## Sunday, January 01, 2012

### The twelve puzzles of Xmas

Behting, 1908

White to play and win

[Thanks to Sean]

Anonymous said...

1.e8=Q f1=Q 2.Qd8+ Ke6 3.cxb4 would appear to be winning as well as the intended solution?!

ejh said...

I don't know that it is, after 3...Qf4+, since I think both White pawns may drop off. I grant you that this isn't remotely as clear as one would like (and is not given as a line in my source, which provides a rather truncated solution without any tries) so I'll leave this one up, and people can see if they think it's sound or not.

The "intended solution" is, of course? (And no naughty computer use!)

rgh said...

1.Ng6 fxg6 2.e8=Q with mate in 3 or
1.Ng6 f1=Q 2.e8=N+ Ke6 3.Ng7 mate ?

ejh said...

Not as such

Matt Fletcher said...

rgh I think Ng6 is right but in the first line I don't see mate in 3 - I get 2. ... f1=Q 3. Qf8+ Ke6 4. Kg7 Kd5 5. Qa8+ Kc4 6. Qa6+ winning the queen, or 5. ... Ke6 6. Qg8+ Ke7 7. Qf8+ Ke6 8. Qf7 mate. But I could be missing something.

In the other line, 3. Ng7 isn't mate (3 ... Ke5) so you need to go to 3. Nc7+ Kf6 4. Nd5+ Ke6 5. Ngf4+ picking up the queen with check.

Anonymous said...

1.Ng6 fg6 2. e8(Q) f1(Q) 3. Qf8+ Ke6 4. Kg7 Kd5 5. Qa8+ Ke6 6. Qg8+ Ke7 7. Qf8+ Ke6 8. Qf7#

1... f1(Q) 2. e8(N)+ Ke6 3. Nc7+ Kf6 4. Nd5+ Ke6 5. Ngf4+ Qxf4 6. Nxf4+ Ke7/f6 7. cxb4 +-

Richard

Anonymous said...

It was hard for this Anon to spot Qa8+, as it appeared that the king escaped after 4. ... Kd5.

The c3-pawn is important: defends d4 in one line, takes the bishop in the other. Would it be more aesthetically placed on e3? There's still Nd5+ at the end to win the bishop.

Campion said...

Spotted the intended continuation in the underpromotion line, but in the other line I kept trying and failing to trap the Black king in a mating net.

If White has a pawn on e3 rather than c3, then in the line with (temporarily) two knight, can Black try to draw by playing ...f4 and going after the d-pawn?

Anonymous said...

I am sure this is correct, but do not know how to type in technical terms:
Pushing the white pawn up to get a knight (e7-e8=Kn), placing the king in check would only leave one spot for the king to retreat to. Afterwards sealing the deal by placing said knight on c7.

ejh said...

I think everybody had some of it, but Matt Fletcher and Richard have basically all of it.

It was found by Sean Terry in his copy of Kasparyan, Domination in 2545 Endgame Studies (I don't know the precise edition details, but will check with Sean) which book gives the study as coming from Bohemia, 1908.

However....this attribution appears to be wrong, as (I am informed) the study certainly appeared in Deutsche Schachzeitzung in March 1907. Meanwhile the van der Heijden Endgame Study Database apparently gives "Bohmeia 1906" although I neither have access to that database nor know from where it sourced its information.

Too many uncertainties, you might think. You might be right.

rgh said...

@Matt Fletcher -

In the other line, 3. Ng7 isn't mate (3 ... Ke5)
In that line White still has a knight on g6.

ejh said...

And indeed a pawn on d4, but it's possible Matt meant 3...Kd5?

ejh said...

Sean informs me that the edition of Kasparyan from which the study comes is that published by Progress, in 1980, in the old Soviet Union. Pages 267-268.

(Minor correction: that should of course have been "Bohemia", above.)