Another puzzle from Dragoslav Andric’s 1981 book “Matni Udar”.

**White to play and mate Black**

Vidmar v Adam, corres 1037

**Solution**

With mate threatened on b2, White has to do something drastic, meaning checks, and with the chance of Black’s King slipping away to the d file and then c8 or even c6, something has to be done.

After examining the normal checks like 1 Qg7+ and 1 Nf5+, and finding them wanting (and Fritz confirms that all moves lose except for the solution) I found the decoy 1 Re4+! and saw immediately that ‘it worked’. And a bit of analysis on each of Black’s several defences proved it.

1…Re4 2 Qd5! is elegant: hitting the LPDO Re4, threatening 3 Qf7 mate, and if 2…Re6, the Q still reaches f7 via 3 Qd7+;

1…Be4 2 Qg7+ Ke6[] and either 3 Qg4+ or 3 Qf7+ start a check-check-check king hunt;

1..Kd7 or 1…Kd8, 2 Qd5+, and the best Black can do is 2…Qd6, when 3 Qb5+ or 3 Qf7+ win the Queen.

**FEN**

4r3/ppp1k3/5p1N/n2b2Q1/1qp3R1/8/PP2rPPP/2KR4 w – – 0 1

Another puzzle from Dragoslav Andric’s 1981 book “Matni Udar”.

**And not just another puzzle, but a gem of a puzzle, and really hard: or I found it so. But all the more enjoyable when solved.**

**White to play and mate Black**

Efim Korchmar v Abram Poliak, Ukraine 1931

**Solution**

After looking and looking, and putting the pieces out on the board, I tried most moves, without success. I then did a Purdy based analysis, also noting John Nunn’s LPDOs, thinking (correctly) that the first move must be a **double attack**. See my anecdote about what Levon Aronian told me about double attack.

So, 1 Nb4! and then it started to come into place.

Black can avoid mate by for instance 1…Rd7, but is clearly lost after 2 Na6 ba and 3 c4 or 3 Bf4; but the fun begins after 1…ab, when 2 Qd6! is the point, removing the defender of e8: the N was tied to defending e8 in the opening position.

Black’s best is 2…Qd7, since taking the Queen loses to 3 Re8+ Rf8[] 4 Rg7+ Kh8[] 5 Rf8 mate. After 2…Qd7 I was delighted to find the magical move 3 Qd5!!, exploiting the fact that the Qd7 is now tied to e8.

The fun doesn’t stop here. White threatens 4 Rg7+, and if 4….g6 5 Rge3! 1-0, so 4…Kf8 when the lovely 5 Rg7! breaks through nevertheless.

The Black Q is LPDO if 5…Rg7, so instead 5…Qd5 but then 6 Rg8+! and 7 Re8+ mating.

Lovely: great piece play.

Probably my favourite puzzle in the book so far.

The game isn’t in Megabase 2012 (my latest version of Megabase) but I have uploaded it here. It is a nice game to play through, to see how White’s attack in the Ruy Lopez Steinitz defence developed so ‘easily’.

**FEN**

3b2k1/1pp2rpp/r2n1p1B/p2N1q2/3Q4/6R1/PPP2PPP/4R1K1 w – – 0 19

Another puzzle from Dragoslav Andric’s 1981 book “Matni Udar”.

**White to play and mate Black**

Znosko-Borowsky v unknown, 1937

**Solution**

Elementary today (but: forewarning: tomorrow’s will be far harder). 1 Qd4! and if 1…Nd4, 2 Nf6+ and 3 Bh6 mate.

Best to take a rest, and get ready for tomorrow’s puzzle, which I found extremely tough, but very satisfying when solved…

**FEN**

r2qk2r/pppbnp1p/2np2p1/3N2B1/2BbP3/8/PPP2PPP/R2QK2R w KQkq – 0 1

Another puzzle from Dragoslav Andric’s 1981 book “Matni Udar”.

** White to play and win **

**Solution **

Straightforward today. 1 Rh7+ Kh7 (else 2 Rff7) 2 Rf7+ Rf7 3 Qg6+ Kh8[] 4 Qf7 and the fact that Black’s Queen is LPDO means the Be7 is fatally pinned, and if e.g. 4…Ng5, 5 e6+ wins.

**FEN**

2r2r2/3qbpkp/p3n1p1/2ppP3/6Q1/1P1B3R/PBP3PP/5R1K w – – 0 27

Another puzzle from Dragoslav Andric’s 1981 book “Matni Udar”.

**Black to play and mate White** (though White can prolong the game, with a lost position)

Zuravljev v Koskin, SSSR 1936

**Solution**

Fairly standard Dragon stuff: 1…Nf3! 2 Nf3 (2 Qg2 say would prolong the game, but 0-1 with two pawns down) Ne4 (or 2…Rb2+ 3 Kb2 Ne4 and the pin decides)

3 Ne4 Rb2+ 4 Kb1, and now the pretty move 4…Rb1+! 0-1

**FEN**

1r1q1rk1/p2bppbp/3p1np1/4n3/3NP1P1/2N1BP2/PPPQ3P/1K1R1B1R b – – 0 13

Another puzzle from Dragoslav Andric’s 1981 book “Matni Udar”.

**Black to play and mate White**

Kotov v Bondarevsky, Leningrad 136

**Solution**

I find such positions, where the pieces are all tangled and confused, difficult. Therefore, mainly by elimination, looking at all other moves (mainly captures and checks) I found the correct move 1…f4! when 2 Nf4 is forced, letting the Queen in, 2….Qf2+ when White again has a forced reply, 3 Kd3.

I then found the prosaic, and winning, 3…Ne5+ 4 Be5[] Be5[] , and also noted that 3…Ndc5+ 4 Bc5[] Bc5 is overwhelming, but the move I missed is 3…Qd4+! and mates: 4 Kd4[]

4…Bc5+ 5 Kd3 Ne5 mate.

**FEN**

r1b3k1/pp1n3p/2pbp3/3pNpP1/2PBn2q/1P1NKP2/P1Q1P1B1/R5R1 b – – 0 1

A break from my daily puzzle from Matni Udar. Today, an occasional one: I have long (decades) had a habit of cutting out, tearing out, photocopying…puzzles I can’t solve, and saving them for a rainy day, dog walk, hill walk…

I came across today’s problem a while ago, and it really stumped me. I eventually solved it, but only by setting the pieces out and moving them around: I would never have found the win in a game.

**White to play and win**

B Socko v S Brunello, Italian Team Championships, 2015

**Solution**

I looked at the problem many times, each time my first thought being 1 Nf5 gf 2 ef, but not being able to crack 2…Rd6

I also looked at 1 Ng6, finding it insufficient, and 1 f4, which I thought was strong.

**1 f4**

In fact, 1 f4 leads to some White advantage, but not much. The key is noting that the Rd8 is LPDO.

1…ef 2 Qf4+ Kh7

3 Ng6! Ng6 4 Rg6 Qg6 (4…Kg6 5 Rg1+ Kh7 6 Rg4! mates, a computer line)

5 Qc7+ with a double attack on the King and LPDO Rd8: the game is equal, with White a pawn up but his King exposed after 5…Kh6 6 Qd8 Qe4 7 Qd2+

So, my judgement on 1f4 was sour: maybe White is a pawn up, but the opening up of his King should dissuade White from entering the line.

Back to 1 Nf5!

**1 Nf5!**

In a game, I would have tried, tried and tried again to make 1 Nf5 work; I would have failed, and then reluctantly made a lesser move. But this is what I should have seen:

1…gf 2 ef Rc6 (the move played in the game: 2…Rd6 or 2…Rb6 are the same).

3 Ra4! – as far as I could see, but I felt that Black had sufficient resources.

In fact, I was wrong: Black is lost, with White having multiple threats.

The easy one first 3…Qh5?? 4 Qg7 mate.

The harder one next: 3…Rg8. Evenually, by setting up the pieces in this position, I found 4 Qh4+ Qh5[] 5 Rg8! Ng8 6 Qg3!

and, to my surprise, found that it is 1-0: Black can’t stop 7 Rh4 winning the Queen: if 6…Qg5 then 7 Rh4+ Kg7 8 Rg4 wins the Queen all the same.

But there is even better after 3…Rg8: 4 Qg8!!

Black’s Queen is tied to defending h5 (against Rh4+) so 4…Ng8[] 5 Rh4+ Qh5[] and now a move I missed, 6 Rg6+!

The prettiest one: 3… Rd2, played in the game, 4 Qg7+!! and after 4…Qf7, 5 Rh4 mate: the LPDO Queen makes this tactic work.

So, wonderful play by Socko, and, alas, an example of how the depth of calculation and evaluation was beyond me. I hope I have learned the lesson of how defenceless Black was with his King on the open g and h files.

**FEN**

3r4/1pp1nq2/p3rppk/4p3/4P1QN/1PP4P/1P3P1K/R5R1 w – – 0 30