Free online Chess server. Play Chess now in a clean interface. No registration, no ads, no plugin required. Play Chess with the computer, friends or random opponents.
Create a game Tournament Simultaneous exhibitions
Chess basics Puzzles Practice Coordinates Study Coaches
Lichess TV Current games Broadcasts (beta) Video library
Players Teams Forum Questions & Answers
Analysis board Board editor Import game Advanced search
Sign in
  1.  Atomic Chess Theoreticians
  2. Forum
  3. Play atomic chess in real / test new rules


[1. intentionally in french, targeting only people in Paris]
1) S'il y a des joueurs en région parisienne, ce serait amusant de se retrouver pour jouer aux échecs atomiques sur des vrais échiquiers ^^ Des volontaires ?

2) What do you think about this simple rule to (try to) balance the game : "no knight move in the first three moves" ? If anyone wanna try (in unrated correspondance games), welcome.

J'habite à Palaiseau, à 40 minutes de Paris par le RER B. De plus, ça tombe bien, c'est toujours avant la rentrée, du coup je serai disponible ce week-end et en semaine la semaine prochaine. :)

I find the game balanced enough as is, in that black has excellent practical chances in most lines. There are lines that are more annoying to face, but if black is sufficiently booked up it takes some heroic efforts by white to get a winning advantage purely from opening play.

Your suggested change seems to me like it'd lead to games with central pawns like 1. e4 d5 2. d4 d5 3. c3 c6 or similar. Probably balanced, but I'm too terrible at that kind of position to tell.

1.e4 d5 2.d4 e5 3.Bh6!? I think would be dangerous, as Black doesn't have Nxh6.

@Illion: you are right about the central pawns games (I also never play that as white ^^), and black has practical chances in some lines. But black generally fight not to lose instead of fighting to win (a draw with black in atomic is more or less considered a win :/ )

@Erdmunr: indeed, I didn't think from this point of view (the rule was mainly thought for white attacks, not preventing black counterplay :) ). However there is 3...Bg4 followed by gxh6 possible, should be even.

I'm working on an online chess app currently and I have variant Atomic. I modified it so that there can be no knight move in the first 3 moves. In theory you can play a rated game based on this rule now online. Of course you won't find any opponent there, but you can test it against an AI ( which just makes a random legal move ). Anyway my app is very beta so I welcome any feedback ( with respect to can you register / log in / does the board work etc. ).

Good initiative :)
However, after registration and game start against AI, the pieces don't move.
I also plan to upload a website dedicated to chess variants soon, but it's still in development.

Thanks for this feedback.

I would like to know where the thing went wrong.

When you press "Play" for white ( AI can only play black ) the top left corner of the board should look like this:

and the seek like this:

In this state the game is not ready yet, because you have to seat the AI as well ( what you see is just a button that seats the AI, if an other players opens the same board, there will be a Play button instead of the AI button ).

When you press the AI button then the top left corner should change to:

and the seek:

I wonder if there is a problem with moving the pieces when the AI is properly seated.

Ah, sorry I didn't understand how to start a game (I didn't click on "AI" to terminate intialisation). Now the game starts, but pieces don't move well : the yellow arrow points toward the initial square instead of destination, and the piece is "pushed backward" like if it was attached to a spring. I played 1.e4 and the pawn went back to the first rank, then black played b5 which advanced further until rank 2, pushing my b pawn and making the knight disappear :/

What did you use for drag and drop ? I choosed jquery.UI on a former website, heavy but easy to use. chessground by lichess creator looks good too.

EDIT : I use chromium (~ last version) under Linux.

I'm using Scala.js for the client which compiles Scala to Javascript. So under the hood the client is also is written in Scala. Now, I have problems wiht the JQuery facade in Scala.js ( only the D3.js facade works as expected ). So I have written the drag and drop at low level: when the user drags a piece, I immediately disable HTML's own drag handler ( which is a pain ) using "preventDefault" and then rely on the mouse move and mouse up events and simply adjust the CSS top and left properties of the piece accordingly.

If you want real fireworks with the Flick variant, use a bullet time control because the flick speed depends on time control. So in bullet you can smash half of your opponent's pieces in one shot.

OK, I didn't understand that this was normal in "Flick" variant ^^ Choosing atomic or standard works well (randomly, but gameplay is OK). However the "flip" button is ineffective - but it's a detail :)