Related to Behting: John Nunn's testposition

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

Moderators: hgm, Harvey Williamson, bob

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
User avatar
Eelco de Groot
Posts: 3909
Joined: Sun Mar 12, 2006 1:40 am
Location: Groningen

Related to Behting: John Nunn's testposition

Post by Eelco de Groot » Sat Aug 11, 2018 7:30 am

I was just wondering, can any chessprogram see a big score here, with or without 7 men tablebases? With them, do you get a Mate number (or would you need additional DTM)? Johhn Nunn wrote, in a ChessBase article: 'The Behting study is sound' By John Nunn

11...Kg2!! This is the move I had missed when I first looked at the position. Black’s surprising triangulation, which in addition unpins the knight, leaves White without a good reply:
Johhn Nunn continues with a list of possible lines with as last item
  • 5
    12. Kd5 Kg3 13. Kxc4 is perhaps the most surprising line. Black simply gives up his pawn in order to activate his pieces. Although White doesn’t appear too badly off, he cannot save the game and the tablebase shows that Black can now mate in 50 moves.
Not having much luck here. At depth 72, with a Kaissa Multi PV = 10 and no tablebases at all, at least the first moves with Kg2 were still okay as per the 7 men database from Marc Bourzutschky, but then it lost even those.


73 249:44 -4.21 1...Qh8 2.Kd5 Qg8+ 3.Kc5 Qc8+ 4.Kd5 Qc7 5.Ne5 Qa5+ 6.Ke4 Qa2 7.Ng4+ Ke1 8.Nf3+ Ke2 9.Nge5 Qb1+ 10.Kf4 Qb6 11.Kf5 Qb4 12.Ke6 Kf2 13.Kd5 Qb5+ 14.Ke4 Qb7+ (48.049.702.515) 3206
73 249:44 -4.21 1...Qh6 2.Kd5 Qa6 3.Kc5 Qc8+ 4.Kd5 Qc7 5.Ne5 Qa5+ 6.Ke4 Qa2 7.Ng4+ Ke1 8.Nf3+ Ke2 9.Nge5 Qb1+ 10.Kf4 Qb6 11.Kf5 Qb4 12.Ke6 Kf2 13.Kd5 Qb5+ 14.Ke4 Qb7+ (48.049.702.515) 3206
73 249:44 -4.21 1...Qh7+ 2.Kd5 Qc7 3.Ne5 Qa5+ 4.Ke4 Qa2 5.Ng4+ Ke1 6.Nf3+ Ke2 7.Nge5 Qb1+ 8.Kf4 Qb6 9.Kf5 Qb4 10.Ke6 Kf2 11.Kd5 Qb5+ 12.Ke4 Qb7+ 13.Kd4 Qb3 14.Ke4 Kf1 (48.049.702.515) 3206
73 249:44 -4.21 1...Qb1+ 2.Kd5 Qb4 3.Ne5 Qa5+ 4.Ke4 Qa2 5.Ng4+ Ke1 6.Nf3+ Ke2 7.Nge5 Qb1+ 8.Kf4 Qb6 9.Kf5 Qb4 10.Ke6 Kf2 11.Kd5 Qb5+ 12.Ke4 Qb7+ 13.Kd4 Qb3 14.Ke4 Kf1 (48.049.702.515) 3206
73 249:44 -4.21 1...Qa1 2.Kd5 Qa6 3.Kc5 Qc8+ 4.Kd5 Qc7 5.Ne5 Qa5+ 6.Ke4 Qa2 7.Ng4+ Ke1 8.Nf3+ Ke2 9.Nge5 Qb1+ 10.Kf4 Qb6 11.Kf5 Qb4 12.Ke6 Kf2 13.Kd5 Qb5+ 14.Ke4 Qb7+ (48.049.702.515) 3206
73 249:44 -4.21 1...Qg2 2.Kd5 Qg8+ 3.Kc5 Qc8+ 4.Kd5 Qc7 5.Ne5 Qa5+ 6.Ke4 Qa2 7.Ng4+ Ke1 8.Nf3+ Ke2 9.Nge5 Qb1+ 10.Kf4 Qb6 11.Kf5 Qb4 12.Ke6 Kf2 13.Kd5 Qb5+ 14.Ke4 Qb7+ (48.049.702.515) 3206
73 249:44 -4.21 1...Qh5 2.Ne5 Qh1+ 3.Nef3 Qg2 4.Kd5 Qg8+ 5.Kc5 Qc8+ 6.Kd5 Qc7 7.Ne5 Qa5+ 8.Ke4 Qa2 9.Ng4+ Ke1 10.Nf3+ Ke2 11.Nge5 Qb1+ 12.Kf4 Qb6 13.Kf5 Qb4 14.Ke6 Kf2 (48.049.702.515) 3206
73 249:44 -3.99 1...Kg3 2.Kd5 Kf4 3.Kxc4 Qg2 4.Kc3 Qf1 5.Kb4 Qd3 6.Kc5 Qa3+ 7.Kb5 Ke4 8.Kc4 Qa4+ 9.Kc5 Qa5+ 10.Kc4 Qd5+ 11.Kc3 Qh5 12.Kc2 Qg4 13.Kc3 Qc8+ 14.Kb4 Qa6 (48.049.702.515) 3206
72 249:44 -3.70 1...Kg2 2.Kd5 Kg3 3.Kxc4 Qf1+ 4.Kd5 Kf4 5.Ne5 Qg2+ 6.Ndf3 Qg8+ 7.Kd6 Qd8+ 8.Kc5 Ke4 9.d3+ Kf4 10.Kc6 Qa5 11.Kd6 Qb6+ 12.Kd5 Qc7 13.Kd4 Qd6+ 14.Kc4 Qe6+ (48.049.702.515) 3206
72 249:44 -3.26 1...Qd1 2.Kd5 Qa4 3.Kc5 Qa2 4.Ne5 Qa5+ 5.Nb5 Qxd2 6.Nxc4 Qg5+ 7.Kc6 Ke2 8.Nc7 Kd3 9.Nb6 Qf6+ 10.Kc5 Qd4+ 11.Kc6 Ke4 12.Ncd5 Kf5 13.Nc7 Qe4+ 14.Kd6 Qd3+ (48.049.702.515) 3206
Debugging is twice as hard as writing the code in the first
place. Therefore, if you write the code as cleverly as possible, you
are, by definition, not smart enough to debug it.
-- Brian W. Kernighan

peter
Posts: 1482
Joined: Sat Feb 16, 2008 6:38 am

Re: Related to Behting: John Nunn's testposition

Post by peter » Sat Aug 11, 2018 8:31 am

Hi Eelco!
Eelco de Groot wrote:
Sat Aug 11, 2018 7:30 am
I was just wondering, can any chessprogram see a big score here, with or without 7 men tablebases? With them, do you get a Mate number (or would you need additional DTM)? Johhn Nunn wrote, in a ChessBase article: 'The Behting study is sound' By John Nunn

11...Kg2!! This is the move I had missed when I first looked at the position.
I don't understand, why do you need any engine, having complete 7MOB already stored at Losomosov's server since quite a while?

The position coming from your dia with 11...Kg2 is



1.Kd5 Kg3 2.Kxc4 Qf1+ 3.Kd5 Kf4 4.Ne5 Qg2+ 5.Ndf3 Qe2 6.Kc5 Qa6 7.Kd5 Qb6 8.
Kc4 Ke4 9.d3+ Kf4 10.Kd5 Qc7 11.Kd4 Qd6+ 12.Kc4 Ke3 13.Kb5 Qd5+ 14.Kb4 Kf4 15.
Kc3 Qc5+ 16.Kb2 Ke3 17.Kb3 Ke2 18.Kb2 Kf2 19.Kb3 Kg3 20.d4 Qc1 21.Kb4 Qc2 22.
Kb5 Kf4 23.Ne1 Qb3+ 24.Kc6 Ke4 25.N1f3 Qe6+ 26.Kc5 Qh6 27.d5 Qe3+ 28.Kc6 Qc3+
29.Kd6 Qc1 30.Ke6 Qc8+ 31.Kd6 Qd8+ 32.Kc5 Qxd5+ 33.Kb4 Qd6+ 34.Kb3 Qb6+ 35.Kc3
Qb5 36.Nc4 Kxf3 37.Kd4 Kf4 38.Kd3 Qc5 39.Nd2 Qe3+ 40.Kc2 Qh3 41.Nc4 Ke4 42.
Nd6+ Kd4 43.Nb5+ Kc4 44.Nd6+ Kc5 45.Nc8 Qg3 46.Ne7 Kc4 47.Nd5 Qd3+ 48.Kb2 Qxd5
49.Ka2 Qb5 50.Ka1 Kc3 51.Ka2 Qb2# 0-1

and the soonest 7some in Nunn's line is first reached after

8/8/7p/3KNN1k/2p4p/8/3P2p1/8 w - - 0 1

1. Ng7+ Kg5 2. Nf3+ Kg4 3. Nf5 h3 4. Ke4 g1=Q 5. Nxh6+ Kh5 6. Nxg1




1...h2 2.Nf3 h1Q 3.Nf5 Kg4 4.Ne3+ Kg3 5.Nf5+ Kg2 6.N5d4 Kf2 7.Kd5 Kg3 8.Kxc4
Qf1+ 9.Kd5 Kf4 10.Ne5 Qg2+ 11.Ndf3 Qe2 12.Kc5 Qa6 13.Kd5 Qb6 14.Kc4 Ke4 15.d3+
Kf4 16.Kd5 Qc7 17.Kd4 Qd6+ 18.Kc4 Ke3 19.Kb5 Qd5+ 20.Kb4 Kf4 21.Kc3 Qc5+ 22.
Kb2 Ke3 23.Kb3 Ke2 24.Kb2 Kf2 25.Kb3 Kg3 26.d4 Qc1 27.Kb4 Qc2 28.Kb5 Kf4 29.
Ne1 Qb3+ 30.Kc6 Ke4 31.N1f3 Qe6+ 32.Kc5 Qh6 33.d5 Qe3+ 34.Kc6 Qc3+ 35.Kd6 Qc1
36.Ke6 Qc8+ 37.Kd6 Qd8+ 38.Kc5 Qxd5+ 39.Kb4 Qd6+ 40.Kb3 Qb6+ 41.Kc3 Qb5 42.Nc4
Kxf3 43.Kd4 Kf4 44.Kd3 Qc5 45.Nd2 Qe3+ 46.Kc2 Qh3 47.Nc4 Ke4 48.Nd6+ Kd4 49.
Nb5+ Kc4 50.Nd6+ Kc5 51.Nc8 Qg3 52.Ne7 Kc4 53.Nd5 Qd3+ 54.Kb2 Qxd5 55.Ka2 Qb5
56.Ka1 Kc3 57.Ka2 Qb2# 0-1

So his continuation from 6th to 11th already gives away some moves to mate.

And SF with full Syzygys gives Bingo- Eval (132+) rather soon here, DTM probably would be got earlier with Backward from mate- output without tbs.
Peter Martan

User avatar
Eelco de Groot
Posts: 3909
Joined: Sun Mar 12, 2006 1:40 am
Location: Groningen

Re: Related to Behting: John Nunn's testposition

Post by Eelco de Groot » Sat Aug 11, 2018 5:53 pm

Hi Peter,

Thanks very much! For all the extensive info. I had not really thought about the Lomonosov(sp?) and I also don't have Aquarium so I probably would have to pay. I don't have a mobile phone.. I do have Shredder but there is no 7 men Shredder yet. Probably Stefan will update Shredder bases one day because it requires much diskspace to have it privately. If he still is in business and ambitious. I just wanted to be sure that it actually is not seen by Kaissa :( I think it has/could have to do with triangulating that Johhn Nunn mentions. If not mistaken, that is three King moves getting back to the same position, assuming the other player does nothing or does not want to move, with different player to move. So that would be three times a (semi-) tempo loss and it does not help that King moves are low on the list and the position also does not improve because opposite player to move. On the other hand maybe you could do consecutive King moves in verification search only. That is just not impossible for my skill I think although I would not know where to start. But concept is clearly defined.

Thank you!
Debugging is twice as hard as writing the code in the first
place. Therefore, if you write the code as cleverly as possible, you
are, by definition, not smart enough to debug it.
-- Brian W. Kernighan


Vinvin
Posts: 4076
Joined: Thu Mar 09, 2006 8:40 am

Re: Related to Behting: John Nunn's testposition

Post by Vinvin » Sat Aug 11, 2018 7:17 pm

Same here : https://lichess.org/analysis/standard/8 ... _b_-_-_0_1
Click on the book button under the move list to see the 7-man EGTB eval.

Uri Blass
Posts: 8116
Joined: Wed Mar 08, 2006 11:37 pm
Location: Tel-Aviv Israel

Re: Related to Behting: John Nunn's testposition

Post by Uri Blass » Sat Aug 11, 2018 8:41 pm

something seems to be wrong with this site
I looked at different positions and got the following link

https://syzygy-tables.info/?fen=8/8/8/q ... _w_-_-_0_1

White is losing with DTZ 101
Kxc4 is a loss that is practically a draw by the 50 move rule.

After choosing Kxc4 I see that
Black is winning with DTZ 121

How is it possible that DTZ increase after the correct move?

Edit:or Maybe DTZ 101 does not really means 101 plies to conversion but also 1 ply to a capture that lead to a cursed win.
In the last case the tablebases do not always suggest the best move because if I have a cursed win I may prefer not to have as soon as possible a capture that lead to a cursed win but to have a move that could win with the 51 move rule if I cannot win with the 50 move rule and win with the 52 move rule if I cannot win the 51 move rule...

syzygy
Posts: 4310
Joined: Tue Feb 28, 2012 10:56 pm

Re: Related to Behting: John Nunn's testposition

Post by syzygy » Sat Aug 11, 2018 11:50 pm

Uri Blass wrote:
Sat Aug 11, 2018 8:41 pm
something seems to be wrong with this site
I looked at different positions and got the following link

https://syzygy-tables.info/?fen=8/8/8/q ... _w_-_-_0_1

White is losing with DTZ 101
Kxc4 is a loss that is practically a draw by the 50 move rule.

After choosing Kxc4 I see that
Black is winning with DTZ 121

How is it possible that DTZ increase after the correct move?

Edit:or Maybe DTZ 101 does not really means 101 plies to conversion but also 1 ply to a capture that lead to a cursed win.
Yes, 101 either means 101 plies to a winning zeroing move (or mate) or 1 ply to a zeroing move into a cursed win.
In the last case the tablebases do not always suggest the best move because if I have a cursed win I may prefer not to have as soon as possible a capture that lead to a cursed win but to have a move that could win with the 51 move rule if I cannot win with the 50 move rule and win with the 52 move rule if I cannot win the 51 move rule...
What is the best move if the position is a draw anyway?
The tables contain enough information to convert the (cursed) win if the 50-move rule is ignored.

It is not possible to have a scalar metric that encodes the 50-move rule, 51-move rule, 52-move rule, etc.

Jouni
Posts: 1821
Joined: Wed Mar 08, 2006 7:15 pm

Re: Related to Behting: John Nunn's testposition

Post by Jouni » Sun Aug 12, 2018 3:30 pm

I tried some 7 piece TBs with latest SF. I don't see draw/win score in any 7 piece position. But I don't have full 6 piece set. Are they totally useless if not full 6 piece set also?
Jouni

Jouni
Posts: 1821
Joined: Wed Mar 08, 2006 7:15 pm

Re: Related to Behting: John Nunn's testposition

Post by Jouni » Sun Aug 12, 2018 5:08 pm

Addition: 6 piece tablebase seems to work OK even without 5 piece set. Is there now difference in 7 piece?
Jouni

syzygy
Posts: 4310
Joined: Tue Feb 28, 2012 10:56 pm

Re: Related to Behting: John Nunn's testposition

Post by syzygy » Sun Aug 12, 2018 8:58 pm

Jouni wrote:
Sun Aug 12, 2018 5:08 pm
Addition: 6 piece tablebase seems to work OK even without 5 piece set.
They don't.

Post Reply