360. 13 October 2007: Precision bombing
Ton Sijbrands (57), former draughts world champion and greatest player of all time, who beat his own blind simul record this week (25 games, 28½ hours, 92 %) in today's NRC/Handelsblad: "Draughts is so stressful that I never liked to play. I was always hoping the Americans would precision bomb the hotel where the tournament was held, so it would be postponed."
359. 10 October 2007: Knight moves
Theo Hommeles sends me a move that he has always admired since it was played against him, over 20 years ago - even if it cost him the game.
White to play
Van Mil - Hommeles
Eindhoven, Dutch team championship 1986
I must confess that when I saw what White played here, I thought for an instant Hommeles had made some mistake with the position. But after the amazing 24.Nxf7, all Black could do was resign. After the forced 24...Qxg3 25.Nxd8 Qh4 (Black cannot guard e6) 26.Nxe6+ Kf6 27.Nxd4 Qxd4 28.Rd1 White remains a piece to the good. The game does not seem to be in any databases.
The Knight's triumphant tour reminds Hommeles of a multicapture in draughts/checkers. It is much like that indeed, but it's also like a carom in three-cushion billiards - a comparison I already made in item 183 of this Diary, where I showed two such Knight's sweeps in Kubbel studies.
Jeff Caveney showed me another sort of brilliant Knight move.
Black to play
Barten - Read
Quarterfinals 19th Correspondence World Championship, 1999
With the witty 30...Ng1!, cutting off the Rh1, Black attacked the Bf1 and, when it moves, the Rh1 with Nf3+. There followed: 31.Bc4+ Kh8 32.Bc3 Rd8 33.Kg5 Nh3+ 34.gxh3 Rxh1 35.Kxf5 Bd4 36.Bxd4 Rxd4 37.Ke5 Rd8 38.Kf4 Rhd1 and White resigned.
And this Knight move was sent to me by Odd Øivind Bergstad.
Black to play
Holm - Bergstad
Ostlandserien, Norway 1987
Here, Bergstad felt that Ra2, to lure the Queen away from f2, had to be decisive, but he didn't see how - no smothered mate on f2. He was already considering other moves when, "after a 40 minutes think I pulled myself together and firmly said to myself that there had to be a combination! Then suddenly I saw it." 24...Ra2! 25.Qxa2 Nd1+! The simple, but hard to see idea: again a Knight move cutting off a Rook. White resigned; 26.Kf1 Bd3+ 27.Re2 Qf2 or 26.Kh1 Rxe1+ etc. is mate.
358. 26 September 2007: New Torre Zwickmühles
As I said in the item below, Chess Today has a keen eye for the sort of thing that I like. A few days ago, they had this position in their Test Yourself! quiz.
Black to play
Mueller - Malakhatko
Monarch Assurance Open, Isle of Man, 22 September 2007
29...Rd2 The main threat is now g3, when the Torre-Zwickmühle (of Torre - Lasker fame) becomes deadly 30.axb5 axb5 30...g3 was already possible (31.c6 gxf2+ 32.Kh1 axb5 33.cxb7+ Kxb7 and Black wins) but Black isn't in a hurry. 31.g3 Or 31.Rad1 Bxf2+, or 31.Rae1 Bxc5 31... hxg3 32.hxg3 f4! 33.Rae1 33.gxf4 g3 33... Bxc5 34.Bxc5 Allowing the Zwickmühle, but White had nothing. 34...Rg2+ 35.Kh1 Rxg3+ 36.Kh2 Rg2+ 37.Kh1 and after using the Zwickmühle to remove the Pg3, Black played 37...g3 himself. White resigned - there is nothing against Rg2-h2-h1 mate.
I keep a constant CQL-vigil for Zwickmühles and recently there were a few other ones of the Torre type.
Black to play
Bancod - Russell
Cebu City, Philippines, 19 September 2007
Another devastating Zwickmühle - which Black wasn't fully aware of, it seems; the game ended strangely.
31...Rxf2+ 32.Kg1 Rg2+ 33.Kh1 Rf2+ The Zwickmühle is great for repetitions in time trouble; but Black could have won outright with 34...Rd2+, when 35.Rf3 is forced because 35.Kg1 h2+ 36.Bxh2 Rg2+ 37.Kh1 Rg8 is mate. 34.Kg1 Rg2+ 35.Kh1 axb6 Before he did that, he could have snatched the Pa2 and repeated twice more. 36.exd6 Rxa2+ 37.Kg1 Rg2+ And again he misses h2+ 38.Kh1 Rd2+ 39.Kg1 Rg2+ And again. But now the game becomes really strange, because after 40.Kh1 Rc2+ 41.Kg1 the score gives 0-1. Of course Black is winning, but not seeing the mate with h2+ a few times, he might have played 41...Rg2+ again, when White could have claimed a draw. Or did White overstep the time?
White to play
Ghader Pour - Paridar
Tehran, Asian women's championship, 3 September 2007
22.Bf6 Qxh5 23.Rxg7+ Kh8 24.Rg5+ Kh7 25.Rg7+ draw
Not an interesting case in itself, but it's a rare lookalike of the UrZwickmühle.
White to play
Torre - Lasker
Moscow 1925
25.Bf6 Here, this wins. 25...Qxh5 26.Rxg7+ Kh8 27.Rxf7+ Kg8 28.Rg7+ Kh8 29.Rxb7+ Kg8 30.Rg7+ Kh8 31.Rg5+ Rxa7+ leaves the Pa2 en prise. 31...Kh7 32.Rxh5 Kg6 33.Rh3 Kxf6 34.Rxh6+ Kg5 35.Rh3 Reb8 36.Rg3+ Kf6 37.Rf3+ Kg6 38.a3 a5 39.bxa5 Rxa5 40.Nc4 Rd5 41.Rf4 Nd7 42.Rxe6+ Kg5 43.g3 and Black resigned.
I took the term Zwickmühle from Nimzowitsch, who devoted a little chapter to it in his 1925 Mein System. About Torre - Lasker he wrote: "In spite of its elegance, I cannot reproduce this game but with inward revolt, so much does Lasker's truly great mind inspire me with awe."
357. 21 September 2007: Chess Today, the Tour de France and Pape - Roth
Shivers ran down my spine when I realized that it is half a century ago when in the evenings after Tour de France stages, the Amsterdam daily Het Vrije Volk had a special edition that I would buy from a paperboy for 12 cents. Guilder cents, but it came back to me when I calculated what one issue of Chess Today costs: 12 Eurocents. That would have been a quarter then - you'd think prices would have gone up more.
I'm a few days late, but I can still congratulate Alex Baburin and his team on the 2509th issue of their wonderful cyber daily. Chess Today has become a part of my mornings - when they don't have one of their small delays, my day begins with opening their mail, looking at the latest news, the quiz positions, the endgames, discussions of chess politics, the deeply analyzed game. It must be a huge amount of work that goes into each issue. And most of their GM contributors share a healthy predilection for the artistic and the bizarre - quite often, on these pages, I have shown games that I had found on their pages.
Cheers, CT, may you celebrate your 5009th issue in equally good form.
It is certainly not to spite CT when I go into a well-known hoax that Alex Baburin fell into, a few issues ago - in fact, choosing that position for the Quiz section showed CT's leaning toward the artistic.
In CT-2493, this position from "Pape - Roth, FRG 1972" had the caption: "Is it time for White to resign?" and the solution was: No, he has 1.d6 exd6 2.Kd3 Bxg3 3.a5 d5 4.a6 Bb8 5.a7 Bxa7 stalemate.
A nice finish, but a few issues later a reader, Alexander Ponomarev, commented that "such a game was never played and the position is a slightly modified study," as had been revealed by the Latvian magazine Sahs. One issue later however, René Olthof hit the nail on the head when he mentioned that the Pape - Roth mystery "was solved a long time ago by Tim Krabbé."
Indeed it was, in my Dutch book Schaakkuriosa from 1974.
The diagram is not "a slightly modified study", but an exact copy of a study by G. Bernhardt, first published in the Neue Wiener Schachzeitung, 1923. A little later, in late 1924 or early 1925, the position surfaced in the Leipziger Tageblatt as the finish of "a game Pape - Roth, played in October at the chessclub of Löbau."
When this "game" was published in February 1925 in Deutsches Wochenschach, and the chess player / composer / writer Joseph Krejcik saw it, he pointed out the theft. Notwithstanding, "Pape - Roth, Löbau 1924" made it into countless tactics books and magazines - and it began a second life in 1972, in the endgames section of Informant no. 15; this time as "Pape - Roth, FRG 1972."
And now, 35 years later again, "Pape - Roth" raises its head a third time. A successful hoax.
Remarkably, that little Bernhardt stalemate didn't only become a persistent spurious game; it has always been popular with the plagiarists among the study composers, too. In 1974, I gave this example in Schaakkuriosa.
White to play and draw
G. Andersson
Ceskoslovensky Sach, 1934
This is a slightly modified version - mirrored too, the earmark of the plagiator.
1.e6 dxe6 2.Ke3 Ba2 3.h5 e5 4.h6 Bg8 5.h7 Bxh7 stalemate.
PS 21 October: Initially, I misspelled this composer's name as Anderson, with a single s - it also appears that way in Harold van der Heijden's database, and my 1974 book Nieuwe Schaakkuriosa. This made Steven Dowd wonder whether he could be G. F. (Gerald Frank) Anderson, "a very well-respected British composer, known for his innovative ideas. (...) I find it hard to
believe he intentionally plagiarized." Dowd adds: "It is such an appealing, yet easy-to-find idea that many composers could have unwittingly recomposed it." Dowd had even composed something very similar himself, but found out in time. "It would have placed me square in the group of plagiarists if I had published it (and it was a mirror!)"
Harold van der Heijden solved this little mystery. He has the original source; Ceskoslovensky Sach no. 2-3, of February / March 1934, where this study is diagram #560. The composer/plagiator is one G. Andersson (with double s) from Ljusne, Sweden.
G. F. Anderson (South Africa, 1898 - England, 1983) was an aviator and diplomat, and a prominent composer of mainly selfmate, Fairy and Kriegspiel problems.
Later, there have been other reinventors - with the Van der Heijden database, combined with
CQL (Chess Query Language), it is easy to spot admirrorers.
White to play and draw
M. Shablinsky
Fizkulturnik Belorussii, 1984
1.e6 dxe6 2.Ke3 Bxb3 3.h5 e5 4.h6 Bg8 5.h7 Bxh7 stalemate. An exact, mirrored, Bernhardt.
White to play and draw
A. Sedletsky
Shakhmaty i Shaskki v BSSR, 1985
A slightly embellished one.
1.Kd4 Bd3 2.Ke3 Bxb1 3.e6 dxe6 4.h4 e5 5.h5 Ba2 6.b3 Bxb3 7.h6 Bg8 8.h7 Bxh7 stalemate.
White to play and draw
L. Kekely
Nedelna Pravda, 1991
This is not a straightforward case of plagiarism. The author's solution is given as 1.d6 exd6 2.Kd3 Bh2 3.a5 d5 4.a6 Bb8 5.a7 Bxa7 stalemate, but there is also the try 4.Kxd4, when Black saves his winning last pawn with 4...Ke2 5.Kxd5 Ke3 6.a6 Bb8 7.a7 Bxa7 8.Ke5 Bd4+
However, there is a blemish; with 2...f5 3.a5 f4 4.a6 f3 5.a7 f2 6.a8Q f1Q+ 7.Ke4 Qg2+ Black wins.
But even if the Van der Heijden database marks a few of Kekely's studies with {pl} for plagiarism, I'm not sure this is an example. In the same issue of Nedelna Pravda, Kekely also published a version with a black pawn on g7 instead of on f7, when the Bernhardt manoeuvre works as usual. He may have intended that first position as a demonstration of some extra tactics - although adding 'after Bernhardt' in the caption would have been chic.
PS 23 September: Harold van der Heijden informs me that the Kekely studies were not in the same issue, and gives this link.
On 5 April (click 1991, and then 2408 on the left) Nedelna Pravda published the above study, with the pawn on f7. On 26 April (click 2413) the solution was given, with the variation 4.Kxd4 etc., but without 2...f5. Two months later, on 28 June (2432), the refutation 2...f5! was published, found by Štefan Todek. Kekely corrected his study in that same issue by moving the f7-pawn to g7.
In other words, this is not only a straightforward example of plagiarism, it's a stupid one at that - a thief punching a hole in a stolen picture.
PS 26 September: Noam Elkies remarks that moving the pawn to g7 introduces a new imperfection. It makes the black Bg1 "obtrusive"; the term for a piece in an initial position that must be a promoted pawn. An obtrusive Bishop is considered to be a minor flaw in a study, and a major one if it could have been avoided at no cost. Poor Kekely - he just can't get it right.
Elkies suggests putting the pawn on g6 instead - but I had barely posted that when Harold van der Heijden pointed out that it then becomes a mirror of Anderson's mirrored rip-off above.
Plagiators keep honoring Bernhardt to this day.
White to play and draw
V. Lukov
Shakhmatna Misl, 2003
1.Kd3 d5 2.a5 Bh2 3.g3 Bxg3 4.f4 Bxf4 5.a6 Bb8 6.a7 Bxa7 stalemate. Many transpositions of moves are possible.
But a database/CQL search doesn't only expose plagiators, it also reveals how an idea was developed by bona fide composers. Bernhardt wasn't by any means the inventor of this stalemate; that was - of course - the Founding Father.
White to play and draw
A. Troitzky
Novoye Vremja, 1898
1.Nd3+ But not 1.Re5+ Kf1 2.Rf5+ Kg1 3.Nb3 cxb3 4.Rd5 and now 4...a1Q? 5.Rd1+ Qxd1 would be stalemate, but with 4...Kf1! (and probably 4...a1R too - Elkies), Black wins. 1... cxd3 2. Re5+ Kf1 3. Rf5+ Kg1 4.Ra5 d2 5.Rxa2 d1Q 6.Rg2+ Kf1 7.Rg1+ Kxg1 stalemate.
And another pre-Bernhardt example:
White to play and draw
A. Selesniev
Tidskrift för Schack, 1921
1.Nh5 g2 2.Nf4 g1N 3.Ne2+ Kd2 4.Nxg1 Ke3 5.Nh3 Bxh3 stalemate.
White to play and draw
L. Kubbel
The Chess Amateur, 1926
1.Ng3 Kxh2 2.Nxh5 g3 is a simple win for Black, but 1.Ke3 Kxh2 2.Kf2 Kxh1 3.Kg3 draws. Black must make a Bishop move and lose his last two pawns, or play 3...Kg1, which is stalemate.
Bernhardt's study is just one of almost 40 based on Troitzky's idea - here are two more elaborate ones.
White to play and draw
K. Runquist
Third Prize, Moscow Tournament, 1926
1.Ra6 g2 2.Rg6 d2 3.Rxg2 d1Q+ 4.Rc2+ Kb1 stalemate - surprising, with so many white pieces. After 1...d2 the same motives appear: 2.Rc6+ Kb1 3.Rc2 g2 (3...d1Q stalemate) 4.Rb2+ Kc1
(Kxa1 5.Rb1+ and stalemate) 5.Ka2 d1Q 6.Nb3+ Qxb3+ 7.Kxb3 g1Q 8.Rb1+ Kxb1 stalemate.
White to play and draw
A. Wotawa
Österreichische Schachzeitung, 1952
1.Bb5! axb5 2.Nc4! And not immediately 2.Ra7+ Kb1 3.Nc4 in view of b4+ and Black wins. 2...bxc4 (2...f1Q 3.Ra7+ Kb1 4.Nd2+ Kc1 5.Ra1 mate) 3.Ra7+ Kb1 4.Rb7+ Kc1 5.Rf7 Bxf7 and after sacrificing his entire army, White is stalemate.
A nice feature in this study is that both stalemating pawns and the stalemating Bishop reach their destinations during the course of the solution.
And did it ever happen in a game? A few years ago, a reader sent me this position.
Black to play
NN - Hondebrink
ICC, 1 0 blitz 2004
In this 1-minute blitz game, Black played 56...Bf8 and White found the only move that doesn't win: 57.Kxf8 stalemate.
But there is now an example from a serious game at the highest level.
Black to play
Krasenkow - Bacrot
Team championship, France 2006
I'm not sure Black can win this, but what he tried didn't work - or perhaps it was his way to offer a draw: 62...h4 63.Rh8 Ke5 64.Re8+ Kd4 65.Rd8+ Kc3 66.Rc8+ Kd2 67.Rd8+ Ke1 68.Re8+ Kf1 69.Re1+ and a draw was agreed.
356. 19 September 2007: The computer as an artist
Robert Åström, Swedish champion of 1996, sent me an interesting mail.
One day that same year, he writes, he was analysing a game from an earlier Swedish championship, "together with my friend Genius, the strongest program at that time when suddenly, the program started to make amazing moves."
Engqvist - Jonsson, Swedish championship, Haparanda 1994
1.Nf3 g6 2.c4 Bg7 3.d4 Nf6 4.g3 O-O 5.Bg2 d6 6.O-O Nbd7 7.Nc3 e5 8.h3 Re8 9.e4 exd4 10.Nxd4 a6 11.Re1 Rb8 12.Bf4 Ne5 13.b3 Bd7 14.a4 Qc8 15.Kh2 h5 16.Nd5 b5 17.cxb5 axb5 18.a5 c5 19.Nxf6+ Bxf6 20.Ne2 Qc7 21.Nc3 Be6 22.Nd5 Bxd5 23.Qxd5 Ra8 24.Bd2 Nc6 25.e5 Nxe5 26.Qxa8 Rxa8 27.Bxa8 d5 28.Kg2 Nc6 29.Ra2 c4 30.bxc4 dxc4 31.a6 Bd4 32.Bf4 Qd7 33.Be3 Bxe3 34.Rxe3 b4 35.Kh2 b3 36.Rae2 Na7 37.Re8+ Kg7 38.R8e7 Qd4 39.R7e4 Qc5 40.Re5 Qb6 41.Re7 b2 42.Bd5
see diagram left
Here, Genius looked at what could happen after 42...b1Q, coming up with 43.Rxf7+ Kh6 44.Ree7 Qc2 45.Rh7+ Kg5 46.h4+ Kg4 47.Re4+ Qxe4 (see diagram on the right) and now that amazing move: 48.Rf7 - but regular readers of this page had already recognized the breathtaking Åström & Ornstein study that I gave below in item 349. Genius concluded that after 48...Qc6 (48...Qbe3 49.fxe3 Qc2+ 50.Bg2 g5 51.Rg7 Kf5 52.Rxa7 gxh4 53.Rf7+ Ke6 54.a7 gives White some winning chances.) 49.f3+ Qxf3 50.Rxf3 Qd6 51.Rf4+ Qxf4 52.gxf4 c3 53.Be4 Kxf4 54.Bxg6 the game would be a draw.
In reality, the game was also a draw after 42...Qf6 43.Rb7 c3 44.Ba2 Qf3 45.Rxa7 Qxe2 46.Rxf7+ Kh6 47.a7 Qa6 48.Bb1 Qa4 49.h4 Qa6 50.Bd3 Qa2 51.Be4 Qxf7 52.a8Q Qxf2+ 53.Kh3 Qf1+ 54.Kh2 Qf2+ 55.Kh3 Qf1+ 56.Kh2 Qf2+ 57.Kh3.
"During the Swedish championship," Åström continues, "I showed the analysis to the skilled composer Axel Ornstein, a seven time Swedish champion." A few days later, Ornstein came up with the prizewinning study. He had added a beautiful stalemating idea, and he had added the a-pawns, but he wasn't sure they were really necessary, and asked Åström to test that with Genius. "But obviously I didn’t do my job" - the study was published without those pawns. John Nunn later added them in Endgame Challenge, his 2002 collection of the 250 greatest endgames of all time.
Of course, when Ornstein sent the study to Tidskrift för Schack (it won First Prize for 1997) he named Åström as a co-composer, but one wonders if they shouldn't have named Genius, too. As far as I'm concerned, they were right when they didn't; artists do not have to pay tribute to the reality that inspired them.
So it turns out, perhaps not unexpectedly, that this gamelike masterpiece did indeed sprout from a game.
"Some years ago," Åström concludes, "I read something about the fears for weak composers in the computer era. I can only laugh. That man must have been thinking about me!"
355. 10 September 2007: An outrageous mate finally come true
"Out of carelessness," writes Max Wahlund from Stockholm, "I as black moved the wrong knight at my 7th move in this ICC blitz game. The result was a quite nice-looking mate 10 moves later."
Denethor - Inegrepus, ICC blitz 3 1, 10 September 2007
1.e4 e6 2.d4 d5 3.Nd2 c5 4.exd5 Qxd5 5.Ngf3 cxd4 6.Bc4 Qd6 7.Nb3 Nf6 8.Nbxd4 Nc6 9.Nb5 Qxd1+ 10.Kxd1 Kd7 11.Ng5 Nd8 12.Ke2 a6 13.Rd1+ Ke7 14.Nc7 Rb8 15.Bf4 Nc6 16.Bd6+ Kd7 17.Nxf7 Rg8 18.Bxe6 mate.
Nice indeed. But: this outrageous mate is both a mirror of an old famous game, and possibly the first time it really occurred.
Dodge - Houghteling, Chicago 1900 (1904 and 1906 are also sometimes given.)
1.d4 d5 2.c4 e5 3.e3 exd4 4.Qxd4 Nf6 5.Nc3 Nc6 6.Qd1 Bf5 7.f3 Nb4 8.Qa4+ Qd7 9.Qxd7+ Kxd7 10.e4 dxe4 11.fxe4 Nxe4 12.Rb1 Nc2+ 13.Kd1 Nf2+ 14.Ke2 Bc5 15.Nf3 Bd3+ 16.Kd2 Be3 mate.
However, this game is often considered to be a hoax. There are no other games known by a Houghteling, nor any games by a Dodge of that period. So perhaps the already famous mate just popped up for the first time in an ICC blitz game this morning. (But see the PS of 10 April 2008 below.)
PS 11 September: Derek Pugh shows me a game he found in Graham Burgess's 1992 book The Complete Alekhine - one of Burgess' own games, which does not seem to be in the big databases.
Clark - Burgess, Weymouth Open, 1989
1.e4 Nf6 2.e5 Nd5 3.d4 d6 4.c4 Nb6 5.f4 dxe5 6.fxe5 c5 7.d5 e6 8.Nc3 exd5 9.cxd5 c4 10.d6 Nc6 11.Bf4 g5 12.Ne4 gxf4 13.Nf6+ Qxf6 14.exf6 Be6 15.Nf3 O-O-O 16.Be2 Bxd6 17.a3 Rhg8 18.Qc2 Bc5 19.Rd1 Rxd1+ 20.Kxd1 Nd5 21.Kc1 Ne3 22.Qe4 Rxg2 23.Re1 Na5 24.Nd2 c3 25.b4 cxd2+ 26.Kxd2 Nb3+ 27.Kc3 Bd4+ 28.Kd3 Bc4 mate.
PS 19 September: Martin van Essen draws my attention to a related mate.
Malinin - Savinov, Leningrad 1988
1.d4 Nf6 2.c4 c5 3.d5 b5 4.cxb5 a6 5.bxa6 Bxa6 6.Nc3 d6 7.Nf3 g6 8.g3 Bg7 9.h4 O-O 10.h5 Nbd7 11.hxg6 hxg6 12.Bh3 Re8 13.Qc2 Rb8 14.Be3 Nxd5 15.Qxg6 fxg6 16.Be6+ Kf8 17.Nxd5 Rxb2 18.Ng5 Nf6 19.Nf4 Qa5+ 20.Kf1 Bxe2+ 21.Kg1 Reb8 22.Bf7 Rb1+ 23.Kh2 Ng4+ 24.Kh3 Rxh1+ 25.Rxh1 Nxf2+ 26.Bxf2 Bg4+ 27.Kxg4 Rb4 28.Kf3 Qa3+ 29.Be3 Qa8+ 30.Bd5 Qa5 31.Nfe6+ Kg8 32.Nc7+ e6 33.Bxe6+ Kf8 34.Nh7+ Ke7 35.Bg5+ Bf6 36.Bxf6 mate.
It seemed familiar to me somehow - and indeed, I had already given it as no. 48 in part 7 of my series The 110 Most Fantastic moves ever played. A CQL-search for this mate yielded no other examples; it might be unique.
As Van Essen remarks, White could have resigned after 30...Rxf4+ - but a little computer check reveals that both players made several gross errors; 26.Kg2! would have won; 28...Rxf4+ and 29...Rxf4+ would also have won, to name a few. Zeitnot, probably.
I also don't think I would include 15.Qxg6 in my list now. It's daring, it's brilliant, but most chessplayers would at least have a fleeting look at what might happen after that move.
PS 10 April 2008: A communication by Frederick Rhine makes it clear that speculations of Houghteling - Dodge being a hoax, are unjustified. "The Houghtelings were a prominent and rich family in Chicago in the late 19th and early 20th centuries," as Rhine writes. He gives one other game by (J.R.) Houghteling:
Houghteling - Cornell, Chicago 1902
1.e4 e6 2.d4 d5 3.Nc3 Nf6 4.e5 Nfd7 5.Qg4 c5 6.Be3 cxd4 7.Bxd4 Nc6 8.Nf3 a6 9.Bd3 Qc7 10.0-0 Nxd4 11.Nxd4 Be7 12.f4 Bc5 13.Ne2 Qb6 14.c3 g6 15.b4 Be7 16.Kh1 Qc7 17.h4 h5
see diagram
18.Qxg6 Nf6 19.Qg7 Rg8 20.exf6 Rf8 21.fxe7 Qxe7 22.Rae1 and Black resigned.
354. 9 September 2007: Corner themes
White to play
Alekhine - Chajes
Karlsbad 1923
Some time ago Noam Elkies showed me this position he'd come across - there followed 63.Rh1 Rd7 64.Ra1 and Black resigned.
"It must already be uncommon for a Rook," wrote Elkies, "to go from one corner of the board to the opposite corner in two moves without capturing anything along the way, and more unusual yet when it goes backwards (a1-h1-h8 would be more natural for White); is there another example of this, even without immediate resignation? Has a single Rook or Queen ever visited all four corners in a single game? (That's a common problem theme, naturally.)"
A Rook crossing the entire board diagonally in two moves is not very rare: it happens in around 1 out of 2000 games. Without captures: in 1 out of 4000. Backwards without captures: in 1 out of 7000 - in my database, there are around 300 of such games. Backwards, without captures, resulting in an immediate resignation? Even that happened four times, including Alekhine - Chajes. (One was a pawnless endgame, when 'backwards' does not really exist.)
Has a single Rook ever visited all four corners? That happens all the time; in 1 out of 350 games - but perhaps a Rook that starts on a8 cannot be said to have visited a8. Rooks that go to all four corner squares are of course less frequent, but this still happens in 1 out of 1500 games.
An obvious next question is: has a Rook ever managed a homerun - visited all four corners on consecutive moves? I found two games where that happened - and both homeruns were without captures.
White to play
Joukl - Lamac
Czech Republic, 1993
59.Rh1 Kb6 60.Rh8 Bc7 61.Ra8 Kb7 62.Ra1 Bb8 63.Rf1 Kb6 64.Rf8 Bc7 65.Ke6 Kc6 66.Rc8 and Black resigned.
Black to play
Ernst - Van der Marel
Groningen 1999
This homerun was a swan song: 75...Ra1 76.Kh5 Ra8 77.g6 Rh8+ 78.Kg5 Rh1 79.Re2+ and Black resigned.
Once in 40.000 games, a Queen visits all four corners, the quickest tour taking 6 moves. An honorable mention goes to Vidarsson - Hjartarson, ch Iceland 1994 (it can be played over on the left) where a Queen needed over 100 moves for her Grand Tour, White playing 38.Qa1, 84.Qa8, 104.Qh8, and finally 142.Qh1. The game was drawn in 180 moves.
Most of these games are not very exciting, but the oldest Queen's Tour that I found, was.
Landau - Van den Bosch, Amsterdam 1930
1.d4 Nf6 2.c4 g6 3.g3 Bg7 4.Bg2 d5 5.cxd5 Nxd5 6.e4 Nb6 7.Ne2 c5 8.d5 Bd7 9.Nbc3 Qc8 10.h4 Na6 11.h5 O-O 12.hxg6 hxg6 13.Bg5 Re8 14.Qc1 Nb4 15.Kf1 Nc4 16.b3 Ne5 17.Qf4 f6 18.Bh6 g5 19.Qd2 Bxh6 20.Rxh6 Kg7 21.Rh1 Rh8 22.Rxh8 Qxh8 23.a3 Nbd3 24.f4 gxf4 25.gxf4 Qh4 26.Nd1 Bh3 27.fxe5 Bxg2+ 28.Kxg2 Rh8 29.exf6+ exf6 30.Qxd3 Qh1+ 31.Kf2 Rh2+ 32.Ke3 Qe1
see diagram
33.Nf2 The only winning move. After 33...Qxf2+ 34.Kd2 Black's attack would be over, so: 33...Qxa1 34.d6 Qe1 35.d7 The same sacrifice is again the only winning move. 35...Rg2 36.Nh1 A third sacrifice of this Knight is again the only winning move. 36...Qxh1 37.Kd2 Qh6+ 38.Kd1 Qh1+ 39.Kd2 Qh6+ 40.Kc2 Rxe2+ 41.Qxe2 Qh8 42.Qd3 Qd8 43.a4 b6 44.Kc3 Kf7 45.Qd6 Kg7 46.Kc4 Kf7 47.Kd5 Qa8+ 48.Qc6 Qd8 49.Qe6+ Kf8 50.Kd6 Qb8+ 51.Kc6 Qd8 52.Kb7 and Black resigned.
The Bishop, being color-challenged, has its own Grand Tour. In four games (1 in 500,000), a single Bishop visited a7, a1, g1 and h8, or its negative. There is no game where all four corners were visited by any Bishop, white or black.
The same goes for Knights: there is not a single game where all four corners had a Knight on them, white or black. (PS 14 September: Christian Rieseneder reminds me of entry 128 in my Diary, where there are two games with all the Knights occupying the corners at the same time. I was worried for a second that I might have used a wrong CQL script, but I checked it again, and what I said stands - for the Bishops too. The explanation is that those games were informal ones, sent to me personally, and that they are not in the databases. I might have remembered them though - but I didn't.)
As to single Knights - no Knight ever visited more than two corners - and even two corners only happens once in 30,000 games. Sixteen Knights (1 in 150,000 games) reached both corners on the other side of the board, six of them singlehandedly capturing both enemy Rooks.
Yudin - Garagulya, Novosibirsk 2001
1.e4 c5 2.c3 d5 3.exd5 Qxd5 4.d4 Nf6 5.Nf3 Bg4 6.Nbd2 Nc6 7.Bc4 Bxf3 8.gxf3 Qd7 9.dxc5 e6 10.b4 a5 11.Rb1 Ne5 12.Qe2 Nxc4 13.Nxc4 Nd5 14.O-O Nxc3 15.Qc2 Nxb1 16.Nb6 Qb5
see diagram
17.Nxa8 Qxb4 18.Bg5 Nc3 19.Qd2 f6 20.Nc7+ Kf7 21.Qd7+ Kg6 22.Nxe6 fxg5 23.Qe8+ Kh6 24.Nd8 g6 25.Nf7+ Kg7 26.Nxh8 Qf4 27.Re1 Qxf3 28.Re3 Qd1+ 29.Kg2 Qg4+ 30.Kf1 Qc4+ 31.Kg1 Nxa2 32.Qe5+ Kg8 33.Rf3 Qg4+ 34.Rg3 Qf5 35.Qe2 Nb4 36.Rf3 Qg4+ 37.Kh1 Kxh8 38.Rxf8+ Kg7 39.Qxg4 and Black resigned.
No King ever reached more than three corners, apart from two games where the players apparently collaborated to get one King on all of them. A King reaching three corners happens surprisingly often; once in 70,000 games. They are long Queen's or K+R vs. K+B endgames - but most of them are K+B+N vs. K matings where the stronger side doesn't seem to be sure about the procedure.
PS 11 September: Sven Mühlenhaus reminds me of a famous example of Alekhine's manoeuver which was not immediately followed by resignation, but preceded by it.
Black to play
Karpov - Taimanov
Leningrad 1977
After 38...Ng3+ White resigned, as 39.hxg3 is forced, but loses to Ra1-a8-h8 mate.
353. 5 September 2007: How to forget from one's experiences
Hauke Reddmann sent me a funny game he recently played.
Reddmann - Schütt, Hamburg 2007
1.d4 d5 2.Bf4 c5 3.dxc5 Nc6 4.e4 dxe4 5.Qxd8+ Kxd8 6.Nc3 e5 7.O-O-O+ Bd7 8.Be3 f5 9.g3 Nf6 10.Bh3 Attacking f5.
see diagram
10...g6 defending it - but losing a different pawn. Black should have moved his King out of the pin. 11.Bg5 Be7 12.Bxf6 Bxf6 13.Nxe4! The attack was also a pin. 13...Be7 14.Nd6 Kc7 15.Nf3 After 15.Nxb7 Nb4 Black has equalized. 15...e4?? 16.Nxe4 and after winning two e4-pawns in four moves with the same pin, White didn't have too much trouble winning the game: 16...Rad8 17.Nd6 Rhf8 18.Rhe1 Rf6 19.Nd4 Bf8 20.Bg2 Nxd4 21.Rxd4 Bc6 22.Bxc6 Kxc6 23.c3 Bxd6 24.cxd6 Rdxd6 25.Rxd6+ Kxd6 26.f4 Kd7 27.a4 Ra6 28.b3 Rb6 29.Kc2 Rc6 30.Re5 Rd6 31.c4 Rc6 32.Kc3 Rf6 33.Kd4 Rf8 34.c5 Rf6 35.Kd5 Ra6 36.Kc4 Kd8 37.Rd5+ Ke7 38.Re5+ Kd8 39.Kb5 Kd7 40.b4 Kd8 41.a5 Rc6 42.Kc4 a6 43.b5 axb5+ 44.Kxb5 Ra6 45.Rd5+ Kc7 46.Rd6 Ra8 47.Rf6 Rh8 48.Rf7+ Kb8 49.Kb6 Ka8 50.a6 bxa6 51.Ra7+ Kb8 52.c6 and Black resigned.
I vaguely remember doing something like that myself in a blitz game once. Against Black's Kh7 + Pg7, I had a Bf8, and I attacked his pawn with Rf7 - only to be chased away with Kg8. A few moves later, when the King was back at h7, I happily attacked g7 again with Rf7, only to stumble on Kg8 a second time.
I found two other examples of a repeated blunder in my files, unfortunately both between unnamed players.
White to play
This is an anecdote from Eduard Gufeld, who alleges the (stylized and quite likely apocryphal) position occurred in a women's game. There followed: 1.Nc5+ Rxc5 - oops, but White saw a chance to redeem her blunder by winning back an exchange: 2.Nb3 Rb5 3.Nc5+ only to be shocked by 3...Rxc5
White to play
Oxford, 198?
From an old column by Gert Ligterink in the Dutch daily De Volkskrant. This is supposed to have happened in an Oxford City team competition, where the University team, headed by John Nunn, won all their matches 4-0, except for one.
Here, on 4th board, White blundered a pawn with 1.Rd1 Qxa2 His hopes to get some compensation in the form of the Bishop's pair with 2.Nc1 Qa5 3.Na2 evaporated when Black played 3...Qxa2
PS 6 September: Martin Stichlberger showed me this variation on the theme.
White to play
Vercauteren - Noviks
XVI NATO championship, Kolobrzeg (Poland), 2005
29.Rc4 Rc8? He doesn't see the pin (Rxd6). 30.f4? Neither does he. 30...Re7? 31.Rc6? Re4? But finally, after missing Rxd6 a total of five consecutive times between them, White sees the light: 32.Rdxd6 There followed: 32...Rce8 33.Rd3 R4e7 34.Rdc3 Rc8 and now White showed that he had learned from the experience: 35.Rxb6 Only, he didn't notice its first use had made the pin unusable. 35...Rd8+ and Black won.
PS 9 September: Bader al-Hajiri, a regular to these pages, sends me a game he played tonight, at a chess cafe in Salmiya City, Kuwait.
White to play
al-Hajiri - Ali
Blitz (chess960), Al-Salmiya City, Kuwait, 8 September 2007
There followed: 1.Qxe4 The Queen is of course untouchable, but it's a mistake. 1...Bxc3+ 2.Kb3 Bxa1 3.Qc2 Re8 4.Ka2 Bc3 5.b5 Re4 He must have felt all his pieces were safe at e4 6.Qxe4 and Black resigned.
PS 11 September: A further example from Eric Roosendaal.
Black to play
NN - Roosendaal
Rapid, Amsterdam, around 1990
Black has some compensation for the exchange, certainly after 1...Nxd5, which White may have thought to be impossible because of the counterpin 2.Rhd1 Now, 2...c6 fails to 3.c4 2...Qc6 White must have seen that as resignation: 3.Rxd5 However, the pin still works: 3...Qxd5 White's advantage is now gone, but he optimistically continued with 4.Rd1 Qa5 5.Rd5 There followed 5...Qxd5 and, "having no further Rooks to hang on d5, White resigned," as Roosendaal writes.
352. 28 August 2007: A strange pair of helpless Queen's pairs
Mark Thornton, who started the thread about helpless Queen's pairs (see items 344 and 349 below) mentioned he had seen, but not bookmarked, a game at chessgames.com, where: "Rook and Pawns drew against 2 Queens. White (with the Rook) had 3 connected passed pawns on the 7th rank, so Black (with the 2 Queens) had to take perpetual check."
Using Chess Query Language (see here or here) it is easy to follow such scents - I found two games that met Thornton's description, one of which I turned out to have already published on the Dutch side of this website, in a piece about the Labourdonnais Monsters.
Birnbaum - Relange, Cappelle Open 1995
1.d4 d5 2.c4 e6 3.Nc3 c6 4.Nf3 dxc4 5.a4 Bb4 6.e3 b5 7.Bd2 Bb7 8.axb5 Bxc3 9.Bxc3 cxb5 10.b3 a5 11.bxc4 b4 12.Bb2 Nf6 13.Bd3 Nbd7 14.Qc2 O-O 15.O-O Qc7 16.e4 e5 17.c5 exd4 18.Bxd4 Ng4 19.Bb5
Nde5 20.Nxe5 Nxe5 21.f4 Nc6 22.Bb2 Ba6 23.Bxa6 Rxa6 24.Qe2 Qa7 25.Rf2 a4 26.Qg4 f5 27.exf5 a3 28.Be5 b3 29.f6 b2 30.Rd1 Nxe5 31.fxe5 Qf7 32.Qe2 Qb3 33.Rff1 a2 34.f7+ Kh8 35.Qxa6 b1Q 36.Qd6 Qb8 37.e6 Qxd6 38.cxd6 a1Q 39.e7 Qb6+ 40.Kh1 Qa8 41.d7
see diagram
And all Black could do was hastily force a draw.
41...Qxg2+ 42.Kxg2 Qg6+ 43.Kh1 Qe4+ and a draw was agreed.
At chessgames.com, this game is given as "Birnbaum Detlev" vs. "Eloi Relange", but Detlev and Eloi are the first names.
The other game I found was:
Kokolias - Sigalas, Acropolis Open, Athens 2005
1.c4 c6 2.Nf3 d5 3.d4 e6 4.Nc3 dxc4 5.a4 Bb4 6.e3 b5 7.Bd2 Bb7 8.axb5 Bxc3 9.Bxc3 cxb5 10.b3 a5 11.bxc4 b4 12.Bb2 Nf6 13.Bd3 O-O 14.O-O Nbd7 15.Qc2 Qc7 16.e4 e5 17.c5 exd4 18.Bxd4 Ng4 19.Bb5 Nde5 20.Nxe5 Nxe5 21.f4 Nc6 22.Bb2 Ba6 23.Bxa6 Rxa6 24.Qe2 Qa7 25.Rf2 a4 26.Qg4 f5 27.exf5 a3 28.Be5 b3 29.f6 b2 30.Rd1 Nxe5 31.fxe5 Qf7 32.Qe2 Qb3 33.Rff1 a2 34.f7+ Kh8 35.Qxa6 b1Q 36.Qd6 Qb8 37.e6 Qxd6 38.cxd6 a1Q 39.e7 Qb6+ 40.Kh1 Qa8 41.d7 Qxg2+ 42.Kxg2 Qg6+ 43.Kh1 Qe4+ draw.
No diagram, because this is, apart from slight move order differences between moves 1-4 and 13-15, the same game. Coincidence or copycats? Probably the latter, even if the opening is well analysed: the position after 16...e5 occurred in 70 games, and the one after 23...Rxa6 in 2 (other) games. Strangely, this game too, is commented upon at chessgames.com, without the commentators of either game being aware of the other one.
Thornton also sent me an idea for a study, based on his own game vs. Horrocks (item 344) and the Åström & Ornstein study in item 349.
"I woke up at 6am," he writes, "with the idea in my head, and a mental picture of files f to h. It then took me 30 mins to work out an arrangement of pawns and queens on the left-hand side of the board."
White to play and win
Mark Thornton
Idea for a study, 2007
1.g8N, and the two Queens are helpless against the mating threats on h6 and f6.
It's just the crowning move of a study yet to be composed - anyone?
Finally, Oded Ross showed me this wonderful study with the theme.
White to play and win
Y. Dorogov
Shakhmaty v SSSR, 1982
1.e7 Nf4+ 2.Qxf4 White's main threat is 3.Qxa4+, followed by 4.e8Q+ Ka5 5.Qa8+ Kb5 6.a4 mate. Therefore, Black has no time for 2...h1Q and 2...Qe6 is simply met by 3.Qf3. And after 2...g2+ 3.Kd2 White still threatens Qxa4+ - 3...Qd7 4.e8Q Qxe8 5.Qxh2 Qg6 6.Qg1 is then a technical win.
2...Qh7+ 3.Kd2 h1Q (else Qxa4+ or Qb8) but even now: 4.Qxa4+!! Kxa4 5. e8Q+ Ka5 6.Qa8+ Kb5 7.Nc7+! Qxc7 8.a4+ Kc4 9.Qg8+ Qd5+ (9...Qf7 10.Qxf7+ Qd5+ 11.Qxd5+!) 10.Kc2 Qxg8 What else? 10...Qdf7 is met by 11.e4, and 10...Qh7+ 11.Qxh7 is still mate, for instance after 11...Qd6 12.Qe4+ and b3 mate.
see diagram
11.e4!! and it turns out that here too, the Queen's pair is helpless against 12.b3 mate.
351. 12 August 2007: 47 checks - a new record by Sampsa Lahtonen
Sampsa Lahtonen 47 consecutive checks (legal position, promoted pieces allowed) Original, 2007 |
Two days after sending me his record breaking series of 46 checks (see ante scriptum below) and telling me he would now let the matter rest, Sampsa Lahtonen sent me this position with 47 consecutive checks.
"The thing about me is that mostly, I'm a very honest and trustworthy person. But when I say something like 'I'll leave this record alone for a week' it's not worth the paper it's not printed on. Thanks to the very irritating fly that kept me awake for hours at night I had time to add an extra check to my composition..."
"As can be seen," he adds, "this is very nearly the same setup [as the 46 - TK] (albeit with reversed colors); it didn't take much of a change."
Lahtonen didn't add a proof game this time, noting that he doesn't see any objections to the legality of the position - neither do I.
PS 14 August: Bader al-Hajiri sent me this proof game:
1.d4 Nc6 2.d5 Ne5 3.d6 Nc4 4.dxc7 d5 5.g4 d4 6.g5 d3 7.g6 Bf5 8.gxh7 g5 9.f4 Bg6 10.f5 g4 11.fxg6 f5 12.h4 f4 13.h5 g3 14.h6 Kd7 15.a4 Kd6 16.a5 Kd5 17.Rh5+ Kd4 18.a6 Bg7 19.axb7 f3 20.Bh3 g2 21.Bf4 Be5 22.Ra6 Bd6 23.Rc6 a5 24.Bc8 a4 25.Bh2 a3 26.Kf2 a2 27.Re5 a1Q 28.Kg3 f2 29.Nh3 f1Q 30.b4 Qf7 31.b5 Nf6 32.g7 Ne3 33.g8Q g1Q+ 34.Kf4 Qe1 35.Qg1 Rg8 36.h8Q Rg2 37.Qe8 d2 38.Qc1 d1Q 39.b6 Qd3 40.b8R Qda3 41.Ba6 Qfb3 42.Qh5 Qf8 43.Re8 Qeb4 44.Nd2 Rd8 45.Qgd1 Nf5 46.b7 Ne4 47.c8Q Ng5 48.b8R Nf7 49.Rbb6 Bb8 50.h7 Rd6 51.h8Q Rdg6 52.Ng5 N5d6
Sampsa Lahtonen 46 consecutive checks (legal position, promoted pieces allowed) Original, 2007 1.Rf8+ Bf7+ 2.Rxf7+ Qxf7+ 3.Qe6+ Rxe6+ 4.Rf3+ Qxf3+ 5.e4+ Rxe4+ 6.Qe6+ Rxe6+ 7.Qe4+ Rxe4+ 8.Qe6+ Rxe6+ 9.Kd4+ Nb5+ 10.Rxb5+ c5+ 11.Nxc5+ Re4+ 12.Ncxe4+ Bc5+ 13.Nxc5+ Qd3+ 14.Ncxd3+ Qe5+ 15.Nxe5+ Nc2+ 16.Bxc2+ Rd3+ 17.Nexd3+ Qd5+ 18.Rxd5+ e5+ 19.Rxe5+ Nxe5+ 20.Ng4+ Qf2+ 21.Qxf2+ Nf3+ 22.Qxf3+ Qf4+ 23.Qe4+ Qxe4+ |
"This is a very difficult type of composition." Lahtonen writes, "and has required a lot more effort than coming up with 14 discovered checks did. At first I couldn't really understand it and, to get feeling of the idea, focused on trying to directly improve Frangen's setup. To begin with I couldn't make any progress but as I grew more and more familiar with the subject I started to get quite good ideas. Finally, I was able to reach 45 checks with one piece less - but I couldn't put the piece back and keep the position legal."
After applying more and ever bigger changes to the Frangen setup, but always ending up with 45 checks, Lahtonen started from scratch with an entirely new setup - again coming up with one tie after another of the 45-checks record.
"Then, this night, I failed to catch sleep, and played out the setup in my mind. I came up with an improvement - a way to turn some queens to rooks - which could theoretically have solved my legality problems. It did help, but not quite enough. I made yet more modifications but nothing helped. I was frustrated, totally stuck at 45 checks. [...] I looked at the unfriendly board once more, the pieces that cooperated against my will to create 47 consecutive checks. And then I saw it, in a sudden moment of clarity. This is the moment when people jump out of their bath and run naked in the street shouting "Heureka!" - there now was a very simple way to add, quite legally, the 28th piece, the light-squared Black bishop, into the position. It wouldn't give the two checks normally achieved by the addition of a piece, but what the hell."
"Forty-six checks (unless I've miscounted), the result of all the trouble I had gone into. I had done it! As can be seen, this is essentially my own setting, although it does still use one of Frangen's ideas. Given the high number of pieces in the final position, and the fact that 41 checks is possible with four pieces less, I very much doubt this is optimal. It will have to do for now, though, as I'll take my well-earned good week's sleep and close my eyes to chess. But stay tuned, I may try to improve it one day."
I asked him how he composes: with token extra Queens, Rooks and Knights (sugar lumps, matchboxes?) - or does he borrow extra sets?
Lahtonen: "I composed in many ways. When I was composing on a board, I just used pawns to play the part of promoted pieces, but I didn't have to do that too much. A lot of composing necessarily took part only within my mind, without any board at all - I was surprised at how easy this was given my total lack of skills in blindfold chess, then again I've never had any problems in working out Bricks ideas that way. But most of the time I used a computer program (namely the FICS interface BabasChess) as my board - no problem in adding a lot of promoted pieces there, and with a movelist I didn't have to count the checks myself all the time."
Lahtonen also added a proof game for the legality of the position:
1.h4 g5 2.a4 b5 3.c4 d5 4.cxd5 bxa4 5.Ra3 gxh4 6.g4 f6 7.g5 h3 8.g6 h2 9.d4 a5 10.Bg5 fxg5 11.f4 g4 12.Nh3 g3 13.Nf2 g2 14.g7 h5 15.b4 Nh6 16.g8R g1Q 17.Kd2 Rh7 18.Rh8 Kd7 19.f5 Nc6 20.Rg3 a3 21.Nc3 a2 22.d6 a4 23.b5 Ra5 24.b6 Re5 25.b7 Na7 26.b8Q Kc6 27.d7 Qe8 28.d8Q Qf7 29.d5+ Kc5 30.d6 Kd4 31.d7 Bb7 32.Qdc8 h4 33.Bg2 Re3 34.d8Q+ Ke5 35.f6 Rg7 36.Be4 Rg4 37.Bb1 Qh7 38.f7 Bg7 39.f8Q a3 40.Nfe4 a1Q 41.Kc2 h3 42.Kb3 a2 43.Kc4 Qb2 44.Qfe8 Bc6 45.Qa6 Kf5 46.Kc5 Bd5 47.Qb4 Bf7 48.Qdb8 Bh5 49.Qf1+ Qf2 50.Qg2 a1N 51.Rd1 h1Q 52.Rd3 Qhc1 53.Qba3 h2 54.Qbb4 h1Q 55.Qba5 Qb8 56.Q5b6 Qh4 57.Qec6 Qfg1 58.Rd4 Qgh1 59.Nf2+ Rge4 60.Kd5 Ng4 61.Ra4 Qg6 62.Ncd1 Bb2 63.Qab4 Ba3 64.Ra5 Rb3 65.Nb2 Qh7 66.Qf1 Qg6 67.Nbd3 Qh7
350. 9 August 2007: Blood sport
When Magnus Carlsen showed such tired, plodding chess in the first rounds of the Arctic Chess Challenge in Tromsø, and his father Henrik (2089) rose above himself, it was inevitable that sooner or later some sort of Murphy's Law would apply, and they would be paired against each other. Yesterday, they both reached 3½ out of 5; today's the day.
It is difficult to imagine the psychology of that game. Magnus might be unable to suppress the thought: Fi fån, what has become of me, now I even have to play this patzer. Henrik might be angry with fate, that in the one game he gets to play with a world class grandmaster, he stumbles upon his son and cannot just play chess. He will be mortally afraid to hurt him. What if Magnus hangs a Queen? Will Henrik grab it?
Lots of siblings have played each other in strong and even international tournaments; Louis and Wilfried Paulsen (they had a strong sister Amalie, too), the Portisch brothers, the Polgar and Kosintseva sisters, the Steiner, Timman and Piket brothers, the Cramling and Mamedjarov(a) brother/sisters.
But father and son? Apart from some domestic games (Morphy!) - not much is known. Alejandro and Gaston Needleman have recently played each other in a few minor international tournaments in Argentina, but the only real example I can think of are father and son Vidmar. They played two official games; one in an international tournament in Ljubljana in 1938, the other in the 1949 Yugoslav championship, also held in their home town Ljubljana. And although father Vidmar (1885 - 1962) is generally seen as the inventor of the grandmaster draw, both were fighting games. In 1949 father won, but the 1938 game, a draw, was more interesting.
Vidmar sr. - Vidmar jr., Ljubljana 1938
1.d4 d5 2.c4 e6 3.Nc3 Nf6 4.Bg5 Be7 5.e3 h6 6.Bh4 O-O 7.Nf3 Ne4 8.Bxe7 Qxe7 9.cxd5 Nxc3 10.bxc3 exd5 11.Qb3 Rd8 12.c4 Nc6 13.cxd5 Qb4+ 14.Qxb4 Nxb4 15.Rc1 Nxd5 16.Bc4 c6 17.O-O Be6 18.Ne5 Rac8 19.Nd3 b6 20.Rfd1 Kf8 21.Ba6 Rc7 22.e4 Ne7
see diagram
23.d5 Bc8 24.Nb4 Bxa6 25.d6 Be2 26.dxe7+ Kxe7 27.Re1 Rd2 28.Nd5+ Kd7 29.Nxc7 Kxc7 30.a3 b5 31.f4 a5 32.e5 Bc4 33.Red1 Re2 draw
PS: The Carlsen family can heave a sigh of relief: Magnus won.
Magnus Carlsen - Henrik Carlsen, Tromsø (6), 9 August 2007
1.d4 Nf6 2.c4 e6 3.Nc3 d5 4.cxd5 exd5 5.Bg5 Be7 6.e3 c6 7.Qc2 h6 8.Bh4 O-O 9.Bd3 Re8 10.Nge2 Ne4 11.Bxe7 Qxe7 12.Bxe4 dxe4 13.Ng3 f5 14.O-O Na6 15.a3 Nc7 16.f3 exf3 17.Rxf3 f4 18.exf4 Bg4 19.Rff1 Rad8
see diagram
20.f5 Rxd4 21.Qf2 Qe3 22.h3 Rd2 23.Qxe3 Rxe3 24.Rf2 Rxg3 25.Rxd2 Bxf5 26.Re1 Kf7 27.Rf1 Kg6 28.Rd6+ Be6 29.Re1 Kf7 30.Ne4 Rg6 31.Nc5 b6 32.Rxc6 bxc5 33.Rxc7+ Kg8 34.Rc6 Rxg2+ 35.Kxg2 Bd5+ 36.Kg3 Bxc6 37.Rc1 1-0
349. 5 August 2007: Some more helpless Queen's pairs
Yesterday, in the first round of the Arctic Chess Challenge in Tromsø, Norway, 2034-rated Brede Hagen held Magnus Carlsen to a draw in a spectacular game. (But see the PS.) There was an even more unexpected result.
White to play
De Firmian (2540) - E. Thingstad (1893)
Tromsø, 4 August 2007
Any reasonably obvious move, like 47.Ka2 or 47.Qb7+ would win easily, but the most obvious of them all, 47.h8Q, allowed 14-year old Thingstad a great swindle - after 47...Qd1+ 48.Ka2 Qb3+! White had to resign in view of 49.Rxb3+ cxb3+ 50.Kb1 Rd1 mate (see diagram right.)
Thanks to Manny Rayner for showing me this.
A few days before, Martin van Essen had called my attention to an astonishing study, also based on the theme of the helpless pair of Queens of which I already gave a few (game) examples below, in item 344.
White to play and win
R. Åström & A. Ornstein
1st Prize, Tidskrift för Schack 1997
A quite gamelike position.
1.Rf8 White was threatened with material disasters on c1 and d8 1...c1Q Or 1...Kh6 2.Rxf7! transposing to the main line, because 2.Rxc2 Qd6(c5) does not win 2.Rxf7+ Kh6 3.Ree7 Threatening mate in a few moves with Rh7+, h4+ en f3 3...Qc2 4.Rh7+ Kg5 5.h4+
Kg4 6.Re4+ Now Kf5 7.Rf7+ Qf6 8.Rf4+ Ke5 9.R4xf6 Kxd5 10.Rxd7+ is a technical win, but of course Black has 6...Qxe4!
see diagram right
Everything seems to lose for White, but he has the breathtakingly quiet move 7.Rf7!! Not only is it amazing that this move could be the best White has - it even turns out that Black, with his two Queens against Rook and Bishop, is helpless. The threat is f3+ and mate, but there's also the Rook mate on f4 to be reckoned with. 7...Qbe3 The most thematic move. 7...Qxf2+ 8.Rxf2 Qe5 9.Rf4+ doesn't help, nor does 7...Qc6 8.f3+ Qxf3 9.Rxf3! etc. 8.fxe3 Qc2+ 9.Bg2 g5 10.e4! Another quiet move; now Black has nothing against Rg7 and Rxg5 mate, because after 10...gxh4 there follows 11.Rg7 mate.
But why not 1.Rxd7? The answer reveals an additional subtlety of this masterpiece. After 1...c1Q 2.Rxf7+ Kh6 3.Ree7 we are in the main line - without the Pd7. That enables Black, after 10.e4, to draw with 10...a4 because now 11.Rg7 Qxg2+ leads to stalemate.
Originally, the study was published without the pawns a3 and a5. But then (1.Rf8) Qa5! is awkward for White; after 2.Rxf7+ Kh6 3.Ree7 is answered by Qxd5 and 3.Re8 by Qxc3! Adding the Pa5 took that away, which called for a wPa3 to retain the stalemate after 10.e4; 10...a4! 11.Rg7 Qxg2+
The study is reminiscent of Thornton - Horrocks of item 344. Quite a few pieces are in the same places, some of the mates are the same. It might even have been inspired by that position, which was published, as Thornton writes, "in the early 1990s, in Dragon, the magazine of Cambridge University Chess Club."
Van Essen also reminded me of one of my favorite games with this theme, which I have often published before, but never on this website.
Franz - Mayet, Berlin 1858
1.e4 e5 2.Bc4 Nf6 3.Nf3 Nxe4 4.Nxe5 d5 5.Bb3 Be6 6.O-O Bd6 7.d4 Qf6 8.f4 c5 9.Ba4+ Ke7 10.c4 dxc4 11.Qc2 Bf5 12.Qxc4 cxd4 13.Qb5 b6 14.Re1 Bc5 15.b4 a6 16.Qc4 b5 17.Qd5 Bxb4 18.Rxe4 Bxe4 19.Qxe4 Rc8 20.Qb7+ Kd8 21.Nxf7+ Qxf7 22.Qxf7 Rxc1+ 23.Kf2 Nd7 24.Bb3 Kc7 25.Qxg7 Bc5 26.f5 Re8 27.Be6
Re7 28.Qg3+ Kb7 29.Qd3 Kb6 30.a4 Ne5 31.a5+ Kc7 32.Qe4 d3+ 33.Kg3 Bd6 34.Qa8 Nc6+ 35.Kh4 Be5 36.Qc8+ Kd6 37.Nd2 Bxa1 38.Ne4+ Ke5 39.f6 Rxe6 40.f7 Rh6+ 41.Kg5 Rg6+ 42.Kh5 Kxe4 43.f8Q
see diagram
43...Be5 White is still winning, but he goes astray quickly - this strange game is another good example of the clumsiness of the pair of Queens. 44.Qf3+ Kd4 45.Qcf5 Rc3 46.Q3e4+ Kc5 47.Qf8+ Rd6 48.Qf2+ Rd4 49.Qf8+ Now Black is winning. 49...Bd6 50.Qf2 d2 51.Qxc6+ Kxc6 52.Qxd4 Rc5+ 53.Kh6 Rd5 54.Qb6+ Kd7 55.Qb7+ Ke6 56.Qc8+ Ke5 57.Qh8+ Kf4 58.Qf6+ Ke4 59.Qf3+ Kd4 60.Kxh7 Kc4 61.Qe4+ Rd4 62.Qc2+ Kd5 63.Qf5+ Be5 64.Qf3+ Kc4 65.Qd1 Rd3 66.h4 Bc3 67.h5 Re3 68.Qf1+ Kb3 69.Qd1+ Kb2 and White resigned.
PS 9 August: Several readers advised me (and it was shown on ChessNinja and perhaps on other blogs and sites) that in that game Hagen - Carlsen, the theme of the helpless Queen's pair could have easily occurred, too.
Hagen (2034) - Carlsen (2710), Arctic Chess Challenge, Tromsø, 4 August 2007
1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6 6.g3 e5 7.Nde2 Be7 8.Bg2 Nbd7 9.O-O b5 10.h3 Bb7 11.g4 b4 12.Nd5 Nxd5 13.exd5 O-O 14.Ng3 Re8 15.a3 a5 16.Be3 Qc7 17.Qd2 Nb6 18.b3 Bf8 19.Rfd1 bxa3 20.Rxa3 a4 21.Qb4 Qxc2 22.Rc1 Qb2 23.Rca1 Bxd5 24.Qxb6 Rab8 25.Qxb8 Rxb8 26.Bxd5 axb3 27.Ra7 Qc2
see diagram left
Now 28.Rxf7 was completely winning. There could have followed 28...b2 (28...Kh8 29.Rc1) 29.Rb7+ Kh8 30.Rxb8 bxa1Q+ 31.Kh2
see diagram right
and with two Queens against none, Black will be mated or worse. It would, together with De Firmian - Thingstad, have been the second occurrence of the theme in one round of a tournament.
After 28.Rc1 Qg6 29.Bxf7+ Qxf7 30.Rxf7 Kxf7 31.Rb1 b2 32.Bd2 d5 33.Bc3 Rc8 34.Bxb2 Rb8 35.Rd1 Rxb2 36.Rxd5 Ke6 a draw was agreed.
348. 2 August 2007: Polgar's stalemate
White to play
Grischuk - J. Polgar
Biel, 1 August 2007
Yesterday in Biel, Judit Polar saved herself with a wonderful stalemate trick.
According to the databases, White has only one winning move in the diagram: 61.f5. After 61.Kf3, Polgar found the only drawing move: 61...Ng4! and White cannot extricate himself. There followed: 62.Nd3 Nh2+ 63.Ke4 Ng4 64.Ne5 Nf6+ 65.Kf3 Ng4 66.Nc4 Nh2+ 67.Ke4 Nf1 68.Ne3 Nxg3+ 69.Ke5 Kh2 70.Kd6 Nh5 71.f5 Ng7 72.f6 and a draw was agreed.
I found one other example of this trick in a game:
White to play
Simacek - Murdzia
Czech Republic, 2006
This is not an easy win, if it is a win at all. But White thought he could finish the game right away: 85.Rg7+ Bxg7 86.hxg7 - which was true after 86...Rg8! 87.Nxg8 stalemate.
The idea hasn't been used very extensively in studies - this is the wittiest example I found.
Black to play, White to draw
V. Pachman
2nd Honorary Mention, Thèmes-64, 1978
1...Ng4+ After 1...Nc4+ 2.Kf6 Rf1+ 3.Kg7, White promotes. 2.Ke6! And not 2.Kd6 Rd1+ 3.Ke7 Rd7+! etc. 2...Re1+ 3.Kf7 Rf1+ A trap of sorts, Re8 is really the move. 4.Ke7 And not 4.Kg7? Ra1 and Black wins because, other than in the variation after 1...Nc4+, f6 is inaccessible. 4...Re1+ 5.Kf7 Re8 6.Kxe8 Nf6+ 7.Kf7 Nxh7 8.Kg7 Black is a piece ahead, and with the move, but he cannot win. 8...Ng5 9.g4+ Kxg4 10.Kxh6 Kf5 Only move to keep his last pawn on the board. But now: 11.g4+ Kf6 with the Polgar stalemate.
347. 31 July 2007: The mysterious man on the Mali stamp
In my story from 1999 WILLI SCHLAGE - The only unknown to become immortal twice, elsewhere on this site, I wondered how the minor master Willi Schlage could have ended up on a stamp in a 1979 Mali series "Grands Maîtres des Échecs", along with Alekhine, Janowsky and Bogoljubow.
Now Jim Kulbacki informs me that the Mali stamps were designed by the French designer Claude Andreotto (http://www.andreotto.com) and that on the Bogoljubow stamp, his name is misspelled as Bogoljulow. Not too surprising; I have at least a dozen chess and cycling stamps with misspelled hero's names, and I'm not even a collector.
This mistake however, Kulbacki writes, "leads me to believe that the designer made another error when he engraved "SCHLAGE" on the stamp. The portrait, in my opinion, bears an uncanny resemblance to Tartakower."
I'm not sure I would call it uncanny, but I agree there's a resemblance between the face on the stamp and this well-known photograph of Tartakower. Of course Tartakower would belong in a series "Grands Maîtres des Échecs", but if it is him on the stamp, then that just heightens the mystery of how Schlage's name ended up there.
I could only find one picture of Willi Schlage on the internet, which you can see at the
ChessBase site (you have to scroll down a bit). We don't see much of him there, but he does look a bit like Tartakower, and he appears to be wearing his Mali stamp glasses.
PS 1 August: Thanks to Edward Winter of Chess Notes, who sends me a clearer picture of Schlage.
PS 2 August: I contacted Claude Andreotto, who replied that he just creates the stamp, and that the subjects are chosen by the country. With the many stamps that he designs, the fields are too diverse for him to become an expert in every one of them.
The mystery remains.
PS 9 August: Harold van der Heijden was intrigued by the funny glasses on the stamp - and found this picture of Tartakower on Chessgames.com. The resemblance is striking indeed - at least, between the glasses.
PS 27 March 2008: Thilo Küssner informs me: "There is a booklet by Helmut Wieteck, 'Schach-Mekka Berlin in the roaring 20s', which provides interesting information not only about chess but about many other facets of cultural life of that time. Schlage is already mentioned on one of the first pages and characterized as a Schachoriginal und Spaßvogel. [Something like: a chess character, a joker - TK.] According to that booklet, Schlage has been chess teacher in Africa. Wieteck writes 'before the war', presumably this should be understood as before the first world war. In any case, this seems to be a plausibe explanation how Schlage made it on that stamp."
346. 24 July 2007: The Ponziani mate
Some time ago Rob Valkeneers asked me who composed the following "extremely beautiful problem" that his club, O'Kelly from Hasselt, Belgium, have as a puzzle on their website.
Mate in 4
J. Marquez
Ruy Lopez, 1897
1.Ra4+ bxa4+ 2.Kc4 b5+ 3.Kc5 b4 4.axb4 mate.
It's of course a cute little gem if you have never seen it before, but it should not be overestimated - even in 1897, it was just a step in the development of an idea that had first been shown over a century earlier.
Black to play and win
D. Ponziani
Il giuoco incomparabile degli Scacchi, 1769
1...g6 2.b6 h6 3.g3+ Kxf3 4.g4 Kf4 5.g5 f(h)xg5 mate.
Another early example, with a Rook's sacrifice:
Mate in 4
A. d'Orville, 1842
1.Rb6 cxb6 2.c3 b5 3.Kc5 b4 4.axb4 mate.
Note that in both of these early Ponziani mates, other than in the Marquez problem, one flight square is guarded, rather than blocked.
I also found a modern version of D'Orville's Rook sacrifice where the Rook is not yet present in the initial position. (I don't know how modern, as I did not date my photocopy. The print makes clear it's from 64, and the caption says it's an original publication - probably from the 70's.)
Mate in 6
A. Kolesnikov & A. Rosljakov
"64", 197?
1.d8R 1.d8Q is stalemate, and 1.Kf3 Kg5 2.d8Q is too slow. 1...f5 2.Rg8 f6 3.Rg5 fxg5+ 4.Kxf5 g4 5.Kf4 g3 6.hxg3 mate.
A wonderfully subtle Ponziani mate, with both a different Zugzwang and a different blocker, by the 17-year old Wizard.
Mate in 5
Sam Loyd
New York Albion, 1858
1.Rb7 Be3 If 1...Bg1, then 2.Rb1 Bh2 3.Re1 Kh4 4.Kg6 and 5.Re4 mate 2.Rb1 Bg5 3.Rh1+ Bh4 4.Rh2 gxh2 5.g4 mate.
4.Rh2 is a nice Zugzwang (reminiscent of the famous 'Morphy problem'), but there is a much subtler Zugzwang hidden in this problem, connected to the question why only 1.Rb7 works and not, for instance, 1.Rc7 or 1.Rd7 or 1.Re7.
1.Rc7 fails because after 1...Be3 the square c1 is guarded and 1.Rd7 is met by 1...Bd4, but the refutation of 1.Re7 is not obvious. The hidden flaw is that after 1...Bg1 2.Re1 Bh2 White is in Zugzwang. He must make a Rook move along the first rank, but 3.Rd1 to Ra1 are refuted by 3...Kh4 4.Kg6 Bg1! and the mate takes one move to many. And 3.Rf1 fails to 3...Kh4 4.Kg6 Kg4 and the Rook is too close to the King. Only 3.Re1 meets all the requirements, which means Rook cannot be already there.
There are many artistic renderings of the Ponziani mate in endgame studies.
White to play and win
F. Lazard
L'Echiquier de Paris, 1949
1.Kb7 Threatening mate. 1...b4 2.Kc6 Be4+ (2...bxa3 3.Kc5 Be4 4.e7 a2 5.e8Q a1Q 6.Nb7+ or Nc6+ and mate.) 3.Kc5 b3 4.e7 b2 5.e8Q b1Q and now only 6.Nc6+ Bxc6 7.Qd8+ Qb6+ 8.Qxb6+ axb6+ 9.Kxc6 b5 10.Kc5 b4 11.axb4 mate.
White to play and win
A. Kakovin
Shatmatyi v SSSR, 1940
First, a struggle to stay ahead materially: 1.Rd8 Ba5 2.Rd5 bxa4 3.Rxa5 b5 But now the Rook is trapped. 4.Kc3 Kb7! (Kb6 5.Kb4) 5.Kd4! And not 5.Kb4 Kb6 and White is in Zugzwang. 5...Kb6 Now the Rook is lost. But: 6.Kd5 Kxa5 Or Kb7 7.Kc5 etc. 7.Kc5 b4 8.axb4 mate. This is probably the best known Ponziani mate.
A double setting of the Ponziani mate was shown by a young (19) Kasparyan.
White to play and win
G. Kasparyan
Shakhmatyi Listok, 1929
1.Nd1 1.Rxb4+ is even losing: axb4 2.Nd1 b3 3.axb3 Kb4 4.Nc3 Kxb3 etc. 1...Rxa4 2.Nc3+ Kb4 3.Kd4 b5 After 3...d6 4.Nxa4 Kxa4 5.Kc4 b5+ 6.Kc3 b4+ 7.Kc4 b3 8.axb3 mate we see a first, standard Ponziani mate. 4.Ne4! And not 4.Nxa4 bxa4 5.Ke5 Kc4 6.Kd6 Kd4 7.d3 Kc3 8.Kxd7 Kb2 9.d6 Kxa2 and Black draws. 4...d6 5.Nc5! dxc5+ 6.Kd3 c4+ 7.Kd4 c3 8.dxc3 and the Ponziani mate has shifted one file to the right.
There are quite a few closely related forms of this smothered Pawn's mate, all lacking the Zugzwang that is essential to the Ponziani mate. Perhaps some day I will come back to those.
345. 20 July: Three times a gentleman is a fool once
Rob Bertholee drew my attention to the following position.
White to play
Fine - Maroczy
Zandvoort 1936, 1st Round
Immediately after playing his last move 38...Rb2-b7, Black resigned. He must indeed be losing, but most players would have waited for White to spot the stalemate trap after i.e. 39.Kf2 a5 40.Ke3 a4 - however simple it would have been to avoid that.
Leafing through that tournament book I saw that Maroczy, at 66, played a wonderful tournament, beating 20-years old Keres and sharing 5th/6th Prize with Bogoljubow, behind 1.Fine, 2.Euwe, 3/4.Tartakower and Keres, ahead of Grünfeld, Landau and Spielmann - and I also noticed he could have done even better.
A few rounds later, Maroczy again refrained from setting a trap.
Black to play
Van Doesburgh - Maroczy
Zandvoort 1936, 5th Round
Here Maroczy played 59...Qg6 and the game ended with 60.g8Q Qxh6+ 61.Qh7 Qxh7+ 62.Kxh7 draw. Most players who saw the trick 59...Qf6, would have waited for their opponent to see it too; White then only has one drawing move, 60.Kh7, because 60.h7?? Qe5 is mate next move.
Maroczy was probably just being a gentleman, but he took chivalry even further in the 8th round.
Tartakower - Maroczy, Zandvoort 1936, 8th Round
1.f4 d5 2.Nf3 Nf6 3.e3 g6 4.Be2 Bg7 5.O-O O-O 6.d3 c5 7.Qe1 Nc6 8.Bd1 Qc7 9.e4 dxe4 10.dxe4 e5 11.fxe5 Nxe5 12.Nxe5 Qxe5 13.Nc3 Bd7 14.Bf4 Qe6 15.Bf3 Bc6 16.Rd1 Rad8 17.Rxd8 Rxd8 18.Qh4 Re8 19.Re1 Nd7 20.Bd2 Ne5 21.b3 Qd7 22.Re2 f5 23.Be1 Qf7 24.Qh3 b5 25.Qg3 b4 26.Nd1 Nxf3+ 27.gxf3 fxe4 28.fxe4 Rxe4 29.Rxe4 Bxe4 30.Ne3 Bd4 31.Bf2
see diagram
In this position, a draw was agreed. Black is completely winning, so what was the matter? According to the tournament book, Maroczy had offered a draw after 22...f5, in an already superior position. Tartakower had refused, but after 31.Bf2 he wanted that draw after all, claiming Maroczy had said: "Sie können remis haben wann Sie wollen" (You can have a draw whenever you want) and not, as Maroczy said he had said: "Sie können remis haben wenn Sie wollen" (You can have a draw if you want.)
Of course Maroczy was right; he couldn't have offered Tartakower the chance to play for a win without the risk of losing. But, again being a gentleman, Maroczy settled for Tartakower's version.
PS 24 July:
It turns out that Fine, in his game against Maroczy, did miss the stalemate. As Steve Wrinn informs me, Fine admitted as much in his book Lessons From My Games. He gives the move 39.Kf2 as actually having been played before Maroczy resigned, and then writes: "At this point an amusing interlude occurred. Bogoljubow came over and suggested 39...a5 40.a4??? Rb2+ 41.Ke3 Rb3+ 42.Kd4 Rd3+!! with a draw by perpetual check, since Black is stalemated. Maroczy replied that he was sure I would have seen through such an elementary trap. Actually I had not noticed the stalemate position, but after 39...a5 40.Ra8 White avoids it and wins easily enough."
344. 19 July 2007: The helplessness of the pair of Queens
Yesterday's Chess Today had this interesting position:
Black to play
Kislinsky - Slugin
Voronezh Open 2007
White is a piece up, but Black seems to have good compensation.
42...Nf4 (perhaps Qe7 is better) 43.Qh4 a2(?) and here CT-analyst Deviatkin gives
43...Nh5 44.Nf6+ Nxf6 45.Qxf6 c4+ 46.Kh2 Qf7 and Black has some chances to hold) 44.Qd8! a1Q+ 45.Kh2
see diagram on the right
and Black is helpless with his two Queens. 45...Qf7 is answered by 46.Nf6+, so: 45... Q1a4 46.Qxd6 Qf7 47.Nf6+ Kh8 48.Qxe5 threatening to remove the Bishop and mating with Qb8 48...Bb5 This is mate in three, but 48...Qa8 49.Nxe8+ Kg8 50.Nf6+ Kh8 51.Ng4+ Kg8 52.Bxf4 was also completely lost. 49.Nd7+ Kg8 50.Qb8+ and Black resigned.
Korchnoi has said: "The board is too small for two Queens of the same color. They're in each other's way; they're much less strong than one would expect on the grounds of their material value."
He was thinking of this game:
White to play
Karpov - Kortchnoi
Dortmund 1994
61.f8Q According to Karpov, 61.f8N+ Kh6 62.gxf4 gave drawing chances. 61...Bxe3+ 62.Kh1 Bh6 63.Qf2 Bg7 64.a6 Rf3 65.Qe1 Bxa6 66.Be2 Rf7 67.Qc5 c3 68.Qcxc3 Bxe2 69.Qxe2 Qf6 70.Qc1 Bh6 71.Qb1 Qf5 72.Kg1 Rc7 and White resigned
In Seirawan's Inside Chess some years ago, Nikolay Minev devoted an article to this phenomenon, in which the best example was this.
Schmidt - Enders, Jena 1981
1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.Bg5 c5 5.d5 h6 6.Bh4 d6 7.e3 Bxc3+ 8.bxc3 e5 9.Bd3 Nbd7 10.Ne2 Qe7 11.Bf5 g5 12.Bg3 Nb6 13.Bxc8 Rxc8 14.Qb3 Qd7 15.a4 Qg4 16.f4 exf4 17.a5 fxg3 18.axb6 gxh2 19.bxa7 Qxg2! Very deep. 20.Rxh2 Qxh2 21.a8Q O-O 22.Qaa2
see diagram
Black has a Rook where White has a Queen, but each side rules his half of the board - White a ghost town on the left; Black a bustling city where Things Happen on the right. Even the computer realizes Black is already better. 22...Rce8 Now e3 cannot decently be guarded. 23.O-O-O Rxe3 24.Qbc2 Rfe8 25.Rd2 25.Ng1 or even Nd4 are met with the nice little crosspin Rxc3 25...Qe5 26.Qab2 h5 The Queens are utterly helpless against the progress of this pawn, however slow. 27.Qd1 After 27.Qb6, Black does not win a piece on e2 which would be a great relief to White, but plays 27...h4 with the possible continuation of 28.Nd4 h3 29.Qf5 h2 30.Rxh2 Rxc3+ 31.Kb2 Qxh2+ 32.Kxc3 Re3+ 33.Qd3 Ne4+ and mates. 27...Ne4 28.Rd3 Nf2 29.Rxe3 Qxe3+ and White resigned.
Some time ago, Mark Thornton showed me another astonishing example.
Kosten - Zelcic, Bozen 1992
1.d4 d5 2.Nf3 c5 3.e3 e6 4.c4 Nf6 5.Nc3 Nc6 6.cxd5 exd5 7.Be2 cxd4 8.Nxd4 Bd6 9.O-O O-O 10.Nf3 Bg4 11.b3 Rc8 12.Bb2 Bb8 13.Nb5 a6 14.Nbd4 Re8 15.Nxc6 Rxc6 16.Nd4 Bxe2 17.Qxe2 Rc8 18.Rac1 Qd6 19.g3 Qd7 20.Nf3 Ne4 21.Rfd1 Rxc1 22.Rxc1 Qf5 23.Nh4 Qe6 24.Qf3 g6 25.Qe2 Ba7 26.Ng2 Ng5 27.h4
Ne4 28.Rd1 Qf5 29.g4 Qd7 30.Nf4 Nxf2 31.Nxd5 Nxg4 32.Nf6+ Nxf6 33.Rxd7 Nxd7 34.Qf3 Bxe3+ 35.Kf1 Re6 36.Qxb7 Nf8 37.Qc8 Bf4 38.a4 Bd6 39.Bc3 Be7 40.b4 h5 41.a5 Rd6 42.b5 axb5 43.a6 b4 44.a7 bxc3 45.a8Q
see diagram
The c3-pawn will fall, after which the unusual constellation of two Queens versus Rook, Knight and Bishop arises. Despite this huge advantage, it turns out White cannot take Black's fortress, which he needed 35 moves to recognize: 45...Re6 46.Qxc3 Bf6 47.Qcc8 Bg7 48.Qab7 Re5 49.Kf2 Re6 50.Qcc7 Rf6+ 51.Ke3 Rf5 52.Qbc8 Rf1 53.Ke2 Rf6 54.Qc4 Re6+ 55.Kd3 Rf6 56.Kc2 Rf2+ 57.Kb3 Rf6 58.Ka4 Rf2 59.Kb5 Rf6 60.Qd8 Rf3 61.Kb6 Rf6+ 62.Kc7 Rf3 63.Qd2 Rf5 64.Kc8 Rf6 65.Qd8 Rf5 66.Qe7 Rf6 67.Qcc5 Re6 68.Qd8 Bh6 69.Qf2 Bg7 70.Qg5 Re8+ 71.Kb7 Ne6 72.Qd5 Re7+ 73.Kc6 Nf8 74.Qd8 Re6+ 75.Kb5 Bh6 76.Qa8 Bg7 77.Qf4 Re7 78.Kc6 Re6+ 79.Kd5 Re7 80.Qf3 Be5 81.Qf1 and a draw was agreed.
Just as Black's fortress cannot be cracked, the whole position cannot be cracked by the computers. In the final postion, they all agree White has an advantage of around 5 pawns - the great Rybka 2.3 is even worse than its collegues: it says White is 6.63 pawns better.
This game is extensively commented at chessgames.com.
PS 20 July: As readers Phil Salathe and Christopher Hartleb note (and as was already noted in the chessgames.com commentaries), White missed a chance in 76.Qc7, when f6 is forced, and the fortress crumbles. 75...Bh6 was probably a mistake. After 75...Rf6 76.Qxf6 Bxf6 77.Qxf6 Ne6 the fortress would not hold but as Salathe remarks, Black should not have too much trouble drawing after 75...Bh8.
PS 24 July: Mark Thornton, who pointed out the Kosten - Zelcic game to me, now shows me a wonderful Schwindel based on the same theme, that he was once able to pull off himself.
Black to play
Thornton - Horrocks
Northumberland League, 1984
"At this point," Thornton writes, "the League rules said we should adjourn, so I offered to fetch an envelope for his sealed move. This seemed to annoy him, and he played his next two moves quickly."
Small wonder it seemed to annoy him, with his two Queens and the pinned white Rook. 37...Qag7?? He must have thought this forced the exchange of one pair of Queens, or won the Rd5. 38.Qe6! With a devilish threat that Black misses. Amazingly however, that doesn't matter: Black is already lost. That came as a shock to Thornton too; initially, he assumed there had to be a defence. Only a week later, when he demonstrated the finish at his club, he discovered there isn't any; White wins in all variations. 38...Qgf7 That's what he must have had in mind, but after 39.Qe4+ the horrible truth was revealed: 39...fxe4 40.Rg5 is a one-Rook mate against two Queens. Black resigned.
As said, other 38th moves wouldn't have saved Black:
- 38...Qbe7 39.Rd4+ Qxd4 40.Qxg6+ and mate
- 38...Qxd5+ 39.Qxd5 Qc3 40.Qf3+ Qxf3+ 41.exf3 mate.
- 38...Qf6 39.Qe3 (or 39.Qxf6 Qxd5+ 40.Kh2 etc.) 39...Qxd5+ 40.Qf3+ Qxf3+ 41.exf3 mate.
The best defence is 38...Qh6, after which White has to see a nice Rook's sacrifice to win:
There follows 39.Kh2 (threatening Rd4+) and now:
- 39...Qbg7 40.Rxf5 gxf5 41.Qc4+ f4 42.Qc8+ and mate
- 39...Qhg7 40.Qe4+
- 39...Qa7 40.Rxf5 gxf5 41.Qc4+ f4 42.Qc8+ and mate
- 39...Qb4 40.Rxf5 gxf5 41.Qg8 and mate
- 39...Qb2 40.Rc5 and Rc4+ followed by mate
- 39...Qxd5 40.Qxd5 Qe3 41.Qg2 and mate
38.Qe6 then, wasn't really a Schwindel by White - 37...Qag7 was a self-Schwindel by Black.
As it should be in cases like this, Thornton's win won the match for his team - and at the end of the season, the League, by a single point.
PS 26 July: Martin van Essen and Bernhard Koenig drew my attention to a recent game from the (still ongoing) Montreal Empresa Tournament.
White to play
Harikrishna - Bluvstein
Montreal, 19 July 2007
Black has just played b1Q but here too, the two Queens are helpless. 57.h7 Qbe1 58.Nxd5 Kxd5 59.Qf4 e5 60.dxe5 Qh1 61.Qf6 Qxh7 62.Qd6+ Kc4 63.Qxd2 Qg7+ 64.Kf4 Qf7+ 65.Kxe4 Qb7+ 66.Kf4 Qf7+ 67.Kg3 Qg6+ 68.Kf3 Qh5+ 69.g4 Qh3+ 70.Kf4 Qh2+ 71.Kf5 Qh7+ 72.Kg5 and Black resigned.
343. 6 July 2007: 76 year old record broken
In item 308 of this Diary, I gave a few records for the construction task of consecutive checks. One of them was for consecutive discovered checks, in a legal position, with promoted pieces allowed. A. Lapierre had published a series of 13 in L'Échiquier in 1931 but now Sampsa Lahtonen (whom we already met in item 340) from Finland goes one better.
Sampsa Lahtonen
14 consecutive discovered checks
Original, 2007
1.Nf7+ Ngxh5+ 2.gxh3+ Bf3+ 3.Be2+ Nxd2+ 4.Nxa3+ Nb3+ 5.Rxc1+ Rb7+ 6.Bc5+ Nxd8+ 7.Ne5+ Nfe6+ Bravo. And in a new setting.
Lahtonen also gives a proof game for the legality of this position:
1.h4 c5 2.d4 g5 3.dxc5 gxh4 4.f4 e5 5.fxe5 b5 6.a4 bxa4 7.b4 d5 8.c4 d4 9.e4 d3 10.Rh3 f5 11.Kf2 f4 12.e6 f3 13.Ke3 f2 14.Nf3 a5 15.c6 Ke7 16.c7 Ba6 17.c8B a3 18.b5 a2 19.Ba3+ Kf6 20.e7 Qd6 21.e5+ Kg6 22.e6 Kf6 23.c5 h5 24.Rg3 d2 25.Bd3 h3 26.b6 Nc6 27.Bc2 Be2 28.Bd3 h2 29.e8B a4 30.e7 Na5 31.Bc6 Nb3 32.Nc3 h1R 33.Qg1 h4 34.Qh2 Re1 35.Qg1 h3 36.e8B h2 37.Bd5 h1R 38.b7 d1N+ 39.Ke4 f1N 40.b8R Nde3 41.Rd1 Ra7 42.Rd2 a1B 43.Nb5 Bb2 44.c6 Bc1 45.Bc5 a3 46.Rg7 R1h3 47.Bb7 Nd4 48.Ba8 a2 49.Rc7 a1B 50.Ba2 Nb3 51.Bb4 Bd4 52.Rcb7 Ra3 53.c7 Nd5 54.Bf7 Nh6 55.c8R Nf5 56.Bg8 Ng7 57.Rc6 Nf4 58.Rd8 Nge6 59.Rh7 Ba7 60.Rh5 Rh7 61.Ne5 Qe7 62.Nd6 Qe8 63.Ndc4 Rb7 64.Bd6 Rb6 65.Bh7 Ng7 66.Bb7 Ng3+ 67.Ke3 Ke6 68.Ba8 Kd5 69.Bg8+ Nge6 70.Qh1 Bh6
PS 19 July: Bader al-Hajiri, the Wizard from Kuwait, sent me a quicker proof game:
1.h4 d5 2.h5 d4 3.h6 d3 4.hxg7 dxc2 5.f4 b5 6.f5 b4 7.f6 b3 8.fxe7 bxa2 9.e4 a5 10.e5 a4 11.d4 a3 12.d5 c5 13.d6 f5 14.b4 f4 15.b5 f3 16.b6 c4 17.Bf4 Kf7 18.Nd2 Ke6 19.Rb1 Kd5 20.b7 Bg4 21.Rh6 a1B 22.Rg6 a2 23.Bd3 Bd4 24.e8B Ra6 25.e6 Ba7 26.e7 Rb6 27.Ba4 Na6 28.e8B Nc7 29.Beb5 h5 30.Kf2 h4 31.Ke3 f2 32.Ngf3 h3 33.d7 h2 34.Bd6 h1N 35.Qg1 c3 36.Rd1 Ng3 37.Ne5 Ne7 38.Ndc4 Be2 39.g8B+ Ne6 40.Bb3 Rh3 41.b8R f1R 42.Rc8 a1R 43.Ba6 Bh6+ 44.Rg5 Ng6 45.Rc6 c1N 46.Bb7 Ra3 47.Ba2 c2 48.Ba8 Ngf4 49.Rh5 Nb3 50.Rd2 Qe8 51.d8R c1B 52.Qh1 Re1
PS 1 August: Suomen Tehtäväniekat, the Finnish problem magazine, was so impressed by this achievement of their compatriot - who was, until now, completely unknown in the Finnish problem world - that they made his problem the coverstory of the latest issue.
342. 4 July 2007: Steel King goes all the way
Johan Hut sent me the following nice example of a Steel King (winnende wandelkoning) from a recent tournament in the Netherlands.
Smits (2281) - Vermeulen (1902), Utrecht, June 2007
1.e4 e6 2.b3 d5 3.Bb2 dxe4 4.Nc3 Nf6 5.Qe2 Qd4 6.O-O-O Bd7 7.g4 Bc6 8.g5 Nfd7 9.Bg2 Na6 10.Bxe4 Bxe4 11.Nxe4 Ba3 12.c3 Bxb2+ 13.Kxb2 Qb6 14.d4
O-O 15.Nf3 c5 16.d5 exd5 17.Rxd5 Qc7 18.Rg1 Rfe8 19.Qd3 Qf4 20.Nd6 Ne5 21.Rxe5 Rxe5 22.Nxe5 Qxf2+ 23.Ka3 Qxg1 24.Qc4 Qc1+ 25.Ka4 b5+ 26.Kxb5 Rb8+ 27.Kc6 Qh1+ 28.Kd7 Rb7+ 29.Kd8
see diagram
Black could, and should perhaps, have drawn here with 29...Rb8+ 30.Kd7 Rb7+ etc. 29...Qd1 30.Nexf7 Kf8 31.Qe4 Better simply 31.Qxa6 Rxf7 32.Qc6 Re7 33.Qxc5 with good winning chances. 31...Rb8+ 32.Kd7 Qh5 The computer gives 32...c4 as Black's last straw: 33.Qxh7 Qg4+ 34.Nf5 Rb7+ 35.Kc8 Tb8+ 36.Kd7 Rb7+ with a draw. Now it's over: 33.Qe7+ Kg8 34.Qe6 Kf8 35.Ne5 c4 Black has no moves. 36.b4 Ra8 37.g6 Nb8+ 38.Kc7 Na6+ 39.Kb7 and Black resigned.
Smits was awarded the Brilliancy Prize for this game.
341. 13 June 2007: Heavy backrank artillery
In today's second tiebreak game Grischuk - Rublevsky, Elista 2007, after the moves
1.e4 c5 2.Nf3 e6 3.d4 cxd4 4.Nxd4 Nc6 5.Nc3 a6 6.Be2 Qc7 7.O-O Nf6 8.Be3 Be7 9.f4 d6 10.a4 O-O 11.Kh1 Re8 12.Bf3 Bf8 13.Qd2 Rb8 14.Rad1 e5 15.Nde2 b5 16.axb5 axb5 17.f5 b4
18.Nd5 Nxd5 19.Qxd5 Ba6 20.Qd2 Nd4 21.Bxd4 exd4 22.Rfe1 Bxe2 23.Qxe2 Be7 24.Rxd4 Bf6 25.Rc4 Qa5 26.c3 bxc3 27.bxc3 d5 28.Rc6 Bxc3 29.Rd1 Bf6 30.Qc2 Qb4 31.e5 Bxe5 32.Bxd5 Qh4 33.g3 Bxg3 34.Bxf7+ Kxf7 35.Qa2+ Kf8 36.Qa3+ Re7 37.Qxg3 the position on the left arose.
There followed: 37...Qe4+ 38.Qg2 Rb1 39.Rcc1 Qe1+ 40.Qg1 (see diagram on the right) with a wonderfully filled backrank.
After 40...Qe4+ 41.Qg2 Qxg2+ 42.Kxg2 Rb2+ 43.Kg3 Rb3+ 44.Kf4 Rb4+ 45.Kg3 Re3+ 46.Kf2 Re5 47.Rc8+ Ke7 48.Rc7+ Kf6 49.Rd6+ Kxf5 50.Rf7+ Ke4 51.Rxg7 Rf5+ 52.Ke2 Rb2+ 53.Rd2 Rxd2+ 54.Kxd2 Rf2+ 55.Ke1 Rxh2 56.Kf1 Ke5 a draw was agreed.
Thanks to Yngvar Hartvigsen and Ashton Anderson for showing me this.
It reminded me of an old Interzonal game.
Kotov - Matanovic, Stockholm Interzonal 1952
1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.e3 c5 5.Bd3 d5 6.Nf3 Nc6 7.O-O O-O 8.a3 Bxc3 9.bxc3 dxc4 10.Bxc4
Qc7 11.a4 Rd8 12.Ba3 cxd4 13.cxd4 Nxd4 14.Nxd4 Qxc4 15.Be7 Rd5 16.Bxf6 gxf6 17.Qf3 Bd7 18.Qxf6 Qc7 19.Rab1 b6 20.Nf3 Qd8 21.Qf4 Rc8 22.e4 Ra5 23.Ne5 f6 24.Nxd7 Qxd7 25.Qxf6 Rf8 26.Qh6 Rxa4
27.Rb3 Qe7 28.Rg3+ Kh8 29.e5 Rh4 30.Qc1 Rhf4 31.Rc3 a5 32.Rc7 Qb4 33.Qe3 a4 34.g3 R4f7 35.Rc6 Rb7 36.Rxe6 a3 37.Rf6 Ra8 38.e6 Qb2 39.Qf3 Rg8 40.Rf8
(see diagram left)
although in that case, the relevant position only occurred in analysis. In the game, Kotov won quickly after 40...Ra7 41.Rxg8+ Kxg8 42.Qd5 and Black resigned; there is nothing against e7 and/or Qd8+.
In the tournament book, Kotov showed a more interesting defense: 40...Rbg7 41.Rd1 a2 42.e7 a1Q 43.e8Q Qaa2 44.Rd8 Qab3 45.Qa8 Q2a2 (see diagram on the right) when seven heavy pieces are attacking and defending g8. White then wins with 46.Rxg8+, and Black has the choice between 46...Qxg8 47.Qxa2 and 46...Rxg8 47.Qe5 mate.
PS 15 June: As Charles Milton Ling remarks, White has a funnier, be it slower win in the second diagram by a massive swap on g8.
© Tim Krabbé 2007