Discussion of chess software programming and technical issues.
Moderators: hgm , Rebel , chrisw
Sven
Posts: 4052 Joined: Thu May 15, 2008 9:57 pm
Location: Berlin, Germany
Full name: Sven Schüle
Post
by Sven » Mon May 25, 2015 5:30 pm
Henk wrote: 10.
Code: Select all
Depth Value Time
1 -15870.00 0.018 64 b7c6
2 -16370.00 0.022 236 d1c1 e3c1
3 -16460.00 0.024 537 d1c1 e3c1 b7c6
4 -18993.00 0.029 1543 d1c1 e3c1 b7c6 b6c6
5 -18145.00 0.040 3985 d1c1 e3c1 b7c6 b6c6 f3e4
6 -18283.00 0.062 9887 d1c1 e3c1 b7c6 b6c6 f3e4 c6a6
7 -18149.00 0.145 30104 d1c1 e3c1 b7c6 b6c6 f3e4 c6a6 f6f5
8 -5581.00 0.275 63336 d1c1 e3c1 d8d1 g3g4 b7c6 c1e3 d1h1 h2g3
9 -5017.00 0.495 120903 d1c1 e3c1 d8d1 g3g4 b7c6 c1e3 d1h1 h2g3 f3e4
10 9999991.00 6.489 1463192 d1c1 e3c1 d8d1 g3g4 h5h4 b6b5 a6b5 c6b7 d1h1
Is that with or without nullmove enabled?
In any case, the PV above is not optimal (although I have to say the same applies to PVs of other engines as well). After 4.Rb5 black can mate immediately with Rh1 while in your PV black plays axb5 first. So 2.Bxc1 is not the "best defense" for white, that would be 2.Rb1 Rxb1 and then 3.Bc1.
jordanbray
Posts: 52 Joined: Mon Aug 11, 2014 3:01 am
Post
by jordanbray » Mon May 25, 2015 5:31 pm
For reference, my stockfish finds this at depth 12.
Code: Select all
Stockfish 6 64 POPCNT by Tord Romstad, Marco Costalba and Joona Kiiski
position fen 3r1k2/1p4p1/pRB2p2/P1P1p2p/4P3/2P1BbPP/5P1K/2Rr4 b - - 0 1
go wtime 10000 btime 10000
info depth 1 seldepth 1 multipv 1 score cp -88 nodes 93 nps 13285 tbhits 0 time 7 pv b7c6 c1d1 d8d1 b6c6 f3e4 c6a6
info depth 2 seldepth 2 multipv 1 score cp -88 nodes 138 nps 19714 tbhits 0 time 7 pv b7c6 c1d1
info depth 3 seldepth 3 multipv 1 score cp -88 nodes 195 nps 24375 tbhits 0 time 8 pv b7c6 c1d1 d8d1
info depth 4 seldepth 5 multipv 1 score cp -88 nodes 354 nps 44250 tbhits 0 time 8 pv b7c6 b6c6 f3e4 c1d1 d8d1
info depth 5 seldepth 5 multipv 1 score cp -89 nodes 500 nps 62500 tbhits 0 time 8 pv b7c6 b6c6 f3e4 c6a6 d1c1
info depth 6 seldepth 6 multipv 1 score cp -89 nodes 730 nps 81111 tbhits 0 time 9 pv b7c6 b6c6 f3e4 c6a6 d1c1 e3c1
info depth 7 seldepth 9 multipv 1 score cp -100 nodes 1478 nps 147800 tbhits 0 time 10 pv b7c6 b6c6 d8d3 c1d1 d3d1 c6c8 f8e7 c8c7 e7d8
info depth 8 seldepth 10 multipv 1 score cp -61 nodes 3810 nps 254000 tbhits 0 time 15 pv b7c6 b6a6 d1c1 e3c1 f3e4 a6b6 d8d3 c3c4
info depth 9 seldepth 12 multipv 1 score cp 10 nodes 11317 nps 419148 tbhits 0 time 27 pv b7c6 g3g4 d1c1 e3c1 d8d1 c1b2 h5g4 h3g4 f3e4
info depth 10 seldepth 17 multipv 1 score cp 113 nodes 53105 nps 1041274 tbhits 0 time 51 pv d1c1 g3g4 c1h1 h2g3 b7c6 g3f3 h1h3 f3g2 h5g4 b6a6 f8e7 a6c6 d8a8 a5a6 a8a7
info depth 11 seldepth 19 multipv 1 score cp 118 nodes 68257 nps 1137616 tbhits 0 time 60 pv d1c1 g3g4 c1h1 h2g3 b7c6 g3f3 h1h3 f3e2 h5g4 b6a6 d8c8 a6b6 h3h1 e3d2
info depth 12 seldepth 19 multipv 1 score mate 5 nodes 150220 nps 1430666 tbhits 0 time 105 pv d1c1 g3g4 h5h4 b6b1 c1b1 e3c1 b1c1 c6b7 c1h1
Roman Hartmann
Posts: 295 Joined: Wed Mar 08, 2006 8:29 pm
Post
by Roman Hartmann » Mon May 25, 2015 7:06 pm
The new engine I'm working on needs 8 ply to see the threat. To see it's mate it needs another ply. This is with nullmove (with research) and with LMR. There is no hashing, no killers or anything else yet. There is not even a real eval actually (just counting material).
Code: Select all
roce: setboard 3r1k2/1p4p1/pRB2p2/P1P1p2p/4P3/2P1BbPP/5P1K/2Rr4 b - - 0 1
roce: analyze
D score variation
1 -128 d1xc1 [36 moves]-[37 qs-nodes]
2 -483 d1xc1 e3xc1 [126 moves]-[140 qs-nodes]
3 -141 d1xc1 e3xc1 b7xc6 [466 moves]-[504 qs-nodes]
4 -282 d1xc1 e3xc1 b7xc6 b6xa6 [2032 moves]-[1904 qs-nodes]
5 -141 d1xc1 e3xc1 b7xc6 b6xa6 f3xe4 [5775 moves]-[5554 qs-nodes]
6 -282 d1xc1 e3xc1 b7xc6 b6xa6 f3xe4 a6xc6 [19757 moves]-[18837 qs-nodes]
7 -282 d1xc1 e3xc1 b7xc6 b6xc6 f3xe4 c6xa6 g7-g5 [78448 moves]-[74423 qs-nodes]
8 851 d1xc1 g3-g4 h5-h4 b6-b1 c1xb1 e3-c1 b7xc6 g4-g5 [496153 moves]-[492407 qs-nodes]
9 1991 d1xc1 g3-g4 h5-h4 b6-b1 c1xb1 e3-c1 b1xc1 c6xb7 c1-h1 [2577807 moves]-[2592571 qs-nodes]
10 1991 d1xc1 g3-g4 h5-h4 b6-b1 c1xb1 e3-c1 b1xc1 c6xb7 c1-h1 b2xa3 [10155575 moves]-[10181933 qs-nodes]
11 1991 d1xc1 g3-g4 h5-h4 b6-b1 c1xb1 e3-c1 b1xc1 c6xb7 c1-h1 f7-f8 g4xh5 [44408879 moves]-[44359687 qs-nodes]
Roman
JVMerlino
Posts: 1357 Joined: Wed Mar 08, 2006 10:15 pm
Location: San Francisco, California
Post
by JVMerlino » Tue May 26, 2015 5:16 am
Myrddin sees it at depth 9:
Code: Select all
1 113 0 13 d1c1 e3c1 b7c6 b6c6 f3e4 c6a6
2 113 0 114 d1c1 e3c1 b7c6 b6c6 f3e4 c6a6 (22 KNPS)
3 113 1 531 d1c1 e3c1 b7c6 b6c6 f3e4 c6a6 (48 KNPS)
4 113 1 1757 d1c1 e3c1 b7c6 b6c6 f3e4 c6a6 (103 KNPS)
5 113 2 5823 d1c1 e3c1 b7c6 b6c6 f3e4 c6a6 (200 KNPS)
6 97 4 13564 d1c1! (294 KNPS)
6 81 5 15613 d1c1! (294 KNPS)
6 63 6 20723 d1c1 e3c1 d8d1 g3g4 b7c6 b6b8 f8e7 (300 KNPS)
7 52 11 41217 d1c1 e3c1 d8d1 g3g4 h5g4 h3g4 b7c6 (355 KNPS)
8 36 20 77331 d1c1! (371 KNPS)
8 20 23 87469 d1c1! (367 KNPS)
8 -12 27 104420 d1c1! (379 KNPS)
8 -60 30 114047 d1c1! (380 KNPS)
8 -124 32 123389 d1c1! (377 KNPS)
8 -204 36 140706 d1c1! (386 KNPS)
8 -300 40 160254 d1c1! (400 KNPS)
8 -412 42 172650 d1c1! (406 KNPS)
8 -540 45 188597 d1c1! (416 KNPS)
8 -684 47 193515 d1c1! (409 KNPS)
8 -1178 56 253426 d1c1 g3g4 h5h4 b6b1 c1b1 e3c1 b7c6 c1e3 f3e4 (447
KNPS)
9 -1194 67 311633 d1c1! (463 KNPS)
9 -1210 78 388612 d1c1! (497 KNPS)
9 -1242 78 390111 d1c1! (494 KNPS)
9 -1290 79 391611 d1c1! (491 KNPS)
9 -1354 80 393482 d1c1! (488 KNPS)
9 -1434 81 394552 d1c1! (485 KNPS)
9 -1530 82 395600 d1c1! (482 KNPS)
9 -1642 82 396648 d1c1! (480 KNPS)
9 -1770 83 397696 d1c1! (478 KNPS)
9 -1914 83 398744 d1c1! (476 KNPS)
9 -32758 87 425031 d1c1 g3g4 h5h4 b6b1 c1b1 e3c1 b1c1 c6b7 c1h1 (485
KNPS)(Mate in -5)
10 -32758 98 490343 d1c1 g3g4 h5h4 b6b1 c1b1 e3c1 b1c1 c6b7 c1h1 (498
KNPS)(Mate in -5)
11 -32758 113 578961 d1c1 g3g4 h5h4 b6b1 c1b1 e3c1 b1c1 c6b7 c1h1 (509
KNPS)(Mate in -5)
jm
Henk
Posts: 7220 Joined: Mon May 27, 2013 10:31 am
Post
by Henk » Tue May 26, 2015 8:39 am
Sven Schüle wrote: Henk wrote: 10.
Code: Select all
Depth Value Time
1 -15870.00 0.018 64 b7c6
2 -16370.00 0.022 236 d1c1 e3c1
3 -16460.00 0.024 537 d1c1 e3c1 b7c6
4 -18993.00 0.029 1543 d1c1 e3c1 b7c6 b6c6
5 -18145.00 0.040 3985 d1c1 e3c1 b7c6 b6c6 f3e4
6 -18283.00 0.062 9887 d1c1 e3c1 b7c6 b6c6 f3e4 c6a6
7 -18149.00 0.145 30104 d1c1 e3c1 b7c6 b6c6 f3e4 c6a6 f6f5
8 -5581.00 0.275 63336 d1c1 e3c1 d8d1 g3g4 b7c6 c1e3 d1h1 h2g3
9 -5017.00 0.495 120903 d1c1 e3c1 d8d1 g3g4 b7c6 c1e3 d1h1 h2g3 f3e4
10 9999991.00 6.489 1463192 d1c1 e3c1 d8d1 g3g4 h5h4 b6b5 a6b5 c6b7 d1h1
Is that with or without nullmove enabled?
In any case, the PV above is not optimal (although I have to say the same applies to PVs of other engines as well). After 4.Rb5 black can mate immediately with Rh1 while in your PV black plays axb5 first. So 2.Bxc1 is not the "best defense" for white, that would be 2.Rb1 Rxb1 and then 3.Bc1.
With nullmove enabled.
Henk
Posts: 7220 Joined: Mon May 27, 2013 10:31 am
Post
by Henk » Tue May 26, 2015 9:11 am
But depth is not that important. More interesting is given the hardware it used the time it needed to find the mate, because depth is artificial and not decreased with 1 each time and there is also depth < 0.