Comments by SamTrenholme

I think the best way to come up with reasonable piece values is to have a computer program play itself hundreds or thousands of games of a given chess variant, and use genetic selection (evolution) to choose the version of the program with piece values that win the most games.

You can even have mating (Sex! Can we say that on the Chess Variants site?): Two sets of known piece values can mate and the resulting piece values will be an average of the piece values of the two 'mates' (with some randomization; the 'child''s piece values will be a random mix of the two 'parents' piece values).

I could do it myself, but I need a chess variant engine that I can set, from the command line, white's and black's values of the pieces independently, and then have the variant play itself a game of the chess variant.

- Sam

Cute, very cute

Game courier is broken: It is no longer possible to resign from games

My thought, as someone who has only seriously 'invented' one chess variant (which is only a minor variation of a 400-year-old variant), is that I don't like proliferation that much.

I think the joy of inventing a Chess variant is the joy of being able to develop opening, midgame, and endgame theory for the new game and new rules.

This is why I only have invented a single chess variant, but I made it one I extensively tested using Zillions before making public, one where I developed some opening theory, and one that I spent hours having the computer play against itself in computer-vs-computer games (usually two different programs playing each other) to creating interesting mating positions.

I personally prefer quality over quantity; 90% of everything is crud, but I think it's better to make just a single variant where it's fully fleshed out: The game includes a game courier preset, a zillion's implementation, in addition to a clearly written description of the rules. Ideally, the game should have some theory established, such as the value of the pieces in the variant, some opening theory developed, and even some mating problems.

Wow, this guy has invented a lot of variants! I guess he's in a contest with Mats Winther, Ralph Betza, and Charles Gilman to invent the most variants.

I'm the opposite. I prefer quality over quantity (see the recent proliferation thread for my viewpoint on all this).

- Sam

I think Mr. Winther puts more effort in to designing the games than just the pieces. Different games use different boards and different ways of adding the pieces to the standard 'FIDE' arrangement; I remember him saying that he gives the opening arrangements some thought for a given set of new pieces he creates.

In addition to pieces, he has also brought the Gustav board back, something I never heard of until seeing it mentioned on his page. I think the Gustav board is a good way of introducing new pieces to FIDE chess without having the new board affecting the game too much, and without somewhat clunky ideas as gating (Gating makes sense when you want the game to be just as much like FIDE chess with new pieces as possible; but the Gustav board is more intuitive and makes for simpler rules).

One can argue 'Why design games that no one plays', just as one can argue 'why analyze games no one plays', and in both cases the answer is the same: Because it can be an enjoyable way to pass the time. If the act of creating a game brings pleasure to someone, it doesn't matter if that game is never played by anyone.

You know, in order to make the politically correct police happy, instead of calling that piece a 'Swastika', we can call it a 'Flywheel'.

Interesting thought. Lets take the 'Swastika'/'Flywheel' and 'Shuriken', and remove the ferz/wazir move from the piece. Now we have this, which I will call the 'Spinner':

. X . . . 
. . . . X 
. . Z . . 
X . . . . 
. . . X . 

This piece, as it turns out, is 5-way colorbound ('Colorbound' is a Betza-ism that means 'this piece can not reach all of the squares on the board'); each side needs five of these pieces to reach the entire board. Now, since the colorboundness is somewhat unusual, if you add another unrelated colorbound move, such as the move of a ferz (Our 'Flywheel'), the piece is no longer colorbound, but can reach every square on the board. Heck, if you add the pawn move to this piece (The piece can move, but not capture, one square straight ahead), the piece is no longer colorbound.

Just glancing at Axiom real briefly, one issue is that it uses a stack-based language (Forth). I don't think it makes much sense to use a stack based language these days except for ultra-embedded applications like robotics and what not. I have never really tried learning a stack based language; I mean I do get reverse polish notation, but it just doesn't feel right to me.

I think the issue is that a stack-based language is what a linguist calls an 'Object Subject Verb' (OSV) language, which are extremely rare in human languages. Languages are usually 'Subject Verb Object' (SVO) languages (English; Spanish; 'a = 2' in programming languages), which is what most programming languages use. Function calls emulate the form of 'Verb Subject Object' (VSO, such as Irish; 'f(a,2)' in programming languages) languages, however, as it turns out.

So, yeah, I think Axiom might get more users if it used a more common language than Forth.

- Sam

Edit: Looking at it a little more, it's nice to finally see a Zillions implementation of Tanbo, which Axiom made possible. I don't see any Chess variants, however.

I like this variant; it's a classic 'How close can we be to normal FIDE chess while throwing out the opening book?' variant. Another variant along the same lines that Mr. Winther has proposed is chess 256, where we randomly move a pawn up one square for the eight squares on both sides of the board.

Another idea I like along the same lines is to keep black pieces where they are, but swap White's queen and king. Now, in White's OO move, the king moves right two squares (from the d file to the f file) and the h rook moves left three squares, and in White's OOO move, the king moves left two square (from the d file to the b file) and White's a rook moves right 2 squares to the c file.

Any other ideas for modest chess variants that keep the game the same as much as possible, while throwing out the FIDE opening book?

Mr. Smith:

How would you change the opening arrangement to address these issues you brought up?

- Sam

Mr. Smith:

Do you have Zillions? If you do, you can see in the Zillions file alternate opening setups with pieces, as I recall, akin to the 'Marshall' and 'Cardinal' in this game.

- Sam

I love it when people break the mold and come up with an alternate tessellation for a chess variant (such as Parachess).

Speaking of which, is there any interest in my inventing a variant using an alternate tessellation. I have an idea that has been bouncing around my head for over a decade which I should make a variant out of, but only if people would be interested in looking at it.

- Sam

OK, I think many people look at variants with different goals in mind. Some people have dreams and fantasies of becoming multimillionaires from inventing a variant; this is a ridiculous fantasy. A chess variant inventor has less change of making money from their variant than a conlang creator has of making money from their language.

Other people enjoy inventing new pieces and making a variant based on those pieces. Betza enjoyed this; he also enjoyed finding a mix of pieces just as strong as the FIDE pieces so one could have balanced games with different Chess armies. Other people enjoy combining themes of various variants to create something using a new theme.

For me, I like a variant where we quickly get out of the opening book and in to the 'street fighting' of trying to do tactics better than your opponent. I also like opening analysis of a variant, for the sake of opening analysis (not that said analysis is useful; then again opening analysis was not really useful in FIDE chess until the 20th century).

This is why I like Capa/Grand Chess variants; with two more pieces almost as powerful as the queen on the board, the games get very tactical very quick. Just like 'mad queen' chess before people discovered boring defenses like the Sicilian defense.

And, there are a lot of Capa opening setups one can choose from making it so there is never a chance of the opening getting stale. But that doesn't stop me from having done some opening analysis of my particular Capa openeing setup.

So, I generally don't invent variants because I find more joy in playing and studying variants already invented, and because there are already a lot of possibilities, even with the modest Capa variants.

- Sam

General comment for Mr. Smith's prolific contributions: I think it would be a good idea to do more than submit an idea for a game here. Non-ASCII graphics would be nice, as would be Zillions implementations and game courier presets.

I'd also like to see some opening theory and mating problems, but I think I'm the only chess variant inventor who has bothered developing a chess variant that much.

Of course, my comments on the proliferation thread show that I'm definitely not a proliferator.

- Sam

The Chiral rook is a very creative piece, and a very original and creative idea.