Why you SHOULDN'T post new variant suggestions + More endgame considerations on Chaturanga

Hello once again.

Last time I exchanged a couple of words with the mods of this forum. It was about Chaturanga and why we need to adjust the ruleset to make the variant enjoyable and reveal its full potential. While I'll be talking about that as well, I want to voice an opinion:

WHY ARE SO MANY PEOPLE OBSESSED WITH POSTING NEW VARIANTS?

Seriously. Why?
Because they look cool? They're wacky and epic? Maybe.
The problem is twofold though:

1) Most of the suggested variants are BAD. Really bad. A variant should be interesting, offering new ways of playing that aren't found elsewhere, and balanced. Most ideas that are posted here are about very weird variants that would be difficult to program and, most importantly, haven't been sufficiently playtested. Who wants to play an overly complicated Frankenstein variant that, in the end, is almost always a win for White/Black?

Controversial gray-areas would be Atomic (a variant with a pretty big initial advantage for White) and Antichess (a mathematically proven win for White). The reason I say this is that these two have resisted the test of time and have a small-medium playerbase who play consistently, proving that the variants are playable in the long run and remain interesting. Even if the two are extremely unbalanced, there are practical chances for a counter-attack (Atomic) or a very complex net of sidelines that make a win hard to attain from a human perspective (Antichess).

I personally believe that it should be Chess.com's call on whether to add these two. Maybe they want to have on their site only fairly balanced variants. Or maybe they don't mind having a couple of unbalanced but classic variants. I'd respect either decision, really.

But a "Fog of War Atomic Seirawan Gravity Chess for Three Players" or something like that probably won't be balanced. The current variants available on Chess.com were tested by a team of researchers, a World Champion and AlphaZero, the strongest chess enitity in existence, and then revised by Chess.com's team. Don't tell me you can do better.

That being said, there absolutely nothing wrong with playing wacky variants. That's actually quite fun to do. But Chess.com probably wants to implement only "serious stuff" and for good reason. I hope the variants will soon exit the Beta phase and we will be able to do tournaments and have rankings. That'd be awesome and it would put this site on the spotlight of the chess-variant comunity. If you want to have some fun though, I suggest you visit fishrandom.com and Pychess. You'll find some nice variants there.

2) There already scarcely played/in-need-of-revision variants. Even if the programmers go through the trouble of making a new variant, how many people do you think would play it? Most players pass their time playing Fog of War. That's great, and it's just such a fun variant. But what about the other variants? Sideways pawns? Capture anything? No castling? Torpedo? These are all great games that make you think about your moves from the very start of every match and let you play beautiful games with new tactics, strategies, and motifs. And, they have been proven to be balanced. So why are they so impopular? Should we just let these variant die?

Mostly I think that Fog of War is stealing the show because it feels like a perfect mix of skill and gambling. That's what makes is so exciting. But people are eventually going to move on and find something new. We don't need to litter the site with new variants that no one will play. We should give a chance to other variants already here and make them at least as popular. I guarantee you, you'll have a great time.

Other variants, on the other hand, have somewhat serious balance issues: Automate and Chaturanga/Shatranj. Automate has already been covered has already been covered. I have also started a thread about Chaturanga.

We should use the forum to make the existing variants GOOD and without bugs. Only once these variants are popular and ready to leave the beta stage we should suggest new ones.

What about Chaturanga?

Yeah, Chaturanga isn't abondoned like Sideways or Torpedo, but it soon will be if it doesn't get fixed. This is such a cool variant and has historical significance to it but the issue remains: without the bare-king rule, a LOT of games will end in a draw. For a discussion on why, I once again link the previous thread on the matter. Now on some endgame considerations for this variant of Chaturanga:
- 2 Knights vs King: WIN for the Knights (they can force stalemate)
- 2 Alfils vs King: obvious DRAW. Alfils seem to be almost useless in the endgame, except in particular situations where you need to control one particular square.
- 2 Opposite color Ferz vs King: Probably a DRAW. At first it felt like they should be able to mate or stalemate the king since they're like two "little bishops". But I don't think that's the case, since the attacking side has to move 3 pieces to mate, while the defending side only has one (the King, who's running for his life). Since the Ferz can't really trap the enemy King like bishops do, he will always be able to run away. So yeah, although I can't prove it, it looks like a draw to me. Just like a 2N vs K, though, if the defending side does some dumb moves, he can get stalemated or mated.
- 2 Same-colour Ferz vs King: DRAW(???). It looks like it should just be like the 2 opposite ferz scenario, but worse. I didn't run any tests on it though.
- 1 Knight + 2 Ferz vs King: Almost surely a WIN.
- 2 Knights vs Ferz: Depends on how the Kings are placed on the board.

These are my first considerations on "Chaturanga" endgames. pls ckess.net fixx bayre king pliz

Did you really have to add the endgame consideration?

A lot of people know nothing or almost nothing about chaturanga endgames. Me included. These are my (unimpressive) findings. It was more to start a conversation about endgames. Share your knowledge, folks!

You could have just done the 1st part and post the endgame consideration in another forum, the 1st part is helpful, but the 2nd part is off-topic to the 1st part.

Meh, I don't think there will be a discussion about whether or not is right to post dumb variant suggestions. So I may as well use the thread to start a convo about endgames, or more importantly, make people conscious we need the bare-king rule. And you know, I didn't want to spam the homepage. But I'm glad you agree with my main point, lots of people just post weird suggestions instead of testing the already balanced variants.

Well, for atomic, white actually does not have an advantage. If they play nf3, black can play f6. It kinda turns out black has an advantage... then white can go e3, black and also play e6. Now it is balanced.

ok, so apart from copying that from lichess, white still has an advantage

with perfect play from both sides, atomic is a win for white.

I think the majority of people will say that white has a pretty big advantage, often throughout the whole game. I have played my fair share of atomic and I am pretty sure that white has a stable advantage. In fact: it's common courtesy to offer a rematch to your opponent when you play white to even things out.

??? Really? I asked REALLY good players and they told me that it's not a proven win. Can you give me a link for proof?

I can't really give you a link, go to lichess analysis and play the best moves in the analysis board

While it's a pretty widespread belief that atomic is a win for white, lichess' analysis board isn't proof, really. When I said antichess was a win for white, I meant mathematically. Scientists actually looked through EVERY line and determined that white can always win. This hasn't been done for atomic, it isn't a solved game yet.

Learn from the best computer: Stockfish. It claims that with its play (the best in the world) in both sides, white wins. And i don't think scientists actually looked if antichess is a win no matter what ngl

>widespread belief that atomic is a win for white,
are there any known rule changes that make it more balanced?

>And i don't think scientists actually looked if antichess is a win
i read a few years ago that some aussie guy solved antichess as win for white (exhaustive, with lota cpu, and published the tree (which was a ton of gigabytes to download..)

I'd be for adding bare king rule as well, if only for historical accuracy. So far most of us play chaturanga like children in the fog, and because of that endgame is of relatively minor importance. But if game community grows, it will change. When we will have a sizeable group of 1800+ players, endgame will become important, and we need to get this right.

As a self-appointed theorist I begin to notice that lack of bare king rule changes piece values: it is seriously downgrading Alfil and Ferz, especially later in the game. It is less of an issue for Ferz, which still is great middlegame piece (I have already won, lost and seen many games decided by an early Ferz raid). Unfortunately, as things are now, Alfil on an open board is just a piece of wood with additional ability to defend one piece.

The nature of Atomic makes it so that initiative is extremely important: every single piece becomes an area denial tool and being able to control parts of the chessboard faster than your opponent is a priority in almost every kind of chess variant. There actually are openings in Atomic that give much more balanced and relatively quiet games, mostly pawn openings. White can opt for these kind of openings if he dislikes the tactical nature of a Knight opening and/or wants to avoid theory (for example: a 2000 atomic player w/ white vs a 1800. The 2000 knows that the other player knows opening theory as well. So that's what he might do, if he is sure he can outplay his opponent relying on his better understanding of Atomic middlegame/endgame).

I'm thinking that due to Atomic's area-control nature , a big-board variant would be a lot of fun and probably much more balanced: think of an Atomic Grand Chess. But to have an equilibrium similar to standard chess we'd have to use a pretty big board, like 12x12 or something.

I agree with your points. I know a couple of really good Chaturanga players already, maybe we should gather up some theory ourselves if we can't find something online. I wasn't able to find ancient theory, maybe someone has better luck?

@Arma55, I have created a thread for chaturanga theory: https://www.chess.com/clubs/forum/view/what-do-we-know-about-chaturanga-strategy

He's still allowed to post his considerations, don't you think?