Comments by SamTrenholme
This .zrf file plays WOTN well, but has two noticable bugs: * The color scheme of the board is hard on my eyes. * A king can capture the other King; this makes it impossible to checkmate the person whose king was captured. I have resolved both of these issues, and will place a fixed version on my web page (http://www.samiam.org) momentarily. Here is a UNIX-style patch which resolves the king-catpures-king issue, basically remove lines with a '-' at the beginning of them and add lines with a '+' at the beginning of them: --- wotn.zrf.orig Thu Dec 26 17:24:00 2002 +++ wotn.zrf Thu Dec 26 17:30:17 2002 @@ -25,6 +25,7 @@ (define CQ-add (high-king-capture) (if (and enemy? (not-piece? Pawn) (not-piece? Squire) (not-piece? Deacon) (not-piece? Knight) (not-piece? Bishop)) (add Paladin) else add)) (define Pal-add (high-king-capture) (if (and enemy? (not-piece? Pawn) (not-piece? Squire) (not-piece? Deacon) (not-piece? Knight) (not-piece? Bishop) (not-piece? Jarl) (not-piece? Vicar)) (add Hero) else add)) (define Hero-add (high-king-capture) (if (and enemy? (not-piece? Pawn) (not-piece? Squire) (not-piece? Deacon) (not-piece? Knight) (not-piece? Bishop) (not-piece? Jarl) (not-piece? Vicar)) (multi-king) (add King) else add)) +(define King-add (high-king-capture) add) ;***** this macro lets you find out whose turn it is in the middle of a move block! ***** @@ -55,7 +56,7 @@ ;***** special movement macros ***** -(define King-shift ($1 (available) (set-attribute never-moved? false) add)) +(define King-shift ($1 (available) (set-attribute never-moved? false) (King-add))) (define O-O ( (verify never-moved?) @@ -513,4 +514,4 @@ (piece (name Three-Kings) ) -) \ No newline at end of file +)
My own take on this idea is to allow one to drop pieces anyone on their half until they capture an opponent's piece or give check to the opponent's king. Once they catpure a piece, or give check, the player who has done so can no longer drop pieces.
- SamPues, me parece que neceitamos encontrar alguien que tiene ambos el tiempo y la voluntad para traduccir este al español correctamente. Desgrasiadamente, mi español no basta para hacerlo. -- Looks like we need someone with the time and willingness to translate this to Spanish correctly. Alas, my Spanish is not good enough to do this. - Sam
I do not think all of the pawns need to be defended in the opening array to make for a good game of chess. For example, I think ANBQKBNM would work as an opening array, even though the B pawn in undefended. The problem with Fischer random chess is that some arrays favor White unfairly, and some will be too drawish.
I think one way we may be able to test the quality of variants is with computer analysis. So, if we have a shuffle variant with, say, 100,000 possible permutations, we can have the computer play itself all of those permutations, In order to find one that is playable. Of course, computers play Chess differently than humans, so a computer's idea of playability may be different than a human's. And, computers are generally programmed to find the best move in a given position, so translating that to playest a game may be tricky. Greg Strong is much better qualified to say whether this is feasable or not. :-) - Sam
Since this comment is for a page that has not been published yet, you must be signed in to read it.
Right now, Vladimir Kramnik is the world chess champion. Ever since Kasparov left FIDE, FIDE has not been considered the determiner of the world chess champion. Of course, it is only a matter of time before a computer becomes the world champion.
Some clarifications: * Fischer is no longer in jail. He is in Iceland. * The Washington Post has an editor who doesn't seem to like this fact.
Hey, Greg, I just asked the editors to delete the image of ChessV playing Gothic Chess and removed the Gothic Chess mention from the ChessV page there, since you do not want it. If other editors try to restore this stuff; well, I've been in edit wars before.
My main concern, just looking at this, is the color imbalance; one player controls the white squares; the other player controls the black squares. It is a maxim in traditional FIDE chess that bishops of opposite colors are draws; I am concerned that this game may be drawish. I haven't tested it, of course. - Sam
This game looks to be very strategic; I would venture to say that this game has about the same strategy/tactics balance as FIDE Chess, while being richer in both aspects. The main disadvantage appears to be that this variant will probably result in longer games than FIDE chess; a blitz game is probably game/10 or game/15 instead of game/5; a tournament game would probably take four to six hours instead of two hours. I especially like Greg Strong's method of coming up with this opening setup. Perhaps a similar heuristic can be designed so that a random chess variant makes for a playable game (the pieces are chosen randomly or semi-randomly, then the opening setup is chosen at random until we find one with a good balance).
I have a black and white version of the board here: http://www.samiam.org/new-ivorytower9x10-bw.bmp - Sam
The zip file is not available here. Look here for this excellent game:
ftp://ftp.zillionsofgames.com/games/IvoryTower.zip
- Sam
P.S. How do I change my password?
I must confess my favorite lines in ortho-chess are the somewhat dubious gambits--the Evans, the Latvian, the Fried Liver Attack, to name three examples. Anyway, I wish your newsletter the best of luck.
The problem with the Dunst Opening is 1... d5 followed by 2... d4, forcing white to move a piece twice. The problem with 1. Nc3 d5 2. d4 is that White now can not do the Queen's gambit. These disadvantages may be offset by getting black out of his book, and playing a line where one knows the traps far better than the other player.
As a chess variants enthuiast, I also see the appeal of having an offbeat opening result in something that doesn't quite feel like chess. Ralph Betza once pointed out that if you want FIDE (modern western) chess to feel like a chess variant, play the Boden-Kieseritzky gambit (1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Nc3 Nxe4 5.O-O).
The way I see it, anything that is reasonable in the first ten moves of chess has already been analyzed and studied by someone. FIDE Chess is a very well-trodden area, where interesting novelties are hard to find. However, there are literally an infinite number of chess variants (Chess is probably more malleable than any other abstract game) and very few of the variants have been explored at all. As just one tiny example, the variant that I just published (shameless plug) shares many pattens with FIDE chess openings, but is a almost completely uncharted territory for people looking for new ground to explore.
Just my two cents.
- Sam
I notice that the Chess variant pages are also inheriting another problem of Usenet: Poor variants (as in, with bad names, or with poorly written rules) are getting a lot more comments and attention than good variants (cough, cough shameless plug cough, cough).
I find it ironic that, seven years after this variant was proposed, people finally make a big fuss over its name. This is not the only Grand Chess variant, and won't be the last Grand Chess variant. Unlike 'Grand Chess 2', which implies the variant was invented by the same person who invented the original game, 'Grander Chess' is more clearly seen to be something invented (or fine-tuned) by someone else.
The whole 'protect all of the pawns' business comes from two things:
- Sam |
Gary, I think I am going to have to invite you to play an informal (FIDE) Chess game with me, with you taking the white pieces. I've never played a Game Courier game before, so this will be new for me.
Two questions: Can we use books and computer listings of other games using a given opening (I presume yes)? Can we use computers to help us with the tactics (I presume no)? Why do I get the feeling this game will start off with 1. Nc3? You're the first person I have known to like the Dunst; I have always read that it is weak. I researched it a little since Zillions likes opening with the Dunst (Zillions really likes moving the knights out early when playing FIDE chess). Now I have to figure out how to get a Game Courier account. - Sam |
Where do these beautiful graphics come from? I've never seen the 'Galatic' graphics before. Are these graphics copyrighted? What license are the graphics released under?
Thanks for your time. - Sam |
My impression: This game will probably need a 'bare amazon' rule (if all pieces except the amazon are taken, the person with just the amazon loses).
I am also very leery of games where the royal piece becomes very powerful; my worry is that this will result in more drawish games, since it is not practical to mount an attack against the opponent's royal piece. Strategys like sacrificing a piece to get a strong attack against the opponent's king (think FIDE Chess world champions Morphy and Tal) just don't work because the amazon can too easily retreat. - Sam |
The idea of having two royal pieces, where checkmating either piece wins (or forking both pieces with a defended piece), I think makes for a game with more tension than FIDE Chess. I especially like one of the royal pieces being very powerful; an Amazon usually is too powerful, but making it royal is a brilliant way of restraining its power.
- Sam |
I like this idea because I like the idea of having a chess variant template which makes for a huge number of playable games; I'm not just talking about the 960 games of Fischer Random Chess or the 252,000 possible games using a 8x10 Carrera setup where the bishops are on opposite colors; I'm talking a Chess variant that allows a number of games with a number like 6,670,903,752,021,072,936,960 (the number of possible Sudoku solutions).
One idea: Each pawn can be one of nine different pawn types:
For the pieces, any of the pieces, except the king, can have any of the 15 combinations of rook, knight, bishop, and camel movements. The king exists in three forms: Can move as a ferz, can move as a wazir, and can move as a FIDE chess king. For an 8 * 8 board, this results in 512,578,125 possible setups; combine this with the pawns above and our 8x8 board now has 22,064,807,537,578,125 possible opening setups. The corresponding 8x10/10x10 board has 402,131,117,372,361,328,125 possible opening setups. Now we're starting to get what looks like a variant template with a decent number of possible starting setups. :) As a practical matter, this template for the pieces probably usually results in arrays where white has a considerable advantage because there is so much force on the board, but this is a thought experiment, not a practical Chess variant design. This might work a little better: Make the atoms Betza's crab (leaps from e4 to d6, f6, c3, and g3), a fers, a wazir, and a camel. But that probably makes most setups too weak. Perhaps if we add a randomizing factor with these weak atoms whick randomly strengthens one of the atoms (makes the ferz atom a bishop atom, a wazir a rook, a crab a knight, and a camel a camel + dabbah). This causes each piece to have one of 32 possible forms; for an 8x8 board this results in a grand total of 4,437,222,213,480,873,984 possible setups; for a 10x8 or 10x10 board, this results in 368,040,959,274,957,611,728,896 possible setups. |
- Sam
Is it OK if I add this fool's mate to the Schoolbook description page, using the words 'David Paulowich found the following 4-move fool's mate that mates with a bishop: 1. e4 g6 2. f4 Kg7? 3. Bf2 Kh6?? 4. Bi5#'?
Quite clever. Note that, since the pawns are more defended in the Schoolbook opening setup than they are in FIDE Chess, we need to pull out the King before we can mate him with the bishop. This stronger defense does not, IMHO, make the game more drawish--having a lot more force on the board more than compensates. - Sam |
As an afterword, the Wikipedia editor (-Ril-) that tried to delete the Zillions of Games article, and who was so rude to people from the ChessVariants server voting to save ZoG, is now being held accountable for being a disruptive editor. |
I can see why it is that you consider the Schoolbook opening setup interesting (note that the Schoolbook setup actually predates this comment, if barely). I also consider something you thought up of on your own a very interesting opening setup for 8x10 chess setups.
I think the original carrera setup for everything besides the very powerful pieces is the best 8x10 setup; the Grand Chess world champion seems to agree.
- Sam
h7 looks the be the weakest pawn in this setup (the knight-side pawns are a good deal weaker than the bishop-side pawns). The corresponding fool's mate that takes advantage of this weakness is 1. c4 Ak6?? 2. Axh7#. With two archbishops, which make for rather powerful midgame attackers, this will be be more intensely tactical than the various 8x10 chess setups out there.
I'm a bit worried that white may have an overwhelming advantage here; basically, white can use the bishop-moving pieces to attack the right hand pawns with moves like c4, d4, and e4 to put pressure on the right hand side and make it very difficult for black to develop his pieces. There may be a way for black to sucessfully defend his knight side, but this will make it much harder for black to equalize, since black needs to deal with the issue of defending his knight side instead of engaging in normal development. Note that I haven't actually playtested this game. - Sam |
I see two problems:
This will make the game more tactical and have more flare than standard FIDE chess. |
Here is what I would do to lower the number of draws:
- Archers, in the interest of minimizing friendly fire, will not fire (capture) when on the friendly side of the river.
- Archers can leap on to or off of the Troy horse any time, as part of their normal move.
- Bridge builders can not capture nor be catpured. Any piece, friendly or enemey, can slide through the bridge builder as if the bridge builder was not there; it is illegal, however, to land in the square that the bridge builder occupies.
- A bridge builder can not destroy a bridge that is adjacent to the opponent's bridge builder.
- Bridge builders move like chess queens.
A lot of draws often times indicates that a game is unbalanced; basically, a weak player can force a draw against a strong player. As it turns out, it's actually harder to fix a drawish game than it is to fix a game where white always win; a 'white always wins' game can usually be fixed with the pie rule; a drawn game needs to be fixed by changing the game to be less drawn (usually by making attack stronger and defense weaker) [1].
Here is some empirical evidence:
In order to make sure this is an apples-to-apples comparison: I have included two other games from the same game server that the Catapults games were played on. I have also included statistics from a real-time server, BrainKing, since this server has a large number of games, and since you mentioned that correspondence games will have more draws. I'm sure you won't do this, but if you ever change your mind and incorporate my ideas, you can still keep the same copyright on Catapults of Troy (then again, you can't copyright ideas, only artistic expression). :-p - Sam [1] Chess-like games can usually be made less drawish by adding Shogi drops to the game. |
Over three years ago, I pointed out that this game is probably too drawish, based on looking at the game mechanics (the king is too hard to checkmate). I'm not surprised that both games of Glinski's Hexagonal Chess played on the game courier server ended in draws.
My Russian isn't good enough to look at all of the games played in tournaments on the Russian page, so I can't get a sense of the draw percentage there, but I will bet you it's a lot bigger than the corresponding FIDE Chess draw percentage. Here are three ways to make this game less drawish:
|
Looks like you're right.
Here is how I got to see all six games:
- Sam |
Since I'm being critical of other people's variants, it's only fair that I be a little critical of my own variant.
In my playtesting, the biggest problem I have found is that it is too easy to swap off the marshalls. What usually happens is that the kingside bishop moves off of the G file near the beginning of the game. This causes the H pawn to be undefended. Next, one plays Mh3/Mh6 threatening the h2/h7 pawn; the only reasonable defense to this is for the other player to move out their own marshall. If they move the marshall to the I file, this results in less marshall mobility, so the best move is to have the two marshalls face each other. This is usually followed by the two marshalls being exchanged. This particular motif makes it so marshalls frequently do not make it to the endgame. One possible solution to this problem is to have the opening setup be one where black's pieces are reflected (the marshall on b1, king on e1, archbishop on f1, and queen on i1) relative to the white pieces. Another idea is to swap the marshall and the kingside rook in the opening setup. - Sam |
Hey, I wasn't planning on inventing another variant for 2006 (I prefer quality over quantity), I have some ideas for a wargame variant inspired by Catapults:
|
I agree that six games is not an exhaustive sample. However, it's the best data I have available. Now, I don't have enough of a background in statistics to give you a error percentage with these figures; however, it may be telling that we have played 31 FIDE Chess games on Game Courier and not one of them ended in a draw.
Then again, according to this Russian page (translated via the fish here) King + Rook vs. King (and King + Queen vs. King) is a mate against a bare King. I wish I had more game results to look at to see just how drawish this game is; the results over at Game Courier don't make this game look very good. Perhaps Glinski's Hex Chess with Shogi drops? - Sam |
Of course, we need a Mormon (non-drinking) version of this game. How about dice chess; roll a dice to determine what you move. One means pawn, two means knight, three means bishop, four means rook, five means queen, and six means king. Goal is capture of the opponent's king. Castling can be done if either a king or rook is rolled. If you can not move any piece, re-roll.
This idea is hardly original; indeed, here is an ancient dice version of Chess. I have heard it said that Shantraj was once a dice game, and the reason Shantraj stopped being a dice game and became an abstract game is because it was against Muslim law to gamble. - Sam |
The Wikipedia article for chess variants, at one point said 'The broad definition of chess variants is so universal, it may include nearly any abstract battle game or war game'. That said, at which point does a variant become a wargame? Good question. The answer depends on who you ask.
I think we're crossing the line once we add dice rolls. I also think that line is crossed when the opening setup is asymmetrical or when there is hidden information ('fog of war'). The reason for having one move all of their pieces (in any order they choose) in a single turn is to make the game impervious to traditional computer analysis. - Sam |
I think Pre-Chess (a variant where each player places a piece anywhere on the first rank until the first rank is full before moving pieces noramlly) is a very interesting variant. This variant is mentioned, as I recall, in New Rules for Classic Games, but only exists as a Java applet on this server with no discussion of the variant. As I recall (I got rid of this book years ago when cleaning out storage), Pre-Chess was mentioned in a 1970s issue of Chess Life magazine.
The Carrera chess variants is a crowded field, but with good reason: This setup makes for a very nice game. I think I mentioned this before, but the big reason there are so many Carrera variants is because there were some serious problems with one of the more famous Carrera opening setups. I said this before, but it's interesting that what you settled on as being the best 10x8 setup is very close to my own Schoolbook setup (which I came up with in the summer of 2004). Indeed, your placment of the rooks may make for a better game.
- Sam
I just want to let you know that I liked the ideas presented in Shangai Palace, and that the game looked like a fun one to play. I think it is a shame that Zillions plays it so dang poorly, and that people were so hard on it. I really like the idea of a game where some pieces can drop and other pieces can not be dropped. Perhaps we can have a tame chessgi where only rooks and bishops can be dropped. Of course, it's very trickly to compare the value of a piece that can be dropped with a piece that can not be dropped.
Another idea: Have it so that a piece can only be dropped with certain pieces capture the piece in question. I think I have some interesting ideas in 'Crossing the Rubicon', but I think I want to get several dozen mating positions for schoolbook, and a computer program that can play it better than Zillions--my current plan is to take the guts out of ChessV and make it a Schoolbook playing engine, and hack up Winboard and Xboard to play Schoolbook instead of FIDE Chess (I once hacked xboard to play Grand Chess, so I know this can be done). Once that is done, maybe I'll consider a variant besides Schoolbook. There are a zillion variants out there; just not enough what I would consider fully-developed variants. It's a shame there isn't as much interest in board games as there used to be in Ohio. I think this is because people are playing online games instead. I never went for online games; if I am playing someone at Chess and losing, I want to be 100% certain that my opponent is not cheating so I can feel that he is winning fairly. I can never do that on the internet. I think an essential human element is lost when people interact with computer screens instead of real live people. - Sam |
First of all, I like the name of the variant. Second of all, I like the name Unicorn. These pieces have traditionally been named Camels, as you point out. A unicorn is also David Paulowich's monkier for the bishop + knightrider piece. Since the Bishop + Knightrider is been called a 'Paladin' in the Way of the Knight ZRF file, I prefer the name 'Unicorn' for the strong but colorbound Camel + Bishop piece.
I also like having the knights and bishops closer to the center. How about: p p p p p p p p p p r c n b q k b n c r - - a - - - - m - -Where 'c' is the camel (what you called the 'unicorn' in this variant), 'a' is the archbishop (knight + bishop), and 'm' is the Marshall (rook + knight). - Sam |
I'm working on building a list of every single Zillions file that fell through the cracks when the big move happened.
- Sam
I have determined that the following 31 zrf files, which were once on the server (and, indeed, still have index pages pointing to a now-non-existant .zip file), are no longer here:
- Sam |
Would it be helpful for me to whip up a Zillions rule file for this game (and all of the variants we have mentioned) to get a feel for how it plays?
- Sam |
I have, via a combination of going through my own backups and getting a few files from the Web archive, managed to recover all but four of the missing Zillions' games which were lost during the big move last year.
The four games that are still missing are: I have added a link to the recovered file in the comments section for all 26 other missing files.If anyone has copies of these still-missing files, let me know by replying to this comment. - Sam |
Larry,
There's a pretty good chance that some chessvariants.org editor has this file somewhere on a backup tape. It's also possible some other reader of the site has a download on their hard disk. I'd wait a while before resorting to re-writing the .zrf file. - Sam |
I only rate a game poor if the game's rules do not adequately describe how to play the game in question. Sometimes, someone has a half-baked idea for a variant, and they don't take the time or energy to even fully describe the variant, much less make a Zillions ZRF or Game Courier preset for the variant. If a game is fully fleshed out, or has a Zillions rule file, or even a game courier preset, I will plain simply not rate the game poor.
If I think there is an issue with the mechanics of a fully described game, I will discuss my issue in an unrated game comment. This gives me an opportunity to discuss whether my concern is really a legitimate concern with other chessvariant.org editors and the game's inventor. I will usually suggest how I would fix the game when bringing up the concern. As just one example, I feel Hex chess may have problems. The consensus, however, is that these are probably not legitimate concerns. Speaking of Game Courier, could someone please index my game courier preset for my take on Carrera chess. - Sam |
Here is some information I have on the value of the Carrera pieces which I came up with when designing my own Carrera-esque opening setup:
Here is a table of four different derived values for the pieces, obtained from three different chess variant playing computer programs and one other source.- Sam
Piece ChessV SMIRF Zillions Aberg Pawn 1.000 1.000 1.000 1.000 Knight 2.500 3.056 2.362 3.000 Bishop 3.250 3.690 2.859 3.300 Rook 4.700 5.604 4.262 5.000 Queen 8.750 9.558 7.060 9.000 Archbishop 6.500 6.838 5.127 6.800 Marshall 8.250 8.832 6.659 8.700 The ChessV numbers were obtained by looking at the source code for ChessV. The SMIRF values, derived by Reinhard Scharnagl for his SMIRF chess computer program, were obtained from this web page. The Zillions of Games' values were obtained by looking at the values of pieces by right-clicking on them after loading a fresh Schoolbook zrf file, and before moving any pieces. Aberg's figures come from right here on the variants server.
All four agree on the following:
The verdict is still out on some other exchanges:
- A bishop is about a half-pawn more valuable than a knight.
- Two knights are worth more than a rook.
- An archbishop is worth more than two knights.
- A marshall is worth more than an archbishop.
- A queen is worth more than a marshall.
- Two rooks are worth more than a queen.
- A marshall is worth more than a rook and knight.
- A marshall is worth more than two bishops.
- A rook and knight are worth more than an archbishop.
- A rook and bishop vs. a queen.
- A rook and bishop vs. a marshall.
- A bishop and knight vs. an archbishop.
- Two bishops vs. an archbishop (Two bishops are probably worth more).
100 comments displayed
Permalink to the exact comments currently displayed.