In creating these rules, I started with the very general and vague rules outlined in the Star Trek Technical Manual published in 1977. Here I give you the reasoning for why the moves probably need to be the way I have defined them. This section contains my reasoning why I arrived at the rules above. There is very little “wiggle room” in the rules. As far as I can tell, the rules have to be the way I laid them out if you want the game to play similar to chess.
Let us consider each square on both the classic chess board and a tri-d chess board and determine for each piece how many squares it can move to.
The challenge is to figure out how pieces should be able to move such that their power and move characteristics are similar to classic chess. For example, a bishop can be in a position to take a rook but that rook can not take the bishop.
Tri-D Chess rules above
I first thought that the knight should be able to move vertically only in knight-like moves. For example, to move up one level, the knight would have to move forward 2 then up 1, or to move up 2 levels, forward 1 and then up 2. This does not work for several reasons. 1) moving forward 2 up 1 would let the knight and rook be able to take one another, 2) moving forward 1 up 2 puts the knight on the same color square as it started, this move would not allow it to escape a rook or a bishop. Lastly, if you only allow a knight to move up/down like this, the knight will not be able to move into the corner (rooks’s home) square of an attack board. We therefore abandon this line of thinking.
In order to let a knight move vertically, a knight moves in its classic L shape but allowing it to change level. But how many levels? Should a knight be able to jump from the top of the board all the way to the bottom? If we suppose a knight can only change 1 level per move, a night can access on average 5.8 spaces. If you allow a knight to jump to any level during a move, its average number of spaces it has access to goes up to 6.6. I opted to allow the knight to change only one level in a move because this allows for better 3 dimensional play because there are more reasons to move to a space if a knight cannot swoop up 2 levels and take you. The difference between 5.8 and 6.6 did not add very much relative weakness between the knight and the bishop.
Allowing the bishop to move like a ‘ray’ in 3 dimensions (i.e. if it started moving up and over, it must continue to move up and over and not continue along the same level) does not yield enough spaces for it to move to. The piece would end up being weaker than the knight. Even if you allow vertical movement, it is still vastly weaker than the knight. To fix problem, I permitted the bishop to move up or down as long as there was a clear path but never return to the same level. This allows the bishop to use the board as if it was many stacked layers (like a shopping mall) or said another way, as if there were more than one way through some path to the space it wanted to go to.
Since the board is aligned with the same color vertically, it seemed logical that a bishop should be able to move vertically and take vertically as well. This added an average of 0.3 spaces a bishop could move to. It also adds a natural extension of the game into the third dimension where the bishop can capture a piece, for example a rook, which cannot capture it back. A similar situation exists in the plane. Removing the vertical rule from the bishop (queen and king) might not adversly effect the game, but allowing these straight up and down moves seemed to me to make the game less restrictive and play better.
On a 3 dimensional board, there should be a reason that a piece would not just take a piece and move to an above or below space. However, sitting above or below a piece should not always be a safe haven.
As in classic chess, the rook should have the capability to move from one end of the board to the other. The only way to do this was to again go down the thinking that the board was like a multi layer space. This also allowed the piece to be roughly as powerful as its counterpart in the classic game.
The queen is the combination of the bishop and rook’s moves. This nicely matched the classic game of chess.
The king being like the queen but limited only to moving one space in any one direction at a time. Again this nicely matched the classic game.
This leaves only the pawn to reason out. The pawns on the edge of the attack boards seem a little stuck. However, if we look at the board, there are essentially 4 files through the board but split into 2 levels, thus a total of 8 virtual files. We have 8 pawns. It’s just a matter of making sure that the pawns on the attack boards are funneled into the 4 available files. So logically, we just add a special move for their first move to get them into their proper file. The solution works very elegantly. The reverse of this rule allows a pawn to get back up on an attack board so that the attack board can be moved around.
I considered a more complicated move for a pawn to allow it to take as if it had really started its move from the equivalent place the other pawn is on the main board. Though workable as a rule it was very complicated to explain. I went therefore with the more simplified version of my rule which allows a pawn to capture on the diagional move coming off an attack board. This also seemed more in the sense that this was an attack board and it should allow attacks.
Castling should accomplish 2 things: 1) it should move the king to a more protected spot on the board and 2) it should get the castle out of the corner. If you put the pieces in the same place as the classic game of chess, this seems to accomplish this very nicely on the king’s side but not on the queen’s side. On the queen’s side the rook would end up where the queen was (still up on the attack board and behind the king). So instead of choosing the actual piece locations from classic chess, I chose to move the rook on the other side of the king like in classic chess. The final reason for liking this method best is that it required moving the bishop and knight because the rook and king move to those locations. You can imagine needing to move the queen out of the way was to get the rook off the attack board without stepping on the queen.
Other castling moves I considered:
To castle on the king’s side (O-O), the king’s bishop and knight must have been moved and those squares open. The king moves to the rook’s square and the rook moves to the king’s kinght’s square. The rook ends up one level down and diagonal from king.
To castle on the queen’s side (O-O-O), the queen and the queen’s bishop and knight most have been moved and those squares are open. The king moves to the queen’s square and the rook moves to the queen’s bishop’s square. The rook ends up one level down and diagonal from the king.
The reason I did not go with this version was I felt it was better to get both pieces off the attack board freeing up the attack board sooner to move around. Furthermore leaving the king on the attack board felt more dangerous because it could be more easily trapped there (you will note that not even in the classic game of chess is the king shoved all the way into the corner).
Another obvious possibility is to swap the king and the king’s rook, or swap the king and the queen’s rook. However, this seems like the worst of both worlds. On a king’s side castle move, the king ends up in the corner and the rook possibly trapped behind a kinght (unless you require the knight be moved), and on a queen’s side castle move, you end up with 2 rooks on one side, one rook blocking the other.
However, I do not feel strongly about the version currently in the rules and this version, if someone a presented a better reason for one over the other or a good reasoning for something completely different I would happily accept it.
Attack boards start above white and black’s level. If there is some advantage to being above or below in the direction of which way a piece starts moving, this could give advantage to one side over another. I therefore wrote the rules so that it did not give any advantage to whether a piece started moving down off a board and could not move up again. It also made sense to me that the attack boards were attached to a specific corner, they should be somehow extensions of that corner and not somehow associated with other places on the board.
Here is what the Star Fleet Technical Manual by Franz Joseph says on the matter:
“16 of the 64 squares are movable in 4 groups of 4 squares each. They may be move [sic] to one adjacent pin position at a time provided they are either vacant or occupied by only one o the player’s pawns, and such action constitutes a move in regular turn. An occupied attack board cannot be moved to an inverted pin position”
I modified these rules which frankly disturbed me because I tried very hard to extend and not modify what Franz Joseph wrote.
I did not see the attack boards on their own as pieces, however, when they are occupied they certainly act like pieces. The problem I faced was that moving an attack board, and in particular inverting it, made no tactical change to the board, at least, not any I could see. I suppose in retrospect you can argue that if you have so few pieces at the end of the game and there was an empty attack board a king might move onto that the opponent could move that attack board in their turn thus depriving the king of being able to use it as an escape. Seems like a very bad move anyway!
In my current thinking, attack boards can only be moved when occupied by a single pawn. The pawn can drive the attack board forward, backwards, sideways, as long as it is moved to an adjacent pin. A pin on the same board is adjacent, but not diagonally. The pin at an angle up or down on the same side of the board (like moving along the sides of a triangle) is also adjacent.
You never spin an attack board. They stay oriented the same way. This does pose a problem getting the pawn off the attack board when it is on one of the forward facing pins because it has no board to move forward to.
Attack boards get inverted only when putting a piece on it. I had originally said also taking a piece off but realized this made no difference. The reason for inverting a board only when putting a piece on it was that by demonstrating ownership of the board (by about to be putting a piece on it), the owner could orient it how they desired. I felt that only owners of their own attack boards should be able to move and invert them.
I did not see any value consuming an entire move or invert an attack board on it’s own. Remember, in Chess, every move is important. There has to be some definite value in moving/inverting an attack board for it to consume an entire move which I could not find. If you can demonstrate or describe more than a few situations where moving/inverting an empty attack board as a move is on par with the value of moving a piece, I will gladly reconsider!
Furthermore in my rules there is not a large tactical advantage to a board being inverted on any given pin, I saw no reason for that to consume a move. There is an advantage of being able to move it then on a subsequent move to a pin that has an upright board on it, but the number of moves it would take to do this far out weighed the value.
If you allow empty attack boards to be moved around, you need to be careful not to cause a situation where you are in a stalemate and you are forced to move so you move an empty attack board instead or where you would otherwise be in checkmate so you move an empty attack board so not to have to move your king and end the game. This would remove a way of causing mate.
If you allow empty attack boards to be moved by any player (hence an empty attack board is unowned), then one player could move the board, the next player could move it back or out of the way and you get into a fight over moving the board around! This seemed useless and saying a board can only be moved when it has a single piece on it to drive it around really started to make sense. The same logic applied to inverting it.
In trying to think this through, if you had a situation where your king was pinned (not in check but couldn’t move to any surrounding square). You could imagine being able to move an empty attack board so your king might escape on it. Unfortunately, if it’s empty, nobody owns it, you could move it, the opponent could move it back or away. So this sort of “rescue the king” type of move in my view only works when you have a pawn on the attack board.
Summary of the wiggle room
In the above, I took some decision on things which I could have argued one way or the other, this is the short list that remains:
- Whether the knight should be able to move more than one vertical level in a move.
- Whether all (or some) pieces should be allowed to move and capture straight vertically or not (i.e. up or down without moving to a different square in the horizontal plane).
- Whether pieces should be able to return to the same level and use the upper level to jump over a piece and capture a piece behind it.
- Whether to consider an attack board an extension of the corner or some intermediate level above, below, or sandwiched between levels
- The way to castle
I am not the only person to have thought up rules for this board. Andrew Bartmess also has a different set of rules. Please see his website at yestercade.net. The rules I have detailed above are different from Andrew’s. He took a different set of assumptions than I did when he designed his rules. I encourage you to go get a copy of his rules and let me know which ones you like best!
MGrant’s Star Trek Tri-D Chess Rules by Michael Grant is licensed under a Creative Commons Attribution 3.0 Unported License.