lichess.org
Donate

Experimental area to explore new chess variants

People in this forum often request the inclusion of some new chess variant.

As proposed some time ago in passing -- fr.lichess.org/forum/general-chess-discussion/legal-computer-assistance?page=4#34 -- I think it would be great to have an experimental area on Lichess in which new variants could be tried.

I'd like to refine the proposal with the idea that the variants should be definable interactively, without intervention of staff. At first the choice would simply be the size of the board, and the pieces, chosen from the existing set. Then some set of fairy chess pieces en.wikipedia.org/wiki/Fairy_chess_piece could be added. Then, with time, the definition of arbitrary pieces could be allowed, and further parameters and options (reintroduction of taken pieces, randomness, number of moves by turn, etc.) might be introduced based on requests.

But just the basic parameters (board size and pieces) would already be fun and cover many known variants, and the rest could come with time, when people feel like doing the effort.

Right now Lichess has only implemented a few well-known variants. But Lichess would be the perfect environment to experiment much more deeply than has ever been done before. If a suitable environment is created, fun variants would probably emerge through natural selection (some variants would be played more, forks could be done to make them even better, etc.). And of course experimenting itself would probably be fun.
I plan of creating a chess variant with the following initial drafts.

1. Played in 8x8 board
2. As much as possible will use existing normal chess pieces
3. No e.p
4. No castle
5. Increase winning possibilities
6. Something tactically dangerous, like in crazyhouse to make it interesting to play
7. Simple rules that is easy to understand
8. Something that will enhance your play in FIDE chess
9. Avoid drops if possible, if not then limit it to number of pieces or at specific squares

If there are people here who are interested in inventing a new variant we can create a team as variant designers. A variant for Lichess created by Lichess players.
#2 You might enjoy Shogi which is quite tactical and limits the squares upon which drops can occur.
@3

Does not suit well in criteria numbered (2) and (9). Also this is already existing.

I may add then.

10. Something that is not existing yet.

I think Arimaa matches all of your criteria except for 8 and 10.
Preliminary diagrams/thoughts.
* No e.p
* No double step pawn push
* No FIDE chess castling
* No drop moves as in zh

http://i.imgur.com/rbbT0VE.png
This doesn't sound like it has any value besides being just a fun thing. It also doesn't seem to target the goals of the Lichess project.

Variant "proposals" are often dropped because they are unlikely to be successful, unpracticed to be practical, or are difficult to implement (on this aspect, what people (try to) claim doesn't matter, but the realistic difficulty). None of these problems are addressed by having a new experimental area. There are also no signs that any "designed" "variants" from this area would likely lift off as a standalone variant worth tracking with a rating.

It should probably be its own web service away from Lichess. If you want to design a variant then do it by writing code, not a proposal.
@8

May I know the target goal of Lichess project that is related to chess variant?
Features should effectively deliver a goal, with aims in mind for the Lichess project, such as:

1. Helping to raise interest and provide a platform for what hobbyist chess players would be interested in. Adding Chinese chess (xianqi) or shogi, for example, would be out of scope - They are too distinct from chess. Unique board sizes, pieces and scoring, for instance, makes all variants of the kind too different. They are different games altogether.

2. Engaging chess enthusiasts with new mechanics to give unique, compelling, and (or?) sophisticated gameplay which gives the mechanic exposure rather than the variant itself. Every worthwhile variant (for example: ja.lichess.org/variant) should have [1] a mechanic, and [2] a set of gameplay patterns from said mechanic which, even if straightforward, leads to sophisticated gameplay.
Variants which are a mesh and combination of several mechanics are hard to teach and play. Variants that don't deliver the structure of a playable game won't sustain itself. Variants that aren't interesting won't be played enough to justify having them.

3. Attracting strong players or learners onto the site. Crazyhouse, for example, is in popular demand among internet chess servers. Atomic, on the other hand, also has quite a following since it was implemented on several servers. The presence of these variants help enrich the Lichess atmosphere and allow strong players to commit to using them, and for learners to have a way of trying the game out in many perspectives. Of course, not every variant is that important (and the amount of important variants are limited).

This topic has been archived and can no longer be replied to.