[HN Gopher] 4-Die Chess
       ___________________________________________________________________
        
       4-Die Chess
        
       Author : memorable
       Score  : 40 points
       Date   : 2022-08-13 09:11 UTC (2 days ago)
        
 (HTM) web link (erinbern.com)
 (TXT) w3m dump (erinbern.com)
        
       | mark_l_watson wrote:
       | Not for me, but each to their own.
       | 
       | I like to play Stockfish, with the playing level set so I win or
       | lose about half the time.
       | 
       | For dice games, Backgammon is just about a perfect game, and
       | skill matters.
        
         | timbit42 wrote:
         | No game with random elements (eg. dice) can ever be perfect.
        
           | bena wrote:
           | I don't think he's using it in the same manner you may be.
           | 
           | A game can have random elements and still be "perfect".
           | Because his perfect game may be a game where he can lose due
           | to variance. There is something to seeing if you can play
           | around variance. That way there's always a way the game can
           | be challenging.
           | 
           | Unlike games with perfect information and no variance. Those
           | games can be completely solved (given enough resources). And
           | I think that's what you're thinking of, games with perfect
           | information.
        
       | marcofiset wrote:
       | I would've liked an explanation on why each of these rules were
       | added to the game. As an advanced-beginner chess player, most of
       | those new rules add needless complexity to the game and seem to
       | serve absolutely no purpose.
       | 
       | > Each player rolls the die, and the number is the amount of
       | turns each player has.
       | 
       | I'll have to think about this. My knee-jerk reaction was to call
       | this "absolutely ridiculous". It may hold some potential upon
       | further thought.
        
         | backpackviolet wrote:
         | It's for fun. It takes playing pieces most people have
         | available, rules many are familar with, and makes a new game
         | with just a few more rules. For fun.
        
           | bena wrote:
           | But what if it's not fun? "For fun" is trying to dodge valid
           | criticisms of the ruleset provided.
        
             | flyaway123 wrote:
             | I agree that more clarification is necessary for this to be
             | adopted mainstream, however that didn't seem to be the
             | intention of the article. i.e. 'fun' for 2 persons knowing
             | each other is highly subjective.
             | 
             | Think of it the other way - some people find it fun just by
             | throwing the dice in turns and whoever has the larger
             | number wins. In this case, there's even a chance for
             | whoever has the huge disadvantage (unlucky) to make a
             | comeback. In cases where it's fatal, either those are added
             | as exceptions e.g. re-roll, or accepted as auto-win. Not
             | unlike some gambling games.
             | 
             | My friends and I had fun adding different rules to existing
             | way of playing various games. We might find out later on
             | that the rule might be incomprehensive, which we could
             | either discard or adjust. It might also never be perfectly
             | balanced. Either way we definitely had fun.
             | 
             | I take the article as more of a "story sharing" than a "new
             | specification" for chess.
        
               | bena wrote:
               | The bulk of the post is a description of the new rules
               | with a small anecdote about why they were created.
               | 
               | It seems like the intention of the article was to share
               | the rules. As such, we should be able to discuss those
               | rules. And this person would like to know the reasoning
               | behind the rules as that does give insight. Maybe the
               | rule is counter to the actual goal of the rule and would
               | be better served by another rule. Or to get rid of it
               | entirely.
               | 
               | Because even chess hasn't always been chess. Chess has
               | been developed over years and settled into its state
               | after a lot of refinements.
               | 
               | "It's for fun" is a thought terminating phrase.
        
         | tetha wrote:
         | > I'll have to think about this. My knee-jerk reaction was to
         | call this "absolutely ridiculous". It may hold some potential
         | upon further thought.
         | 
         | I'm not sure, but my gut feeling is that every roll above 3 and
         | maybe even equal to 3 is an immediate checkmate or at the very
         | least a crushing advantage. Like, rolling a 4 on the first move
         | wins the game, rolling a 3 wins the queen, ... that much tempo
         | is devastating.
         | 
         | However, it does remind me of a rather funny chess variant. In
         | that variant, each player has three D6 and each face value
         | corresponds to a piece - 1 for a pawn, 2 for a knight, and so
         | on. Each turn, you roll your 3 dice and you can make one move
         | with either of the pieces you rolled.
         | 
         | This ends up being pretty instructive, because you really have
         | to activate all pieces because you might not roll your most
         | active pieces on a turn.
        
           | backpackviolet wrote:
           | > Like, rolling a 4 on the first move wins the game
           | 
           | No, that and your checkmate concerns are almost certainly why
           | they came up with this rule:
           | 
           | "The rest of a piece's moves are forfeited if a check is
           | obtained."
           | 
           | So you cannot move and strike the King is the same moveset,
           | and your opponent will have 1-4 moves to counter/flee.
        
             | Hasu wrote:
             | 1. e4 2. Bc4 3. Qf3 4. Qxf7 (or the same with turns 2 and 3
             | reversed, the final move can also be the bishop) wins with
             | the first check being a checkmate.
        
             | xnorswap wrote:
             | E4 Bc4 Qf3 Qf7#
             | 
             | There is no check there until checkmate is delivered.
             | 
             | If the enemy is then allowed to play 4 moves to capture the
             | queen while the king is in check then that implies illegal
             | moves are allowed in the intermediate moves.
             | 
             | But if illegal moves are allowed then white can open with
             | 1. Qxd8 Qxa8 Qxh8 Qd1 because once illegal moves are
             | allowed then the whole game devolves a step further.
             | 
             | Accurately encoding the rules of games in a manner that
             | does not allow for loopholes is a really hard problem. Just
             | look at golf. A simple game in which you hit a ball into a
             | hole in the shortest time yet the "official" (R&A) rules
             | run to 230+ pages and has frequent updates to fix issues.
             | 
             | Cricket appears to be a far more complex game yet the laws
             | of cricket as published by Wisden actually comes in at a
             | slightly shorter 208 pages.
        
               | zimpenfish wrote:
               | > you hit a ball into a hole in the shortest time
               | 
               | Technically the "fewest number of strokes" rather than
               | "the shortest time", I think?
               | 
               | (apologies if there's a speed-golfing community)
        
               | Someone wrote:
               | There is, with each stroke counting for 60 seconds
               | (https://en.wikipedia.org/wiki/Speed_golf)
        
               | jjnoakes wrote:
               | > that implies illegal moves are allowed in the
               | intermediate moves
               | 
               | Seems simple enough to define illegal moves as moves that
               | end your turn in check, which is basically the same
               | definition as regular chess. I don't immediately see the
               | problem there.
               | 
               | > if illegal moves are allowed then white can open with
               | 1. Qxd8 Qxa8 Qxh8 Qd1
               | 
               | This isn't the same thing at all...
        
               | xnorswap wrote:
               | But the rule in chess is actually:
               | 
               | > No piece can be moved that will either expose the king
               | of the same colour to check or leave that king in check.
               | 
               | And you're right that you can re-write that rule, but my
               | point is that it needs to actually be done here, which it
               | hasn't been.
               | 
               | I'm not trying to shit on these rules, I'm just
               | highlighting the difficulty in actually writing
               | consistent rules.
        
               | jjnoakes wrote:
               | Right, I didn't say the rule in chess was exactly that,
               | but "basically" that - which effectively is the same.
               | 
               | The rules could certainly be written out in more
               | legalese, but I found them easy to understand as-is
               | without resorting to suggesting that the queen could also
               | start ignoring and moving through pieces at will.
        
           | dev_tty01 wrote:
           | >I'm not sure, but my gut feeling is that every roll above 3
           | and maybe even equal to 3 is an immediate checkmate or at the
           | very least a crushing advantage. Like, rolling a 4 on the
           | first move wins the game, rolling a 3 wins the queen, ...
           | that much tempo is devastating.
           | 
           | That advantage is held in check, so to speak, by the check
           | limited turns. Once a piece has put an opponent in check, it
           | can't move again. Other pieces cannot put the opponent in
           | check on the same move. That still creates the possibility of
           | checkmate in one turn, but remember that the opponent will
           | likely also get multiple moves to get out of the situation.
           | Assuming that the in-check player can use all of their moves
           | to get out of check, they might get 3 or 4 moves with which
           | they can wipe out the advantage gained by the other player on
           | the previous move.
           | 
           | Of course, if the in-check player only happens to get one
           | move then it is probably over. Basically, this injects a
           | significant level of chance into gameplay and it speeds up
           | the game. Its really a different game. The more I think about
           | it the more interesting it becomes.
           | 
           | To make it more even handed and less dependent upon chance,
           | it could be done with only 2-3-4 move turns. Being allowed
           | only a single move after an opponent had 4 moves is a
           | disaster. Or maybe a limit on the delta between move counts
           | for successive turns.
        
           | xnorswap wrote:
           | Oh absolutely, just 1 or 2 moves would be more than enough to
           | really shake things up and make players far more defensive.
           | 
           | Three moves are generally crushing and four is totally
           | unplayable. There aren't many positions where four moves for
           | your opponent won't get you checkmated.
           | 
           | There is general chess advice of "Give you opponent three
           | free moves, what would they do?"
           | 
           | But even that advice implies those moves would be sensible
           | moves, i.e. not just hang pieces in the intermediate moves.
           | 
           | Perhaps if pieces couldn't move en-prise in the intermediate
           | moves it would be improved but I think 3 or 4 moves would
           | still be crushing.
           | 
           | Even a single tempo loss is a big concession, to the point
           | where in some opening lines players will happily sacrifice a
           | pawn to be a tempo up with a good attack (e.g. Smith-Morra).
           | 
           | It feels like there are missing rules about not being able to
           | capture in intermediate moves (or a capture ending the whole
           | turn). Given 4 moves a queen could trivially capture three
           | enemy pawns/pieces then move to a safe square. Even just
           | capturing two pieces and moving to safety would be a decisive
           | advantage.
        
         | twright wrote:
         | I mostly agree but I think the complexity is explicitly the
         | purpose and some people like that.
        
         | seunosewa wrote:
         | A good simplification: Toss a coin before each play. If it
         | lands on a tail, you lose your turn.
        
           | egypturnash wrote:
           | Alternately: heads, two turns; tails, one turn. Having a
           | "turn" where all you do is say "I don't have anything I can
           | do" sucks all the fun out of a game at very, very high speed.
           | 
           | "I keep getting tails and you keep getting heads, this sucks"
           | is still a problem but it's not going to be quite as bad as
           | "dude I never even got to move and you won, let's play
           | something else".
        
       | twright wrote:
       | If the player going first rolls 4 or more they can checkmate? e4
       | Bc4 Qf3 Qf7# (for White with a mirror equivalent for Black) With
       | closer reading I think there's a missing rule where you just move
       | a single piece that many times?
        
         | jjnoakes wrote:
         | I assume it is only checkmate if black's reply can't get him
         | out of check by the end of his turn.
        
         | backpackviolet wrote:
         | No, there is a rule in there that pieces causing Check forfeit
         | further moves.
        
           | vlovich123 wrote:
           | There is no check in op's stated list of moves
        
       | olddustytrail wrote:
       | This article is not intended to be serious. Read the rules
       | carefully: none of them make sense.
        
       | dmos62 wrote:
       | Cheers. I find it stimulating when people rethink something
       | that's seemingly set in stone. Rules for chess in this case.
        
         | karmakurtisaani wrote:
         | There are actually dozens of chess variants out there. Yet, the
         | classic version seems to remain the most popular.
        
           | ZetaZero wrote:
           | Today's "classic version" is just a variant that allows pawns
           | to move two squares on their first move.
        
             | karmakurtisaani wrote:
             | You know, at this point I can't even imagine HN without the
             | all encompassing pedantry.
        
             | cyrialize wrote:
             | Is this in reference to how chess was played in India
             | (Chaturanga)? I've been reading a book on Sultan Khan [0],
             | and it talks about the many different ways Chess was
             | played.
             | 
             | Interestingly enough, the only common rule amongst all the
             | different versions of Chess (since it changed depending on
             | where you lived) was that the pawn always went one space.
             | 
             | [0]: https://www.amazon.com/Sultan-Khan-Servant-Champion-
             | British/...
        
       | Imnimo wrote:
       | >The opposing player has 1-turn to defend, or remove, the king
       | from danger.
       | 
       | Does this mean that if I am put in check, regardless of what I
       | roll on the die, my king must be out of danger after my first
       | move?
        
       | xnorswap wrote:
       | Very few of these rules make sense without more clarifying rules.
       | 
       | Never-mind the impossibility of Kings playing en-passant, there
       | is also the concept introduced of "higher piece" without defining
       | an order. Can a bishop upgrade to a knight? Can a knight upgrade
       | to a Bishop? Can a rook upgrade to a bishop?
       | 
       | Despite the popular "valuing" of pieces, they aren't actually in
       | the rules of chess and need to be stated.
       | 
       | > If a king is within its starting row, it can castle with both
       | rooks.
       | 
       | What does it mean to castle with _both_ rooks? Is that a typo for
       | "either"?
       | 
       | Also it seems to suggest that you can move the king along the
       | starting rank and still castle later. How would that work if your
       | king was on b8 for example?
       | 
       | This is a strange kind of nonsense. It shows the difficulty of
       | translating some home-brew rules into actual rules that others
       | can follow.
        
         | bena wrote:
         | With a 4 sided die, the game will end on the first turn about
         | 6% of the time. The Scholar's mate takes 4 moves. And if the
         | other player rolls a 1, then it's a win for white.
         | 
         | Yeah, kings playing en-passant, rule 2, seems useless. The king
         | moves in all eight directions, you can just like, take the
         | piece. If the two kings try to pass each other, the passing
         | king will check the opposing king, end the turn, then you just
         | take the piece. Unless you're automatically checking the piece
         | when your turn starts, in that case, it's just a stalemate as
         | each king is checking the other at the start of the turn and
         | ending the turn.
         | 
         | Third rule is just promotion but for all pieces. This is just
         | saying "and everything becomes a queen".
         | 
         | Fourth rule is just a more relaxed version of castling.
         | 
         | Fifth rule is to just get new pieces on the board, because
         | otherwise, the game ends in a draw. No one is going to
         | willingly put their king next to the other king. Also "starting
         | position" is kind of nebulous. There are 8 of them. Do I get to
         | choose which column to start the pawn on? After that, it's just
         | a race to promotion.
         | 
         | The rule about allowing another turn after check just extends
         | the game. And unless you roll a 1, you should be able to get
         | out of most checkmates.
         | 
         | This is adding more rules, not adding complexity. Every rule is
         | to make the game easier, removing restrictions. The die lets
         | you get around the 1 move per turn rule. Rule 2 is pretty much
         | for decoration. Rule 3 gets you faster promotions. It's about
         | who can get to the opposite row first. Rule 4 makes castling
         | easier. Rule 5 is about not wanting a draw. Because with Rule
         | 1, sometimes the king can take a queen with no assistance. This
         | is a game the creator thinks is brilliant and thinks shows how
         | smart they are because they are virtually undefeated because
         | all of their opponents eventually get tired of playing because
         | it's just so tedious.
         | 
         | It's like all the rules added to Monopoly. They're all to make
         | the game "more fair" or "better", but all they wind up doing is
         | making the game take longer because no one wins.
        
           | pc86 wrote:
           | I think the en-passant rule is a bit more clever than you're
           | giving it credit for. At the root of en-passant is that a
           | pawn moving into the space the opposing pawn "skipped" allows
           | you to take it, but only if done immediately.
           | 
           | So we're playing this version, down to just two kings, I roll
           | a 4 and move from E6 to E2. If you're at G3, and you roll at
           | least a two, you can move over to E3 and capture me en-
           | passant. Or at least that's my understanding of the rule. You
           | moved into a square my king was occupying in the turn[s]
           | immediately following mine.
        
             | bena wrote:
             | That just seems... bad.
             | 
             | Also easy to circumvent by moving along diagonals. You can
             | even zig-zag your move so you can wind up where you want. I
             | could go E6-D5-C4-D3-E2, then you couldn't en passant to
             | E3. Not that you could do it on E3 as that would put you in
             | check and isn't allowed. But G4 to E4 is a legal move with
             | the same outcome in your scenario.
             | 
             | I really can't see a way that you'd be able to en passant a
             | king that's not easily routed around or just illegal.
             | 
             | It also violates the concept that you're taking multiple
             | turns. You can capture a pawn en passant because the pawn's
             | double move is a relatively recent rule. It's solely to
             | make sure you can't avoid capture by using the double move.
             | But it's still a single turn.
             | 
             | The die roll lets you move that many pieces. As N distinct
             | turns. If the turns would have played out one-by-one, it
             | wouldn't be allowed as a player can't put their own king in
             | check, and even if they were allowed, the opposing player
             | could just take the king in response.
             | 
             | And it's no longer "in passing" if it's from several
             | columns away.
        
           | Someone wrote:
           | > Also "starting position" is kind of nebulous. There are 8
           | of them. Do I get to choose which column to start the pawn
           | on?
           | 
           | I think you would have to. If the location is predictable, my
           | king would be next to the place that pawn spawns at, ready to
           | take it.
           | 
           | If you can choose the column, my king would be in one of the
           | four center fields, and still would win the race to take it,
           | but if you're smart, you place one in column A first, force
           | me to run there, and then place one in column H.
           | 
           | On the other hand, once you're on the last row, you can spawn
           | a new pawn every second move, but so, likely, can I. This
           | might be interesting to analyze.
        
           | ulkesh wrote:
           | This assessment is exactly what went through my brain when
           | reading the list. It's great if it's fun for the author and
           | their gaming partner, but I, for one, would never play this.
           | It's way too ambiguous and adds chance in a game that is
           | designed not to have chance in the game (aside from who plays
           | white).
           | 
           | I'm probably an elitist board gaming snob, but I prefer games
           | that have a lot of player agency and as little luck involved
           | as possible. Chess, Go, Diplomacy (though you may lose some
           | friends), and Scythe (which has some instances of luck, but
           | has mitigation mechanics) all come to mind. Granted most
           | popular games will have elements of luck because it can add
           | to replay-ability.
        
         | anonymoushn wrote:
         | > What does it mean to castle with both rooks? Is that a typo
         | for "either"?
         | 
         | The king moves two squares in both directions (you have two
         | kings now) and the rooks move to the other sides of the kings,
         | so you finally have KR_RK where you had __K__ before.
        
         | dev_tty01 wrote:
         | It is for fun. If any thing needs to be clarified, you and your
         | opponent make a decision and then play the game. Maybe you do
         | multiple games using different variants and see what works
         | best. I don't think the author is proposing a new game
         | standard. They are just trying to describe a variant that they
         | found enjoyable.
         | 
         | I agree that some things need to be clarified, but that's part
         | of the fun. You are correct that this does point out the
         | difficulty in converting home-brew to a formalized game.
         | However, that doesn't make it nonsense. Just think of it as a
         | good starting point.
        
         | vlovich123 wrote:
         | I agree but not for the reasons you've stated. The rules you
         | have complaints about seem to have straightforward
         | interpretations.
         | 
         | On reaching the last row it's the same promotion rules as
         | pawns.
         | 
         | Re castling what they're saying is that you can keep castling
         | with rooks provided your king has never left the home row (ie
         | the rook would have to leave to let you castle with the other
         | one). A7 implies the king has left the home row and can no
         | longer castle.
         | 
         | I do agree that there is ambiguity to clarify though:
         | 
         | * where is the castled position when the rooks aren't in their
         | starting position?
         | 
         | * what on earth does en pessant mean for kings?
        
           | xnorswap wrote:
           | Sorry, I meant b8 not a7.
        
           | zuminator wrote:
           | I thought the castling section meant, not that they could
           | castle multiple times, but that unlike normal chess they
           | could still castle after moving the king as long as it
           | remained in the home row.
           | 
           | You might very well be correct though. So I agree with the
           | previous poster that there's too much ambiguity here.
           | 
           | One thing I'd like to know is the ELO rating of the variant
           | creator. Is this a game that someone who understands standard
           | chess very well came up with and play-tested, or is it a game
           | that someone with passing familiarity with chess thought of
           | off the cuff? If the latter, the chances that the game is a
           | well-balanced variant are not so high.
           | 
           | Also, is "baring" a typo for "barring"?
           | 
           | Finally it should be noted that dice chess is already a
           | thing: https://en.m.wikipedia.org/wiki/Dice_chess
        
         | cactacea wrote:
         | > Never-mind the impossibility of Kings playing en-passant,
         | 
         | Players can make more than one move:
         | 
         | "Each player rolls the die, and the number is the amount of
         | turns each player has."
        
           | anonymoushn wrote:
           | Can you explain your interpretation of the rule so that other
           | people who think it means something can compare notes?
           | 
           | For pawns, en-pessant means "a pawn may, during the move
           | immediately following another pawn's move, capture another
           | pawn that has chosen to move two spaces in one move as if it
           | had only moved one space." I think for kings you would like
           | to say "a king may, during the sequence of many consecutive
           | moves by the same player, capture another king that
           | immediately before has made many moves in a row by attacking
           | a space it moved into in one of those moves but no longer
           | occupies."
        
       | rawling wrote:
       | Why do kings need en-passant?
        
         | clan wrote:
         | It seems like they play until you have a winner. With only
         | kings left (with normal chess rules) you would have a remis
         | (draw). Kings are simply not allowed to move into a threatened
         | position.
         | 
         | Edit: No it does not make much sense. Re-read the rules. I'll
         | be the old grump who prefers the old vanilla boring rules.
        
           | rawling wrote:
           | Yeah, but I don't see how en passant helps with this. If you
           | can get them en passant by going diagonally, you can get them
           | normally by going sideways, but they'd have just taken you
           | instead of passing in the first place...
        
         | marcofiset wrote:
         | I don't know and it's not even possible. One of the kings would
         | need to get next to the other, which is an illegal move.
        
         | jjnoakes wrote:
         | One king can walk two spaces on a turn without getting too
         | close to the other king. Then the other king might roll high
         | enough to land on a square that the first king passed through
         | on the previous turn.
        
       | blowski wrote:
       | I really like the challenge of taking a well-known set of rules
       | and adding some new ones. I'd like the author to clarify a few of
       | the questions, but the spirit of this is great.
        
       | isolli wrote:
       | This is my favorite chess variation:
       | https://en.wikipedia.org/wiki/Bughouse_chess
        
       | m_eiman wrote:
       | One of my ideas for a toy project is Schrodinger's Chess or
       | perhaps Quantum Chess:
       | 
       | when you move a piece, you can move it to any and all possible
       | squares at the same time - your piece is on all of the squares
       | you've placed it, so in your opponent's next move they can
       | eliminate your piece by attacking any of the positions it's
       | currently in.
       | 
       | In your own next move, you can move the piece from any of the
       | positions it's currently in and place it in a new set of
       | positions.
       | 
       | Eliminating or moving a piece removes it from all positions it's
       | currently in.
        
         | dfan wrote:
         | You may be interested in Quantum Chess
         | (https://quantumchess.net/how-to-play/).
        
       | j_m_b wrote:
       | To determine who goes first, you can just have one player select
       | even or odd, then both role dice. The even/odd of the sum of both
       | dice determines who goes first.
        
         | aloisdg wrote:
         | or go for the coin throw directly
        
       | atilimcetin wrote:
       | My favorite chess variant is duck chess: https://duckchess.com/
       | (and HN discussion:
       | https://news.ycombinator.com/item?id=31429290)
        
       | jstanley wrote:
       | What's the benefit of rolling a 4-sided die to decide who goes
       | first, over flipping a coin? Each player has 50% chance of going
       | first, except with a single coin flip there can't be a tie.
        
         | wang_li wrote:
         | How does one decide who flips the coin. Whoever flips is at a
         | disadvantage since the coin is flipped on a tie of die roles.
         | The outcomes of the flip are another die roll or the other
         | player goes.
         | 
         | Also quite strange that you can't quit until there are 3 pieces
         | per player left. How is that supposed to be enforced? Domestic
         | violence or something?
        
         | Karliss wrote:
         | I guess this might be due to using the same dice to decide how
         | many moves you can do in each turn.
        
         | verytrivial wrote:
         | I think they're non-binary.
        
         | tromp wrote:
         | It fits in with the overall theme of adding unnecessary
         | complexity:-?
        
         | backpackviolet wrote:
         | Uses the D4 motif that this variant is based on? If anything,
         | why have a coin flip to resolve the tie? Just re-roll the D4
         | and stay in-universe
        
           | jstanley wrote:
           | In that case you could roll a single d4 and give one player
           | the odd numbers and the other player the even numbers.
        
             | samatman wrote:
             | With the additional advantage that the original rules need
             | 2d4, but only once, at the very beginning.
        
       | yewenjie wrote:
       | After something like 1.e4 (or basically any move) if black gets 4
       | moves, they can just checkmate white and the game ends.
        
       ___________________________________________________________________
       (page generated 2022-08-15 23:02 UTC)