Check out Symmetric Chess, our featured variant for March, 2024.

This page is written by the game's inventor, Ralph Betza.

The Rules of DemiChess

This game was introduced in The Crab. It is called DemiChess because each piece is worth at most half as much as the corresponding piece in FIDE Chess. Given the names of the pieces, one might also call it "chess on the half-shell"!

In the rules below, I changed the name of one piece, and added a few clarifying words here and there. The original rules are basically unchanged.

  1. The rules of FIDE Chess apply except as follows
  2. The Rook is replaced by a piece called the Snail, which moves one square Rookwise in any direction or leaps two squares straight forwards. (The word "leap" implies that this piece can jump over obstacles, and therefore can play 1. (WfD)a1-a3.)
  3. The Knights are replaced by Crabs, ffNbsN, which move forwards as narrow Knights or rearwards as wide Knights.
  4. The Bishops are replaced by Lobsters, bFfA, which leap either one square diagonally to the rear or two squares diagonally forwards.
  5. The Queen is replaced by the Oyster, WfDffNbsN, which combines the powers of the Crab and of the Snail.
  6. In addition to checkmate, the game can be won either by stalemating the opponent or by "bare King": if it is your move, and you have any other pieces in addition to your King, but the opponent has only the King, you have won the game.

Gearshift Scotch DemiChess

A new type of Progressive chess is called Gearshift (or Cambiamarcia).

Each turn, you can play one more move than your opponent just played, or one move less, or the same number of moves. When "check ends the series", it also changes the number of moves (so you can cut back from 10 moves per turn to one move, by giving check). Of course, you have to make at least one move, other rules of scotch chess apply, and if you are stalemated when you have played one move less than your opponent, it's your choice whether the game is over or if it's your opponent's move (it's stalemate if you choose to play "more moves", the game continues if you choose to play "fewer moves".) (In DemiChess, of course you wouldn't choose stalemate!)

Here is a game of Gearshift Scotch DemiChess:

White: Joao Neto
Black: Ralph Betza

  1.   c2-c4
  1... Lc8-e6
  2.   c4-c5,   f2-f4
  2... Cb8-c6,  f7-f5
       (3. e2-e4, e4:f5, f5:e6; 3...Cc6-d4, Qd8:e6, Qe6:f4, Qf4-e2+
        demonstrates the dangers of speeding up.)
  3.   Cb1-c3, Lf1-d3
  3... Lf8-d6, Ld6:f4          (gives the King a flight square)
  4.   Cc3-d5, Cd5:f4
  4... g7-g5,   g5:f4
  5.   Cg1-f3,   O-O
  5... Rh8-h6,  a7-a5          (try to develop as many different
        things, so when it speeds up I will hope to find something good)
  6.   Ra1-a3, Ld3:f5          (this made me say ouch)

   r . . q k . c .
   . p p p p . . p
   . . c . l . . r
   p . O . . L . .
   . . . . . p . .
   R . . . . C . .
   O O . O O . O O
   . . L Q . R K .

  6... Le6-g4, Lg4:f5,  a5-a4
  7.    g2-g4,  g4:f5, Rf1-f2  (? perhaps "Q-d3-d4:c6:d8+" ?)
  7... Qd8-d6, Qd6-d4, Qd4:d2, Qd2:d1
  8.   Rf2-f1, Rf1-e1, Re1:d1, Kg1-g2
  8...  b7-b5,  b5-b4,  b4:a3,  a3:b2,  b2:c1=3D(WfDffNbsN)
  9.   Rd1:c1, Cf3-e5, Ce5:d7, Cd7-b6, Cb6:a8, Ca8:c7 !
  9... Cc6-d4, Cb4:e2, Ke8-d7, Kd7:c7,  e7-e6,  e6:f5, Kc7-d7
 10.   Kg2-f2, Kf2:e2, Ke2-f3, Kf3:f4, Kf4:f5, Rc1-c3, Rc3-c4
 10... Rh6-h4, Rh4:h2, Cg8-f6, Cf6-e4, Ce4:c5, Kd7-d6, Kd6-d5, Kd5:c4
    resign
 (or 10..Kd7-c7-b7-a6-b5:c4:c5, Rh6-g6-f6+! (fortress))
This is the first Gearshift (CambiaMarcia) game ever played, so who knows? We both followed the timid strategy of limiting our opponents' moves by playing fewer ourselves. I hoped this would be possible... When you don't feel like thinking much, "if I play 3, I have to worry what he'll do with 4"; but if you were playing for a world championship, you would calculate more -- and there's more to calculate than in normal Progressive.

Pure DemiChess

It took me a while to get around to it, but I finally sat down and played a few games of pure DemiChess, one move per turn. An average game takes more moves than an average game of FIDE Chess, and most people don't seem to like "Chess with weak pieces", so I simply played both sides.

In the 1960s, I played a few hundred games of five-minute Shatranj with Alex Dunne (who now runs the USCF Postal Chess division), and I learned to like "Chess with weak pieces".

I was surprised to find that DemiChess seems to be more tactically involved than Shatranj. Perhaps it was merely because I am unfamiliar with DemiChess, but time and time again, I surprised myself by making a slow threat (for example, "the Crab will retreat to there, and then there, and then advance twice, and attack that pawn with its retreating move", giving the other side 5 moves to defend it) only to find there was no defense.

The pieces are awkward, and if you don't look far ahead you can be blindsided dreadfully.

In one game, after the symmetrical opening 1. e4 e5 2. Cg1-f3 Cg8-f6 3. Od1-e3?! Od8-e6 4. d3 5. Lh3 and so on, the following position was reached:

 . . . . s . k .
 p l s . c p p p
 . . . . o . . l
 . . c p P . . .
 . . . . . . . .
 . . S . O C . L
 P L C . S P P P
 . . . . . . K .
After 1. Cc2-d4, the Oyster is attacked and can only go to f4; then on 2. g3, Of4-h5 is forced, and 3. Kg1-g2 hopes to trap the Oyster.

 . . . . s . k .
 p l s . c p p p
 . . . . . . . l
 . . c p P . . o
 . . . C . . . .
 . . S . O C P L
 P L . . S P K P
 . . . . . . . .
I replied to 3...f7-f5 with the blunder 4. Cf3-d2; instead, 4. Lh3:f5! Ce7:f5! 5. Oe3:f5 g7-g6 6. Of5-f6+ (interesting is Of5-h4, but I'd rather give the Pawn back than create a doubled pan and a hole on f4) Oh5:f6 7. e5:f6 Kf7 8. Cf3-e5+ Kf7:f6 9. Ce5-g4+ Kf6-g7 seems to be a slight advantage for White: the Lh6 is attacked, has no moves, and is defended by the King that occupies the Lobster's retreat square (Aha! look more closely! It is possible for the King to move away, threatening Lh6-g7, and if Cg4:h6 the King simply returns to g7 and the Crab is trapped!); the Pd5 is isolated, and Cg4-e3 can attack it; and White has control of the interesting center square d4.

Alternatively, 4. e5:f6 en passant would be an attempt at getting a simple advantage. I think it would fail because Cc5-e6 is a good defensive resource, and Lh6-g7 with Lg7-e5 is strong.

After the actual 4. Cf3-d2 f5-f4 5. Oe3-f3, g7-g5! surprised both of me, so to speak. White must lose a piece.

 . . . . s . k .
 p l s . c . . p
 . . . . . . . l
 . . c p P . p o
 . . . C . p . .
 . . S . . O P L
 P L . C S P K P
 . . . . . . . .
The continuation was 6. g3-g4 Oh5-h4+ 7. Kg2-f1 Oh4:h3 8. Of3:g5; losing a piece is no reason to give up, since a Lobster is not worth terribly much more than a Pawn, and White has hopes of trapping the Black Oyster.
 . . . . s . k .
 p l s . c . . p
 . . . . . . . l
 . . c p P . O .
 . . . C . p P .
 . . S . . . . o
 P L . C S P . P
 . . . . . K . .
For example, after 8...Oh3:h2? 9. Og5-h5! Lh6-g7 10. Kf1-e1, White wins.

Probably the best move would have been 8...Oh3-h4, in order to prevent 9. Og5-h5 (the O at h5 dominates the O at h3, which cannot then be extricated).

You can see that 8...Lh6-g7? 9. Og5-h5! Lg7:e5 wins an important Pawn and attacks the undefended S at c3 -- but then simply 10. Sc3:c5 Sc7:c5 11. Kf1-g2 is a very clear win for White.

Instead, there followed 8...Lh6-g7? 9. Cd4-f3? (defends e5 and h2, but loses an important opportunity) Oh3-h4!

 . . . . s . k .
 p l s . c . l p
 . . . . . . . .
 . . c p P . O .
 . . . . . p P o
 . . S . . C . .
 P L . C S P . P
 . . . . . K . .
Now White's Oyster is trapped. Although it cannot be won, its exchange can be forced.

After 10. Lb2-d4 h7-h6!, White would lose the Oyster after 11. Og5-g6+ Kg8-f7, and therefore White is forced to play 11. Og5-h5.

Now Black has another chance to lose by 11...Oh4:g4?? 12. h2-h3!

Instead, the correct move, 11....Oh4:h5, was played, and the game was soon abandoned as a Black victory (...Ce7-c8-b6-c4 will be very strong).

I thought it was very interesting how both sides had their "Queen" trapped, move after move.

Ruminations about Pure DemiChess

I like the game, although I realize it may not be to everybody's taste. You must realize that a complex tactical passage such as shown in the example above is a rare event. Most of the moves are "boring". (Boring to read, but not to play: while you're making all the boring moves, you're struggling to think of a way to unbalance the position and get an advantage, and this struggle is interesting.)

The "bare King" rule is necessary, and makes most endgames easier. However, you can have really good endgames:

 . . . . . . . .
 . . . k . . . .
 . . . . . . . p
 K . P . . . p .
 . . . C l . P .
 P . . . . . . P
 . . . . . . . .
 . . . . . . . .
In this critical position, the Lobster threatens to equalize the material by eating h3 and g4. If White now plays Cd4-f3 Le4-g3 Cf3-h2 Lg2:h3, the Crab is trapped. Therefore White should probably try Ka5-b6, and if Kc7-c8, now it is good to play Cd4-f3 (Le4-g2 Cf3-e5! with the idea of stalemating Black's King, waiting for the Lobster to move away, and playing Ce5-f7 and grabbing h6).

A position like the one above led me to something like this:

 . . . . . . . .
 . . . . . . . .
 . K . . . . O .
 . . . . . . . .
 P . . . . . . .
 . . . . . . . .
 . . . . . k . .
 . . . . . . o .
and if you think about this position for a while, you'll realize that it isn't easy.

Meanwhile, in Pure DemiChess, it would seem logical to adopt the ancient Shatranj rule that winning by checkmate is twice as good as winning by bare King. In modern terms, this would mean that winning by stalemate or by bare King scores 0.75 (and the loser gets 0.25), whereas checkmate scores 1.0 (and the loser gets a zero).

In addition, you'd have to say that winning by bare King must be claimed at the first opportunity, and if you pass it up, you can no longer win by bare King nor by stalemate (if you're winning by bare King, you can often force a stalemate).

 . . . . . . k .       . . . . . . . .
 . . . . . . . .       . . . . . K . k
 . . . . O . K .       . . . . . . . .
 . . . . . . . .       . . . . . O . .
 . . . . . . . .       . . . . . . . .
 . . . . . . . .       . . . . . . . .
 . . . . . . . .       . . . . . . . .
 . . . . . . . .       . . . . . . . .
In the above positions, should White claim a win by bare King, or try for checkmate? Since the Oyster moves differently forwards and backwards, what if we rotate the positions by 90 degrees clockwise, so that the Black King is in the lower right corner of the board?

In the opening, it seems to be a problem that Black can simply play symmetrically. Theoretically, symmetrical defenses are bad; but in this game, the reduced tactical resources make it a bit more difficult for the first player to profit. However, the second player can never do better than a draw by symmetrical play, and the first player will win at least sometimes. Therefore, the symmetrical defense is not much of a problem.

It seems to me that the tactical complexity of this game arises from the fact that the pieces move differently in advance or retreat. Every time you advance to the attack, you have to wonder whether you can ever get back!


Other Links In these Pages

This is a Mailme.