Comments/Ratings for a Single Item

The pages for Kyoto Shogi and Paco Shako have no information on online play or any record of past games.

Kyoto Shogi has recorded history of play at least here.

The Jocly page on Cavalier Chess does not have that... Or Grand Cavalier Chess. Or Grotesque Chess. Or Univers Chess

These do all have a written description of the rules, which Tenjiku Shogi does not have. They also have fewer pieces, and they use a more familiar piece design that makes it easier to tell what is what. With a bunch of unfamiliar pieces in Kanji, a western player will normally need an illustrated piece-by-piece guide to the pieces.

My iPad is not letting me copy the rest. But one of the ways of studying the rules of a game is to consult a cheat sheet while playing a computer opponent like the one on Jocly.

The problem is that a written description of the rules for Tenjiku Shogi would be so long that it would only be accessible through extensive scrolling. And it that case having it on the same page is worse than having it on a separate page to which you can easily switch back and forth.

Mind you, I don't want to argue that the Jocly presentation of Tenjiku Shogi cannot be improved. But it is just not in the same class as 8x8 variants with 6 piece types. It has 37 piece types, many with unusual moves (double captures and such). To be of any use an aid for the uninitiated should be a lot more advanced than a conventional cheat sheet.

Western players would probably best play with a pictogram representation rather than kanji. (And the 2d representation in Jocly will have to be fixed to provide that opportunity.) And the Tenjiku article should have an Interactive Diagram that can be switched between several piece representations, including a 2-kanji representation. And perhaps there should be a Jocly 3d representation with mnemonic pieces.

The problem is that a written description of the rules for Tenjiku Shogi would be so long that it would only be accessible through extensive scrolling.

The Game Courier preset has an illustrated piece-by-piece guide that spans four screens on my HD monitor. But it wastes a lot of space using only two columns when many piece descriptions are short. Using flex or grid, it could fit the information into more columns. In fact, the first screen is mainly pieces that don't promote, and these could be handled more compactly as a group of pieces without promoted versions.

Western players would probably best play with a pictogram representation rather than kanji.

Certainly. I normally play Shogi with the Motif or Symbolic pieces, which use pictographic images from sets for Chess on Shogi wedges. Large Shogi variants like this one would benefit from a set with Alfaerie images on wedges.

This leaves Ultima, which has now been seconded, as the most eligible game with a nomination and a second. So, I will feature Ultima for January, 2024.

If it’s featured, it should have a banner image:

Opulent Lemurian Shatranj has only been programmed as an interactive diagram. It has not been programmed for online correspondence play.

I found the Game Courier preset for OLS here, by the way, with logs for past games here.

Opulent Lemurian Shatranj has only been programmed as an interactive diagram. It has not been programmed for online correspondence play.

I found the Game Courier preset for OLS here, by the way, with logs for past games here.

I didn't say there wasn't a preset for it. I said it hasn't been programmed for online correspondence play. This preset is unprogrammed.

Among the games that have been seconded, none qualify for being featured. Would someone like to nominate or second some qualified games?

If Symmetric Chess qualifies, I'll nominate that CV.

If Symmetric Chess qualifies, I'll nominate that CV.

It does, and I've added your nomination to the page.

In the meantime, I've been considering Marseillais Chess. I fixed up the sample games to work with Game Courier's new JavaScript-aided way of navigating through the moves of a game, and I think these would be good for drawing attention to what Game Courier can do now.

I'll wait until morning in case our European members, who may be asleep now, want to add some nominations or seconds.

Maybe Kyoto Shogi?

Maybe Kyoto Shogi?

You already seconded it, and I did already say that none of the seconded games qualify. Its page is lacking anything about being programmed for online play, having a game history available, or even just playing it online.

I'll second Marseillais Chess and feature it for February, 2024.

I did already say that none of the seconded games qualify.

Does anyone want to second anything before the month is over? Of the games that have been nominated and seconded, each has problems.

@Fergus,

Perhaps some improvements, even minor ones, to the criteria could be made and/or are in order, if for no other reason than the fact that you said we were running out of eligible variants to be featured when you featured Hectochess.

One improvement we could easily make (which I vehemently recommend) is, for the computer program, replacing the preference for Zillions of Games with a preference for a free-to-play program.

SInce Zillions is a pay-to-play program (meaning it costs money), and you already have it and are accustomed to it, it is safe to say that the Zillions of Games requirement is a personal preference on your part. Unless you are a connoisseur in the types of games that Zillions is made for, you won't be likely to pick it up, especially with the price tag. It makes no sense to have a preference for a pay-to-play program for making to-be-featured Chess Variants when there are plenty of free-to-play options for doing the exact same thing, especially with the world in its current state.

If you want something with rule enforcement options, there are several options, such as ChessV, Ai Ai, Jocly, or Ludii. I would highly recommend listing H. G. Muller's Interactive Diagrams as well, as it is extremely flexible, and has become at least somewhat well-known and adopted by the site's contributors (for me this was a godsend). Each has its own advantages and drawbacks, but these would likely better serve as examples while having the preference as "a free-to-play program" in general.

Indeed, availability for Zillions means nothing to me. I don't have it, and I don't intend to buy it.

It must be said that alternatives like Ai Ai, the Interactive Diagram or Jocly are not on par with it regarding playing strength. The ID and Jocly are written in JavaScript, which is much slower than naturally compiled languages like C. Both evolved from projects that primarily focused on the user interface, and don't use the most efficient code even by JavaScript standards. For the ID I did not care, as it was never intended to be of super-human strength. And especially the Jocly AI uses a very poor engine design for Chess (probably as a compromise to also work for entirely different games). Ludii I know absolutely nothing about.

ChessV is a strong engine, and in general stronger than Zillions. But it is not as versatile. The configurable multi-variant engines for WinBoard (Fairy-Max, Sjaak II, Nebiyu) are also much stronger than Zillions, and although perhaps somewhat more versatile than ChessV not very easy to configure.

Perhaps I should start a project for rewriting the AI of the Interactive Diagram in C, so it could be used as a strong engine in one of the available user interfaces (WinBoard or ChessV). It would be interesting to have Greg Strong's opinion on this.

@Fergus: I second Symmetric Chess, I like its Bishop's conversion rule.

Zillions-of-Games is mentioned as a minimal way of meeting this requirement, because it is the most convenient option for someone to get his game programmed for play against a fairly strong computer opponent. For a small cost, which is currently whatever you want to donate, because it is no longer being sold commercially, you can get a key that unlocks the free Zillion-of-Games demo. With the unlocked demo, someone with some programming skill can program his own game. However, I did mention that it is not the only option. So if a game is programmed for ChessV or Fairy-Max, that’s going to count too. It’s just that for these programs you’re more at the mercy of the developer and can’t so easily do it yourself. It’s also going to count if you do a program in some more general-purpose language like C or C#. For most people though, this option will be more difficult or require the much greater cost of hiring a programmer to do it for you. So I mentioned the easiest option first but did not rule out other options.

I never implemented anything in Zillions, but I had the impression that it is orders of magnitude more difficult than configuring Nebiyu or Sjaak II for games these can handle. You have basically to write your own move generator for the involved pieces in some cumbersome, LISP-like programming language, rather than just specify a FEN and a list of participating pieces as in Sjaak or Nebiyu.

As the developer of Fairy-Max, you do have the luxury of not needing to care about Zillion-of-Games, because you can program your games without it. For most of us, though, Zillion-of-Games is the best DIY option we have for getting our games programmed.

Through all this, I'm still hoping to figure out what not-for-pay software would be best suited for Short Sliders (and the Leapers Who Love Them). (Preferably but not necessarilyialso able to handle some of my 3D games.)

I have no familiarity with Nebiyu or Sjaak II. If all you do for these is specify a FEN, they are much less capable than Zillions-of-Games. Zillions-of-Games is very easy to program for new games, as the work of programming a game engine has already been done by the developers, and all that's left for someone wanting to program his own game is to specify the setup, how pieces move, and any additional rules. This is why we have close to 1000 Chess variants programmed for Zillions-of-Games here, and the Zillions-of-Games website has many more games we don't have here.

You need to specify more than just a FEN, and youcan specify moves. And they are certainly less capable than ZoG. But I think they would not have much difficulty implementing the variants that we typically feature hee. (I could be wrong, though, as I haven't configured them for any games either; just looked at how the config file for some of the preconfigured games looks.

But there is room for a serious engine that is as versatile as the Interactive Diagram, and as easily configurable. In fact, the Play-Test Applet could be used as a front-end for compiling the XBetza, and producing a move table similar to the one it already produces as GAME code. Which you could then copy into a config file with the name of the variant for use by the engine. (In fact it would be rather trivial to make the negine directly read the GAME code that now goes into the Pre-Game section from such a file; it is little more than a large sequence of numbers separated by whitespace.)

The hardest part would be to equip it with a sufficiently powerful user interface.

I have updated the text of this page to make it clearer that Game Courier and Zillions-of-Games are being offered as the easiest options, and not as the only options, for meeting some of the requirements.

@Fergus: I have also seconded Symmetric Chess.