Forums

3 Dimensional Cubic Chess 8x8x8 Board

Sort:
EndgameEnthusiast2357

My previous thread on this is long dead and I have changed my mind about some of the ideas, so would like to share this again:

Real 3d chess is not what you see on star trek, it is a cubic 8x8x8 board with 512 cells, and potentially 256 pieces in total. The initial set up would be all of each sides pieces on their 1st "plane", such as the following:

Something to this effect, and a mirror image for black (but unsure exactly where the kings would be). Both sides would need all the extra pieces to account for all the extra cubes pieces have available. The 2nd plane would be 64 pawns. Now for piece movements:

King: This would be the simplest, the king could move to any adjacent cell horizontally, vertically, up or down, and along 3 dimensional diagonals 1 cube away as well. This yields 26 different cubes a king could move to from the center. Think of a rubiks cube, the king is in the core of it, it can move to any of the cubes on the surface from the center.

Queen: Can move in the same 26 directions as the king, but any number of cubes away obviously. This totals 86, which will become clear once you calculate the rook and bishop type moves..

Rook: Very easy, Just like normal chess, except it could also move up and down, so 21 total moves. There are no 3 dimensional moves when it comes to rook moves, they all can be represented in a single plane, so it only has those 21 (7 in each of the 3 perpendicular straight lines).

Bishop: The bishop could move along any of the 6 2d diagonals. There are 2 perpendicular diagonals in each of the 3 two-dimensional planes it is in, and since there are 13 squares in a normal 2d plane it could move to, x3 = 39. Now here's where it gets a little tricky..there are 3 dimensional diagonals the bishop could move within as well. Think of the cube in the lower-front-left, and the cube in the upper-back-right, a line connecting those two through the cubes constitutes a 3 dimensional diagonal. From the center, these can be summed up by 2 slanted planes with 13 possible moves each = 39 + 26 = 65 total moves. Point being all pieces can make both normal 2 dimensional moves, 2 dimensional moves in perpendicular planes, and 3 dimensional diagonal moves. The bishop is actually more powerful than a rook in 3d chess.

Queen = Rook + Bishop Moves Combined = 86.

Knight: Calculation of all the 2d moves is simple. 6 directions to choose from for the initial 2 square straight line movement, and 4 possible 1 square movements perpendicular to that, which is 24. But remember, 3d diagonals moves have to be taken into account as well, a knight could also make a 1 cube diagonal move that's still perpendicular to the initial 2 cube move, so that's another possible 24 moves. So the number is even higher but haven't calculated exactly what it would be. Once again, more than twice as powerful as the rook. As weird as it sounds a 3 dimension knight can fork 2 pieces that are right next to each other. In fact, a knight in say, the cube "d-5-5" could fork all of these pieces assuming this is plane 7:

Pawn: Each side would have 64 pawns covering the entire 2nd rank, or 2nd plane. Same first move 2 cube ability rule applies. However, a pawn could capture on 8 different cubes. It could capture a piece on any cube that's one cube above it, except the cube directly above it (4 2d diagonals moves upward, and 4 3d diagonals moves upward. Pawns could only make moves that advance it one rank. They couldn't move back and forth or side to side within the same rank, only up toward the top. En passant could theoretically be done 2 different ways on the same pawn, depending on how you interpret the pawn capture in 3d, but for simplicity sake, we'll say a 3d en passant capturing move would land the pawn directly above it vertically, not 1 cube away from it on any diagonals. So the pawn on a starting square, not on the edge, could potentially have 10 different moves.

What's very hard to calculate is if a king and queen could force checkmate. A checkmate position would be a basic support mate with the king in the corner, a queen 1 cube diagonally away, and a king 1 cube away from the queen, but it may not be forcible, even with 2 rooks or a queen and rook it may not be forcible. All of this would have to be calculated from scratch to determine what is a draw or not!

Ilampozhil25

interesting

comparing to other stuff

the planar piece arrangement feels different, but makes much more sense

the rook movement is (x,0,0) yes? 

the knight is (2,1,0) + (2,1,1) i think

bishop is (x,x,0) or (x,x,x) right

queen is rook + bishop

pawns go only towards the other side in movement and in capture the pawn goes (1,1,0) and (1,1,1) in the forwards direction?

if this is right, you can rephrase pawn capture as "single step bishop" which is better imo

castling!?

and en passant...

lets say a white pawn is on 5e7 (white pieces start on plane 1)

and a black pawn goes from 7d7 to 5d7, then the white pawn can capture to 6d7?

bit confused on what "no 3d diagonal" means here coz that makes me think en passant requires a pawn to go on top of another pawn...

EndgameEnthusiast2357

The king could castle with any of the 4 rooks in a straight line away on the edge. Pawns can capture in any of the 8 upward diagonal directions ahead only 1 cube. A pawn could fork as many pieces in 3d as a knight could in 2d. Yes a pawn could take en passant that way. All 3d diagonals are included.

Ilampozhil25

hmm

so a white pawn on 2d2 (by this notation i will be using) can capture on 3d1, 3c1, 3e1, 3c2, 3e2, 3c3, 3d3, and 3e3

an upshifted king, basically

the major problem is how will people play this

EndgameEnthusiast2357

Yes, as weird as it is, a pawn is basically an upshifted king, which makes sense, as that's what it is on a regular chess board. Pawn controls the 3 squares in front of it (can only move not capture on one of them). The bishop though actually becomes more powerful than a rook on a 3d board because there are more diagonal directions available than vertical and horizontal moves. The 3d knight is also similar to the king as both control 8 squares on a 2d board, but it has even more than a king in a 3d board because the L shape can also be slanted. This also makes it consistent with the pawn, as a knight and pawn on a 2d board control 2 squares upward of it, so in 3d they control the same 8 squares upward of it. Consistently involves including the 3d diagonals in all pieces. Actually FIDE rules define the knights movement as not an L shape, but rather "it can move to the closest square that's not on the same rank, file, or diagonal" which forms the 2x1 L shape. Therefore slanted L shapes are also possible in 2d chess as they are still 2x1.

Ilampozhil25

recommended reading for anyone who is interested in the idea of 3d chess

https://www.chessvariants.com/d.betza/chessvar/3d/index.html

EndgameEnthusiast2357

I've been looking for stuff online about 8x8x8 chess and haven't been able to find much, thanks for that! But his interpretations of it are illogical. Yes, eliminating all 3d diagonal moves for every piece is consistent, but does that really make sense? That would mean a king and queen couldn't actually move in every direction available to them. That's why I chose all 3d diagonal moves being included for every piece. However, he also interprets a 2d rook move like a bishop (1a1 to 8a8) which is absurd! The board being 3d doesn't all of a sudden make 2d bishops moves rook moves! 2d diagonals are still diagonals, even if they are less mobile than 3d ones. Also, the alternating pattern of colors would continue in 3d. The entire tower of a1 cubes on every level wouldn't all be black! 1a1 would be black, 2a1 would be white..etc. It would be like mirror image chess boards stacked on top of each other in an alternating pattern. Bishops would be able to access every cube on the board as a result, rendering colors meaningless, which is interesting! In fact, restricting bishops to only 3d diagonals would mean despite being able to change the color of the cube it's on, it still couldn't access half the cubes on the board. But a bishop capable of both 2d and 3d diagonal moves could. This makes a bishop more mobile than a rook, and why it makes sense to have more rooks than bishops in the initial set-up.

Ilampozhil25

he found the "bishops change color" and "rooks move in what looks diagonal to the players" unchess

(or instead of bishops change color, "g5 is a light square" but imo that cant apply here coz the pieces are on PLANES 1 and 2 not ranks)

but "to the players" actually has a different meaning here!

you have white at planes 1 and 2; he had white at ranks 1 and 2

(i prefer planes, it gives the army more of a cohesive feel and the pawns make purely more sense to me)

Ilampozhil25

is it just my clients problem or are people adding 1/4 of their comment after posting today

been noticing this a lot

EndgameEnthusiast2357

Yeah, having the same 16 pieces wouldn't make sense. Not even sure if checkmate would be even possible with the standard 16 pieces. Think about it? A king in the center can move to 26 different positions, is a queen and king enough to force the king to an edge-edge of the board? Is 2 or 3 queens even enough? It's hard to visualize because you couldn't see all the cubes from all these angles at once.

EndgameEnthusiast2357

Let's say hypothetically bishops we're restricted to only 2 dimensional diagonal moves. The bishops staying on the same color wouldn't hold in a 3 dimensional alternating cube pattern. The cube (1,1,1) is dark, and the cube (2,2,2) is light, and colors alternate along 3d diagonals. These bishops would still be restricted to cubes whose coordinates sum to an odd number ( 1+1+1 =3, 3 +1 + 3 =7)..etc, and bishops originating on cubes whose coordinates sum to an even number, will never have access to a cube whose coordinates sum to an odd number. Now some may suggest restricting bishops to 3d diagonals to limit its power this way, but then it would seem stupid that a queen could attack a bishop along a 2d diagonal, but the bishop wouldn't be attacking back?

Ilampozhil25

well, he had the 16 pieces; on every single plane for each side

Ilampozhil25

just realised that putting all pieces on planes instead of ranks changes nothing except for a 90 degree rotation

it still looks more cohesive tho, instead of just a bunch of 2d chess games stacked lol

wait that might be an actual good idea...

yours is also more elegant in not requiring non royal kings

heres the "problems about the old setup" page

https://www.chessvariants.com/d.betza/chessvar/3d/oldmap.html

https://www.chessvariants.com/d.betza/chessvar/3d/myboard.html

the bottom of the second link shows a possible gui for a 3d chess variant online

Ilampozhil25

Instead of stacking 8 normal chess sets vertically, you could put 28 Rooks on the outermost places, 20 Knights within, 12 Bishops, 3 Queens and a King. I prefer the stack of 8 standard sets.

i copied this from that directory

seems like everything that could be invented has been invented

yolosolo123
Seems complicated but fun at the same time. I hope I can try it! It also feels like a nice variant for the achievement, cubic chess(play a game of 3D chess), cubed brain(play 10 games of 3D chess) and 3D chess 100( get a rating of 100 in 3D chess)
EndgameEnthusiast2357

It would be hard to play as you couldn't reach the pieces enclosed in the middle of a 3d board. Maybe the cubic board would have to have internal rotation systems, similar to a rubiks cube, as a way to move the pieces. On this site it could be digital.

Ilampozhil25

but its 8 by 8

and the center pieces would be very hard to get to

besides youre rotating the board with the pieces on? wont they fall? (it has to be open in order for the pieces to be movable)

those betza links have a prototype board, might be interesting (both physical and digital)

ofc theres also blindfold 3d chess

EndgameEnthusiast2357

Maybe they would be holographic, or magnetic so piece wouldn't fall.

yolosolo123

Do we need a better notation system?

EndgameEnthusiast2357

Cubes would just be labeled A11 to H88, or even just 111 to 888.