The Chess Variant Pages

Check out Marseillais Chess, our featured variant for February, 2024.

This page is written by the game's inventor, Jens Nielsen.

Darkness Chess

by Jens Bæk Nielsen

Torben Osted and I have played a correspondance game of chess with changed rules.
It is inspired by the version "Are there any?" (Kriegsspiel), and like that version the goal of the game is to come as close as possible to a real war. To do this, the most important is that you do not exactly know where the opponents pieces are! We have called this variant for "darkchess", as you feel like moving in the dark, when you play the game.
The startposition is as usual. The rules are easy if you imagine, that a piece is able to "see" the squares it attacks and can go to (Torben invented this essential rule for this variant). If you open e2-e4 you are told what the pawn can see on d5,e5,f5, and what the bishop and queen can see on their opened diagonals. For each single move both black and white are told what they can see.
We have discarded the en-passant move, as it is difficult to handle and does not influence the game anyhow. If you have an uncovered piece which becomes captured, you are only told that it has disappeared.
The goal is not to mate the opponents king, but to capture him! This means you are not told if your king is in check or you put him into check yourself. During castling the king is allowed to jump over an attacked square.
The game is not so serious as normal chess, but quite amusing and exciting!
To run the game we used a third part. It was my wife who acted as gamesupervisor (the task could be done by a computerprogram with passwords for black and white. Who makes it?).

Now to the game which was played from 7 nov. 1989 to 30 may 1992, but first an explanation to the notation of the information given from the gamesupervisor:

  1. The empty squares a moved piece can see are not notated at all (blacks first 2 moves).
  2. A "-" denotes, that no information was given for the opponents move (whites first 6 moves).
  3. Any information is shown in brackets ().
  4. If an opponents piece moves to an attacked square, the piece is shown in brackets (whites 7. move). Similar notation is used for which opponents pieces a moved piece can see (blacks 3. move). Several pieces are separated with a "," (blacks 14. move).
  5. If a move for either player opens for a row/line/diagonal of a piece, the information is given after a "/" as shown in whites 22. move where he played Ne2-c3. The move opens the view of blacks Ba6, which can see an empty square at f1 (empty squares are always notated here to assure, that this information was not forgotten).
  6. If an uncovered piece is captured it looks as whites 15. move.

To give the right feeling of the game, it is only reported from blacks view (the whole game follows).

White: Torben Osted
Black: Jens Baek Nielsen


Hoping that white has not played g3 and Bg2.
Still hoping the same! But with the look at g2 I would now welcome the pawn to move.

I have made unusual moves to make it harder for my opponent to guess my moves. I tried to use the rook as a spy and now lost the exchange. The loss of material is probably not so important in this variant of chess, where information is important. But I had with -,cxd6 got a cramped position, that requires many moves to develop.
15.(Ne4 captured),d5(Pd4,Be4)

It is not covered, but white did not know!

Something had to happen...

I feared a capture at d5, as white probably had a rook at the e-file. The move also cleared the 8. row and gave an unusual placement for my king which might be hard for white to guess.

I had to see if the white queen should arrive at b5 giving check. I did not bother to cover Na6.

The white queen actually came to b5! I had not expected this after whites safe play so far and realised I could have won with the trap 25....,Qa8 and 26....,Rb8.
At this point I felt I had a reasonable position with the plan h4 and g3 (hoping white had castled kingside), but white now forced exchange of queens and knew where my king and knight were positioned.
28.(Qb5 captured/Pa5-a4),Nb4

Would lose if white played Rxf7+. I had let my king stay at d7, as white probably had assumed I had replaced it after 27. Qb5+. He probably assumed the king covered f7. My plan was now to play the bishop to h6 to support the pawn to d2. 41.(/Bg7-Pb2),Rc2(Pb2,Ph2)
I assumed white played the knight to d1 and dropped the plan to promote the d-pawn. But I better move the rook if white had played 41.Ne4.
White moved his h-pawn (and has earlier moved the g-pawn), and I cannot stop them. With this move I took a riscy (white could have a rook at f4 or more likely e4) look at g1. But of course whites king was not on a black square.
I had a strong feeling, that whites king was at h1. If this chance had to succeed, it required that
1) h2 was only covered by the white king
2) the g-pawn was not at g3 (or unlikely a rook at f4)
3) white did not know if my bishop covered g1-a7 or h2-b8.
44.(Rh2 captured)
White captured the rook and said he would play his king to h1 next move, so I resigned. He had found it most likely, that I most would have liked him to play 44.Kg1 and have had the pleasure to play 44....,Bxg1 winning the game and have fooled him. He is right! But it is still a fifty-fifty chance, as I could have thought he would think so, and have played my bishop to e5 etc.
42....,Be5 had won the game.

Here is the complete game.
1.d4,Nf6 2.Bf4,b5 3.c3,Bb7 4.Qc2,a5 5.e3,Ra6 6.Be2,Re6 7.Bf3,Qc8 8.Nd2,Rd6 9.Bxd6,cxd6 10.a4,b4 11.Ne2,Ba6 12.Rc1,d5 13.e4,dxe4 14.Nxe4,Nxe4 15.Bxe4,d5 16.Bf3,e6 17.0-0,Qd8 18.Kh1,Bd6 19.Rfe1,g5 20.Qb1,h5 21.cxb4,Bxb4 22.Nc3,Kd7 23.Be2,g4 24.Bxa6,Nxa6 25.Re3,Qb8 26.Qd3,Bd6 27.Qb5+,Qxb5 28.axb5,Nb4 29.Ne2,Rg8 30.Ra3,h4 31.R1c3,g3 32.fxg3,hxg3 33.Nxg3,Rb8 34.Ra1,Rxb5 35.Ne2,e5 36.dxe5,Bxe5 37.Rf3,d4 38.Raf1,d3 39.Nc3,Rc5 40.g4,Bg7 41.Nd1,Rc2 42.h4,Bd4 43.h5,Rh2+ 44.Kxh2,resign

© 1997: Written by Jens Bæk Nielsen; edited by Fabio Forzoni; copied from Fabio Forzoni's WWW pages with his permission.

This game can be played via email on Richard's Play-By-eMail Server.

WWW page created: June 11, 1997.
Mar 2000: D. Howe added link to Richard's Play-By-eMail Server.