The Chess Variant Pages

[ Help | Earliest Comments | Latest Comments ]
[ List All Subjects of Discussion | Create New Subject of Discussion ]
[ List Latest Comments Only For Pages | Games | Rated Pages | Rated Games | Subjects of Discussion ]

Comments/Ratings for a Single Item

Later Reverse Order EarlierEarliest
Raumschach. The classical variant of three-dimensional chess: 5 by 5 by 5. (5x5x5, Cells: 125) (Recognized!)[All Comments] [Add Comment or Rating]
Jean-Louis Cazaux wrote on 2021-12-12 UTC

Excellent! Thank you very much

octarinebean wrote on 2021-12-12 UTC

I found the book from the inventor describing the game in 1919:

He gives multiple versions of the starting position and rules for pawn movement, as well as also describing some 4x4x4 and 7x7x7 variants. The version this article covers is the 10 pawn version (C3) of type C "Neue", which Maack calls the normal way to play. There is also the type A "alte" with pawns only on the γ level, and the type B "vier Einhörnern" where each side has four unicorns, but only one rook and knight, in order to span the whole board with each piece. All of the diagrams in the book show point reflection symmetry, not rotation symmetry, between the two sides, contrary to this article. The type B "reduzierte" pawn movement which the author prefers is disliked by Maack, in favor of the type C "neue" pawn which includes the forward-and-vertical capture.

Please make good use of this primary source, which, it seems, no one until now has found.

Ben Reiniger wrote on 2019-07-15 UTC

@Kelvin, I think these are equivalent, because of the current text's requirement "always in the same coordinate plane," or from the later "never step through the corner of a cell."

Also equivalent, I think, is the one-orthogonal one-diagonal-outward if this page treated the unicorn's "triagonal" as non-diagonal.  I think maybe the easiest description would be "like a knight in any of the coordinate planes," but that's probably pretty subjective.

If you disagree with any of these equivalences, could you point out an example move that one has that the other doesn't?

KelvinFox wrote on 2019-07-15 UTC

The description for the knight is incorrect. It should read that the knight moves tw squares in one rook wise direction and 1 in another. The piece's current description, while making the same piece as the first one on 2d boards, gives a different piece on 3d boards

Jim a wrote on 2018-03-09 UTC

What if you changed the unicorn so that in addition to its normal move/capture movement along the corners of cubes, it could move (but not capture) only one square along a diagonal edge of a cube?  That would allow each unicorn to cover all the squares of it's color, without making it too overpowered. 

Kevin Pacey wrote on 2017-01-05 UTC

I'd tentatively estimate the relative piece values in Raumschach as follows:

P=1; U=1.75; R=3; B=3.25; N=5; Q=10, and a K has a fighting value =6.66 (noting it can't be traded).

Kevin Pacey wrote on 2016-08-27 UTC

Two possibly interesting variations on the rules of Raumschach could be based on altering its setup postion, replacing each side's rather weak Unicorns with either Manns (non-royal pieces that move like kings), or alternatively compound pieces, all of Mann & Unicorn movement capabilities combined. I tend to fancy the latter alternative at the moment (keeping the setup position the same in other respects). In either variation of the rules, the Unicorns are replaced with in effect major pieces (joining the queen as one), and perhaps significantly increasing the possibility of delivering an early mate (moreso if the suggested compound pieces are used). Meanwhile, rooks, bishops and knights are still preserved as relative minor pieces, and might be swapped without great loss for a few pawns if required during a game.

I'm not sure either idea for a variation on the rules of Raumschach would be significant enough to warrant 1 or 2 submissions to CVP, but I thought I'd put the variations out there for possible discussion, at least. As an aside, I think Alice Chess is, to date, the ultimate 3D variant as far as being chess-like, but if Raumschach can be improved on somehow that still may be worthwhile as a goal.

H. G. Muller wrote on 2016-02-19 UTC
OK, I see. You cover the two squares with the Bishop from another plane. If you do that from the cell a Unicorn step away from it, it is even forcible to some extent. If the posting system had not deleted my first very elaborate attempt to post this, I would even have given a mate in 4.

Kevin Pacey wrote on 2016-02-18 UTC
H.G. wrote:

K.Pacey> in Raumschach there are possible mating positions with a K & B plus R

"How do you envisage that? Even in a corner the bare King can only be limited to a plane, and the other pieces would have to cover 4 squares in that plane (including the one the bare King is on). A Bishop could check and cover one of those at the same time, but the only way for a Rook to cover the other two would put it in a place that blocks the Bishop from checking.

I think you need at least two Rooks to have a mate position. And I am not sure that is forcible."

Hi H.G.

I imagined the lone K in a corner on the upper level. The superior side's K would be in opposition to it on the same file (say), same level. The superior side's B would be on the square between them on the same file, same level. The superior side's R (the mating piece) would be on the same corner square as the lone king, except one level below it. The R would be protected by the B. Thus, aside from the superior side's K guarding 4 critical squares, the B guards 2 critical squares (including one on the upper level), and the R guards 2 critical squares (including one on the level below the Ks). That's if I've visualized this all correctly.

H. G. Muller wrote on 2016-02-18 UTC
> <i>in Raumschach there are possible mating positions with a K & B plus R</i> <p> How do you envisage that? Even in a corner the bare King can only be limited to a plane, and the other pieces would have to cover 4 squares in that plane (including the one the bare King is on). A Bishop could check and cover one of those at the same time, but the only way for a Rook to cover the other two would put it in a place that blocks the Bishop from checking. <p> I think you need at least two Rooks to have a mate position. And I am not sure that is forcible.

Kevin Pacey wrote on 2016-02-18 UTCGood ★★★★
I may be wrong, but so far I've concluded that in Raumschach there are possible mating positions with a K & B plus R (or an opposite coloured B) vs. lone K, but it seems that they are not forcible, or 'basic' mates (unlike K & Q vs. lone K). I'm wondering if a K plus some combination(s) of 3 pieces (aside from including any Q or P) can force mate against a lone K. Perhaps more than 3 such pieces are required?

Malcolm Webb wrote on 2013-01-30 UTCBelowAverage ★★
I noted an acknowledgement of Mr Pfieffer in pointing out the error. However there is no acknowledgement of David Paulowich for pointing out another error. According to Dickins' book Black pawns promote on A first rank, White pawns promote on E fifth rank.

Malcolm Webb wrote on 2013-01-30 UTCBelowAverage ★★
This page (& the associated Zillions rules file) would be good if the pieces were set up correctly. Mr Pfieffer is right: according to Anthony Dickins "A Guide to Fairy Chess" the positions of the Black pieces on the D-level should be:

Da5-Unicorn, Db5-Bishop, Dc5-Queen, Dd5-Unicorn, De5-Bishop.

That is, each player should see each of their Bishops to the left of each of their Unicorns. Unfortunately the author has not yet corrected this error.

Dickins' book is a secondary source, based on articles by T.R.Dawson in Chess Amateur 1926. The most authoritative source would be Ferdinand Maack's original three books in German; if anyone has access to these books and can show that Dickins was wrong, then I stand corrected. However this page references Dickins' book without correctly implementing it.

There are problems with Raumschach as a game, one being the incomplete coverage of the 3D-board by the Unicorns. However Raumschach has a place in Chess history, and should be correctly presented with all its faults.

Ben Reiniger wrote on 2011-02-28 UTC
In response to the anonymous comment, if the bishops are placed adjacent to the queen, then they are both bound to the same color, which is often considered unfavorable.

Anonymous wrote on 2011-02-28 UTCPoor ★
both bishops should be beside the queen.

Eduardo wrote on 2011-01-04 UTC
Hi. The old ASCII version of this site has a different initial position for Black Unicorn. Which is correct, Db5 or Da5?

Alfred Pfeiffer wrote on 2010-08-20 UTC
The initial setup of the black pieces is wrong. 
For the correct starting array swap the black Bishops with the Unicorns. Then you get the following arrangement at the Level D:

| u | b | q | u | b | 5  Queen Dc5; Bishop Db5, De5; Unicorn Da5, Dd5; 
| p | p | p | p | p | 4  Pawn Da4, Db4, Dc4; Dd4; De4.
|   |   |   |   |   | 3
|   |   |   |   |   | 2
|   |   |   |   |   | 1
  a   b   c   d   e

Unfortunately also the Zillions file in '' copied this error.

The error does not occur at the old page for this variant: ''.

See also the books:
 Anthony Dickins: A Guide to Fairy Chess, and
 D. B. Pritchard: The Encyclopedia of Chess Variants.

David Paulowich wrote on 2010-07-18 UTC

ATTENTION: the following diagram and rules are (c) 2004 Jim Aikin and are taken from his webpage Five Up. Note: Jim's 5x5x5 variant sets up the White pieces on the top two levels, thus requiring the White Pawns to move downwards. Apart from this change, his pawn movement rules are the same as those stated in my comment ten days earlier.

Figure 6. The white pawn shown here, which is advancing toward the A5 row, 
can make capturing moves to the cells marked 'x' and non-capturing moves to the cells marked '+'.

   _ _ _ _ _     _ _ _ _ _     _ _ _ _ _     _ _ _ _ _     _ _ _ _ _  
5 |_|_|_|_|_| 5 |_|_|_|_|_| 5 |_|_|_|_|_| 5 |_|_|_|_|_| 5 |_|_|_|_|_| 
4 |_|_|_|_|_| 4 |_|x|_|_|_| 4 |x|+|x|_|_| 4 |_|_|_|_|_| 4 |_|_|_|_|_| 
3 |_|_|_|_|_| 3 |x|+|x|_|_| 3 |_|P|_|_|_| 3 |_|_|_|_|_| 3 |_|_|_|_|_| 
2 |_|_|_|_|_| 2 |_|_|_|_|_| 2 |_|_|_|_|_| 2 |_|_|_|_|_| 2 |_|_|_|_|_| 
1 |_|_|_|_|_| 1 |_|_|_|_|_| 1 |_|_|_|_|_| 1 |_|_|_|_|_| 1 |_|_|_|_|_| 
A  a b c d e  B  a b c d e  C  a b c d e  D  a b c d e  E  a b c d e 

Joe Joyce wrote on 2010-07-09 UTC
Would setting the preset up sideways help? 
I think it would be pretty obvious where pawns could promote with this setup.

David Paulowich wrote on 2010-07-08 UTC

ATTENTION: the 3-D CHESS FAQ FILE contradicts this Raumschach page with the following information: 'Pawns promote on the back rank of the opponent's end-level. White Pawns promote on the fifth rank of Level E; Black Pawns promote on the first rank of Level A. Dickins notes that in capturing Pawns must move toward their promotion rank. The '3-D' application of this rule means that a White Pawn on Cc3 can capture on Cb4 or Cd4 (like regular chess) or on Db3, Dd3, or Dc4, but not Db2. (Db2 is upward but backwards -- toward White's back rank instead of toward Black's back rank.)'

NOTE: I have a text-file copy of Bruce Balden's two newsgroup posts on October 8, 1990. I believe that he has accidentally reversed the pawn promotion zones and that the rules originally given on this page are also mistaken. Here is one reason for believing so. Everyone agrees that a pawn on level 'C' must either remain on that level or move to an adjacent level: 'D' for a White Pawn or 'B' for a Black Pawn. But this means that a pawn on level 'C' will never promote (according to Bruce Balden). So I accept David Moeser's pawn promotion zones (the starting squares of your opponent's King, Knights, Rooks) as stated in the 3-D CHESS FAQ FILE.

ALSO: my Raumschach Revised preset has the correct alternation of colors on a (slightly wider) Raumschach board. The initial setup and co-ordinate system match the game information page.

David Howe wrote on 2009-04-06 UTC
Hi Alan, why don't you post your designs here, and we'll see if we can integrate them into the Raumschach page? Or you can email them to us.

Alan Kopta wrote on 2009-04-05 UTC
I believe I copied a set of Raumschach design issues from this site, some time back. The issue of cell/cube/square coloring to help indicate triagonal line movers (unicorn/mage) reachable positions like the use of the white and black squares was mentioned but never developed. I just wanted to submit that I believe I have developed a board(s) design which permits indication of the diagonal line mover (bishop) and triagonal line movers. I currently expect to use my design to construct 5 boards using two patterns since the board patterns for the even boards/levels are the same as the boards for the odd boards /levels are the same. So for the game's five levels Levels 1, 3, and five have the same pattern while Levels 2, and 4 have the a pattern different from the pattern of levels 1, 3, and 5. But the same with respect to each other I have bitmap images representing the pattern of the different levels but was not certain if I could or should attempt to present them here.

David Paulowich wrote on 2008-04-14 UTC

Using the Customize button, before sending an invitation, allows you to choose between Colors: 88AA88 FFFFDD (my colors for Raumschach Revised) and Colors: FFFFDD FF8000 (Matthew's orange). You can even bump the scale up from 65 percent to 75 percent, but that makes the board slightly wider than my 1280 x 1024 display. EDIT: Light/Dark squares also depend on your choice of Board: 10.01. or 01.10. My color scheme was originally chosen to go with Roberto Lavieri's elegant 'Galactic Graphics' pieces, in Midgard Chess. I may be using them in a 3D Chess variant someday.

I must confess that I have not paid much attention to PBM optional features, but I believe that the player receiving a move also has full 'Customize' options. Looks like the onlookers are stuck with the options chosen by the two players, though.

Gary Gifford wrote on 2008-04-14 UTC
They are too close to say which is better. In fact, depending on the monitor used, one may look better on Monitor A, but look slightly worse on Monitor B.

Matthew La Vallee wrote on 2008-04-14 UTC
What do you think: better colors? in this one Clearly, this is not an important issue.

25 comments displayed

Later Reverse Order EarlierEarliest

Permalink to the exact comments currently displayed.