You are on the backup site for Any posts, moves, or other changes you make here will not be permanent, because the pages and database from the main site will be backed up here every midnight EST. Additionally, things may not be working right, because this site is also a testbed for newer system software. So, if you are not here to test, develop, or merely read this site, you may want to change .org to .com in the navigation bar and go to the main site.

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 ]

Game Reviews by George Duke

Later Reverse Order EarlierEarliest
Jumping Chess. Pieces capture by jumping. Board has extra edge squares making it 10x10. (10x10, Cells: 100) [All Comments] [Add Comment or Rating]
George Duke wrote on 2018-06-27 UTCExcellent ★★★★★

Jumping Chess originates the edge squares that Rococo uses two years later. Bishop captures like International Draughts diagonally and Rook like Turkish/Israeli draughts orthogonally. Except no plural captures, and in JC the line pieces slide any distance beforehand. But no displacement capture at all here. Jumping concepts are bandied about in 'ECV' a few times, but credit this improvement for the rim accessible only capturing.

JC may create too many defensive positions for most aesthetics.

JC year-2000 date of invention harkens to V. R. Parton's booklet 'My Games for 2000 a.d. and After' published 1972. There the CV "2000 AD" sources pieces for 30 years later great Rococo. Firsthand, Rococo is basically a derivative Ultima (1962).

( Contrariwise, Robert Abbott himself weighed in early Rococo comment that no need for border squares, just get rid of them. ) See next how Rococo draws on both Abbott and Parton. 20th century the chief variantists were Boyer, Parton, Betza and Dawson, but Dawson didn't bother with designing actual CVs.

The Rococo pieces straight out of Abbott's Ultima are Withdrawer, Immobilizer, Long Leaper, Chameleon. And the Rococo pieces straight out of 2000 A. D. are Ximaera and Swapper. Ximaera gets re-named Advancer. Finally, Rococo takes its own inventor's border squares from JC and adds that great novelty Cannon Pawn.

Perimeter-squared JC has little play, but Rococo, when adding its subvariants Push-Pull and Mirror, has the same number 10 rank approximately of near-form Ultima at Game Courier. And several ahead of them are a standard Chess form around hundred(s) years. Or combine play numbers of Ultima and Rococo and they are number 3. So arguably derived-form Rococo is a topmost world-class CV. Thanks to contribution of porous out-migration squares from selfsame JC.

Orwell Chess. Three player variant themed on George Orwell's 1984. (7x12, Cells: 84) [All Comments] [Add Comment or Rating]
George Duke wrote on 2018-05-27 UTCExcellent ★★★★★

J. C. Hallman in 2004 'The Chess Artist' interviews Kirsan Ilyumzhinov at his multi-million Chess City, Kalmykia.  Then the f.i.d.e. President says to the effect:  whatever happens, or even thought of, in the mundane "real world," Chess has already been there, having visited every eventuality.   So this Orwell '1984'  by Overby.

As in '1984', three  players Eurasia, East Asia. and Oceania.  Everyone reviewing loved it except Charles Gilman.  Maybe he is right one or two of the 7 piece-types could be tweaked.  But this is perfectly symmetrical solution to three-player CV.  

Three teams ebb and flow constrained from unfair alliance by cylindrical downwards and up, and by Shifting Alliances rule, and by Perpetual Powers rule.  Variant pieces go back to year 1283 in Gryphon.  King may move into check because the dice may free him.  The '3x1's give where the Berolina pawns promote, and the seventh piece is promotee royal Maharajah.  

Regulator Chess. Game on a 35 square board with a 7 square track on which a piece moves that determines how Knights and Bishops can move. (6x7, Cells: 42) [All Comments] [Add Comment or Rating]
George Duke wrote on 2018-05-08 UTCGood ★★★★

How to compute piece values here?  It's pointless to bother.  Regulator move one forward or backward is determined by play on the small board of 35 squares.  Too bad no one has experimented with the Regulator band on 8x8 with 7-square streak to the side for neat 71 square Chess Variant, a first for the size.

84-square Fourriere's Jacks and Witches lessens Bishop value to Knight by the 16-square hole in the center -- more detrimental to the Bishop now worth 2.5.   Any different board size must affect piece values somewhat.  A Fischer Random Chess array NBKRRQBN should benefit Rook to relative downgrading of Knight, maybe 5.3-2.7.  But adherents to FRC have not gotten that far yet.  So the starting line-up alone affects piece-values.  Also do the Rules in and of themselves, such as obvious thing like Bishop one-time Wazir step conversion in some CVs. 

Take the simple Regulator embodiment here.  Level 6 of regulator makes Knight into Marshall, and Level 3 makes Bishop into Cardinal.  If the board were 8x8, it's not definite which in a given game benefits more, Bishop or Knight,  because there are going to be enough move- and capture-triggers to jockey the calibration up towards level 7 tactically pretty easily, and keep it there.  Over the long haul however, Bishop value is going to show relative increase, on any rectangle 35 to 100.  That is because of its compound to BN being in effect 5/7 of the time and the other to NR only 2/7 the time.

Next, there is room for subvariants, that either side can alter the calibration, let's say forward by agreed-on even number of steps across 7 and back to 1 and onward, in lieu of a move. That can include in the Regulator Band not just moving the Regulator but either of the two trigger levels the same way, as not overcomplicated move addition.

New-Chess. Variant on 10 by 10 board with combination pieces. (10x10, Cells: 100) [All Comments] [Add Comment or Rating]
George Duke wrote on 2018-05-07 UTCGood ★★★★

Fifteen years til the first comment and rating now.

Sava made New-Chess by 1973, in time for the second edition of Gollon's book.  There are Marshall RN (but not Cardinal BN), Amazon RBN and Gnu N-Camel.  Then Betza and Cohen came up with Tutti-Frutti in 1978 with the 'Capablanca two' and Amazon again, this time on 8x8.  So it appears Betza and Cohen got some inspiration from then recent New-Chess and thought it important to put their similar piece mix on little 8x8. New CVs were fewer and further between those decades. In fact, Betza always designed 1970s through 2003 on 8x8 with two exceptions.  His 'Outrigger' article adds files to get 8x10, and Chess on Really Big Board has four 8x8 boards for 256 squares.*

No allowing in New-Chess for Pawn three-step, but it has perfect implementation of modern free castling with the King moving over 2 or more but not past Rook.  Gollon's book has a few dozen CVs exhibited, nothing like the couple thousand of Pritchard 'ECV'  twenty years later.

Aronson calls Complete Permutation Chess more flamboyant 'Tutti-Uti-Frutti Chess' in that one's text, and if I had noticed the above sequence more carefully I would have approved TUF over bland CPC. Complete Permutation adheres to idea of using each possible bi-compound once -- originating in Betza & Cohen.

*Betza's Chessopoly and Race Chess are 64-square 4 x 16.

Neutral Subject Chess. Most pieces start neutral, and players compete to recruit them. (8x8, Cells: 64) [All Comments] [Add Comment or Rating]
George Duke wrote on 2018-04-27 UTCExcellent ★★★★★

Parton made Neutral King in 1953, where player has own orthodox pieces but the King is co-owned and

yet has to be checkmated.  Simple and elegant.  Most of Gilman's CVs are hurt by overcomplications in

piece-moves, odd board sizes, too many special rules, or attempt hybridizing Eastern chesses with forced

templates.  Once in a while he strikes paydirt such as AltOrthHex idea of splitting up the hexagonal Rook

into two, though nobody has really done anything with that either.  

Neutral Subject realizes that Parton's Mutator has wider applicability. Here player only has King and Queen to begin.  Neutral pieces get moved and then assigned to one side or the other.  The criterion to assign is applied at end of each turn according to hypothetical attack of each 'Neutral' on any piece already assigned.  Who wouldn't want more pieces rather than fewer? Many other CVs could be made in this genre of the pieces on board not belonging to either army initially.  

Charles' novel CV invention, expanding on Parton, gets somewhat awkward explanation in his essay. Like Aronson and Howe with Rococo, great idea is not followed up with clear summary fully disambiguating.

Still in all, there could be other ways to set up the bazaar of recruitment to build the forces in subvariants and new CVs this type of possible breakthrough Mutator.



Dream Chess 46. 46-squasre variant played from opposite corners of a FIDE board with the other corners removed. (8x8, Cells: 46) [All Comments] [Add Comment or Rating]
George Duke wrote on 2018-04-27 UTCGood ★★★★

This is a straightforward CV by Gilman.  To suit the constricted board, he decides to use Shogi promotees for Bishop and Rook, adding Wazir and Ferz respectively.  Pawns are Centennial-acclaimed Quadra-pawns.  Some restriction on Knight at inside corners.  That's it.  Should be playable enough short games.  

But the piece in right corner needs correction to King.

Nachtmahr. Game with seven different kinds of Nightriders. (8x8, Cells: 64) [All Comments] [Add Comment or Rating]
George Duke wrote on 2018-03-08 UTCExcellent ★★★★★

There are reams more nightriders mostly unutilized than the ordinary hack one developed by Dawson a century ago. So far they remain in problems and thought experiments.  Classic essay here proposes Straight Wide Crooked, Diagonal Narrow Crooked, Diagonal Wide Crooked, and Straight Narrow Crooked.  Best of all, the essential nightrider Quintessence.  Each one makes better more interesting play than Betzan-tagged 'NN'.  Play of that ordinary Dawson nightrider is inferior because it just duplicates successive Knight moves same direction.  It is no more interesting than "limited" pieces like an up-to-three-step Bishop or Chess Different Armies Short Rook.

Quintessence itself gets play in odd-shaped 84-square Quintessential Chess, adding also  Leeloo compound R + Quintessence.  

Quinquereme takes it up to 12x12 with the same Quintessence.  Each of the various nightriders in combinations, one and two of each together with some of the other 6 or 8 piece-types in the set, on different board sizes can create thousands, well millions easily, of individualized CVs.  Worth exploring in the abstract are the standard boards 9x9, 9x10, 10x10, 10x12, 12x12, 10x16.  All the large sizes should have a variant nightrider species for improved implementations. Even rudimentary Dawson NN of such wide appearance is superior to also-overused Carreran BN and RN, four hundred years beat to death.

Partnership Mitregi. Unthemed 4-player variant with most pieces always moving toward or across the River. (8x8, Cells: 64) [All Comments] [Add Comment or Rating]
George Duke wrote on 2017-03-21 UTCPoor ★

Yes Charles, I think it would be fine to drop this (he asks advice on this in red). "sidewaysmost, 'Halfcamel', 'skewed Dabbabah', 'Colourbound analogue' and 'river-straddling zigzag' are turgid and off-putting without any of Ralph's deadend tongue in cheek. However other CVs that get deleted also lose the scathing review.

Gess. A Chess variant played on a Go board where pieces are collections of go stones. (18x18, Cells: 324) [All Comments] [Add Comment or Rating]
George Duke wrote on 2017-03-13 UTCExcellent ★★★★★

Player must keep a Ring of 3x3 made from the stones, and to win is to destroy opponent last Ring.  Stones move in 3x3s. This appeared first in Spektrum der Wissenschaft.

Influence Chess. Pieces on the top or bottom layer influence which chess pieces may move on the middle layer. (4x7x3, Cells: 84) [All Comments] [Add Comment or Rating]
George Duke wrote on 2017-03-12 UTCExcellent ★★★★★

 A square that a main Middle board piece sits on has corresponding square in Above and Below boards. These locations (departure square) 'influence' whether a move can be considered or not. To make the move, it also must be legal within the Middle board. Sometimes the Above or Below two piece-types move their one- or two-square way, and other times they duplicate a Middle board movement. Rules may very well be interpretable (including moving opponent's piece) in all cases.

Giant-King Chess. Kings take up four squares each, all of which must be attacked to check. (10x10, Cells: 100) [All Comments] [Add Comment or Rating]
George Duke wrote on 2017-03-10 UTCExcellent ★★★★★

Pawns do not promote. The Pawns reaching promotion zone cause other pieces to promote and the Pawn leaves the board. The four-square occupation of King requires all four attacked for mate.

Colour Chess. Pieces paint the squares they leave, allowing other pieces to move as them. (8x8, Cells: 64) [All Comments] [Add Comment or Rating]
George Duke wrote on 2017-03-02 UTCExcellent ★★★★★

Squares increase in power.  Each time a piece leaves it, it leaves a trace, so a square can eventually confer Amazon power even to lowly Pawn when he arrives.

French revolution chess. Advanced pawns threaten the noble pieces. (8x8, Cells: 64) [All Comments] [Add Comment or Rating]
George Duke wrote on 2017-03-01 UTCGood ★★★★

Another advanced pawn starting formation.


Patt-schach (Stalemate chess). Players start with an illegal move from a stalemated position. (8x8, Cells: 64) [All Comments] [Add Comment or Rating]
George Duke wrote on 2017-03-01 UTCGood ★★★★

The first move has to be illegal, so Black Pawn cannot on the first move take a White Pawn that has moved 1 P-a2.  Since it is legal move for Black, he cannot do so on move 1.

Tryzantine Chess. A three-handed form of Byzantine (circular) Chess. (4x21, Cells: 84) [All Comments] [Add Comment or Rating]
George Duke wrote on 2017-03-01 UTCGood ★★★★

Here is the only 3-handed circular chess so far.

Swap Chess. A move can consist of a series of pieces swapping places. (8x8, Cells: 64) [All Comments] [Add Comment or Rating]
George Duke wrote on 2017-03-01 UTCExcellent ★★★★★

Swap Chess allows serial swapping as a move along subsequent lines of attack.  Swap Chess has never been put up in Game Courier like Switching Chess.

Upside-Down Chess. White starts at the upper two rows, black at the bottom. (8x8, Cells: 64) [All Comments] [Add Comment or Rating]
George Duke wrote on 2017-02-28 UTCExcellent ★★★★★

White is going north and Black south as usual in the CV Upside Down Chess.

For OrthoChess problem by Lord Dunsany, solve this:

White is to play and mate in four moves. The position is one that could occur in actual play of F.I.D.E. Chess.

Altair. Altair is a modern game with an oriental flavor. (9x9, Cells: 81) [All Comments] [Add Comment or Rating]
George Duke wrote on 2017-02-13 UTCExcellent ★★★★★

Altair is CV where "piece values are not a good indicator of one side's advantage in chess" to use V's current words, because most of the pieces for a move can also be dropped to empty square in rank nearby of the same color. Also they most of them can slide along their rank unimpeded. So if coming up with guide-values for stronger Mage and Lion and Diamond around 7, 5 and 4 respectively, good use of the board itself makes all the pieces closer to heuristic 3.8-4.2 each with only Pawns in some 2.x range.

Muller wrote up problem theme 3Q v. 7N in "Charge of Light Brigade." If you keep 8 Pawns, the 3 Queens versus 7 Knights may go to 3Q by already 8x10 any array, certainly by 10x10. Board used and Rules interact piece values, and cannot really be safely generalized even as to '<' or '>' for all cases; with special rules (or board) we can think of CV where even N>Q one on one!

For ex., make narrow stair step where Q can only occasionally go 3 or 2, but N leaps cross empty space and get values maybe N4 Q3 as convenient.

Earthquake Chess. An earthquake caused a kind of Z-form in the board. (8x8, Cells: 8) [All Comments] [Add Comment or Rating]
George Duke wrote on 2017-02-03 UTCGood ★★★★

Since the 1820s backranks have been altered to thwart opening theory. Fischer was just as ignorant of chess history as Seirawan in the latter's re-introducing compounds of RN and BN to whatever he calls them in Seirawan Chess. Fischer's re-invention in the recent nineties of up to 960 set-ups led to also variants Slide-Shuffle and Deployment and Free Placement. Randomization of array is Mutator applicable to any CV. Instead at the same time of FRC, Ralph Betza proposed changing connectivity of the once-sancrosanct little 64-board. Besides Betza's two-ranks displacement, here are some other board possibilities:

There are also things like Transcendental, T_Chess where the two sides' initial positions do not have to match, and Chaos Chess in which pieces start dropped to other than only the nearest rank.

Pied Color Chess. Oh no! All the colors on the board have been scrambled -- however will the pieces move? (8x8, Cells: 64) [All Comments] [Add Comment or Rating]
George Duke wrote on 2016-12-22 UTCExcellent ★★★★★

In Ralph Betza's Pied Color Chess, all the pieces except Pawns are changed in their movement by the piing, unless they happen to have normal dark and light (eight)-square surrounding. In the Example the Rook on b1 starting to c2 can stop there or continue -b3-c4-d5-e6-f7-g8. In the coloration exampled there are no, zero, "normal" squares so far as adjacencies let alone two away and beyond. Watch what's beneath you!

Anti-King Chess. Each player has both a King and an Anti-King to protect; Anti-Kings are in check when not attacked. (8x8, Cells: 64) (Recognized!)[All Comments] [Add Comment or Rating]
George Duke wrote on 2016-11-08 UTCGood ★★★★

Here is a CV with two Kings like Muller's example for negative-value piece.  In Anti-King win is by checkmate of regular King or removing check from other's Anti-King.  Two other CVs with two Kings are Two Kings Chess and Double Chess.

Both Aronson's Berolina Pawn version and Anti-King Chess II have strategy to keep the side's Anti-King in check. In AKC-I with Berolina note that Anti-King is initially attacked by four pieces checking, and it will take a while to get them "safely" out of the way. Anti-King Chess II may benefit from changing Anti-King move to Knight move only as subvariant.

How do these relate to negative values? That pieces may want to be removed, if possible, in end game in order to have no forces nearby to attack opponent Anti-King, but their over-all average value would be positive just taking on negative value at end. Player may just settle for checkmating regular King.

Fergus Duniho's insightful strategy for actual game played 13 years ago: Strategy, where few pieces were captured.

The Game of Nemoroth. For the sake of your sanity, do not read this variant! (8x8, Cells: 64) [All Comments] [Add Comment or Rating]
George Duke wrote on 2016-10-31 UTCGood ★★★★

Traditional for Halloween October.

Chessembly. Open Board Setup, Free Placement Chess. (8x8, Cells: 64) [All Comments] [Add Comment or Rating]
George Duke wrote on 2016-10-31 UTCGood ★★★★

Basic Chess is another placement CV:


Another one is Multiple_Formations.

Maxima. Maxima is an interesting and exiting variant of Ultima, with new elements that make Maxima more clear and dynamic. (Cells: 76) [All Comments] [Add Comment or Rating]
George Duke wrote on 2016-10-20 UTCExcellent ★★★★★

Mage of Maxima is another Gryphon -- before Aurelian Florea's Apothecary.

Here Lavieri claims Guard resists accurate valuation: Piece_Value.

Understand that before Muller we used to do these things in more of a ballpark way.

Symmetric Unirexal Chess. Each player has a half of a king. (8x8, Cells: 64) [All Comments] [Add Comment or Rating]
George Duke wrote on 2016-10-18 UTCGood ★★★★

Black has nearly 100 CVs that have been looked at sparingly. He, along with his brother, have some Mutators presented into standard f.i.d.e. that are pretty challenging and logical by second reading anyway. Symmetric Unirexal Chess should be clarified further that each half-piece move is actually by the same rules as ordinary Rook, Knight, Bishop, Pawn, Queen or King. This first comment here is for interpretation of unambiguous rules-set. There is no half move, just restrictions on when a piece is half-piece, where there is initially counterpart half-piece having the same full regular move. The new emergent half-pieces would have to be marked. It appears, if all become half pieces, all 64 squares could be theoretically covered, but that may be impossible to achieve without some tweaking p-t rules of movement. That's for a study or problem, and the actual goal by Black is just a regular CV won by checkmate.

Recent topic was Symmetry or symmetric moves in CV rules, like Frolov's "Reflection Teammate," and it led to this Black CV-Mutator for having the word in title. It reminds me of another CV, drawn up by orthodox defenders Chessbase in 2014 Tandem Pawn, since there Pawns split into two pieces.

25 comments displayed

Later Reverse Order EarlierEarliest

Permalink to the exact comments currently displayed.