Comments by nelk114

The link in the introduction is broken; it should probably point to the author's translation of the Chaturanga page — note that the latter, as well as the Courier Chess page still need converting (the Shogi page oþoh seems intact — it was last edited in 2018 though so perhaps it escaped the corruption). Also is there an old backup of the Chinese page on FIDE? As it stands now it's complete nonsense. EDIT: The Wayback Machine has a copy (and the garbled version seems to have the same update stamp so it probably hadn't changed in the 14 years before conversion to its current state)

I don't see anything wrong with having pages in multiple languages here, and we do indeed have some pages in Spanish, among others. The Alphabetical Index issues queries for English pages only by default, but you can query for non‐English pages too — apparently there's only(!) 108 of them so there's no difficulty getting them to display. Though in any case the Index lacks pages for “Pages beginning with ⟨菲⟩”, for example…

That does indeed look quite plausible (as does the Shatranj page), and concords with my own efforts at making sense of it, as well as lining up, afaict, with what the English page says.

The note at the end looks a little odd (It literally reads, as far as my Chinese gets me: “This is our English page's translation Fischer Random Chess”; the syntax of the Chinese is fine afaict but the link afterward, even with a space separating it, reads strangely). Would it be worth putting the link on the Chinese for “Our English page” (i.e. “我们英文页面”) instead?

Also incidentally what software did you use to convert it to CP1252? All the immediately accessible ones on linux seem to put up a (quiet) fuss about e.g. U+0081 not being available in the target encoding.

The classic answer was to hold a Design contest (the obvious themes in this case being Silver and/or the nr 25), but I'm not sure if we have enough people active here aþm for that to really work — istr the 2017 20th‌‐anniversary one never really went anywhere (though I suppose one could always try recruiting from other Variant fora…).

That said, a tournament would work too. Featuring games from throughout the pages' tenure…

Speaking of both names and M&B, worth noting that Paulowich's Spotted Gryphon is called an Angel there, and Gilman's Nightingale is close to your F-then-W-then-DD piece, though it cannot stop on the F square (and thus makes exactly an even number of steps). The Chainsaw (and a whole class of related pieces) remains unnamed.

It's added to the number of remaining moves before a draw, afaict — i.e. your latter alternative. Though I can see how my wording was ambiguous; sorry for any additional confusion from my part

This one?: https://www.chessvariants.com/index/listcomments.php?id=33008

I happened to be reading this thread recently so it's still freshish in my memory

Three obvious potential improvements stand out, besides the aforementioned grammatical issues:

• The setup diagram is misleading; even though you clarify that the fore‐ and hindmost ranks are not actually part of the board, it's probably better to make that clear in the diagram too
• Similarly, using pieces to denote destination squares in the movement diagrams is extremely(!) confusing. The diagram designer does provide for using coloured circles for this purpose; consider finding out how to use that functionality
• The drop rule is not clear. You say that cannons and two soldiers start in hand and may be dropped only on resp^ly the 1^st or 3^rd rank, but even this information is easily missed and further details (is it Shōgi‐style drops? Seirawan‐style? Sth else entirely?) are completely unspecified. I'd suggest putting a note about it in the Rules section

Also a typo: your paragraph about the Soldier refers to droppable pawns.

You mean how could it be described? With the new E atom it should be simple enough to do exactly one reflection: [B-E-sB] (a slight change from the regular refl. B in HG's comment) For arbitrarily many, but at least one, I think we still need the off‑board interpretation of o as the [] notation lacks arbitrary repetitions afaict. That would give something like yafoabyas(yafoabyas)B (though the sandbox seems to ignore the brackets on the second one, only allowing either one or two reflections… probably I'm doing sth wrong)

I think you've asked a similar question before, and the answer (including re this case) is further down this comment thread ;)

I also have a question of my own: just to clarify, a Dark Spirit or Buddhist Spirit capturing a Deva or Teaching King, or vice versa, causes it, like other pieces, to convert to its victim? The notes clarify that, as expected, one of them would disappear, but don't make clear which one, and one could make a case imo for contageous pieces being immune to contageon themselves.

It's still not clear what this means. A withdrawer captures by making a move, whilst an immobiliser stops others from making moves — the former's effect is on its own turn while the latter's is on the opponent's.

As such there's a couple of interpretations possible:

• It stops things from moving away like a withdrawer. This is just a weaker immobiliser, and is already attested in Euqorab
• It petrifies pieces that it moves away from rather than capturing them. Then the question is whether and if so when pieces can come back into play: never (as with Nemoroth's basilisk)? When the withdrawing‐petrifier moves again (leaving it able to only petrify one piece at a time)? After a larger but still fixed number of moves (turn counting, ugh)? Under some other condition? Are involuntary moves (from Swappers, shepherding pieces, Go Away!s, ⁊c.) counted? Does capturing it release pieces?

Imo the former option is not very interesting, nor necessarily well‐defined (how does it deal with knights?), while the latter is quite complex in principle and perhaps not very immobiliser‐like — though I admit the possibility of ulima‐style pieces with effects besides capture is interesting and not very well explored

‘Idk’ and the others are indeed abbreviations: ‘Idk’ itself is ‘I don't know’; ‘Afaik’ ‘as far as I know’; ‘Ofc’ ‘Of course’; and ‘Iirc’ ‘if I recall correctly’.

As for trivially making up, a pawn on the central rank stops the opponent's pawn from making the same move (as that space is now occupied) and the opponent can't immediately do anything equivalent, and so has won a (potentially) better position with little effort — and since White can force this more easily than Black, the argument goes that normal pawns of an odd‐file board might give White more of an advantage. Here is the formal write‐up of Hutnik's idea, which he first proposed here, and the first few comments on Elven Chess(/Elven Shogi) also touch on it.

I seem to remember the old versions of some of your larger games (Gigachess ⁊c.) had the Ship, before you updated them. And ofc I wasn't accusing you of ‘stealing’ the ship ;) just noting that it also exists where the snaketongue was first named (I expect Eric probably got it from there, though it's certainly possible he came up with it independently).

I believe you want checkaride instead of checkride — the latter checks all directions symmetrically (making a full gryphon plus conditional wazir moves), while the former is asymmetric.

Presumably if your suggestion for the Ship is otherwise correct, the snaketongue would similarly be:

def G fn (checkaride #0 #1 1 1 and empty #0)
    where #0 0 1
    #1
    or fn (checkaride #0 #1 -1 1 and empty #0)
    where #0 0 1
    #1
    or fn (checkaride #0 #1 1 -1 and empty #0)
    where #0 0 -1
    #1
    or fn (checkaride #0 #1 -1 -1 and empty #0)
    where #0 0 -1
    #1
    or checkleap #0 #1 1 0;

def GL mergeall
    leaps #0 1 0
    ray where #0 0 1 1 1
    ray where #0 0 -1 1 -1
    ray where #0 0 1 -1 1
    ray where #0 0 -1 -1 -1;

Chu does have quite a few short‐range pieces, like (Sho) Shōgi; it's not exactly devoid of longer‐range ones though: Rook, Bishop, Queen, as well as Dragon Horse and ‐King are the more conventional ones (and all but the queen in pairs), and it even has, to Western eyes, weird things like side‐/vertical movers and their promotions. And even with the short‐range ones, at first sight the variety of very similar moves might seem confusing just as several long‐range pieces might.

Gross Chess is popular here among CV fans; that speaks, no doubt, to its playability and potential popularity — and may well indicate it as a good candidate for a successor — but says very little imo about how 12×12 might fare among a more lay audience — while Chu demonstrates that it's possible for it to hold that status.

The point about game length is potentially a concern once the board gets bigger (and is almost certainly, alongside tractability, once of the limiting factors for going to e.g. 14×14 and beyond as anything ore than a novelty), though I'd've expected at least games with plenty of long‐range pieces to balance that somewhat. I wonder how long the average game of Gross or Metamachy (of which I've been playing a fair bit against Jocly's AI recently) is, esp. compared to Chu.

Merry (belated) Christmas and a Happy New Year to you all!

the same trouble with the bishop and queen.

It's not entirely clear what the analogous ‘error’ would be. In the REX King's and Glinski Pawns' cases it's using orthogonal moves to the exclusion of hex‐diagonal ones, while this knight apparently just miscounted the diagonal portion, resulting in a piece (which Charles Gilman terms a Student) which is analogous to the square‐cell Zebra.

A queen analogous to the REX king just becomes a rook, but that leaves the bishop completely unaccounted for.

Ofc, there are a few variants which take this version of king and queen as their basis and build the rest of the pieces around them: the oldest is Sigmund Wellisch's 3‐player game (for which this site unfortunately has only a Java Applet, though a more complete description is available e.g. on John Savard's page); the king moves one orthogonally, the knight to any nearest square that the king can't reach (there is a certain logic to calling the hex diagonals ‘leaps’, given that the relevant cells don't actually touch), the rook slides orthogonally, the queen moves as rook or knight (technically a marshal analogue therefore), and the pawn in either of the forwardmost directions (the board being oriented as in Fergus' Hex Shogis).

Alternatively, Gilman's Alternate Orthogonals Hex Chesses do exactly what the name suggests: assign alternate orthogonals as analogous to the square‐board directions, giving a REX king and Glinski pawns together with Wellisch knights, a rook as a ‘queen’, and ‘rooks’ and ‘bishops’ which have each other's move but backwards — albeit this being Charles Gilman, the pieces all have ifferent names. This one had quite a positive reception, and it does preserve some aspects of square‐cell chess that other analogies lack (some of which are touched on in its comments) — it's certainly worth a look

@Max

Game Courier doesn't (currently?) support games for more than two players. Idk how those four‐player variants were done, though it wouldn't surprise me if it's in teams where each player is suppoesd to control both armies in each team. (presumably these are not rule‐enforcing)

I feel like multiplayer games may technically be on Fergus' list of things to Maybe Eventually Add to GC (istr him mentioning it though I wouldn't hold him to that) but there's no support now and iirc probably requires a fairlyfundamental refactor at the very least

that initial legs would behave differently from only legs.

The behaviour is the same if you stipulate that all slider legs are potentially 0‐length but null moves are disallowed unless explicitly specified

Perhaps we have different intuition

May well be :) And fwiw I'm fine with either system in practice

When 0 steps is allowed, you would need [F-fF-fB] for the Tamerlane Picket

Or simply [nA-fB], which to me looks more natural as an alfil extension (istr Gilman classes it that way too). [F-fB] is ofc a bit odd as a Bishop description, but there are always going to be strange ways of notating things

BTW, [W?sfNN], and even [W?sfCC] work now. All through using a new, undocumented (and quite horrible) extension of XBetza.

That's pretty cool :) (and agreed, the repeating ys are… not pretty). I can even get an offset giraffe‐rider (or even zemel‐rider — presumably longer ones work too, if they'd fit on the board), even though normal giraffe‐riders (FXFX?) are apparently unsupported!

But the y extension still fails for e.g. [W?sfZZ] (also shouldn't that be fsNN ⁊c?), let alone pathological things like [C?fsZZ], so if we're making an effort to support direction‐type changes it probably deserves to be more general.

Also speaking of the Z, [Z?sfB] currently gives me Zebra‐then‐Rook, and vice‐versa

a King, a Xiangqi Elephant, a Xiangqi Horse and a two-path lame Dabbabah (XBetza KaFafsW).

From the description (both here and of the Sliding General), shouldn't it be a three‐path lame dabbabah? Including the possibility of two consecutive same‐direction Wazir steps. That'd give KaFafsfW or KaFafsWnD

Afaict, it looks like Sin-yeon-sang-hui is a historical Janggi variant (though my Korean is nonexistent so I can't confirm any of what Daphne posted), and this is a back‐formation (‘dramatised’ is probably Google Translate or equivalent) of an an equivalent Xiàngqì‐derived variant.

Presumably the Korean original has no river and Korean‐style cannons/advisors/generals

Indeed, it seems that either the knight's verbal description or its XBetza move has been exchanged with the one in the Modern game.

@Aurelian?

Indeed, that was my impression from his book as well; I'd initially missed the detail of how recent this variant was and had assumed it significantly older (and I don't trust people to count like we do today(!)).

I enjoyed the other papers you posted here and look forward to reading this one too

Apparently I forgot to add the link to the XBetza page in my previous comment; I've now added it there.

yafsF is indeed the sliding part. a is as you've found, ‘again’; fs for w's and F's is interpreted as for a king, so for an F it changes to a W direction — and ‘forward’ for anything but the first part of the move is interpreted as ‘outward’ (like Alfonso about the rhinoceros); y is a ‘range toggle’ i.e. it switches from being a leaper/stepper to being a slider. Thus, yafsF is one step diagonally followed by a 45° turn and sliding orthogonally.

Complexity is in the eye of the beholder. It's not immediately obvious (especially compared to t[FR]) but it's apperntly easier to describe to a computer (and easier to generalise), which for the interactive diagrams is a definite plus

Ok, having actually gone to find a copy online, I agree that Tressau specifies the Kb/j–Rc/i castle; in principle one could still object that the example games may not be played by the original rules (while he says they're real, rather than constructed, games, it could still be under the influence of a misunderstanding), but it seems upon a cursory reading that for the Sultan's Game in particular his book may in fact be the original source? The Emperors Game is cited in the Spielarchiv, but Tressau explicitly notes (p.80) that a game with a Marshal had been suggested there but not described, rather being rejected due to the necessary odd number of files being unwieldy (in particular due to either same‐colour bishops or transposition of one bishop but not the other with its adjacent knight).

Unfortunately a quick search for the Archiv der Spiele online appears entirely fruitless so I can't confirm that…

The dragon is indeed a t[FR], in Betza's original notation. However, that part of it was never documented on the Betza Notation page (instead languishing on the Chess on a Really Big Board page, though it turns up elsewhere too), and is arguably a little underspecified, so H. G.'s XBetza (which is what the interactive diagram uses) specifies such multi‐leg moves in its own way. In this extension, FyafsF is indeed equivalent to the original t[FR]

The vulture afaict is mainly a longer‐range relative of George Duke's (and more recently Uli Schwekendiek's) Falcon, whose advantage over the bison (from a game design perspective) is its blockability — presumably the same is sought here. Unfortunately, due to the multiple paths to a given destination, it is quite complex to describe. Idk about the extra knight move though, that's perhaps a little gratuitous (presumably to make up for the basic vutlre's lack of maneuverability?)

I agree the birds are quite complex, if potentially interesting to play with? And whether the knight/elephant enhancements are truly necessary may be worth a playtest as well

Originally posted on Pemba, where this piece is called a Crocodile.

I assume JL means this one? His page includes it thrice: that ‘close‐up’ at the beginning of the Rules section, the full page featuring it after the list of volumes in Alfonso's book, and a reproduction on a commemorative stamp at the end of the page

Is that Tressau's interpretation? And do we know how much info the original sourcs gives on the matter?

After all if the latter does specify, as this article does, that both pieces move four spaces each, the Kc/i–Rd/h interpretation would make sense both in terms of preserving usual castling and lining up with the frequent use of inclusive counting (see also paragraph 4 of the Comments in Cazaux' page on Grant Acedrex)

On the other hand, with an initial leap the blocking would only be on a single diagonal, rather than having to check the nearby orthogonal squares as well. A tradeoff really, though in any case due to the ability to avoid nearby blocking pieces the t[HB] is probably too powerful (or, on small enough boards, too awkward) to use in (most?) games anyway.

I believe this page to now be ready for publication

The difficulty with viewing e.g. the 2.griffin/running ostrich as R2‐then‐B is that the obvious reading of that (in line with the obvious reading of full ‘rook then bishop’ — see also the large shogis' ‘rook‐then‐rook’ and ‘bishop‐then‐bishop’ hook movers) suggests that it could also make the Bishop move after only a single Wazir step, becoming effectively a compound of griffon and ostrich — what Gilman called a Fimbriated griffon (after a kind of outline in heraldry). Which is really quite powerful and not what either of us means afaict.

My view here is that the usual Ostrich (and Osprey) have a move along a given path, but the shortest of its moves is two steps — something it has in common with Tamerlane's picket, Alfonso X's unicorn, and indeed Shatranj's and Xiàng Qí's elephant. For the picket and elephant, the 2‐step move is non‐coprime, and so a one‐step move can be trivially interpolated: for the former this gives the familiar Bishop, while the latter gave a piece that was dubbed the ‘modern’ elephant (and of course with a modern dabbaba to match). For the unicorn it is less trivial (the knight has two possible interpolations) but extending the long‐range move backwards suggests orthogonal‐then‐diagonal over the alternative, giving our Manticore.

In the Osprey's and Ostrich's case, the 2‐step shortest move, as with the picket and elephant, is non‐coprime, and so the obvious interpolation lines up with your 2.bent riders. In the Osprey's case, the alternative exists of doing as with the unicorn and extending backwards, giving a ferz‐then‐bishop‐at‐90°, but fsr 90° turns seem (above‐mentioned hook movers notwithstanding) to be less favoured.

I don't disagree about blockability: what I have termed ‘running’, as opposed to the preëxisting ‘modern’, is explicitly blockable — though arguably calling them ‘lame/stepping modern’ os[prey/triche]s is just as descriptive. Fergus' helical pieces also differ from Charles' Proselyte ⁊c in being (by default) blockable, as well as interpolating.

As for 3‐or‐more.gryphons/manticores, it might indeed be interesting to have names for those (though they might begin to veer into being a little too exotic?), but it'd be equally useful imo to have names for the equally unusual threeleaper‐then‐bishop or quibbler‐then‐rook.

Are there any plans to restore this functionality? Not sure whether the ongoing procedures with backups ⁊c make this a more or less opportune time to look into this, so if the latter it can ofc wait, but it'd be good to have it back eventually

when playing the game one is generally attempting to achieve a particular double

Perhaps, but if that's what you meant it might be worth being clearer about that; the way it's phrased aþm suggests that the probability of being able to make any Séj‐dice capturing move (assuming availability of pieces to capture) is 1∕3+1∕36=13∕36, which doesn't really make sense (not least, that'd be likelier than merely being able to move even if the 1∕3 figure were correct). The actual probability is in fact, as dax00 said, 11∕36(=chance of matching the last piece to move)×1∕6(=chance of a double)=11∕216. The chance of any given piece being able to capture is 1∕6 of that again, i.e. 11∕1296.

since the fraction 11∕36 is almost a third, doesn't that in effect equate to 1∕3?

Well 1∕3=12∕36, so… no? It's close, sure, but still an 8.33% difference — if you consider that trivial enough to be discounted fine, but don't expect everyone (especially those of us with a mathematical inclination) to agree.

There is a 66.6% chance that, because each turn the dice MUST match the opponent's piece, thence the game will continue as regular Classical Chess

I also just noticed this remark; even aside from the percentage being wrong — the chance that a given turn will be ‘normal’ is 25∕36=69.44% — it's not clear whether you mean that to apply only to each turn, or (incorrectly) to the whole game. After 2 turns the likelihood of still having a normal game is 25∕36×25∕36=625∕1296 (less than 1∕2) and it keeps going down from there. Having a full game of Séj where the dice do not once allow a deviation from ‘classical’ chess is vanishingly unlikely.

the players start with […] a queen and a king and 8 pawns on their secind rank

All standard chess pieces except the king

The king isn't royal

These statements contradict each other. How can the player start with a king that's not in the game, and how can said king be non‐royal?

Also the rules regarding pawns are a bit unclear — am I right in inferring that a pawn effectively demotes upon reaching the last rank while adding a new ‘pawn’ (which can still move as a queen) into the pocket? thereby making the pawns a potentially infinite source of new pieces?

It looks as if it might mean bringing one's king to a space adjacent to the opponent's)? Would be not unlike Bachelor Chess (and a number of Gilman's) in that respect. Hence the commment about kings meeting on “their” — i.e. ‘one's own’ — half of the board, the case of 4th/5th ranks, and the note about “approaching” the opposing king.

Also the fifth paragraph remains given twice

Back‐of‐the‐envelope translation:

este ajedrez tiene un tablero modificado de 73 casillas .

This chess [variant] has a modified board of 73 squares.

las piezas adicionales a las convencionales son :

The pieces in addition to the conventional ones are:

un canciller : torre + caballo .

A Chancellor: Rook+Knight

el omega que mueve dos en diagonal y salta a la segunda casilla en diagonal ademas ,puede saltar a la 2da. casilla en ortogonal con una pieza de por medio.

The Omega which moves 2 diagonally and jumps to the seconddiagonal square; additionally it can jump to the 2nd square orthogonally with a piece in between.

los peones se llaman columnas y pueden tambien mover una casilla en diagonal hacia atras en ambas direcciones si estas estan desocupadas.

Pawns are called Columns and can also move one square diagonally backwards in either direction if these are vacant

el rey es un Emperador : rey + caballo .

The king is an Emperor: King+Knight

reglas especiales :

Special Rules:

_se mueven dos fichas por turno pero solo se puede comer una vez por turno.

• Two pieces are moved per turn but only one capture may be made per turn

_una ficha no se puede mover dos veces en un solo turno..

• A piece cannot move twice in one turn

_por cada 4 piezas que un bando atrape obliga en el siguiente turno de su adversario a que reinicie una de las piezas que haya capturado como aliada.

• For every four pieces captures by a side, his opponent is obliged on his next turn to reintroduce one of the pieces he has captured on his side

_las piezas iniciadas comienzan en cualquier casilla de la 1era. fila de su bando.

• [Re‐]introduced pieces begin on any square on the 1st file of their side

_cuando un rey está en jaque puede salir de un mate intercambiando 2 piezas de jerarquia () por un cambio de posición hacia una casilla que esté libre. (es opcional).

• When a king [i.e. an emperor?] is in check he can escape mate by exchanging two pieces of hierarchy () [sic] for a change of position to an empty space (this is optional)

_el atacante escoge una pieza y la inicia como aliada , el atacado escoge la otra pieza..

• The attacker chooses a piece and introduces it on his side; the attackee selects the other piece

As to the “two” corner squares, removing two squares from each corner of a 9×9 board does in fact leave 73 squares, though whether the second square is vertically or horizontally adjacent to the square in the corner remains unclear. Presumably the promised diagram will clarify.

The chance of rolling a double is 1/36. https://stackoverflow.com/questions/32313428/understanding-the-probability-of-a-double-six-if-i-roll-two-dice

The chance of rolling any given double (e.g., as in your link, a double 6) is indeed 1∕36. There are, however, six doubles to choose from, so the total probability is in fact 1∕6 of rolling any double.

The chance of rolling a single is 1/6 ⁓ but two dice thence double the likelihood of instances to 1/3

Alas, adding probabilities does not work that way. In effect you've counted the outcome of a double twice. The chance of rolling at least one of a given number with 2 dice is, in fact, 1−(1−1∕6)²=11∕36.

x is already taken sadly (for move‐relayers that ‘eXcite’ other pieces). Afaict the only (ASCII — I'd wager few would support using þ, ß, ⱶ, ⁊c.) letters that aren't yet used in XBetza are t and w — of which the former had a meaning for Betza himself (though iirc the article he introduces it in also features incompatible meanings for a number of other letters). w as a prefix (to ‘widen’ the available options?) sounds plausible I think

Ok so you are indeed including the 1‐step move among the movement possibilities — I suppose your ‘2.manticore’ is different enough from a lame/stepping/blockable osprey/lama (and mutatis mutandis for the ‘2.gryphon’) to merit a different name… maybe. If it's not being distinguished from a ‘true’ osprey/zephyr in a single game I'd be tempted to leave that name as is, though avoiding going too near overlexicalisation is a personal preference.

If the distinction does need to be made within a game (at which point the tradeoff changes imo), then ‘2.manticore’(/‘2.griffon’) is probably fine, if perhaps closer to the (presumably also present in a game featuring both of these) actual manticore/griffin than might be desirable, though I'm not really one to ask for original name suggestions (and M&B13 is apparently lacking names for lama/rook and zephyr/bishop compound with which we could (ab)use M&B8's ‘‐lander’ suffix). Perhaps a name based on lama/zephyr/osprey/ostrich/whatever you want to call them would be more apt? ‘Running Osprey’ isn't altogether without a ring to it…

The Princess' diagram disagrees with the text in omitting sideways moves

Charles Gilman called these Zephyr and Lama respectively, though by his description they cannot make the one‐step move that I'm not sure whether you're including. The latter also turns up as an Osprey in Expanded Chess

While on its own the original diagram on this page was a bit obscure, in conjunction with Fergus' circular diagrams it really clarifies the rationale behind Nadvorney's interpretation of the diagonal move; a bit of thought also reveals why Miller's reasoning in keeping the bishop on its own colour is flawed: if the bishop stays on its own colour you would expect a rook stepping over the pole to change colour as it does on a normal square‐cell board, whereas here (or on any spherical/Klein‐bottle‐shaped board with a multiple of four files) it doesn't. On e.g. a 10‐file board, Miller's reasoning would line up with Nadvorney's.

As for Chess on the Dot, the change in the diagonal's handedness at the poles also keeps it on one colour (on a board of this parity), but isn't stirctly necessary for a closed loop: Nadvorney's version (as can be seen on its diagram) does it just as well, and even Miller's manages, albeit via a much more circuitous route.

Fwiw, here's the original diagram as salvaged from the Internet Archive:

c7  d7  e7  f7  g7  h7  a7  b7  c7  d7  e7  f7
c8  d8  e8  f8  g8  h8  a8  b8  c8  d8  e8  f8
g8  h8  a8  b8  c8  d8  e8  f8  g8  h8  a8  b8
g7  h7  a7  b7  c7  d7  e7  f7  g7  h7  a7  b7
g6  h6  a6  b6  c6  d6  e6  f6  g6  h6  a6  b6
g5  h5  a5  b5  c5  d5  e5  f5  g5  h5  a5  b5
g4  h4  a4  b4  c4  d4  e4  f4  g4  h4  a4  b4
g3  h3  a3  b3  c3  d3  e3  f3  g3  h3  a3  b3
g2  h2  a2  b2  c2  d2  e2  f2  g2  h2  a2  b2
g1  h1  a1  b1  c1  d1  e1  f1  g1  h1  a1  b1
c1  d1  e1  f1  g1  h1  a1  b1  c1  d1  e1  f1
c2  d2  e2  f2  g2  h2  a2  b2  c2  d2  e2  f2 

As for pieces that can only do slide‐then‐step and not the reverse, the only one that comes to mind in a game is the Transcendental Prelate.

Are those now the ‘default’ versions of the Hero and Shaman then? If so what do we call the non‐bent versions in the Chieftan variants: Linear (rejecting ‘straight’ as, like ‘bent’, perhaps overimplying somewhat)?

Abdication could be represented by putting [the prince] on top of 2 checkers

Or just replacing it with the king, since the latter gets removed from the board otherwise

Ditto — thanks

It seems that multi‐capturing two‌‐square leapers are oddly popular as knight enhancements: this game has the multi‐capture apparently mandatory for when going to Alibaba destinations, but still allows leaping otherwise; Larry Smith's Li Qi replaces knights with ‘Young’ Chu‐shogi lions, which cannot move back to the starting square, to match its planar linepieces; and ofc H. G.'s Mighty‐lion Chess replaces one knight with a full Lion.

I wonder why?

Since any parents or children are listed on the page for a tag

Parents are listed, but at present children seem to be absent

@Uli:

Re Duke's variant and patent: whilst it was, indeed, probably an exercise in extending the patent's scope, I'd argue that technically he did describe this variant first (and strictly speaking patents are a different thing from copyright, and I'm not convinced that someone patenting a new set of chords (perhaps in some unusual tuning?) that wasn't in prior use coudn't in fact claim a breach of patent on anything using them, though ofc IANAL) — however you rightly point out that he came to different conclusions, and that merely having described it as part of a set of possibilities doesn't mean that he should be credited for it (H. G. Muller's analogy with integers is apt here). My main point was that it's more interesting (at least for me ☺︎) to acknowledge the commonalities (as you have now, indeed, done) and explore the differences within that than to insist that everything is unique and special in itself.

Also I agree, naming things is hard (which is partly why it was uncanny that you ended up with the same name as Duke did).

@Jörg:

Strictly speaking Kestrel has been used (by Gilman, predictably enough) for a piece — just not in any games. It's the compound of (stepping) Falcon and Kite (the latter moving as falcon but replacing orthogonal steps with ‘nonstandard’ (i.e. √3) diagonal ones) according to M&B13.

I think the rules were clear, just that Betza notation does not describe it unambiguously. Though it seems reasonable to interpret crooked moves as continuing in the outwardmost direction by default, unless otherwise specified, in which case t[WzB] would indeed describe the Harvestman (the other option is then not covered by the original Betza notation, though something like t[WfhzB] would probably be clear enough)

To be fair, Cattle are ungulates too, and I suspect in some ways perhaps more apt for such a weak piece (not only is its leap quite long and awkward, but it's also bound to 1/4 of the board like the Dabbaba) — certainly Stag would suggest to me something stronger. Ofc in Gilman's case he also wants to be able to extrapolate (to Zherolais, Ghirolais ⁊c. for 6:4, 2:8, ⁊c. leapers) so he's constrained in his naming by that.

I seem to recall having read somewhere about a variation on the pawns' initial double move which, contrary to the standard rules, allows not only a single pawn moving two steps, but also two pawns moving one step each. Iirc it was described as being played in India, with a popular opening involving one King's pawn moving two steps and the other side moving both bishops' pawns, though it may have been the knights' pawns and I can't remember which side makes which move.

Does this ring a bell for anyone? I can't seem to find it anywhere myself.

Gilman calls it the Charolais.

I've done one last (hopefully, for now) update to this page, incorporating H. G.'s suggestion about the ‘Contra‐’ prefix, and a caveat about Mideast Chess' Cavalier. I'll henceforth probably leave this page alone for now.

The remark was very much an assessment of my impression of Man and Beast rather than an actual suggestion as such.

True, he was rather fond of, sometimes gratuitously, proliferating names, though in his case it kind of makes sense considering they would often turn up in games together and it'd be a bit of a pain to have several kinds of very different ‘rook’ in a game. A matter of degree I suppose really (after all we don't go around calling things wazir‐/ferz‐/manriders (the reuse of those for shogi‐general extrapolations in M&B10 notwithstanding) — and ofc the large shogis are even more extreme, if not nearly as numerous).

Fwiw at least Boyscout is well‐enough established imo that ‘helical boyscout’ wouldn't generate much confusion — but helical bishop doesn't have any other obvious meaning so… they can be synonyms! (now all that's missing is a prefix combining helical and switchback…)

Your comment said that RennChess was “a follow up of Mideast Chess” — if so then it would make sense that Eric would not keep the Cavalier's name but change the move. Thus his understanding of Mideast probably agreed with Pritchard, which would mean that Gollon also agreed (in both sources, if different). In that case Hans' page has an error introduced either by a typo from Eric, or by Hans.

Given that that move is probably erroneous, it makes sense to either remove the reference, or keep it but with a note that it's probably in error. In the latter case, calling it “Hans' account” could risk reflecting badly on him, even though (of course) that is not the intent. Fwiw, “These pages' account” risks the same directed at ourselves(!), while “one account” or “some accounts” is quite unspecific (and the latter may be incorrect if ours is the only such).

Hope that's clearer — English is very much one of my mother tongues (though sometimes I wonder whether it'd be more interesting if it weren't), as is indirection/terseness it would seem ;)

It's not terribly clear given that the diagram lacks horizontal separators between ranks (instead using one text line per rank), but the pawns start (as written below the diagram) on the 3rd rank, not the second.

Though I'm a bit curious as to why 3 generals rather than the more obvious two?

Thanks, Fergus!

@Jean‐Louis: Hmm, in that case it does seem like Hans was in error, esp. if, as you say, RennChess was a followup to Mideast which would suggest that Greenwood had the (putatively) correct description. It may be worth taking the reference back out then given that the otherwise necessary explicit caveat regarding Hans' account might reflect unnecessarily harshly.

@H. G.: Fair; I'll put that in the notes section. Do you think it's worth generally adopting Gilman's additional ‘Double’ term as well for the Duke?

I also remembered one more variant featuring the manticore as iirc a knight upgrade, but haven't found it again [Edit: just found it], and I also found this one with a lame double‐ski (i.e. at least 3‐square) manticore move as one form of the ‘mutating serpent’. Oþoh I feel like this page may well be more than comprehensive enough as it is(!)

Also since I expect to make at least one more revision of this, I seem to remember there's a preference for relative urls in intra‐site links; am I correct in thinking that those are the same as absolute ones but without the leading https://www.chessvariants.com?

Might it make sense to bump the update time, or some equivalent metadata, when a page is unhidden so that it appears on the What's New page? It seems quite unfortunate that some submissions apparently will miss out on some exposure because they were last updated over a month before publication. And it can be quite confusing to see things turn up even on the most current page as being new the previous week, when it wasn't there back then.

Chushin Shogi (unhidden earlier today but last updated at the beginning of December and thus on the 60–90‐day‐old page) would be the latest example of this.

According to Jean‐Louis' site the king can make an initial two‐square radial leap

Edit: just took a look at the code and saw that those are already there and the question was about knight leaps in addition to those. But yeah there seems to be no indication that those are available

Re names, I much prefer Acromantula over Rotated/Tilted/Complement of/Altered Griffon as I consider neither one more/less basic. As for a generic term for pieces with ortho‐/diagonal components swapped, something with ‘complement’ seems appropriate (suggesting a symmetrical relationship) — perhaps ‘diagonal complement’ or ‘radial complement’? For pieces with only one kind of radial move Charles Gilman uses ‘dual’, but for pieces mixing them that's subtly different. The concept is more complicated on 3D boards so the terminology needn't take that into account.

@Aurelian: fwiw, Daniil Frolov's variant (mentioned on the page) uses Gryphon for its usual referent and Dragon for the t[WB], so that way round wouldn't be without precedent — in fact I'd initially forgotten the name change from Gryphon/Aanca and assumed that the Dragon was the t[WB] when drafting this.

Good point. I took a look and although I don't have Zillions, the code seems to define 24 leaps for the Waterbugs and Mantis and 48 for the Tarantula, which indeed corresponds to the area leaping rather than just radial moves.

Interestingly the header comment in the .zrf calls it Entomology Chess rather than Insect Chess. Perhaps he deemed it too obscure a word?

OK so for the birds (I assume based on previous comments that ‘thunderbolt’ is a typo, though I suppose md or something suggests some kind of energy) I'd go with either modified bird pictures (things like _PA_cd, _PA_wa, _PA_wc, _jc_af, !aaf, !aak, !aaw ⁊c) or something which suggests the move (so the likes of _JG_bgr, _JG_bspgr, _JG_co, _JG_gr, _JG_ha, _JG_raa, _JG_rc, _JG_rcd, _JG_rcflaa, _JG_re, _JG_rspaa, _MLV_si2 ⁊c.).

The humanoids could, given their moves, also draw from the same set as the birds, or you could just go with the various human heads (ge, maybe th or ch, _MLV_ge, _MLV_ma ⁊c) — as far as the cyclops is concerned, you can only see up to one eye at a time(!) Alternatively _MH_ge kind of looks like an eye.

As for the Valkyrie, idk how I'd represent an actual valkyrie so I'd suggest some kind of augmented bishop. _MH_b, _JG_ap2b, or _JG_apb2b perhaps?