8-e8 a3-a4
5 0.1024 0.00 784 Be6-d7
4 0.1024 0.00 501 Be6-d7
3 0.1024 0.00 370 Be6-d7 b2-b3 Rf8-e8
2 0.0896 0.00 542 Be6-d7 b2-b3
1 0.1152 0.00 100 Be6-d7
Henk
Posts: 7261 Joined: Mon May 27, 2013 10:31 am
Post
by Henk » Tue May 10, 2016 5:25 pm
Stockfish 6 evaluation:
Code: Select all
25 +1.37 24.2M 0:16.85 f6 Qd1 Rf7 Nb3 Bf8 Bd2 h5 Nc1 h4 Nc2 a4 Ne2 Rh7 b4 axb3 Rxb3 Bd6 Qb1 Bc7 Rb5 h3 Bh1 Rg7 Re1 Qd7 Qb2
24 +1.33 18.8M 0:13.00 f6 Qd1 Rf7 Nb3 Bf8 Bd2 h5 Nc1 h4 Nc2 h3 Bh1 a4 b4 axb3 Rxb3 Rb7 Ne2 Qc7 Rb5 Bd6 Qb1 Rf8 Re1 Rf7 Qb2 g4 fxg4
23 +1.33 17.5M 0:12.02 f6 Qd1 Rf7 Nb3 Bf8 Bd2 h5 Nc1 h4 Nc2 h3 Bh1 a4 Ne2 Bd6 b3 axb3 Rxb3 Rfa7 Rb5 Qc7 Qb1 Rb8 Qb3 Raa8 Rb1
22 +1.39 14.0M 0:09.62 f6 Qd1 Rf7 Nb3 Bf8 Bd2 h5 Nc1 h4 Ne2 h3 Bh1 Bd6 Qa4 Bd7 Qd1 Qc7 Nc2 Be6 b4 axb4 axb4 Ra2 bxc5 Bxc5
22 +1.34 9.4M 0:06.52 f6!
22 +1.33 7.2M 0:04.97 f6!
22 +1.27 6.9M 0:04.73 f6 Qa4?
21 +1.33 2.4M 0:01.59 f6 Qd1 h5 Nc2 a4 b3 axb3 Rxb3 h4 Rb5 h3 Bh1 Qc7 Qe2 Ra7 Re1 Rfa8 Rb2 Bd7 Rb5 Bf8 Rb1 Bd6 Rd1 Ra4 Bb2
20 +1.33 2.1M 0:01.41 f6 Qd1 h5 Nc2 a4 b3 axb3 Rxb3 h4 Rb5 h3 Bh1 Qc7 Qe2 Ra7 Re1 Bd7 Rb1 Rfa8 Bb2 Bf8 Qf1 Be6
19 +1.22 674462 0:00.44 f6 Qd1 h5 b3 Rf7 Nc2 Bf8 Bb2 h4 a4 Rh7 Qe2 Qd7 Rfd1 hxg3 hxg3 Rh8 Kf1 Rc8 Na3 Rh2 Nb5 Nb4 Re1
18 +1.25 365162 0:00.24 f6 Qd1 h5 b3 Rf7 Nc2 Bf8 Bb2 h4 a4 Rh7 Qe2 Qd7 Rfd1 Ra7 Na3 hxg3 hxg3 g4 fxg4
17 +1.19 185211 0:00.12 f6 Qd1 h5 b3 h4 Nc2 h3 Bh1 Qd6 Bb2 Ra7 Re1 Re7 Qe2 Qc7 Ra1 Rfe8 a4 Qd6 Rad1
16 +1.23 141249 0:00.10 f6 Qd1 h5 b3 h4 Nc2 h3 Bh1 Qd6 Bb2 Ra7 Re1 Rc7 Qe2 Re8 a4 Rf7 Rf1 Ree7 Rbe1
15 +1.20 102004 0:00.07 f6 Qd1 h5 b3 h4 Nc2 h3 Bh1 Qd6 Bb2 Ra7 Re1 Rc7 Qe2 Rff7 Rbd1
14 +1.13 79446 0:00.06 h5 Qd1 h4 b3 h3 Bh1 Qe7 Nc2 Bc8 Bb2 Bb7 Rf2 Qd7 Qe2 f6 Re1 Rad8
13 +1.15 50009 0:00.04 h5 Qd1 h4 b3 h3 Bh1 Qe7 Nc2 Bc8 Bb2 Bb7 a4 Qd6 Re1
12 +1.17 18156 0:00.02 h5 Qd1 f6 b3 h4 Nc2 h3 Bh1 Qe7 a4 Rab8 Bb2 Rf7
11 +1.15 8814 0:00.01 h5 Qd1 h4 b3 h3 Bh1 f5 exf5 Bxf5 Ne4 Bxe4 dxe4
10 +1.15 7536 0:00.01 h5 Qd1 h4 b3 h3 Bh1 f5 exf5 Bxf5 Ne4 Bxe4 dxe4
9 +1.14 4446 0:00.00 h5 Qd1 h4 b3 h3 Bh1 Qe7 Nc2 Bc8 Bb2
8 +1.24 1795 0:00.00 h5 Qd1 h4 Rf2 f6 Nc2 h3 Bf1
7 +1.21 1040 0:00.00 h5 Qd1 f6 b3 Rf7 Qe2 h4
6 +1.23 640 0:00.00 h5 Qd1 h4 Rf2 h3 Bf1
5 +1.28 376 0:00.00 h5 Qa4 Qd6 Qd1 f6
4 +1.28 286 0:00.00 h5 Qa4 Qd6 Qd1 f6
3 +1.28 180 0:00.00 h5 Qd1 f6
2 +1.28 119 0:00.00 h5 Qd1
1 +1.28 43 0:00.00 h5
I also think white's position is hopeless. So a bug or mis-evaluations or bad search algorithm in my chess program. Has something to do with space I think.
PK
Posts: 913 Joined: Mon Jan 15, 2007 11:23 am
Location: Warsza
Post
by PK » Tue May 10, 2016 7:01 pm
Most likely it is a search issue - you need some search to see that white cannot untangle. But if You modify the root position, moving a knight from d2 to e2, then suddenly it becomes almost equal, because f3-f4 works. It is not eval's job to know it.
Henk
Posts: 7261 Joined: Mon May 27, 2013 10:31 am
Post
by Henk » Thu May 12, 2016 9:09 am
I still don't understand why this position is not given a higher score like +20 or so.
Ralf Müller
Posts: 127 Joined: Sat Dec 29, 2012 12:07 am
Post
by Ralf Müller » Thu May 12, 2016 9:44 am
In this cases I check it with depth 0 or depth 1 to understand the evaluation.