Help for this position

[BALAVOINE]

Hello,

We looked at the following position with the club of Compiegne last Saturday.

And Serge Minaux said to me that this position was winner for the white.

After analysis I do not find the position winner.

Could somebody light me with his database?

Thank you by advance.

Eddy

[Ed Gilbert]

Hi Eddy,

At first looking at your board setup, I had two questions:
- Is it white's turn to play, or black's?
- Is square 1 at bottom right or top left?

However, it turns out that the result is the same for all 4 combinations of answers. It is always a white win.

If you want to post some of your analysis suggesting a draw I can show you the correction.

-- Ed

[BALAVOINE]

Hi ED,

Initially thank you for your assistance.

The blacks go to the bottom, and 1 puts it is in top on the left.

It is with the white to play.

I do not find the position winner because the black pawns 45 and 36 are too close to promotion. And the exit of 47 or 50 involves a promotion.

[A.Presman]

I guess that many (older) human players will answer your question.
Because in The Time Before Databases they studied books.



You have no problem to get this position just in 2 moves.
After that due to 44-40 and 50-39 black king has only fields 46 and 5.
If it stands on 5 - white play 47-33 and black has no moves because after 5-46 white catch the King 49-32 X
And if it stands on 46 - all White has to do is so called "win one temp". They play for example 47-15-33.

[Ed Gilbert]

I guess that many (older) human players will answer your question.
Because in The Time Before Databases they studied books.

Indeed you do not need databases to solve this endgame. I am not clever enough to solve it by sight, but if I turn off the endgame databases in kingsrow it can see the end of the game after searching for only a few seconds. The solution proceeds as follows:

[FEN "W:WK47,K48,K49,K50:B36,45,K46."]
1. 48-34 46-5 2. 47-42 5-46(A) 3. 34-29 46-5 4. 42-33 5-46 5. 33-44 46-5 6. 29-33 5-32 7. 49x27 36-41 8. 27-38 41-46 9. 38-32 46x39 10. 44x33 WW

(A) black cannot promote from 36 because
36-41 3. 34-40 45x34 4. 50-39 34x43 5. 49x46 WW

-- Ed

[BALAVOINE]

Thank you for your answer, I had to forget this possibility.

[Wieger Wesselink]

This is a famous position. The theory says that it still wins if one of the kings on 47/48/49 is a piece. The king on 50 is essential, I think this is because white needs to have 44-40 45x34 50-39 34x43 49x... A long time ago someone (Leo Springer?) wrote a nice article about it in the magazine 'Dammen' from Ton Sijbrands.

[Ed Gilbert]

This is a famous position. The theory says that it still wins if one of the kings on 47/48/49 is a piece.

Hi Wieger,

Here I can say that the databases disagree. If any of 47, 48, or 49 is a man instead of king then the position is a draw.

-- Ed

[Wieger Wesselink]

Hi Ed,

My memory turned out to be incorrect: it was just the other way around . The only winning position is when 50 is a piece. With pieces on 47, 48 or 49 the position is indeed known to be a draw.

I looked up the article of Leo Springer in 'Dammen' no. 1 from January 1985. In this article he does the following claims with respect to the position with 4 kings:

a) with an extra piece on 26 white wins (from the key position with white kings on 33/44/49/50 and black on 5 the main variation is 26-31 33-47 5-46 44-39)

b) with an extra piece on 16 it is a draw

c) with an extra piece on 6 it is a draw, since in the key position black can play 5-46

d) with an extra piece on 1 it is a draw

e) with an extra piece on 2/3/4 white wins

f) with an extra piece on 5 white wins (49-16 46-10* 47-15 10-46* 15-10 5x14 48-31 36x27 16x5+)

g) with an extra piece on 15 white wins (from the key position: 48-42 46-5 42-33 5-19 33-44)

h) with an extra piece on 25 white wins by losing a tempo: 48-34 46-5 47-24! 5-46 24-33 46-5 33-47 5-46 34-48

i) with an extra piece on 35 it is a draw

If 47 is a piece and the black king is on 3, then white wins after 49-32 3-12 47-42 12-29 42-37 29-20 32-16! 20-15 48-34! 15-4 34-23 4-15.

[Ed Gilbert]

Hi Wieger,

Kingsrow agrees with most of the conclusions from the Leo Springer article. It disagrees in one place (if I have the position correct).

b) with an extra piece on 16 it is a draw

[FEN "W:WK47,K48,K49,K50:B16,36,45,K46."]

This position is a white win. This is how kingsrow wants to play it:
1. 48-34 46-5 2. 34-1 5-46 3. 47-42 46-5 4. 49-44 5-46 5. 42-33 16-21 6. 33-47 21-26 7. 44-49 46-5 8. 1-6 26-31 9. 6-11 5-46 10. 11-17 46-37 11. 17-12 37-46 12. 12-26 31-37 13. 26x42 WW

-- Ed

[Wieger Wesselink]

Hi Ed,

thanks for checking the positions. After re-reading the text Leo Springer turned out to be also correct in the case with a piece on 16. It was only written down a bit cryptically, since he started by concluding that the key position with an extra piece on 16 is a draw. But then he gives the same variant as Kingsrow, with the 3rd and 4th move exchanged. It's quite impressive that he had all the conclusions right.

[Rein Halbersma]

Hi Ed,

thanks for checking the positions. After re-reading the text Leo Springer turned out to be also correct in the case with a piece on 16. It was only written down a bit cryptically, since he started by concluding that the key position with an extra piece on 16 is a draw. But then he gives the same variant as Kingsrow, with the 3rd and 4th move exchanged. It's quite impressive that he had all the conclusions right.

Another giant of such a type of endgame analysis is Gerrit de Bruijn. IMHO, his book on 5 vs 2 endgames is probably the clearest book on endgames ever written. In particular because of his very innovative use of diagrams to visualize his reasoning. He also wrote a very elegant article about a particular 4 vs 2 endgame that was found to be 100% correct when Stef Keetman first wrote about the 6 pc endgame dbs. It would be extremely nice if the chaotic book by Moser would be updated by someone as systematic like him with access to endgame databases. The chess books by John Nunn come up as another example of using modern technology to disseminate such knowledge.

[TAILLE]

Hi,

By the way the follwing position is still a win for white.



Gérard

[steenslag]

(white to move and win)

This composition by Arjen Timmer was awarded first prize in a problem tournament. The intended solution starts with 44-40, but current analysis shows black is able to draw with 34-39. What do the engines say?

[Ed Gilbert]

The position is a white win.

[FEN "W:W15,31,32,33,38,44,45:B4,9,10,17,19,22,34."]
1. 44-40 34-39 2. 33x44 9-13 3. 32-27 13-18 4. 44-39 ...

If you post the analysis that is suggesting a draw I can show where kingsrow deviates.

-- Ed