allanbeardsworth

Black to play and win

H Steyn v JJ Steenkamp 2001

Solution

If 1…Kh6, white wins by 2 Be3+ forking king and queen. So what else? The interference 1…Bd7! is a natural try, and if 2 Qd7+ the queen is deflected; or if 2 ed black can mate with 2…Qg2+ etc.

Posted with Blogsy

White to play: can he defend?

(Not one to spend much time on: not the best of puzzles today)

GW McElhinny v JG Heldzingen 1950-52

Solution

A really odd puzzle, especially because the rubric says white, in this correspondence game, resigned here, and Reitstein's question was what was Black's threat which caused white to resign.

So, the threat is obvious, Rh6, and one is tempted to find some reason why white was wrong to resign, but it doesn't take long to see that white's position is indeed hopeless.

Let's call it a day off.

Posted with Blogsy

White to play and win

C Hoek v GA Hopkins 2004

Solution

A nice one today. White clearly has to breakthrough before black can muster his defences, but with black's restricted space he has some time. So he needs to bring all the pieces to the party. Two biffs have to be examined, and 1 Bf8!! is clearly better by comparison than 1 Rf8+ (because the latter permits, in fact forces, simplification) and after 1….Qc7 white has the nice rook lift 2 Rf6! aiming for h6 and h7. So 2…Rd8 3 Rh6 Rd7 defending h7 with equal numbers but 4 Bd6! deflects one of the defenders away.

Black could also try 1…Qg6 or 1…Qg8, both met by 2 Qf4, though 1…Qg8 can also be strongly (and better) met by the rook lift 2 Rf3!

Posted with Blogsy

White to play and win

M Katz v E Pfister 1974

Solution

Full credit to white for finding the solution: knowing it is a problem, this was was a 'toughie', so in a game, it would have been far, far, harder. The obvious move to play, 1 Re6, leads to what I suspect could be a drawn position after 1…Qe6[] 2 Qe6 Re6[] 3 Ra8.

And I think black's h pawn means that he might be able to draw this position: although 'further investigation is required'.

However, I am very confident that in a game, as white, I would have played for the line, and hoped to win it (some months back, I saw an excellent video series by Alex Yermolinsky on ICC on similar to endings, and this taught me some of the common themes of this type of ending, or R+3 each on king side/ white having a passes offside pawn.

Read more…

Black to play and win

A du P. Heyns v PJ Foley 1995

Solution

A rest day today: 1…Bf3 is the obvious pressurising move, if nothing else it constricts white's choices (particularly preventing 2 Re2). After a quick inspection it is obvious that white is in zugzwang: and a bad zugzwang, at that: 0-1.

Posted with Blogsy

White to play and win

W Kobese v R Banai 1996

Solution

At first, I looked at the wrong move order: 1 e6, to disconnect black's queen from the kingside, followed up by 2 Bh6, but it gives black one more option (of 2…ef) compared with the better 1 Bh6!.

After 1 Bh6!, the win is trivial: 1…Bh6 2 e6! and the bishop falls (2…Kg7 3 Qg4+) after which white has too many pieces in the attack, with the Ra1 easily activated to boot, compared with all black's pieces being off side.

Posted with Blogsy

White to play and win

JJ Steenkamp v MG Smith 2002

Solution

The first move is fairly straightforward: the body count on the king side, with white having three pieces to black's one defender, makes the sacrifice 1 Bg6! automatic: especially since white remains with a queen and knight, a good attacking combination, plus the Nd8 being a LPDO, plus, for good measure, the easy with which the Rb5 can be 'lifted' ('downed') to b3 and then over to the f file.

So, 1 Bg6! fg 2 Qg6+:

2…Kf8 3 Rb3!; 2….Kh8 then probably 3 Qh5+ (always useful to gobble pawns) 4 Qg5+, 5Qd8+ (perhaps prefaced by Qe7+, depending on where the king is) 1-0.

Black could instead struggle on. 1…c6 is a tougher defence, met either by 2 Bh7+, and white is a safe pawn up, with the advantage to boot: and the availability of the safety check is all that is needed to know the sacrifice is playable. Or, white has (1…c6) 2 Bf7+! Nf7 3 Ra5! overloading the queen; 2…Kf8 3 Qh7…these lines don't really need calculating, being dependent on black's choice of poison.

Posted with Blogsy

White to play and win – forced mate in 5

A Calitz v L Bouah 2000

Solution

Knowing that the problem is a mate in five, one has to look for forced lines. There are a few random checks which need to be looked at, such as 1 Rd4+?? 1 Qc6+?? 1 Re6+??, but one stands out as needing to be signed: 1 Nb5+!

Black's play is forced: 1…cb 2 Qa6+ Kc7[] (2…Bc6 3 Qc6 mate) 3 Qa7+ Kd6[] (3…Kc8 4 Qb7 mate) 4 Qb6+ Bc6[] 5 Qc6 mate.

A nice exercise, good practice for similar cases where you know there is a mate on board, just can't immediately find it.

Posted with Blogsy

Black to play

White has just played 1 Ke1-e2

D Morschel v K Whyld 1974

Solution

1…Ne5! (examine all biffs) and if the N is taken, then 2…Be5 forks both LPDO knights, and white's last move has prevented them from defending one another.

Reitstein says that black has won a pawn, and has the better game. This isn't correct: white retains an advantage after 3 Raf1 hitting the f5 pawn. If black defends it, 3…e6?, then he loses after 4 Bb6! and 5 Rf3- the crafty rook manoeuvre both hits the f5 pawn and prepares this tictac. So, after 3 Raf1, black must take one of the knights, and white captures on f5, and is a pawn up with the better position.

Posted with Blogsy