Dr.D's Check-Up: Mostly Solutions
By Dr.Steven Dowd   
September 18, 2007
problementry.jpgSince so much is going on in the problem world right now (see, for example: http://www.chessfed.gr/wccc2007 , our own FIDE convention; some of you may want to get in on some of the composing tourneys), I am going to give you just a few problems and then we'll concentrate this month on last month's solutions, and some teaching to help you understand the problems better. I've included several comments and observations by readers this month. Be sure to e-mail me with questions as well as solutions at DrDowd@uschess.org.The selection this month is from the 1957 Dutch Jaarboek of problem chess, which I acquired as a birthday present (I hit 50 this month) for myself! 9 ladder points are at stake.

P. Petkow
Probleemblad 1957

White to play and mate in 2



White to play and mate in 3

H.P. Rehm
Probleemblad 1957SeptDrD3.jpg

White to play and mate in 4

About our selection: GM Petkow is well-known in the fairy chess area and in selfmates (I believe he has enough FIDE Album points for two GM titles), but rarely appears in directmates, so this was a welcome find. The Czech master shows his usual form, as does GM Rehm in an early problem – which will relate to one of of our problems from last month. Good luck!

Solutions from last month

Tom Langland provides us with our first four – where I phase in will be in italics. I am also using his model for reporting as the model; it is clean and efficient while documenting his thought processes. I can't stress this enough; recording your thinking processes, no matter how wrong-headed they may seem, provide a lot of illumination into all the details of a problem, and will make you a better solver.

001 Ahues-


1. Bg7
White has a nice discovered check with Ne7+ but the h3B is then hanging.   Trying to move it with 1. Bf4 doesn't work because then we don't have 2. Nxf4 mate if black blocks our discovery with 1. ... Ne4.  1. Bg5 doesn't work either because this blocks the g-file and allows black to play 1. ... Nd3 and white doesn't have 2. Qg2 mate anymore.

002 Gebura-

 1. Qd3

Guarding the d4N threatens 2. Rc6.  If black captures the d4N then the queen goes to the 6th rank and mate follows.   What separates 1. Qd3 from 1. Qf2 or 1. Qg1 (all guard the d3N while preserving the ability to go to the 6th rank), is with the key 1. ... b3 2. Nxd3+, white's Queen either goes to c3 mate or on 2. ... Kb3 it covers c4 so 3. a3 is mate!

Good job, Tom. You identified the most relevant tries and saw the weaknesses white needs to attack. From there, even if a solver thought 1. Qg1 and 1. Qf2 were the key, would be that further probing would show their weaknesses – and thus, the need for a more spectacular key. And yes, it is getting heavy piece(s) to c6 that is important, in most cases.

It is really the second queen sacrifice that sells the problem to me, and prompted me to compose a number with the theme "Sham or actual queen sacrifice as the key, with a follow-up queen sacrifice in the continuation."  Here Gebura puts two actual queen sacrifices together. In problem chess we use the same terminology as Spielmann proposed years ago – a sham sacrifice is one that cannot be taken. For example, if accepting the queen sacrifice leads to a mate in two when the stipulation is mate in three, that is a sham – black can't take it. But 1. .. cxd leads to a neat mate in the stipulated number of moves. 2. Nb3+ Kc4 3. Rc6# - an amusing example of a king trapped by his own pawns.

And of course, actual queen sacrifices are better than sham sacrifices!

003-Le Grand


  1. e4
The key may look obvious;  dealing with the defenses is not.   On 1. ... Qxg3 white must play 2. f5 to threaten 3. Rxd7 now that one of the double-covering pieces on the c8/h3 diagonal is gone.   If 1. ...  Re1 then 2. Bc4 threatens 3. Bxc7 mate, again because one of the double-covering pieces on the c-file is  now gone. As I noted, the pseudo LeGrand theme is shown here, for those who think thematically.

004  Bogdanov-

  1. Bd7

After the key, black has only three moves.   On 1. ..   Ke4 then 2. Nb4 and mate with 3. Bc6 is next, or 1. ... c3 which allows 2. Bb5+ and 3. Nxc3 mate, lastly 1. ... e4 is dealt with by 2. Ba4 with mate by 3. Bc2 or 3. Qe3.
So far, I can see removing f6P, f5P, g2P, g4P, g5P, h1B retains the solution, but I can't see how to remove any more than that.

Thanks Tom – now a few comments of my own. First, the single step of the bishop is a favorite schematic device of composers. To have a piece that is capable of so much movement – one reason why Pf5 is necessary – take a single step is often as impressive as having it traverse the entire long diagonal. And the "short first step; long second step(s)" is very appealing to us.

Very close! The white pawns on f5 and f6 prevent duals and cooks. But all else is not needed. Let's use an example of another problem to look at this, and to do so we will visit Brian Stephenson's website and his database program, Meson. Brian is a tireless volunteer for problem chess, organizing all sorts of solving championships and has a column on studies in the British magazine Chess.  A look in his online database (http://www.bstephen.me.uk/chessproblems/meson/meson.html)  is appropriate here, especially given our topic of "Googling the grandmasters."

For years, I have wanted to create a new Bohemian – we talked about them in a previous column and many of you are familiar with them. There is an almost puritanical emphasis on model mates, echoes, and so on. Of course, the great composers have already mined most of the positions, so every find is a nugget of gold. But each failure  - about 9 of 10! – is painful.

Here is a problem by Mandler:

Ceskoslovensky sach 1960

White to play and mate in 4

Solution: 1.Be7 Kc3 [1...Kxc4 2.Qe3 threatening Bd6(#2) 2...Kd5 3.Bb4 Kc4 4.Qe4#] 2.Bd6 Kxc4 [2...Kd3 3.Be5 Kxc4 (3...Ke4 4.Qe2#) 4.Qd4#] 3.Qe3 Kd5 4.Qd3#

If you take a look at the four solutions, you'll see some wonderfully echoed model mates. It isn't just a packet of nice mates, they all relate to each other.

It is a heavy position, but was full of try-play: 1. Bxh6? (a heavy handed capture) Kc3!; 1. Bb4? Kxc4! and  1. Bd6? Kc3! It should take a bit of time before the refutation to 1, Bd6 is clear.

How does this relate to my quest? Well, every time I achieve a problem with what I wanted, I found myself anticipated, and this is no exception. Although my friend Chris Feather says, with some justification, that there is nothing worse than being anticipated and finding that the original author did it worse than you did, ruining your effort, being scooped, even by the masters, is painful. The problem below was my idea:

Steven Dowd, after Mandler (even though I composed it before I knew of Mandler!)

White to play and mate in 4

Solution: 1.c5 Kd3 [1...Kxd4 2.Qf3 threatening Be6(#2) 2...Ke5 3.Bc4 Kd4 4.Qf4#] 2.Be6 Kxd4 [2...Ke3 3.Bf5 Kxd4 (3...Kf4 4.Qf2#) 4.Qe4#] 3.Qf3 Ke5 4.Qe3#

It looks more attractive, but it loses all the try-play and what does white do besides 1. c5? My original point was that the taking of d6 was not all that obvious; I've been disabused of that notion. So I have the nice mates, but no try play – the master did it better. He achieved the "one step" of the bishop in a very artistic way. Meson saved me from submitting something to a magazine that would have been potentially embarrassing, especially if someone noticed the superior predecessor.
005 Stratikis)Dowd5.jpg

1. Bg3
1. Bg3 fxg3 2. Rh2 gxh2 3. g4#  

Barry Keith notes: I wish I'd composed this!  It has also amused my eight-year-old students.

Barry, I agree! Most amusing; it is one of those simple finds. Tom Langland also notes: The theme of this problem is to get rid of the f4P, so white can mate with 3. g4.  After the key, black has to capture and the then the splendor of the problem is shown with 2. Rh2!   Black is again in zugzwang, and must capture again.  Lovely is how the h4N is kept pinned throughout..

One of the most pleasurable things about this column has been communicating with teachers who use chess problems for my column in their teaching!
For Barry's students and students of chess everywhere, I offer the following guilty pleasure. It can't be new, and it is a Mansube – yes, the ancient Arab manuscript – referring to a type of problem where checks, captures, and so on are fair game. They are not high art but they are fun. Not presented for solution ladder credit, but for the fun of solving. All checks, captures, an ideal mate, and that's all the hints you get.

Steven Dowd
For Barry Keith's students and all chess students!


White to play and mate in 6
OK, maybe one or two more hints: the first move is the same as the last, and white ends up with only one piece. Another way to stump people is to ask, "mate in 6, which piece mates?" ala Sam Loyd!

006 Makaronez-Dowd6.jpg

1.Bg8 and 1.Ka4

Not too many comments here. In (a) 1.Bg8 threatening Rd5#(#1) 1...Kxd3 2.Bd5 Kxd2 3.Bc4 Kc1 [3...Kd1 4.Kc3 Kc1 5.Re1# is the nice echo] 4.Rd5 Kb1 5.Rd1# , a satisfying Rex Solus that then had (b) BPd3 which changes it from an echo on the rank to a single line zugzwang problem: 1.Ka4 Kc4 2.Ra5 Kd4 3.Kb5 Kd5 4.Ra6 Kd4 5.Rd6#.

007 Wispaelare-Dowd7.jpg

1. Kg6

Would you be surprised to know that these kind of zugzwang problems are the most often missed? I am not sure why, but even very good solvers missed 1.Kg6 fxe4 2.Nh5+ Kg4 3.Be3 Kh4 4.Kf5 threatening Bg5+ (#2) 4...Kxh5 5.Bg5 threatening g4# (#1) 5...e3 6.g4#. By "which way" I meant that one had to decide which way to move with the king; one expects the knight will be sacrificed, so the main choice here is which way to move with the king (the withdrawal 1. Sg3, which would be an interesting key if it worked, in my opinion, but I could not see any way to get the king in the corner in time).

008- Kozdon –

1. e8R

Anatoly Goldberg provided the only full version: a very difficult problem until you see the solution, then it is easy! But I imagine most of you were as impressed by the depth of the conception as I was….. The battle of rook versus bishop has always impressed me; I've begun to call these sort of things "exhaustion problems" as the poor rook is hunted until he can do no more. I see them as a step up from the zugzwang type of longer problem; an extended zugzwang to extinction. The black rook is allowed to chase the king for awhile, when he can't the so-called logical combination takes over.
1. e8R  Rg5  2. Bd3  Rg4+  3. Kf5  Rg5+  4. Kf6  Ra5  5. Re4+  Kh5  6. Be2+  Kh6  7. Rh4+ Rh5  8. Rxh5    
Hope you have enjoyed this set of solutions and their discussion. Till next month!