Comments/Ratings for a Single Item

'Why not make different armies for Xiang-qi?'
I think, it's interesting idea. I think, general, pawns, advisors and
elephants should be common for all armies, as they plays special roles in
game, but horses, rooks and cannons must be different.
Shogi with different armies is also interesting idea, but it needs some
headache with promotion... I can suggest this: not only kings and pawns are
common, but gold generals also common, and first rank pieces promotes to
them. Pieces, wich replaces rook and bishop, gets 4 additional moves after
promotion. Another idea - gold generals also different, and they promotes
to normal gold generals (in standart shogi they don't have to promote, as
they are already gold generals).

Invented in 1977, Chess Different Armies initiated salvation for little customary 64 squares. Betza left us thinking it is to be Next Chess, period, 
http://www.chessvariants.org/index/displaycomment.php?commentid=614. After all, every cycle and every life ends, or re-continues, and sixty-four -- vraiment, compare its diminutiveness to big 81 Shogi and big 90 Xiangqi -- were computer-busted by the nineties if not the eighties. Busted to the point ex-player Fischer (1943-2008) proposed junking the set array RNBQKBNR in 1996, ignorant that was done by Frenchman Alexandre as early as the 1820s post-Napoleon. Fischer Random is Alexandre redux.  Now by Fischer we can have BNQRRKNB, so long as Bishops do not compete. Beautiful. Betza saw irony/inferiority there and redoubled towards perfection of CDA, obviating FRC for Next Chess, in armies: Nutty Knights, Remarkable Rookies, Colourbound Clobberers, Forward F.I.D.E.s, Amazon Army, Meticulous Mashers, pitted against each other or against the F.i.d.e. one.  The pointage of that latter, 39, is the benchmark for equality of usually different forces. There have to be subtle differences inexactly achieved, equal-valued forces being impossible when so much as a single piece-type differs side to side. There are contributors' Armies of point value also approximately 39 not created by Betza: Lawson's Pizza Kings, Aronson's Fighting Fizzies, Streetman's Spartan. The latter in current discussion has not a single new piece-type and may be adequate new combination as useful as Lawson's Army or Aronson's Army. The best Armies appear to be ones described in follow-up comment soon.
Does it matter that one Army is 38 points or 40 points? A little. When there is a disparity humans can detect, the challenge then becomes to win with White 38.5-pointed, or win with Black 39.5-pointed.  How many near-39-point Armies can be created? Hundreds. Thousands. Probably thousands of good ones, not beyond that, if computer-generated and having time allotted to devote.  How many active or finished games of Chess Different Armies are indicated at Game Courier? 105. Also for follow-up will be to find the leading player(s) Chess Different Armies over the past ten years.

I have play-tested some combinations using Fairy-Max, and it seems that both the Nutty Knights and the Colorbound Clobberers have a sizable advantage over the FIDE army.

For the Nutty Knights this advantage seems to be slightly over a full Pawn. It seems fully due to the Charging Knights. In a direct comparison all other Nutters perform slightly worse or equal to their FIDE counterparts. But replacing a pair of Bishops by a pair of Charging Knights provides a spectacular advantage. With the standard values 325 for a lone Bishop and 375 for a paired one, the Charging Knight might be 400 or even 425.

This seems unreasonably strong for a piece with only 9 moves, the extra move compared to Knight even being backwards. I guess this is one of the rare move combinations that noys a large bonus over the additive value of the individual moves, like the Archbishop (BN) or the divergent piece that moves as Knight but captures as King. In fact the Charging Knight is also a combination of Knight and King moves. Such pieces combine the speed of the Knight with the manoeuvrability and concentrated attack power of the King/Commoner. The latter endows them with mating potential, and makes them very effective supporters or attackers of FIDE Pawns, as they can protect/attack a Pawn, and at the same time the square in front of it.

The Clobberers are also significantly stronger than FIDE (advantage slightly under 1 Pawn),althogh not as much as you would expect from their individual piece values. A pair of Bedes tests better than a Rook (525 against 500 centiPawn), a pair of FADs as slightly worse (450-475), and thus provides an advantage of more than 2 Pawns over the Bishop pair. This is not dequately compensated by substituting the Queen for an Archbishop, which differs by less than a Pawn from it. (The WA is almost equal in value to the Knight.)

The names used here for the pieces are awful, of course. In the WinBoard / Fairy-Max implementation I use different names. (Or at least different letters to indicate the pieces in FEN and SAN; WinBoard never uses full names of pieces.)

FIDE      Nutters     Clobberers
N Knight  H Horse     E Elephant
B Bishop  U Unicorn   D Deputy
R Rook    T Turret    L Lama
Q Queen   C Colonel   A Archbishop

'Unicorn'seems an applicable name for Knight-King chimera, and WinBoard happens to have a bilt-in bitmap for it. For the Horse I use the WinBoard Nightrider symbol, and for the Colonel the Knight-on-Rook symbol that is popularas a representation of the Cancellor in some 10x8 variants, so that the Nutters army indeed looks quite Knight-like. For the 'Lama' I use the Promoted Bishop symbol, which in WinBoard generically stands for a Bishopwith some extra moves, (in this case the (2,0) teleports), and the WA is an Elephant variation because of the Alfil move.

References

• Index of Ralph Betza's work on piece values
• A Pawn is as Strong as the Hand that Holds It Part I
• A Pawn is as Strong as the Hand that Holds It Part II
• A Pawn is as Strong as the Hand that Holds It Part III

By taking away the known and studied board positions, it takes the chess back to its root as a test of logic and strategy. From pure information overload, players should not -- and generally can not -- rely on memorized openings or endgames, but must instead invent the process as they play.

Really, is that not the point of chess variants, giving chess players a new experience outside of the tried-and-true? I have not, nor do I have any desire to work towards memorizing opening books, beyond two or three moves. My enjoyment of chess comes from working out the best tactic as the game develops.

CWDA is my preferred variant, because of its simplicity and expandability. Really, the game play is chess. Learn the movements of four new pieces, and you can introduce a new army into the game. With just four armies, there are now 15 games you can play (not counting the FIDE vs FIDE match). Starting with a knowledge of chess, and only adding the movement of 12 previously unknown pieces. Simple! Versatile! Elegant! What's not to like?

http://www.chessvariants.org/index/displaycomment.php?commentid=614. This comment of Betza was linked before, but what ''pprilla'' says today is exactly what Betza was saying the last couple years to 2003.  NextChess will take up the C.D.A.(provisional #30) this year.

We tacitly assume that strength can be measured by one number and that the numbers can be compared using a transitive relation like 'greater than'. However, this needs not to be true, see here for a simple game with dice:

http://en.wikipedia.org/wiki/Nontransitive_dice

So here is a new chess variant challenge: Chess with nontransitive armies

Design a chess variant with different armies such that, whatever army your opponent chooses first, you can choose another army having an advantage over your opponent's army. (To avoid the first move problem assume 2 games where either army moves first once)

Muller's solution there cannot be topped, but hypothesize,
http://www.chessvariants.org/d.betza/chessvar/cvda/alice.html,
that Fabulous F.i.d.e. Army beats Maharajah, Maharajah beats Alician Army, and Alician Army beats Fabulous F.i.d.e. Maharajah army would be royal (RNB) alone.  The rules have to be tweaked properly from the several Alice-to-F.i.d.e. choice cvs Betza presents above.  The index to this old part of Betza:

Okay, Cyclic Advantage Armies can be many too. Right or wrong should assure maybe 75%-25%, with any question resolvable by Muller's or other engines. Easier to generate are Pawn armies dispensing with piece, instead 8 pawns and King . That would be the problemists' way; or problemist method would be to try very few piece-types only 1 or 2, or even only 2 or 3 pieces themselves let alone types.  (In fact, the two triads of C.A.A.s so far are that way.) This is a good topic appropriate for C.D.A. ////
Example One of Pawns only: In general, to stay 8x8 and allow flexibility, make the arrays Pawns a1-b1-c1-d1-e1-f1-g1-h1; King e2.
Example One, Pawns only, call it ''R.O.Q.'' Rock, like the Rock/scissors/paper it simulates: Quadra-Pawn army > Rococo Pawn army > Ortho-Pawn army > Quadra-Pawn army....
Ortho-Pawns have 1-, 2-, or 3-step opening option 8x8. Promotion all three types to Queen. Quadra-Pawn is like used in Centennial Chess. As per the suggestion, play this alternately on three-player board for mayhem and indeterminate outcomes, certainly not any 63-19-18, in peculiar way to back-equalize:
http://www.chessvariants.org/multiplayer.dir/three_player/three_player_chess.html. Does any program play three-players yet?
[Incidentally contrast all Pawns here to no Pawns of current fad Chieftain Chess.]

Revision. Cyclic Advantage Army, Pawns only. ROQ, ROCK Armies are the three-fold. Quadra-Pawn army > Rococo army > Ortho-pawn army > Quadra-pawn army...ad infinitum.  What was not noticed, Centennial's two-step for Quadra-pawn has to be eliminated, getting back to normal, since Quadra-pawn is not original with Centennial anyway.  Arrays have to be those 1- and 8-ranked described, promotion to Queen virtually wins, but one more tweak may be necessary.  Ortho-pawn may need 1-, 2-, 3-, or *4*-step opening option, to be sure to subdue Quadra-pawn. And that strengthening of the Ortho-pawn will not change Rococo advantage over her. And Quadra-pawn without two-step will still give Rococo fits, who has to jump-capture. At least all to the tune of 60-40 if not 70-30, best projected.  The Kings will be offensive weapons par excellence.

These Cyclic Armies definitely deserve articles or cvs. Questions like: (1) How about four in sequence? (2) Hexagonal ones. Experimenting there are found 4 Dabbabah bindings, that Gilman has not touched on yet; but I did not finish a tripartite  hexagonal army with any confidence. (3) Try one piece only and King back on squares.  This one is only preliminary for 8x8:    [Pasha/Mastodon + King] >  [(Dabbabante+Wazir)+King] > [(Quadra-leaper 1,5 1,6 2,5 2,6) + King] > [Mastodon+King] ad infinitum. ////   That (15/16/25/26) is okay to call (Ibis-Flamingo-(Satyr or Korsar)-26).

Sphinx of black quartz, judge my vow.
There are shorter ones, but contrived by use of proper nouns.

Some (tiny) steps have been taken towards CwDA opening theory. The page quoted below also has a link to a mini-tournament of CwDA games.

'The Paulowich Plan for playing with the Remarkable Rookies is very interesting.

My own first attempts at playing this army involved taking a cramped closed central position and suffering for a long while before winning; my second idea was to do a Halfduck Dance, which may work even though it goes against the general principle of developing weaker pieces first.

Pushing the b-Pawn so the WD can sit behind it is an interesting and creative idea; but in Paulowich-Aronson I'd instinctively prefer 3...a5-a4.

(My instinct could be wrong, of course.)' -- gnohmon [9 Oct 2001].

I found this comment by looking at a nonindexed page on this site: Recent Ratings and Comments, which actually covers old comments from [27 May 2001] to [31 Mar 2002]. This list is also available in another format: alphabetic by variant name, where you can more easily find the 1st Email Championship Chess w... comments. NOTE: the page name on the left links to the main page, while the three blue dots on the extreme right link to the comments.

To add an army, balance the sides because material must vary. The Immortal AntiClericals versus Fabulous F.I.D.E.s. The new idea for army is Immortal AntiClericals and they go: RNIQKbNR. There should be same-value forces where 'b' is Barrier Pawn of year 1948, http://www.chessvariants.org/large.dir/kristensens.html. 'I' is Immortal mediaeval Germanic Mann, the sub-piece variate of Man/Mamra suggested by Jeremy Lennert. 'Immortal' here moves and captures non-royal King-like and when captured belongs again to the capturee. Capturee later drops I. on line 1 or 2 vacancy. Since all Betzan Armies (unlike atoms) are equal, there can now be playable fair match-ups Immortal AntiClericals v. Pizza Kings and I. A. v. Nutty Nights, and so on, weigh in Betza willing.