Shirov's Bh3, Houdini solves it

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Rebel, chrisw

Vinvin
Posts: 5228
Joined: Thu Mar 09, 2006 9:40 am
Full name: Vincent Lejeune

Shirov's Bh3, Houdini solves it

Post by Vinvin »

On Intel 3 cores @ 3.1 Ghz

FEN: 8/8/4kpp1/3p1b2/p6P/2B5/6P1/6K1 b - - 0 1

Houdini_15_w32:
...
30/56 14:11 2.726.080.589 3.202.000 +1,41 Ff5-e4 Rg1-f2
f6-f5 Fc3-d4 Re6-d7 g2-g3 Rd7-c6 Rf2-e3 Rc6-b5 Fd4-b2 Rb5-c4 Re3-d2
d5-d4 Rd2-e2 d4-d3+ Re2-d2 Rc4-b3 Fb2-d4 a4-a3 Rd2-c1 Rb3-c4 Fd4-e5
a3-a2 Fe5-a1 Fe4-c6 Fa1-b2 Rc4-b3 Fb2-a1 Rb3-b4 Rc1-d2 Rb4-c4 Fa1-f6
Fc6-a4 Ff6-a1 Fa4-c2 Fa1-c3 Rc4-b3 Fc3-d4 Rb3-b4
30/67+ 24:59 4.913.246.933 3.277.000 +1,49 Ff5-h3
30/78+ 1:36:56 21.150.955.106 3.636.000 +5,54 Ff5-h3
30/87 5:12:05 75.495.674.421 4.031.000 +5,37 Ff5-h3
Rg1-f2 Re6-f5 g2-g3 Rf5-e4 Fc3-b2 Re4-d3 Rf2-f3 Fh3-f1 Rf3-f2 Ff1-e2
Rf2-e1 f6-f5 Fb2-a3 Rd3-c2 Fa3-c5 Fe2-g4 Fc5-e3 Fg4-h5 Re1-f2 Rc2-c3
Rf2-e1 d5-d4 Fe3xd4+ Rc3xd4 Re1-d2 a4-a3 Rd2-c2 a3-a2
31/87 5:31:41 80.019.827.765 4.020.000 +5,37 Ff5-h3
Rg1-f2 Re6-f5 g2-g3 Rf5-e4 Fc3-b2 Re4-d3 Rf2-f3 Fh3-f1 Rf3-f2 Ff1-e2
Rf2-g2 Fe2-h5 Rg2-f1 Fh5-g4 Rf1-g1 d5-d4 Rg1-f2 f6-f5 Rf2-f1 Rd3-c2
Fb2xd4 a4-a3 Rf1-f2 a3-a2 Fd4-e5 Rc2-b1 Rf2-e3 a2-a1D Fe5xa1 Rb1xa1
Re3-f2 Ra1-b2 Rf2-e1 Rb2-c3
32/87 6:06:50 87.684.517.651 3.983.000 +5,49 Ff5-h3
Rg1-f2 Re6-f5 g2-g3 Rf5-e4 Fc3-b2 Re4-d3 Rf2-f3 Fh3-f1 Rf3-f2 Ff1-e2
Rf2-g2 Fe2-h5 Rg2-f2 f6-f5 Rf2-g2 Fh5-g4 Rg2-f2 Fg4-e2 Rf2-g1 Fe2-f3
Rg1-f1 Ff3-h5 Rf1-e1 Fh5-e2 Fb2-a3 Fe2-f3 Re1-f2 Ff3-g4 Fa3-b2 Fg4-e2
33/87+ 7:26:21 105.294.949.142 3.931.000 +5,72 Ff5-h3
33/87 7:49:08 110.308.795.417 3.918.000 +5,71 Ff5-h3
Rg1-f2 Re6-f5 g2-g3 Rf5-e4 Fc3-b2 Re4-d3 Rf2-f3 Fh3-f1 Rf3-f2 Ff1-e2
Rf2-g2 Fe2-h5 Rg2-f2 f6-f5 Rf2-f1 Fh5-e2+ Rf1-e1 Fe2-f3 Re1-f1 Ff3-h5
Rf1-f2 Fh5-g4 Rf2-e1 Fg4-f3 Re1-f1

24m59s to find the move but 1h12m more to solve the fail-high.
Robert, is the latest Houdini faster to resolve the fail-high ?


BTW, I prepare a new Hard test set (hardest from hard-CCC, the-nightmare-game, MGV77, 100-brillante Schachzug and hard positions from Rybka forum). I hope to get ready in 1 week.
Last edited by Vinvin on Tue Jul 03, 2012 2:02 pm, edited 1 time in total.
Top
zullil
Posts: 6442
Joined: Tue Jan 09, 2007 12:31 am
Location: PA USA
Full name: Louis Zulli

Re: Shirov's Bh3, Houdini solve it

Post by zullil »

Vinvin wrote:On Intel 3 cores @ 3.1 Ghz

[D]8/8/4kpp1/3p1b2/p6P/2B5/6P1/6K1 b - - 0 1

Houdini_15_w32:
...
30/56 14:11 2.726.080.589 3.202.000 +1,41 Ff5-e4 Rg1-f2
f6-f5 Fc3-d4 Re6-d7 g2-g3 Rd7-c6 Rf2-e3 Rc6-b5 Fd4-b2 Rb5-c4 Re3-d2
d5-d4 Rd2-e2 d4-d3+ Re2-d2 Rc4-b3 Fb2-d4 a4-a3 Rd2-c1 Rb3-c4 Fd4-e5
a3-a2 Fe5-a1 Fe4-c6 Fa1-b2 Rc4-b3 Fb2-a1 Rb3-b4 Rc1-d2 Rb4-c4 Fa1-f6
Fc6-a4 Ff6-a1 Fa4-c2 Fa1-c3 Rc4-b3 Fc3-d4 Rb3-b4
30/67+ 24:59 4.913.246.933 3.277.000 +1,49 Ff5-h3
30/78+ 1:36:56 21.150.955.106 3.636.000 +5,54 Ff5-h3
30/87 5:12:05 75.495.674.421 4.031.000 +5,37 Ff5-h3
Rg1-f2 Re6-f5 g2-g3 Rf5-e4 Fc3-b2 Re4-d3 Rf2-f3 Fh3-f1 Rf3-f2 Ff1-e2
Rf2-e1 f6-f5 Fb2-a3 Rd3-c2 Fa3-c5 Fe2-g4 Fc5-e3 Fg4-h5 Re1-f2 Rc2-c3
Rf2-e1 d5-d4 Fe3xd4+ Rc3xd4 Re1-d2 a4-a3 Rd2-c2 a3-a2
31/87 5:31:41 80.019.827.765 4.020.000 +5,37 Ff5-h3
Rg1-f2 Re6-f5 g2-g3 Rf5-e4 Fc3-b2 Re4-d3 Rf2-f3 Fh3-f1 Rf3-f2 Ff1-e2
Rf2-g2 Fe2-h5 Rg2-f1 Fh5-g4 Rf1-g1 d5-d4 Rg1-f2 f6-f5 Rf2-f1 Rd3-c2
Fb2xd4 a4-a3 Rf1-f2 a3-a2 Fd4-e5 Rc2-b1 Rf2-e3 a2-a1D Fe5xa1 Rb1xa1
Re3-f2 Ra1-b2 Rf2-e1 Rb2-c3
32/87 6:06:50 87.684.517.651 3.983.000 +5,49 Ff5-h3
Rg1-f2 Re6-f5 g2-g3 Rf5-e4 Fc3-b2 Re4-d3 Rf2-f3 Fh3-f1 Rf3-f2 Ff1-e2
Rf2-g2 Fe2-h5 Rg2-f2 f6-f5 Rf2-g2 Fh5-g4 Rg2-f2 Fg4-e2 Rf2-g1 Fe2-f3
Rg1-f1 Ff3-h5 Rf1-e1 Fh5-e2 Fb2-a3 Fe2-f3 Re1-f2 Ff3-g4 Fa3-b2 Fg4-e2
33/87+ 7:26:21 105.294.949.142 3.931.000 +5,72 Ff5-h3
33/87 7:49:08 110.308.795.417 3.918.000 +5,71 Ff5-h3
Rg1-f2 Re6-f5 g2-g3 Rf5-e4 Fc3-b2 Re4-d3 Rf2-f3 Fh3-f1 Rf3-f2 Ff1-e2
Rf2-g2 Fe2-h5 Rg2-f2 f6-f5 Rf2-f1 Fh5-e2+ Rf1-e1 Fe2-f3 Re1-f1 Ff3-h5
Rf1-f2 Fh5-g4 Rf2-e1 Fg4-f3 Re1-f1

24m59s to find the move but 1h12m more to solve the fail-high.
Robert, is the latest Houdini faster to resolve the fail-high ?


BTW, I prepare a new Hard test set (hardest from hard-CCC, the-nightmare-game, MGV77, 100-brillante Schachzug and hard positions from Rybka forum). I hope to get ready in 1 week.
Top
zullil
Posts: 6442
Joined: Tue Jan 09, 2007 12:31 am
Location: PA USA
Full name: Louis Zulli

Re: Shirov's Bh3, Houdini solve it

Post by zullil »

zullil wrote:
Vinvin wrote:On Intel 3 cores @ 3.1 Ghz

[D]8/8/4kpp1/3p1b2/p6P/2B5/6P1/6K1 b - - 0 1
The new Critter finds Bh3 quickly:

Code: Select all

Critter v1.6a 64-bit, by Richard Vida
hardware POPCNT enabled
setoption name Hash value 4096
setboard 8/8/4kpp1/3p1b2/p6P/2B5/6P1/6K1 b - - 0 1
info string Opening book: "book.cbk" - 57853 entries [native]
info string GTB Init OK (5 piece set)
info string Using 8 thread(s), 4096 MB hash LP=off
go infinite
 2/ 4  00:00      185  185000  +1.94 a3 Kf2 a2 Ke3
 3/ 6  00:00      344  172000  +1.93 a3 Kf2 a2 Ke3 Be4
 4/ 7  00:00      550  275000  +2.05 a3 Kf2 a2 Ke3 Be4 g4
 5/ 8  00:00      995  248750  +1.87 a3 Kf2 Be4 Ke3 Bxg2 Kd4
 6/ 9  00:00     2028  289714  +1.69 a3 Kf2 Be4 g3 a2 Ke3 f5 Kd4
 7/10  00:00     3782  315166  +1.58 a3 Kf2 Be4 g3 a2 Ke3 f5 Kd4 Kd6
 8/13- 00:00    16568  436000  +1.48 a3 Kf2 Be4 g3 a2 Ke3 f5 Kd4 Kd6
 8/16  00:00    42642  592250  +1.51 a3 Kf2 Bc2 Ke3 a2 Kd4 Be4 g3 Kf5 Ke3 Bg2 Kd4
 9/16- 00:00    63237  559619  +1.42 a3 Kf2 Bc2 Ke3 a2 Kd4 Be4 g3 Kf5 Ke3 Bg2 Kd4
 9/17  00:00   105318  662377  +1.39 a3 Kf2 Bc2 Ke3 Kf5 g3 Be4 Kd2 Ke6 g4 Bf3
10/19  00:00   135505  691352  +1.47 a3 Kf2 Be4 g4 a2 Ke2 g5 hxg5 fxg5 Ke3 Kd6 Kd4 Bf3 Bb4+ Ke6
11/19  00:00   212635  793414  +1.40 a3 Kf2 Be4 g4 a2 Ke2 g5 hxg5 fxg5 Kd2 Kd6 Kc1 Kc5 Ba1 d4 Kb2
12/21  00:00   391784  994375  +1.37 a3 Kf2 Be4 g3 f5 Ke3 Kd7 Kd4 Kc6 Ba1 Kb5 Kc3 Ka4 Kd4
13/21  00:00   624498   1224K  +1.37 a3 Kf2 Be4 g3 f5 Ke2 Kd7 Ke3 Kc6 Kd4 Kb6 Ba1 Kb5 Kc3 Ka4 Kd4 a2
14/21  00:00   951758   1461K  +1.37 a3 Kf2 Be4 g3 f5 Ke2 Kd7 Ke3 Kc6 Kd4 Kb6 Ba1 Kb5 Kc3 Ka4 Kd4 a2
15/24  00:00    1563K   1788K  +1.43 a3 Kf2 Be4 g3 f5 Ke2 a2 Ke3 Ke7 Kd4 Kd6 Ba1 Bb1 Bc3 Ke6 Ba1 Be4 Bc3 Kf7 Ke3 Ke7
15/24+ 00:01    1978K   1935K  +1.46 Be4 Kf2 f5 g3 a3 Ke2 a2 Ke3 Ke7 Kd4 Kd6 Ba1 Bb1 Bc3 Ke6 Ba1 Be4 Bc3 Kf7 Ke3 Ke7
15/24+ 00:01    2100K   2000K  +1.59 Be4 Kf2 f5 g3 a3 Ke2 a2 Ke3 Ke7 Kd4 Kd6 Ba1 Bb1 Bc3 Ke6 Ba1 Be4 Bc3 Kf7 Ke3 Ke7
15/31  00:01    2582K   2179K  +1.74 Be4 g3 f5 Kf2 Kd6 Bb4+ Kc6 Ba3 Kb5 Ke3 Kc4 Bb2 Kb3 Ba1 a3 Kd4 a2 Bc3 Bf3 Ba1 Kc2
16/31+ 00:01    3864K   2574K  +1.85 Be4 g3 f5 Kf2 Kd6 Bb4+ Kc6 Ba3 Kb5 Ke3 Kc4 Bb2 Kb3 Ba1 a3 Kd4 a2 Bc3 Bf3 Kd3 Bh5 Ba1 Bg4 Kd4
16/31  00:01    4990K   2928K  +1.80 Be4 g4 Bf3 g5 f5 Kf2 Be4 Ke3 a3 Ba1 a2 Bd4 Bg2 Ba1 Bh3 Bd4 Bg4 Ba1 Bh5 Bd4 Bg4
17/31  00:01    6215K   3235K  +1.83 Be4 g4 Bf3 g5 f5 Kf2 Bg4 Ke3 a3 Ba1 a2 Bd4 Bh5 Bb2 Kd6 Bd4 Bd1 Ba1 Kc5 Kd3 f4 Bd4+ Kb4
18/32  00:02    8925K   3825K  +1.83 Be4 g4 Bf3 g5 f5 Kf2 Bg4 Ke3 a3 Ba1 a2 Bd4 Bh5 Bb2 Kd6 Bd4 Bd1 Ba1 Kc5 Kd3 f4 Bd4+ Kb4
19/32  00:03   15508K   4993K  +1.86 Be4 g4 Bf3 g5 f5 Kf2 Bg4 Ke3 a3 Ba1 a2 Bd4 Bh5 Bb2 Kd6 Ba3+ Kd7 Bb2 Kd6
20/38  00:05   43049K   7308K  +1.94 Be4 Ba1 f5 Kf2 Kd6 Bd4 a3 g3 Kc6 Ke3 Kb5 Bc3 Kc4 Bh8 a2 Ba1 Bb1 Bg7 Bd3 Ba1 Bf1 Kd2 d4 Kc2 d3+ Kc1 Be2 Kd2 Kb3 Be5
21/38  00:06   51634K   7494K  +1.95 Be4 Ba1 f5 Kf2 Kd6 Bd4 a3 g3 Kc6 Ke3 Kb5 Bc3 Kc4 Bh8 a2 Ba1 Bb1 Bg7 Bd3 Ba1 Be4 Be5 Bc2 Ba1 Bd3 Bb2 Bb1 Ba1 Bc2 Kd2 Be4 Bc3 Bd3 Ba1 d4 Kd1 Be4 Kd2 Bc6 Kc2 d3+ Kc1 Be4 Kd2 Bf3 Bb2 Bc6 Bg7 Be8 Ba1 Bb5 Bc3 Bc6 Bg7
22/38  00:09   71904K   7932K  +1.95 Be4 Ba1 f5 Kf2 Kd6 Bd4 a3 g3 Kc6 Ke3 Kb5 Bc3 Kc4 Bh8 a2 Ba1 Bb1 Bg7 Bd3 Ba1 Be4 Be5 Bc2 Bb2 Bb1 Ba1 Bc2 Kd2 Be4 Bc3 Bd3 Ba1 d4 Kd1 Be4 Kd2 Bc6 Kc2 d3+ Kc1 Be4 Kd2 Bf3 Bb2 Bc6 Bg7 Be8 Ba1 Bb5 Bc3 Bc6 Bg7
23/38  00:12  100829K   8260K  +1.95 Be4 Ba1 f5 Kf2 Kd6 Bd4 a3 g3 Kc6 Ke3 Kb5 Bc3 Kc4 Bh8 a2 Ba1 Bb1 Bg7 Bd3 Ba1 Be4 Be5 Bc2 Ba1 Bd3 Bb2 Bb1 Ba1 Bc2 Kd2 Be4 Bc3 Bd3 Ba1 d4 Kd1 Be4 Kd2 Bc6 Kc2 d3+ Kc1 Be4 Kd2 Bf3 Bb2 Bc6 Bg7 Be8 Ba1 Bb5 Bc3 Bc6 Bg7
24/42  00:20  177000K   8788K  +1.95 Be4 Ba1 f5 Kf2 Kd6 Bd4 a3 g3 Kc6 Ke3 Kb5 Bc3 Kc4 Bh8 Bf3 Ba1 Bg4 Kd2 d4 Kc1 Bh3 Kd2 Bg2 Kc2 d3+ Kd2 Bc6 Kd1 Bb5 Kd2 Ba4 Ke3 Bc6 Kd2 Ba4
25/43  00:29  268136K   8993K  +1.95 Be4 Ba1 f5 Kf2 Kd6 Bd4 Kc6 g3 Kb5 Ke3 Kc4 Ba1 a3 Be5 Bf3 Ba1 Bd1 Kd2 Bg4 Kc2 Bf3 Kb1 d4 Kc2 d3+ Kd2 Bc6 Kd1 Bb5 Kd2 Ba4 Ke3 Bc6 Kd2 Ba4
26/46  01:11  699471K   9734K  +1.95 Be4 g3 f5 Kf2 Kd6 Bb4+ Kc6 Ba3 d4 Ke2 Kb5 Kd2 Kc4 Be7 d3 Kc1 Kb3 Bc5 a3 Bd4 Kc4 Ba1 Bd5 Bf6 Bf3 Kd2 Bc6 Bc3 Bd7 Bf6 Bb5 Bc3 Bd7
27/51  01:33  906708K   9662K  +1.95 Be4 g3 f5 Kf2 Kd6 Bb4+ Kc6 Ba3 d4 Ke2 Kb5 Kd2 Kc4 Be7 d3 Kc1 Kb3 Bc5 a3 Bd4 Kc4 Ba1 Bf3 Kd2 Bd5 Be5 Bc6 Bf6 Bb5 Bc3 Ba4 Ba1 a2 Be5 Bc6 Ba1 Bg2 Bc3 Bc6 Bg7 Be8 Ba1 Bb5 Bc3 Bc6
27/60+ 02:20    1394M   9905K  +2.07 Bh3 gxh3 Kf5 Kf2 Ke4 Bxf6 d4 h5 gxh5 Be7 Kd3 Bc5 Kc4 Ba3 Kc3 Ke1 Kc2 Be7 d3 Bb4 a3 Kf2 a2
27/60+ 02:42    1627M  10018K  +2.25 Bh3 gxh3 Kf5 Kf2 Ke4 Bxf6 d4 h5 gxh5 Be7 Kd3 Bc5 Kc4 Ba3 Kc3 Ke1 Kc2 Be7 d3 Bb4 a3 Kf2 a2
27/63+ 03:18    2031M  10238K  +2.52 Bh3 gxh3 Kf5 Kf2 Ke4 Bxf6 d4 h5 gxh5 Be7 Kd3 Bc5 Kc4 Ba3 Kc3 Ke1 Kc2 Be7 d3 Bb4 a3 Kf2 a2
27/66+ 04:41    2990M  10618K  +2.92 Bh3 gxh3 Kf5 Kf2 Ke4 Bxf6 d4 h5 gxh5 Be7 Kd3 Bc5 Kc4 Ba3 Kc3 Ke1 Kc2 Be7 d3 Bb4 a3 Kf2 a2
27/70+ 08:12    5560M  11284K  +3.53 Bh3 gxh3 Kf5 Kf2 Ke4 Bxf6 d4 h5 gxh5 Be7 Kd3 Bc5 Kc4 Ba3 Kc3 Ke1 Kc2 Be7 d3 Bb4 a3 Kf2 a2
27/74  15:20   10721M  11648K  +4.42 Bh3 Kf2 Kf5 Ke3 Bxg2 Bb2 Ke6 Kf4 f5 Kg5 d4 Bxd4 a3 Ba1 a2 Bh8 Kd5 Kf4 Be4 Ke3 Kc4 Be5 Kb3 Kf4 Kc2 Ke3 Kb1 Bh8 a1=Q Bxa1 Kxa1 Kd4 Kb1 Ke5 Kc2 Kf4 Kb2 Kg5 Kb1 Kf4 Kc2 Ke3 Bc6 Kf4 Kd3
Top
User avatar
Don
Posts: 5106
Joined: Tue Apr 29, 2008 4:27 pm

Re: Shirov's Bh3, Houdini solves it

Post by Don »

Test sets must be upgraded every few years. A good set for a top program should be difficult enough that a good percentage of the positions would take an impractical amount of time to solve by the top programs.

Also, they must be accurate - in past years sets would come out that had all sort of alternate solutions not considered. Also, the key move (or moves) should be the only one that wins or draws, or the only one that avoids a loss. The reason for that is that computers look deeper and deeper and there comes a point that a clever tactic that wins in an already winning position may not be particularly impressive to a computer program. A future program might even see the tactic and choose a different path that wins material later in the game. You should not be penalized for playing a winning move and it may not be easy for a general problem set to determine if a computer see's the tactic if it's not required to win.

Vinvin wrote:On Intel 3 cores @ 3.1 Ghz

FEN: 8/8/4kpp1/3p1b2/p6P/2B5/6P1/6K1 b - - 0 1

Houdini_15_w32:
...
30/56 14:11 2.726.080.589 3.202.000 +1,41 Ff5-e4 Rg1-f2
f6-f5 Fc3-d4 Re6-d7 g2-g3 Rd7-c6 Rf2-e3 Rc6-b5 Fd4-b2 Rb5-c4 Re3-d2
d5-d4 Rd2-e2 d4-d3+ Re2-d2 Rc4-b3 Fb2-d4 a4-a3 Rd2-c1 Rb3-c4 Fd4-e5
a3-a2 Fe5-a1 Fe4-c6 Fa1-b2 Rc4-b3 Fb2-a1 Rb3-b4 Rc1-d2 Rb4-c4 Fa1-f6
Fc6-a4 Ff6-a1 Fa4-c2 Fa1-c3 Rc4-b3 Fc3-d4 Rb3-b4
30/67+ 24:59 4.913.246.933 3.277.000 +1,49 Ff5-h3
30/78+ 1:36:56 21.150.955.106 3.636.000 +5,54 Ff5-h3
30/87 5:12:05 75.495.674.421 4.031.000 +5,37 Ff5-h3
Rg1-f2 Re6-f5 g2-g3 Rf5-e4 Fc3-b2 Re4-d3 Rf2-f3 Fh3-f1 Rf3-f2 Ff1-e2
Rf2-e1 f6-f5 Fb2-a3 Rd3-c2 Fa3-c5 Fe2-g4 Fc5-e3 Fg4-h5 Re1-f2 Rc2-c3
Rf2-e1 d5-d4 Fe3xd4+ Rc3xd4 Re1-d2 a4-a3 Rd2-c2 a3-a2
31/87 5:31:41 80.019.827.765 4.020.000 +5,37 Ff5-h3
Rg1-f2 Re6-f5 g2-g3 Rf5-e4 Fc3-b2 Re4-d3 Rf2-f3 Fh3-f1 Rf3-f2 Ff1-e2
Rf2-g2 Fe2-h5 Rg2-f1 Fh5-g4 Rf1-g1 d5-d4 Rg1-f2 f6-f5 Rf2-f1 Rd3-c2
Fb2xd4 a4-a3 Rf1-f2 a3-a2 Fd4-e5 Rc2-b1 Rf2-e3 a2-a1D Fe5xa1 Rb1xa1
Re3-f2 Ra1-b2 Rf2-e1 Rb2-c3
32/87 6:06:50 87.684.517.651 3.983.000 +5,49 Ff5-h3
Rg1-f2 Re6-f5 g2-g3 Rf5-e4 Fc3-b2 Re4-d3 Rf2-f3 Fh3-f1 Rf3-f2 Ff1-e2
Rf2-g2 Fe2-h5 Rg2-f2 f6-f5 Rf2-g2 Fh5-g4 Rg2-f2 Fg4-e2 Rf2-g1 Fe2-f3
Rg1-f1 Ff3-h5 Rf1-e1 Fh5-e2 Fb2-a3 Fe2-f3 Re1-f2 Ff3-g4 Fa3-b2 Fg4-e2
33/87+ 7:26:21 105.294.949.142 3.931.000 +5,72 Ff5-h3
33/87 7:49:08 110.308.795.417 3.918.000 +5,71 Ff5-h3
Rg1-f2 Re6-f5 g2-g3 Rf5-e4 Fc3-b2 Re4-d3 Rf2-f3 Fh3-f1 Rf3-f2 Ff1-e2
Rf2-g2 Fe2-h5 Rg2-f2 f6-f5 Rf2-f1 Fh5-e2+ Rf1-e1 Fe2-f3 Re1-f1 Ff3-h5
Rf1-f2 Fh5-g4 Rf2-e1 Fg4-f3 Re1-f1

24m59s to find the move but 1h12m more to solve the fail-high.
Robert, is the latest Houdini faster to resolve the fail-high ?


BTW, I prepare a new Hard test set (hardest from hard-CCC, the-nightmare-game, MGV77, 100-brillante Schachzug and hard positions from Rybka forum). I hope to get ready in 1 week.
Capital punishment would be more effective as a preventive measure if it were administered prior to the crime.
Top
User avatar
AdminX
Posts: 6340
Joined: Mon Mar 13, 2006 2:34 pm
Location: Acworth, GA

Re: Shirov's Bh3, Houdini solves it

Post by AdminX »

Don wrote:Test sets must be upgraded every few years. A good set for a top program should be difficult enough that a good percentage of the positions would take an impractical amount of time to solve by the top programs.

Also, they must be accurate - in past years sets would come out that had all sort of alternate solutions not considered. Also, the key move (or moves) should be the only one that wins or draws, or the only one that avoids a loss. The reason for that is that computers look deeper and deeper and there comes a point that a clever tactic that wins in an already winning position may not be particularly impressive to a computer program. A future program might even see the tactic and choose a different path that wins material later in the game. You should not be penalized for playing a winning move and it may not be easy for a general problem set to determine if a computer see's the tactic if it's not required to win.

Vinvin wrote:On Intel 3 cores @ 3.1 Ghz

FEN: 8/8/4kpp1/3p1b2/p6P/2B5/6P1/6K1 b - - 0 1

Houdini_15_w32:
...
30/56 14:11 2.726.080.589 3.202.000 +1,41 Ff5-e4 Rg1-f2
f6-f5 Fc3-d4 Re6-d7 g2-g3 Rd7-c6 Rf2-e3 Rc6-b5 Fd4-b2 Rb5-c4 Re3-d2
d5-d4 Rd2-e2 d4-d3+ Re2-d2 Rc4-b3 Fb2-d4 a4-a3 Rd2-c1 Rb3-c4 Fd4-e5
a3-a2 Fe5-a1 Fe4-c6 Fa1-b2 Rc4-b3 Fb2-a1 Rb3-b4 Rc1-d2 Rb4-c4 Fa1-f6
Fc6-a4 Ff6-a1 Fa4-c2 Fa1-c3 Rc4-b3 Fc3-d4 Rb3-b4
30/67+ 24:59 4.913.246.933 3.277.000 +1,49 Ff5-h3
30/78+ 1:36:56 21.150.955.106 3.636.000 +5,54 Ff5-h3
30/87 5:12:05 75.495.674.421 4.031.000 +5,37 Ff5-h3
Rg1-f2 Re6-f5 g2-g3 Rf5-e4 Fc3-b2 Re4-d3 Rf2-f3 Fh3-f1 Rf3-f2 Ff1-e2
Rf2-e1 f6-f5 Fb2-a3 Rd3-c2 Fa3-c5 Fe2-g4 Fc5-e3 Fg4-h5 Re1-f2 Rc2-c3
Rf2-e1 d5-d4 Fe3xd4+ Rc3xd4 Re1-d2 a4-a3 Rd2-c2 a3-a2
31/87 5:31:41 80.019.827.765 4.020.000 +5,37 Ff5-h3
Rg1-f2 Re6-f5 g2-g3 Rf5-e4 Fc3-b2 Re4-d3 Rf2-f3 Fh3-f1 Rf3-f2 Ff1-e2
Rf2-g2 Fe2-h5 Rg2-f1 Fh5-g4 Rf1-g1 d5-d4 Rg1-f2 f6-f5 Rf2-f1 Rd3-c2
Fb2xd4 a4-a3 Rf1-f2 a3-a2 Fd4-e5 Rc2-b1 Rf2-e3 a2-a1D Fe5xa1 Rb1xa1
Re3-f2 Ra1-b2 Rf2-e1 Rb2-c3
32/87 6:06:50 87.684.517.651 3.983.000 +5,49 Ff5-h3
Rg1-f2 Re6-f5 g2-g3 Rf5-e4 Fc3-b2 Re4-d3 Rf2-f3 Fh3-f1 Rf3-f2 Ff1-e2
Rf2-g2 Fe2-h5 Rg2-f2 f6-f5 Rf2-g2 Fh5-g4 Rg2-f2 Fg4-e2 Rf2-g1 Fe2-f3
Rg1-f1 Ff3-h5 Rf1-e1 Fh5-e2 Fb2-a3 Fe2-f3 Re1-f2 Ff3-g4 Fa3-b2 Fg4-e2
33/87+ 7:26:21 105.294.949.142 3.931.000 +5,72 Ff5-h3
33/87 7:49:08 110.308.795.417 3.918.000 +5,71 Ff5-h3
Rg1-f2 Re6-f5 g2-g3 Rf5-e4 Fc3-b2 Re4-d3 Rf2-f3 Fh3-f1 Rf3-f2 Ff1-e2
Rf2-g2 Fe2-h5 Rg2-f2 f6-f5 Rf2-f1 Fh5-e2+ Rf1-e1 Fe2-f3 Re1-f1 Ff3-h5
Rf1-f2 Fh5-g4 Rf2-e1 Fg4-f3 Re1-f1

24m59s to find the move but 1h12m more to solve the fail-high.
Robert, is the latest Houdini faster to resolve the fail-high ?


BTW, I prepare a new Hard test set (hardest from hard-CCC, the-nightmare-game, MGV77, 100-brillante Schachzug and hard positions from Rybka forum). I hope to get ready in 1 week.
+1 ,No make that three! :lol:
"Good decisions come from experience, and experience comes from bad decisions."
__________________________________________________________________
Ted Summers
Top
lech
Posts: 1136
Joined: Sun Feb 14, 2010 10:02 pm

Re: Shirov's Bh3, Houdini solves it

Post by lech »

[d]8/8/4kpp1/3p1b2/p6P/2B5/6P1/6K1 b - -
Sting SF PV (32-bit) needs 6 minutes on my slow computer (~36 second on Louis's computer :wink: ).

Code: Select all

info depth 38 seldepth 47 multipv 1 score cp 286 nodes 192936048 nps 1084385 tim
e 177922 pv a4a3 g1f2 a3a2 g2g3 f5e4 c3b2 e4c2 f2e3 e6f5 b2c3 c2d1 c3a1 f5g4 a1f
6 g4g3 e3d2 d1f3 d2c2 f3e4 c2d2 g3g4 d2c1 e4d3 c1d2 d3f1 d2e3 f1c4 e3d2 g4f5 f6b
2 f5e4 b2a1 d5d4 a1b2 c4e6 b2a1 e6d5 a1b2 d5b3 b2a1 b3c4
info depth 39
info currmove a4a3 currmovenumber 1
info nodes 206656953 nps 1083658 time 190703
info currmove f5e4 currmovenumber 2
info nodes 207606257 nps 1084281 time 191469
info currmove d5d4 currmovenumber 3
info currmove f5g4 currmovenumber 4
info currmove e6f7 currmovenumber 5
info currmove e6d6 currmovenumber 6
info nodes 211264009 nps 1085146 time 194687
info currmove g6g5 currmovenumber 7
info currmove f5b1 currmovenumber 8
info currmove e6e7 currmovenumber 9
info currmove f5c2 currmovenumber 10
info currmove f5d3 currmovenumber 11
info currmove e6d7 currmovenumber 12
info currmove f5h3 currmovenumber 13
info depth 39 seldepth 0 multipv 1 score cp 298 lowerbound nodes 345198532 nps 9
35180 time 369125 pv f5h3 c3b2 e6f5 g1f2 f5e4 f2g3 h3g2 g3g2 d5d4 g2g3 f6f5 g3f2
 d4d3 f2e1 f5f4 e1d2 f4f3
info nodes 345198532 nps 935140 time 369141
info currmove f5h3 currmovenumber 1
info depth 39 seldepth 0 multipv 1 score cp 311 lowerbound nodes 468587246 nps 8
39715 time 558031 pv f5h3 c3b2 e6f5 g1f2 f5e4 g2g3 e4d3 f2f3 h3f5 f3f2 f5g4 h4h5
 g6h5 b2f6
info nodes 468587246 nps 839691 time 558047
info currmove f5h3 currmovenumber 1
Maybe, I can't be friendly, but let me be useful.
Top
ernest
Posts: 2041
Joined: Wed Mar 08, 2006 8:30 pm

Re: Shirov's Bh3, Houdini solves it

Post by ernest »

lech wrote:Sting SF PV (32-bit)
I know Sting.2x32_VE and Sting SF 120610, but what is Sting SF PV (32-bit)?

here on dualcore @3GHz:

Code: Select all

8/8/4kpp1/3p1b2/p6P/2B5/6P1/6K1 b - - 0 1
Analysis by Sting SF 2_VE 32bit:
.............
1...Be4 2.Kf2 f5 3.g3 Kd6 4.Bb4+ Kc6 5.Ke3 Kb5 6.Bf8 Kc4 7.Kd2 Kb3 8.Bg7 a3 9.Bf6 Kc4 10.Ke3 a2 11.Ba1 Bd3 12.Bd4 Bb1 13.Bb2 Kb3 14.Bd4 Bc2 15.Bg7 Kb4 16.Ba1 Be4 17.Be5 Ka4 18.Kd2 Kb3 19.Bd4 Kb4 20.Ba1 Kb3 21.Be5 Kc4 22.Ke3 Kb4 23.Ba1 Kc4 24.Be5 
  -+  (-2.54)   Depth: 37/43   00:00:25  60879kN
1...Bh3 2.Kh2 Bg4 3.Kg3 Kf5 4.Bb4 d4 5.Kf2 Ke4 6.Ke1 Kd3 7.Kf2 f5 8.Bc5 Kc4 9.Be7 d3 10.Ke3 Kc3 
  -+  (-2.62 !)   Depth: 38   00:01:26  223mN
1...Bh3 2.Kh2 Bg4 3.Kg3 Kf5 4.Bb4 d4 5.Kf2 Ke4 6.Ke1 Kd3 7.Kf2 f5 8.Bc5 Kc4 9.Be7 d3 10.Ke3 Kc3 
  -+  (-2.70 !)   Depth: 38   00:01:38  256mN
1...Bh3 2.Kh2 Bg4 3.Kg3 Kf5 4.Bb4 d4 5.Kf2 Ke4 6.Ke1 Kd3 7.Kf2 f5 8.Kg1 Kc2 
  -+  (-2.82 !)   Depth: 38   00:02:01  318mN
1...Bh3 2.Kf1 Kf5 3.Ke2 Bxg2 4.Kd2 g5 5.h5 Be4 6.h6 Kg6 7.h7 Kxh7 8.Bxf6 Kg6 9.Ba1 Kf5 10.Bd4 g4 11.Kc3 g3 
  -+  (-3.01 !)   Depth: 38   00:02:24  377mN
Top
lech
Posts: 1136
Joined: Sun Feb 14, 2010 10:02 pm

Re: Shirov's Bh3, Houdini solves it

Post by lech »

ernest wrote:I know Sting.2x32_VE and Sting SF 120610, but what is Sting SF PV (32-bit)?
Sting.2x32_VE and Sting SF 120610 are the same versions but compiled but other persons.
Sting SF PV is my private version (not published yet) . As you can see it can be slower in some positions but it gives more chances to solve many difficult endgames and zugzwang positions.
Belive me, it is not an easy question.
Maybe, I can't be friendly, but let me be useful.
Top
Paloma
Posts: 1167
Joined: Thu Dec 25, 2008 9:07 pm
Full name: Herbert L

Re: Shirov's Bh3, Houdini solves it

Post by Paloma »

Vinvin wrote: BTW, I prepare a new Hard test set (hardest from hard-CCC, the-nightmare-game, MGV77, 100-brillante Schachzug and hard positions from Rybka forum). I hope to get ready in 1 week.
Great Vinvin,

I'm looking forward to it.
Top
User avatar
Houdini
Posts: 1471
Joined: Tue Mar 16, 2010 12:00 am

Re: Shirov's Bh3, Houdini solves it

Post by Houdini »

Vinvin wrote:On Intel 3 cores @ 3.1 Ghz

FEN: 8/8/4kpp1/3p1b2/p6P/2B5/6P1/6K1 b - - 0 1

Houdini_15_w32:
...
30/56 14:11 2.726.080.589 3.202.000 +1,41 Ff5-e4 Rg1-f2
f6-f5 Fc3-d4 Re6-d7 g2-g3 Rd7-c6 Rf2-e3 Rc6-b5 Fd4-b2 Rb5-c4 Re3-d2
d5-d4 Rd2-e2 d4-d3+ Re2-d2 Rc4-b3 Fb2-d4 a4-a3 Rd2-c1 Rb3-c4 Fd4-e5
a3-a2 Fe5-a1 Fe4-c6 Fa1-b2 Rc4-b3 Fb2-a1 Rb3-b4 Rc1-d2 Rb4-c4 Fa1-f6
Fc6-a4 Ff6-a1 Fa4-c2 Fa1-c3 Rc4-b3 Fc3-d4 Rb3-b4
30/67+ 24:59 4.913.246.933 3.277.000 +1,49 Ff5-h3
30/78+ 1:36:56 21.150.955.106 3.636.000 +5,54 Ff5-h3
30/87 5:12:05 75.495.674.421 4.031.000 +5,37 Ff5-h3
Rg1-f2 Re6-f5 g2-g3 Rf5-e4 Fc3-b2 Re4-d3 Rf2-f3 Fh3-f1 Rf3-f2 Ff1-e2
Rf2-e1 f6-f5 Fb2-a3 Rd3-c2 Fa3-c5 Fe2-g4 Fc5-e3 Fg4-h5 Re1-f2 Rc2-c3
Rf2-e1 d5-d4 Fe3xd4+ Rc3xd4 Re1-d2 a4-a3 Rd2-c2 a3-a2
31/87 5:31:41 80.019.827.765 4.020.000 +5,37 Ff5-h3
Rg1-f2 Re6-f5 g2-g3 Rf5-e4 Fc3-b2 Re4-d3 Rf2-f3 Fh3-f1 Rf3-f2 Ff1-e2
Rf2-g2 Fe2-h5 Rg2-f1 Fh5-g4 Rf1-g1 d5-d4 Rg1-f2 f6-f5 Rf2-f1 Rd3-c2
Fb2xd4 a4-a3 Rf1-f2 a3-a2 Fd4-e5 Rc2-b1 Rf2-e3 a2-a1D Fe5xa1 Rb1xa1
Re3-f2 Ra1-b2 Rf2-e1 Rb2-c3
32/87 6:06:50 87.684.517.651 3.983.000 +5,49 Ff5-h3
Rg1-f2 Re6-f5 g2-g3 Rf5-e4 Fc3-b2 Re4-d3 Rf2-f3 Fh3-f1 Rf3-f2 Ff1-e2
Rf2-g2 Fe2-h5 Rg2-f2 f6-f5 Rf2-g2 Fh5-g4 Rg2-f2 Fg4-e2 Rf2-g1 Fe2-f3
Rg1-f1 Ff3-h5 Rf1-e1 Fh5-e2 Fb2-a3 Fe2-f3 Re1-f2 Ff3-g4 Fa3-b2 Fg4-e2
33/87+ 7:26:21 105.294.949.142 3.931.000 +5,72 Ff5-h3
33/87 7:49:08 110.308.795.417 3.918.000 +5,71 Ff5-h3
Rg1-f2 Re6-f5 g2-g3 Rf5-e4 Fc3-b2 Re4-d3 Rf2-f3 Fh3-f1 Rf3-f2 Ff1-e2
Rf2-g2 Fe2-h5 Rg2-f2 f6-f5 Rf2-f1 Fh5-e2+ Rf1-e1 Fe2-f3 Re1-f1 Ff3-h5
Rf1-f2 Fh5-g4 Rf2-e1 Fg4-f3 Re1-f1

24m59s to find the move but 1h12m more to solve the fail-high.
Robert, is the latest Houdini faster to resolve the fail-high ?
Yes, the fail-high trick works very well here: fail high after 2:33 (1.1 billion nodes), complete PV after 3:35 (1.5 billion nodes).
Below the analysis with the current Houdini dev on 3 cores of a standard core i5-750, with 2048 MB of hash:

Code: Select all

8/8/4kpp1/3p1b2/p6P/2B5/6P1/6K1 b - -

Engine: Houdini DEV 3T EGTB (2048 MB)
by Robert Houdart

28/50  0:56   -1.37    1...Be4 2.g3 f5 3.Kf2 Kd6 4.Bb4+ Kc6 
                       5.Ke2 Kb5 6.Bf8 Kc4 7.Kd2 d4 8.Bd6 d3 
                       9.Be7 Kb3 10.Bf6 a3 11.Kc1 Bh1 
                       12.Bd4 Ka2 13.Be3 Bb7 14.Bf4 Bf3 (380.899.736) 6771 TB:480 

29/50  1:23   -1.37    1...Be4 2.g3 f5 3.Kf2 Kd6 4.Bb4+ Kc6 
                       5.Ke2 Kb5 6.Bf8 Kc4 7.Kd2 d4 8.Bd6 d3 
                       9.Be7 Kb3 10.Bf6 a3 11.Kc1 Bh1 
                       12.Bd4 Ka2 13.Be3 Bb7 14.Bf4 Bf3 (569.305.857) 6780 
                       TB:1.996 

29/66  2:33   -1.45++  1...Bh3 (1.067.275.340) 6930 
                       TB:8.194 

29/66  2:47   -1.61++  1...Bh3 (1.168.173.148) 6975 
                       TB:8.765 

26/66  2:52   -1.97++  1...Bh3 (1.204.615.635) 6985 
                       TB:8.829 

26/66  3:05   -2.92++  1...Bh3 (1.298.645.183) 7010 
                       TB:8.938 

26/66  3:35   -3.96    1...Bh3 2.g3 Kf5 3.Kf2 Ke4 4.Ke2 a3 
                       5.Bb4 a2 6.Bc3 d4 7.Ba1 Be6 8.Kd2 Bg4 
                       9.Bb2 f5 10.Ba1 d3 11.Bc3 Bh5 12.Bg7 Be2 
                       13.Ba1 Kf3 14.Be5 Bf1 (1.522.155.628) 7063 
                       TB:9.589 

27/66  3:47   -3.77--  1...Bh3 2.g3 (1.609.053.638) 7083 
                       TB:10.469 

27/66  3:50   -3.40--  1...Bh3 2.g3 (1.630.813.610) 7083 
                       TB:11.215 

27/67  4:00   -4.12++  1...Bh3 (1.710.238.541) 7105 
                       TB:11.626 

27/67  4:02   -4.10    1...Bh3 2.Kf2 Kf5 3.Ke3 Bxg2 4.Bb2 Be4 
                       5.Kf2 Ke6 6.Kg3 f5 7.Bd4 a3 8.Kf4 a2 
                       9.Kg5 Kd6 10.Kf4 Bc2 11.Ba1 Ke6 
                       12.h5 gxh5 13.Kg3 Kd6 14.Bb2 Bb1 (1.723.909.906) 7103 
                       TB:12.051 

28/67  4:17   -4.27++  1...Bh3 (1.838.851.021) 7143 
                       TB:12.780 

28/67  4:33   -4.61++  1...Bh3 (1.965.514.862) 7179 
                       TB:13.432 

28/67  4:56   -5.51++  1...Bh3 (2.135.543.862) 7209 
                       TB:14.422 

28/68  5:38   -7.63    1...Bh3 2.Kf2 Kf5 3.Ke3 Bxg2 4.Bb2 Be4 
                       5.Kf2 Ke6 6.Kg3 f5 7.Bd4 a3 8.Kf4 a2 
                       9.Kg5 Kd6 10.Kf4 Kc6 11.Ba1 Kc5 
                       12.Ke3 Kc4 13.Bb2 d4+ 14.Bxd4 f4+ (2.459.980.802) 7273 
                       TB:17.413
Top

Return to “Computer Chess Club: General Topics”