|
|
It started with an e-mail from Oscar Panno with an urgent request to mail him back. I did, and then, some days later, he rang me. It was about an endgame study made by his friend Francisco Benko -no relation to Paul Benko, the Hungaro-American grandmaster who prides himself on being the best grandmaster-composer of all time. Panno had looked at the study himself, and he thought it was entirely correct. Whence the rush? I wondered. Panno then explained that his friend had already turned 90, and that time was getting short. He would love to live to see the publication of his study in New In Chess, accompanied by my comments.
I vividly remembered the other Benko. I had met him in Oisterwijk during the 7th. Interpolis tournament when he had shown great interest in my endgame studies. There was one study about which he was unhappy, since he had the feeling that it had been anticipated. This is what it looked like:
Jan Timman, Schaakwerk II 1983. White: Kf8, Ra2, Nc2 and pawn f6; Black: Kf5 and Qb6. White to play and draw.
The idea, surely, is clear: the sidelined white pieces must be sacrificed to enable the f-pawn to queen.
1.Ne3
1.f7 won't do in view of 1...Kf6, with a quick mate, and 1Nd4 Kf6
is equally hopeless, as it quickly loses a piece.
1...Ke6
A venomous reply. Black continues to play for mate. After 1...Qe3 2.f7 White would have an easy draw.
2.Nd5!
White continues to offer the knight in order to lure away the king, at the same time covering the f-pawn. 2.Ng4 was insufficient for this purpose in view of 2...Qb4, and wins.
2...Kd5
Now Black is forced to accept the sacrifice, as 2...Qd8 3.Kg7 Qd5 4.Re2, followed by 5.f7, causes no problems for White.
3.f7 Ke6
The situation looks critical for White, as 4.Re2 fails to 4...Kf6,
and 4.Kg8 Qg1 5.Kf8 Qc5 6.Kg8 Qg5 won't do either.
4.Ra6!! Qa6 5.Kg8
Draw. The knight sac was meant to lure away the king, while White
wrong-footed the enemy queen with the rook sac.
A lover of chess studies to whom I showed this specimen doubted the
authenticity of the white idea, but he couldn't help me find an earlier genesis. Lommer's 1357 End-Game Studies gave me the following study, which had won first prize in a tournament sponsored by Ceskoslovensky Sach.
Tigran Gorgiev, Ceskoslovensky Sach 1956. White: Kh8, Rd5 and pawn f5; Black: Kf4 and Qe3. White to play and draw.
Here, too, it is important to leave the rook to its own devices in
order to advance the f-pawn far enough.
1.f6 Qh3 2.Kg7 Qg2 3.Kh6!
The only way, as White would be finished off quickly after 3.Kh8 Qg6!.
3...Qd5 4.f7
The point of the previous move. White exploits the unfortunate position of the black king.
4...Qd8 5.Kg7 Qd7 6.Kh8!
And drawn, according to Lommer. Some elaboration is called for, however. After 6...,Kg5 7.f8(Q) Kg6 8.Qg8 Kh6 9.Qe6 White forces
stalemate. So far my comments in Schaakuerk 2 which appeared in 1991.
This lover of chess studies had been Francisco Benko a vigorous
greybeard, he seemed to me, friendly and with a sharp and critim
cat eye. I did, however, wonder why he had never composed anything
himself? And now, finally, he has managed to pull it off. Panno had
warned me that his friend might send in the explanation in old
Gothic script but I needn't have worried. The New In Chess offices
received a perfectly legible fax in simple and clear German. This is
whatitsaid:
Dear Mr Timman,
I am sure that you remember me. I was in 111burg in 1983, and
we got on well together when you showed me many of your beautiful
endgame studies. Our mutual friend Oscar Panno has already
contacted you, so you have some idea of what I am after. I consider my endgame study as an enrichment of chess theory, as both Cheron and Nunn erred in the Gunst endgame (see below). I would like to ask you the fob lowing favour: if you agree with me that my endgame is of some importance, would you please publish it in an article in New in Chess, together with your comments. It is an original and was composed this year. As I have just turned 90, you will understand that I haven't much time left. But even if you don't agree with my assessment I would like to hear your judgement.
Francisco Benko, Original. White: Kc2 and Qb2; Black: Kf2, Bd8 and pawn h2. White to play and win.
The solution consists of three parts: In the first part the queen
goes to h3 while giving checks (1.Kd3) in order to force the
bishop to c7 when the black king is on f2. Second part: The queen
returns to b2 while giving checks, and now White plays 12.Ke3,
threatening mate. After 12...Bb6 13.Qb6!! (13...Bf4 14.Kf4 or
13...Bg3 14.Qe2 Kgl 15.Kf3) the third part starts, which consists of
a theoretical win with the black queen on hl.
1.Kd3 Kgl 2.Qd4 Kf1 3.Qf4 Kg2 4.Qg4 Kf2 5.Qh3 Be7!! 6.Qe3 Kg2 7.Qe2 Kg1 8.Qdl Kg2 9.Qc2 Kgl 10.Qb1 Kg2 11.Qb2 Kf1 (or 11...Kgl) 12Ke3 Bb6 (12...Bf4 13.Kf4 or 12...Bg3 13.Qe2 Kg1 14.Kf3, and wins) 13.Qb6 h1(Q) 14.Qf6 Kg2 15.QgS and wins.
White could have played Ke3 on the second move, but then 2...Bg5 (not 2...Bb6 3.Qb6, winning) secures the draw. If White plays his queen to h3 first only then playing Kd3, Black can draw with Bf4 or Bg3, stopping my winning manoeuvre. Only when the white king goes to e4 can and must the bishop go to b8, c7 or d6. What is the difference? If the bishop goes to c7, for instance, and the queen appears on b2 as before (or even earlier), the black king can flee to g3 or h3, while the white king blocks the long diagonal and White has to do something against h1(Q) and cannot make any progress. I believe this is another novelty for the theory that is as yet unknown. Cheron is mistaken in his comment after 2.Kc4 in the Gunst endgame in Lehr- und Handbuch der Endpiele, Band III, page 116, No. 1605 (1958 edition). He lets Black play Kgl on move 10, claiming that this leads to a draw, whereas my endgame proves that the opposite is true. Only with Bf4 or g3 can Black keep the draw, as I indicated above. Nunn features the same Gunst endgame (which is very beautiful and correct) in Solving in Style No 127, page 101. Unfortunately, his solution contains the same error as Cheron's. He has probably lifted
the solution straight from him.
Thank you very mucht in advance and best regards.
Francisco Benko
All this is quite clear, but to know exactly what's going on, a bit of background information is essential. First the Gunst study. This looks as follows:
Julius Gunst ist prize Hinds II, Suomen Shakki 1946. White: Kb4, Bf1 and pawn e7; Black: Kh1, Bd8 and pawn h2. White to play and win.
The most impressive aspect of this study is probably that the win
seems quite easy at first sight, but in fact requires great accuracy.
The material forces are equal, but White can queen his pawn by tak-
ing a bishop, which leaves him with considerable material superiority. I wonder how many positions there are in which such superiority fails to lead to a win. The main line goes as follows:
1.e8(Q)!
The main reason not to take the bishop is to avoid stalemate. After
1.exd8(Q) Kgl. White can only achieve an ending of queen v queen and bishop. The Belgian composer Vandiest has done us a service by composing many studies with this material, but I don't think even he would be able to find anything more for White here.
1...Be7
Forcing White to capture the bishop, enabling Black to aim for the endgame referred to earlier. But the white queen turns out to be better placed on e7 than on d8.
2.Qe7 Kg1
Now the trick is to find the right square for the bishop.
3.Ba6!!
The only way for White to weave a motif of total domination of the future black queen into the position.
3...h1(Q) 4.Bb7
This must have been Gunst's starting-position. The black queen
has no squares.
4...Qh3 5.Qe1
And mate in two moves. All well and good, I hear You say, but when will we see Benko's discovery? In my comment I wrote that White has to take the bishop on move 2, which is correct, but it is not totally clear that 2.Kc4 is no dual. Cheron supplies the following variation to show a draw for Black:
2.Kc4 Kgl 3.Bg2
What else?
3...Kg2 4.Qa8 Kg1 5.Qa1 Kg2 6.Qb2 Kg1 7.Qd4 Kg2 8.Qg4 Kf2!! 9.Qh3 Bd6!! 10.Kd3
This is the crucial moment. Cheron now gives 10...Kg1, but this is wrong, as Benko has shown. Only 10...Bf4 or 10...Bg3 will allow Black to draw. With this study, the Argentinean Benko -the other one, who had remained in the shadow- has made an important contribution to a small part of the theoretical endgame theory, this enormous field in which it is extremely diffcult to find anything new without cyberneticassistance.
