At a glance you may think this is a draw. White wins!

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

Moderators: hgm, Rebel, chrisw

ernest
Posts: 2041
Joined: Wed Mar 08, 2006 8:30 pm

Re: At a glance you may think this is a draw. White wins!

Post by ernest »

duncan wrote:if 8 black went kb5, finalgen shows it as unsolved. would you be able to show the win for white. ?
[d]8/5p2/8/PkP4p/1P4p1/8/6K1/8 w - - 0 1
Analysis by Houdini 3 w32 Nlm:
9.Kg3 f5 10.Kf2 h4 11.Kg2 f4 12.Kg1 h3 13.Kh2 f3 14.Kg3 h2 15.Kxh2 f2 16.Kg2 g3 17.Kf1 Ka6 18.c6 Ka7 19.b5 Ka8 20.a6 Kb8 21.b6 Ka8 22.Kg2 f1R 23.Kxf1 Kb8 24.Kg2
+- (3.91) Depth: 19/36 00:00:00 97kN
9.Kg3 f5 10.Kf2 h4 11.Kg2 f4 12.Kg1 h3 13.Kh2 f3 14.Kg3 h2 15.Kxh2 f2 16.Kg2 g3 17.Kf1 Ka6 18.c6 Ka7 19.b5 Ka8 20.a6 Kb8 21.b6 Ka8 22.Kg2 f1R 23.Kxf1 Kb8 24.Kg2
+- (3.91) Depth: 20/36 00:00:00 154kN
9.Kg3
+- (4.07 !) Depth: 21/36 00:00:00 192kN
9.Kg3
+- (4.39 !) Depth: 21/36 00:00:00 193kN
9.Kg3
+- (5.24 !) Depth: 21/36 00:00:00 193kN
9.Kg3 f5 10.Kf2 h4 11.Kg2 f4 12.Kg1 h3 13.Kh2 f3 14.Kg3 h2 15.Kxh2 f2 16.Kg2 g3 17.Kf1 Kc6 18.a6 Kc7 19.b5 Kc8 20.b6 g2+ 21.Kxg2 Kb8 22.Kxf2 Kc8 23.Kf3 Kb8 24.Ke4 Kc8 25.c6
+- (6.21) Depth: 21/41 00:00:00 1356kN, tb=4
9.Kg3 f5
+- (5.95 --) Depth: 22/41 00:00:00 1533kN, tb=5
9.Kg3 f5
+- (5.43 --) Depth: 22/41 00:00:00 1718kN, tb=5
9.Kg3
+- (6.47 !) Depth: 22/41 00:00:00 2999kN, tb=13
9.Kg3 f5 10.Kf2 h4 11.Kg2 f4 12.Kg1 h3 13.Kh2 f3 14.Kg3 h2 15.Kxh2 Kc6 16.a6 Kc7 17.b5 Kb8 18.Kg3 Ka7 19.Kf2 Ka8 20.b6 g3+ 21.Kxg3 Kb8 22.Kxf3 Kc8 23.Ke4 Kb8 24.c6 Ka8 25.Kd5
+- (6.57) Depth: 22/43 00:00:01 4055kN, tb=13
9.Kg3
+- (6.84 !) Depth: 23/43 00:00:01 6088kN, tb=44
9.Kg3 f5 10.Kf2 h4 11.Kg2 f4 12.Kg1 h3 13.Kh2 f3 14.Kg3 h2 15.Kxh2 Kc6 16.a6 Kc7 17.b5 Kb8 18.Kg3 Ka7 19.Kf2 Ka8 20.b6 g3+ 21.Kxg3 Kb8 22.Kxf3 Kc8 23.Ke4 Kb8 24.c6 Ka8 25.Kd5
+- (7.11) Depth: 23/46 00:00:02 8175kN, tb=72
9.Kg3
+- (7.40 !) Depth: 24/46 00:00:03 12547kN, tb=140
9.Kg3
+- (7.99 !) Depth: 24/46 00:00:05 17806kN, tb=180
9.Kg3 f5 10.Kf2 h4 11.Kg2 f4 12.Kg1 h3 13.Kh2 f3 14.Kg3 h2 15.Kxh2 Kc6 16.a6 f2 17.Kg2 g3 18.b5+ Kxb5 19.a7 Kxc5 20.a8Q Kb6 21.Kf1 Kc7 22.Qa6 Kd7 23.Qd3+ Ke6 24.Qxg3 Kf6 25.Kxf2 Ke6 26.Ke3 Kd5 27.Qg5+ Kc4 28.Qa5 Kb3 29.Qd5+ Kb4 30.Kd4
+- (9.42) Depth: 24/48 00:00:08 32002kN, tb=260
9.Kg3 f5 10.Kf2 h4 11.Kg2 f4 12.Kg1 h3 13.Kh2 f3 14.Kg3 h2 15.Kxh2 f2 16.Kg2 f1R 17.Kxf1 g3 18.Kg2 Kxb4 19.a6 Kxc5 20.a7 Kb6 21.a8Q Kb5 22.Qd5+ Kb4 23.Qd3 Kc5 24.Kxg3 Kb4 25.Kf4 Kc5 26.Qe4 Kb6 27.Qd5 Kc7 28.Ke5 Kb6 29.Kd6 Ka6
+- (9.65) Depth: 25/48 00:00:12 45852kN, tb=510
9.Kg3 f5 10.Kf2 h4 11.Kg2 f4 12.Kg1 h3 13.Kh2 f3 14.Kg3 h2 15.Kxh2 f2 16.Kg2 f1R 17.Kxf1 g3 18.Kg2 Kxb4 19.a6 Kxc5 20.a7 Kb6 21.a8Q Kb5 22.Qd5+ Kb4 23.Qd3 Kc5 24.Kxg3 Kb4 25.Kf2 Ka4 26.Qc3 Kb5 27.Ke3 Kb6 28.Qe5 Kc6 29.Kd4 Kb6 30.Qc5+ Kb7 31.Qd6 Ka7 32.Qc6 Kb8 33.Qd7
+- (9.80) Depth: 26/48 00:00:15 59297kN, tb=999
9.Kg3 f5 10.Kf2 h4 11.Kg2 f4 12.Kg1 h3 13.Kh2 f3 14.Kg3 h2 15.Kxh2 f2 16.Kg2 f1R 17.Kxf1 g3 18.Kg2 Kxb4 19.c6 Kxa5 20.c7 Ka4 21.c8Q Kb4 22.Kxg3 Kb3 23.Qc5 Ka4 24.Qc4+ Ka5 25.Qb3 Ka6 26.Qb4 Ka7 27.Qd4+ Kb7 28.Qc5 Ka6 29.Qb4 Ka7
+- (9.90) Depth: 27/50 00:00:20 77477kN, tb=1815
9.Kg3
+- (10.31 !) Depth: 28/53 00:00:28 109mN, tb=2849
9.Kg3
+- (11.13 !) Depth: 28/59 00:00:36 143mN, tb=3257
9.Kg3
+- (13.48 !) Depth: 28/59 00:00:55 218mN, tb=3672
9.Kg3
+- (20.43 !) Depth: 28/59 00:02:01 515mN, tb=4319
9.Kg3 f5 10.Kf2 h4 11.Kg2 f4 12.Kg1 h3 13.Kh2 f3 14.Kg3 h2 15.Kxh2 f2 16.Kg2 f1R 17.Kxf1 g3 18.Kg2 Ka6 19.c6 Ka7 20.b5 Ka8 21.a6 Kb8 22.Kxg3 Kc8 23.a7 Kc7 24.a8Q Kb6 25.Qb8+ Kc5 26.c7 Kd5 27.Qa8+ Kd6 28.c8Q Ke5 29.b6 Kd6 30.b7 Ke7 31.Qf8+ Ke6 32.b8Q
+- (26.33) Depth: 28/61 00:02:17 583mN, tb=5082
9.Kg3 f5
+- (25.24 --) Depth: 29/61 00:02:19 592mN, tb=6071
9.Kg3 f5
+- (23.05 --) Depth: 29/61 00:02:21 599mN, tb=6466
9.Kg3 f5
+- (16.59 --) Depth: 29/61 00:02:23 608mN, tb=6705
S.Taylor
Posts: 8514
Joined: Thu Mar 09, 2006 3:25 am
Location: Jerusalem Israel

Re: At a glance you may think this is a draw. White wins!

Post by S.Taylor »

Robert Flesher wrote:
Uri Blass wrote:
Robert Flesher wrote:A beautiful position that illustrates the advantage that a single tempo can give. Then zugzwang and black is lost. The key to this position is looking at it yourself, before using an engine. It's very instructive! Some very strong engine(s) also struggle here. Komodo CCT after 5 min cannot see the win on my machine.



[D]4k3/5ppp/8/8/8/8/PPP5/3K4 w - - 0 1
Humans cannot calculate all possible lines so how looking at it by myself can help?

seeing the solution does not help to be able to practically win it because black can practically play a different move.

For example 1.Ke2 Ke7 or 1.Ke2 a5 or 1.Ke2 Kd7 2.Kf3 Kc6

Impossible to calculate all lines and if humans can understand that white wins then I would like to read some strategy that tells me what to do in every position and not a variation.
Hi Uri, I did not say calculate it youself. I said look at the position yourself without an engine. You of all people I thought would understand this. LOL, I guess not!. I studied and learned the secrets of this postion without an engine. I wanted people to move the pieces and try to understand this position. I guess this was the wrong place to post this.
Since i can see by myself why 18...h2 was forced, and that it loses, so all i need now is to try to see if there is any other improvement before that.
Obviously you have already done all that, and found it is a forced win in every variation, and therefore it looks like a wonderful bit of endgame technique to be familiar with.

[Oh! and I am refering to the 32 move variation by Critter which someone put up on the first page here.]
Robert Flesher
Posts: 1280
Joined: Tue Aug 18, 2009 3:06 am

Re: At a glance you may think this is a draw. White wins!

Post by Robert Flesher »

S.Taylor wrote:
Robert Flesher wrote:
Uri Blass wrote:
Robert Flesher wrote:A beautiful position that illustrates the advantage that a single tempo can give. Then zugzwang and black is lost. The key to this position is looking at it yourself, before using an engine. It's very instructive! Some very strong engine(s) also struggle here. Komodo CCT after 5 min cannot see the win on my machine.



[D]4k3/5ppp/8/8/8/8/PPP5/3K4 w - - 0 1
Humans cannot calculate all possible lines so how looking at it by myself can help?

seeing the solution does not help to be able to practically win it because black can practically play a different move.

For example 1.Ke2 Ke7 or 1.Ke2 a5 or 1.Ke2 Kd7 2.Kf3 Kc6

Impossible to calculate all lines and if humans can understand that white wins then I would like to read some strategy that tells me what to do in every position and not a variation.
Hi Uri, I did not say calculate it youself. I said look at the position yourself without an engine. You of all people I thought would understand this. LOL, I guess not!. I studied and learned the secrets of this postion without an engine. I wanted people to move the pieces and try to understand this position. I guess this was the wrong place to post this.
Since i can see by myself why 18...h2 was forced, and that it loses, so all i need now is to try to see if there is any other improvement before that.
Obviously you have already done all that, and found it is a forced win in every variation, and therefore it looks like a wonderful bit of endgame technique to be familiar with.

[Oh! and I am refering to the 32 move variation by Critter which someone put up on the first page here.]

Bingo!
Arpad Rusz
Posts: 273
Joined: Sat Apr 17, 2010 2:34 pm
Location: Budapest

Re: At a glance you may think this is a draw. White wins!

Post by Arpad Rusz »

Arpad Rusz wrote:If one sets the critical position's value to DRAW, FinalGen reports a win for 1.Ke2 and 1.Ke1.
Sorry, I made a mistake in my previous post. You should read:
If one sets the critical position's value to WHITE WINS, FinalGen reports a win for 1.Ke2 and 1.Ke1.
duncan
Posts: 12038
Joined: Mon Jul 07, 2008 10:50 pm

Re: At a glance you may think this is a draw. White wins!

Post by duncan »

Arpad Rusz wrote:
Arpad Rusz wrote:If one sets the critical position's value to DRAW, FinalGen reports a win for 1.Ke2 and 1.Ke1.
Sorry, I made a mistake in my previous post. You should read:
If one sets the critical position's value to WHITE WINS, FinalGen reports a win for 1.Ke2 and 1.Ke1.
is this the critical position you put in?

[D]2Q5/8/8/P7/1P4p1/3k4/7K/5n2 w

duncan
Arpad Rusz
Posts: 273
Joined: Sat Apr 17, 2010 2:34 pm
Location: Budapest

Re: At a glance you may think this is a draw. White wins!

Post by Arpad Rusz »

duncan wrote:is this the critical position you put in?

[D]2Q5/8/8/P7/1P4p1/3k4/7K/5n2 w

duncan
No, I set this position:

[D]8/8/8/PkP5/1P4p1/5pKp/8/8 b - - 0 1

But of course one should get the same result by setting that Q/N position to WHITE WINS.
duncan
Posts: 12038
Joined: Mon Jul 07, 2008 10:50 pm

Re: At a glance you may think this is a draw. White wins!

Post by duncan »

Arpad Rusz wrote:
duncan wrote:is this the critical position you put in?

[D]2Q5/8/8/P7/1P4p1/3k4/7K/5n2 w

duncan
No, I set this position:

[D]8/8/8/PkP5/1P4p1/5pKp/8/8 b - - 0 1

But of course one should get the same result by setting that Q/N position to WHITE WINS.
tried your position, for some reason did not get your results. did you do no 50 rules move ?

duncan
Arpad Rusz
Posts: 273
Joined: Sat Apr 17, 2010 2:34 pm
Location: Budapest

Re: At a glance you may think this is a draw. White wins!

Post by Arpad Rusz »

duncan wrote:tried your position, for some reason did not get your results. did you do no 50 rules move ?
Yes, I did used "No 50 moves rule (Depth To Promotion)".

Now I managed to get more precise results using two seeded positions:
[D]8/8/8/PkP5/1P4p1/5pKp/8/8 b - - 0 1 WHITE WINS
[D]8/8/PkP5/1P4p1/5pKp/8/8/8 w - - 0 1 BLACK WINS

From [D]4k3/5ppp/8/8/8/8/PPP5/3K4 w - - 0 1 I've got the following values:

Ke2 White wins in 24
Ke1 White wins in 25
c4 White wins or Draw
a4 White wins or Draw
b4 Draw
c3 Black wins in 25
b3 Black wins in 25
a3 Black wins in 25
Kd2 Black wins in 25
Kc1 Black wins in 25
duncan
Posts: 12038
Joined: Mon Jul 07, 2008 10:50 pm

Re: At a glance you may think this is a draw. White wins!

Post by duncan »

Arpad Rusz wrote:
duncan wrote:tried your position, for some reason did not get your results. did you do no 50 rules move ?
Yes, I did used "No 50 moves rule (Depth To Promotion)".

Now I managed to get more precise results using two seeded positions:
[D]8/8/8/PkP5/1P4p1/5pKp/8/8 b - - 0 1 WHITE WINS
[D]8/8/PkP5/1P4p1/5pKp/8/8/8 w - - 0 1 BLACK WINS

From [D]4k3/5ppp/8/8/8/8/PPP5/3K4 w - - 0 1 I've got the following values:

Ke2 White wins in 24
Ke1 White wins in 25
c4 White wins or Draw
a4 White wins or Draw
b4 Draw
c3 Black wins in 25
b3 Black wins in 25
a3 Black wins in 25
Kd2 Black wins in 25
Kc1 Black wins in 25
thanks. very useful.
how did you get 2 positions in the finalgen. did you use include in file facility ? (I could not get mine to work)

duncan
Arpad Rusz
Posts: 273
Joined: Sat Apr 17, 2010 2:34 pm
Location: Budapest

Re: At a glance you may think this is a draw. White wins!

Post by Arpad Rusz »

After you click the "Set value" button, you can set another position, etc.

If you want to use a file, that should be a .txt file with FENs. The values are set by using the letters W, D, or L (Win, Draw, Lost) before every FEN string followed by '|'. E.g.:

W|8/8/8/PkP5/1P4p1/5pKp/8/8 b - - 0 1
L|8/8/PkP5/1P4p1/5pKp/8/8/8 w - - 0 1
D|8/8/PkP5/1P3pp1/5K1p/8/8/8 w - - 0 1
D|8/8/8/P1k5/1PP3p1/5pKp/8/8 b - - 0 1