Fortress Chess

This page will contain a discussion and complete rules to Fortress Chess, a superlarge variant.
Following is the current preset:

Initial discussion on developing this game:

One of the nice things about this site is that you can get so many different opinions. Sometimes I like a good design challenge, and the superlarge game poses such challenges. To make it more interesting, I want to use the FIDE unlimited sliders in the game, because they are 'too powerful', and I want to design a new composite/compound/whatever piece to be used in the game, too… oh, and the game should be reasonably easy to learn and play, and not take too long. There! Have I left anything out?

How do you get enough piece variety in a superlarge to make the game worthwhile without overloading the player with reams of rules? One way is to establish some basic piece types and modify each of them with a few different movement rules. To make this work, you must have a good, clear, simple, easily understandable symbology to go along with your good, clear, simple, easily understandable and short [for playability] rules. So we start by using David Howe's Alfaerie icons, something that is most likely very familiar to anyone who plays variants and would be reading this, and if not, the info is easily accessible. They are clear, simple, easily distinguishable, and easily modifiable, all great virtues for any game designer. Then we add a few simple symbols to the mix, that modify the piece moves.

What sorts of pieces will we have? … [W]e want a few kinds of leaders. Top dog is the king, but we will also use other leader pieces. Every leader will be allowed to move 1 piece under its command and within its [limited] command range every turn. This should take care of little problems like how we work multi-move turns and how to tame queens that can move 23 squares/turn.

[Initially] I plan[ned] to put small, powerful forces in the corner forts, weak and medium-strength pieces in the corps flanking each army, maybe add a few pieces immediately behind the army on the board, keeping them short and medium range pieces, and put the high king, his marshall, guards, and the elite troops and reserve behind the steward wall. This setup minimizes the initial effects of unlimited sliders, and will have about 80-100 pieces/side, of which about 25-30 or so pieces will be 'fortress' pieces, ie: formations of wazirs and stewards, and their leaders. I will also add an alternate frontline setup, with only one flanking corps per side, on opposite flanks. Finally, the formations of wazirs and stewards are the forerunners of a new type of 'piece', consisting of several mostly shortrange pieces and a leader unit specific to them that they must be in 'contact' with to move. These would be 'Autonomous Multiple Pieces', or AMPs. While the 2 examples I've
discussed so far are simple and slow, if these amps evolve a bit [a 3rd piece would be 6 forward-only ferzes and their leader - to make it a better attack piece, up the number of its components allowed to move each turn], their natural habitat would likely be on boards of side 30-50. I see them evolving specific organs [pieces] for attack, defense, and movement.

One area that I think needs some exploration is the multi-move turn. In the
games I'm aware of, the extra moves are just sort of tacked onto the game
with no real attempt at rationale. Some games you move 2 pieces/turn, some
more, some depend on what your opponent did; sometimes the same piece can
move more than once, and capture, in other games if a capture is made, no
other move can be… Anyhow, no real attempt has been made to explain why
one rule or rule set was chosen over another.

The superlarge I'm contemplating will fall in between … 2 extremes.
The 'high king' will be, like the FIDE king, checkmateable for victory
purposes and have unlimited command control range for any 1 piece per
turn. There will also be 2 more lower levels of leader, generals and
captains. These will command different numbers and strengths/types of
pieces, with command control ranges that would be roughly 5 and 10. I'd
also throw in a marshall, with the same command powers as the king. These
powers would include the ability to activate at least 1 local piece, as
well as the 1 unlimited-range activation.

We've discussed the king, and then lower-level leaders, represented by guard icons. What do the armies they lead look like?

How do we get a good mix of pieces in a superlarge game without giving the player too much to remember? This is a key make-or-break question. If we want an interesting and playable game, we must do this part very well indeed.
I've proposed a 2-part system. It combines a few basic piece types with a few movement patterns to give a range of easily identifiable and usable pieces to complement what we already have. …I'll steal the basics from The ShortRange Project piece builder.
Our first 4 piece types are the Wazir [1 square orthogonal step], the Ferz [1 square diagonal step], the Dabbabah [2 square orthogonal leap], and the alfil [2 square diagonal leap]. Their piece icons are simple, obvious, easy to combine with each other, and it's very easy to understand the resulting pieces. Now, let's strip the knight from the longrange Fides, and put it in with the 4 basic Shorties, where it really belongs. Yes, it's really a combo of wazir and ferz, but the knight icon is all but universally recognized for standing for that 'wazir then outward ferz' move knights make. And it looks so much prettier on a combined icon. (See the High priestess and [Minister] pieces in the Grand Shatranj Alfaerie set and see what you think.)

See: Joe's Strange Notation or Appendix 1 for an explanation of piece types.

Piece types:
a king and leaders [guards];

These are our 11 basic piece types. That's 11 basic icons, but all or most should be familiar to players. There are also 5 Movement Patterns, 4 of which have specific identifying mini-icons to be used on the piece icons.

Medium-range pieces: Use FIDE icons with dots. Let each dot represent [up to] 4 squares of movement.

The superlarge test game design outline is just about finished. After clearing up a few loose ends, it can be taken to the board. One thing not discussed yet is castling. Because we have those big,
beautiful 'forts' in the corners, there is a good place to castle into. So we'll work out some castling options in the rules.

The leader rules are quite nebulous. There are a number of ways we can handle leaders; the simplest way is probably the best for now. 'Leaders have specified 'activation ranges'. Leaders can activate any one friendly piece within range.' More realistic/advanced/difficult rules can be looked at later for play balance if necessary. But the simple requiring of activation for a piece to move gives us a lot in controlling superlarge games' tendencies to get out of control. Any leader can move a queen across the board, but how does the queen get back?

Leaders are a new class, 'semi-royal' pieces. Loss of one does not end the game, but it does penalize the losing player more than just losing a non-royal piece, because the player also loses movement opportunities. I fully expect using several leaders will prevent the superlarge game from
becoming either tedious and boring or going chaotically out of control. [How's that for putting yourself out on a limb?] Now it's time to demonstrate just how well all these ideas will really work.
Jeremy Good has generously given me enough rope to han… um, has created a number of new Alfaerie piece icons that will allow testing of all the ideas presented here. Jeremy, thank you. Without your help, I couldn't have done this. [Maybe not everyone will thank you for this.] I'd also like to thank those who poked, prodded, and contributed to this so far, including, but not limited to, Mats, David and Greg.

Pieces and Moves

The piece set is Alfaerie: Many.
All the standard FIDE pieces are used: King, Queen, Rook, Bishop, kNight,
Pawn; as are the 4 basic ancient pieces: Ferz, Wazir, Alfil, and Dabbabah,
although these 4 are mostly combined with each other. All these pieces have
their standard moves.

Combo pieces: The modern Elephant combines the piece icons for Ferz and
Alfil, and may move like either one. The Warmachine combines the icons for
Wazir and Dabbabah, and may move like either. In general, any combo piece
that contains only basic piece icons moves as any one of the icons.

To generate intermediate-range pieces, 5 easy movement rules [patterns]
are defined. They are represented by simple symbols. These symbols are
combined with the basic piece icons to generate families of pieces. The
symbols are:

*Diamonds: 1, 2, or 3 small black diamonds on FIDE Qs, Rs, and Bs mean
those pieces may move only 4, 8, or 12 squares maximum in a turn.
*Squares: An elephant or warmachine with a square around the central icon
mave move as either or each of its components [in either order] in a
straight line. [Currently found in Chieftain Chess II]
*Circles: An elephant or warmachine with a circle around the central icon
may move as either or both of its compnents, and may change directions
between the steps of its move. [Found in Lemurian Shatranj]
*2 Parallel 'Speed' lines: All these pieces are 2-step linear riders. A
modern elephant with what looks like an equals sign on its right side may
move as an alfil or ferz, then as either of the 2 [not necessarily the
same as the first time] again, in a straight [diagonal] line. [Grand
*Zigzag Speed line: very similar to the 2 parallel speed lines, this symbol
looks like a 'Z' on the right side of the piece. A modern warmachine with
this symbol may move as a wazir or jump as a dabbabah, then do either
again, and may change directions between steps. [Atlantean Barroom
The leader units are Guards, plain and fancy. A leader moves 1 square for
each 8-pointed star on its icon. Leaders with a grey tint may change
directions during their move. If they don't have a grey tint, they are
linear movers. The Marshall, 3 stars with grey tint in the center, moves
up to 3 squares, changing direction as desired, and may leap any adjacent
square to land in the square directly across from the original square.
This counts as moving 2 squares, so the piece may only slide 1 more square
during its turn.
No piece may make a null move.
Before any piece can move, it must be activated by a leader. Each leader,
including the king, may activate 1 piece per turn. Activation ranges: L1 =
2 squares; L2 = 4 squares; L3 = 6 squares; marshall = 12; king = 99.
That's all of it. If anything at all is unclear, let me know.


This is an attempt to push chess right to the edge of wargaming, but still keeping it chess and not a combat simulation wargame. The specific chess features that I think are key here are: perfect information; symmetrical armies; no random events [eg: combat results table]; checkmatable [high] king; directional pawns; and the essential 'chessness' of the pieces [in that they are in theory an 'army', but in practice, each piece has moves that are very non-real-world]. I hope you find it worth the anticipation. It still has to 'play well' to be any good. A game that size that plays
poorly, or merely 'okay', is a catastrophe. I think fort is a bit of a kludge. I think it has too many pieces [100/side] and an awkward starting setup [too deep]. I'm also trying to test several things at once, which is rarely a good idea. But I think it's got a real shot at playing well in its simplest, easiest form, and is also very tweakable, if necessary. I'm ready to find out now, the invite is up.}}

Appendix 1
The Warmachine; Piece Movement and Symbology

The first part of our 'shortrange pieces for longrange boards' discussion
has given us 5 basic piece types with 5 simple symbols for easy
combination. Combine the wazir and dabbabah into the warmachine. The
dabbabah icon is a wheeled tower and the wazir icon is a plus sign, so the
warmachine is a wheeled tower with a plus sign on its side. The generic
piece can be identified like this: 'DW'. This identifies the components
without specifying any particular movement pattern. Now let's define some
movement patterns with the help of this example piece. Then we associate
symbols with these patterns. Again, to keep things simple, we'll use the
basic movement patterns discussed in TSRP.

1] Or. The warmachine may move as either one of its components, that is,
like a wazir or a dabbabah. It steps 1 orthogonally or leaps 2
orthogonally. As this is the simplest movement pattern, it doesn't need
anything extra on the piece icon. So a combo icon with no info other than
the various piece symbols may move as any one of the pictured pieces. This
can be distinguished in writing by the slash '/' symbol. Our piece
appears on the board as a wheeled tower with a plus sign on its side, and
in writing, it looks like this: 'D/W'.

2] And [linear]. The warmachine may move as either or both of its
components, in either order. It may not change direction during this move.
To the basic 'DW' icon, we will add a square around the central symbol.
In writing, we will indicate this by D+W.

3] And [nonlinear]. The warmachine may move as either or both of its
components, in either order. It *may* change direction during this move.
To the basic 'DW' icon, we will add a circle around the central symbol.
In writing, we will indicate this by a 'plus/minus' symbol: D +/- W.

4] And-Or [two-step linear rider]. Our basic DW piece may move as either
of its components, then it may [or may not] move as either of its
components again. It may not change direction during this move. The basic
icon gets 2 'speed lines' on its side. Written, it uses the plus sign
between 2 of the [written] piece symbols: D/W + D/W.

5] And-Or [two-step nonlinear rider]. Our basic DW piece may move as
either of its components, then it may [or may not] move as either of its
components again. It *may* change direction during this move. The basic
icon gets 2 speed lines connected by a crossbar, making a 'Z' on the
piece side. Written, it uses the plus/minus sign between 2 of the
[written] piece symbols: D/W +/- D/W.

Appendix 2
The Elephant series and Promotion

Alright, let's take a look at another piece series that will be on the board, and also promotions. The movement rules we're using allow us to easily create promotion ladders, which are an arranged piece series. So we will lay out a piece series to be used on the board in ascending order of piece power, and that order will be our promotion ladder.

Let's consider the modern elephant, generic 'written' description of AF, or a combination of alfil and ferz. It's lowest 'rank' is modern elephant, written here as A/F, to indicate it moves once only, as either an alfil or a ferz. It's board icon is an elephant with an 'X' on its side.

When it achieves promotion by crossing a line, it becomes a 2-step piece, the [linear] shaman, which moves as either A or F or moves twice, in a straight line, once each as A and F, in either order the player chooses. The basic elephant with X icon gets a square box drawn around the X, and it's written A+F.

The next promo is to the bent shaman, which moves like the shaman, except that it can change direction between the first and second step of its move. The icon is the elephant plus X with a circle around the X, and it's written A +/- F.

The next is to oliphant, a linear, 2-step, modern elephant-rider, symbolized as the elephant plus X with 2 parallel speed lines on its side, written as A/F + A/F.

The final one is to twisted knight, the bent 2-step modern elephant rider. The board icon is the elephant and X, with a 'Z' [2 parallel speed lines connected by an angled crossbar] on its side, written as A/F +/- A/F.

Discussion - 7/6/08

Battlefield promotions do not work properly, they break the game. The person who initiates combat loses, because when that person's capturing piece is taken in turn, the opponent now has a promoted piece out of the exchange. This effectively prevents either player from initiating exchanges. Thus, references to promotions after capture have been removed.

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-Share Alike 2.5 License.