lichess.org
Donate

Lichess computer system error

As i previously stated, on ICC you would have to talk to an admin to turn your result from a draw into a win. I think their site decided whether a position was winable by checking if it could mate a lone king.
So in your example, Dr_King_Schultz



white just ran all the time on their clock to achieve a draw. Unfair. Nice checkmate was not given.

That's really bad rules then, Lichess, I agree with others.

You really should consider to change that rule, accordingly.

Or one more example:

http://en.lichess.org/NLqPsUdB/black#132

Black has forced win in 90 moves.

But again, they have "insufficient material" (2 knights). If opponent just waste all his time it will be unfair draw.

Current rules have no sense.
I will say again that I think it would be a very good idea for lichess to have a document clearly indicating its ruleset, equivalent to the USCF vs FIDE one posted before. That way it can be transparent and obvious for all users exactly what rules have been chosen to be in effect and not.
Yea, I agree that producing a rule book would be a good idea. I think it will be looked into.
I think #30 has a great point here; there is no other reasonable way to divide it other than "possible to win" and "impossible to win". Or else this game should probably also be a draw, since black cannot win by any normal means: http://en.lichess.org/PtTNLOSlTXen.

Even though that game is constructed, there are plenty of similar games played every minute, where one single pawn wins against queens and rooks and knights etc. So there are already loads of games ending in a win when there really is "no way to win by normal means."

Most online chess servers ignore the opponent's pieces when judging whether the game is a win or a draw, because it is easier like that. This way of determining the result of a position when the flag falls is an online thing only, and it is kinda weird that people automatically accept that KB vs KP is draw, while KP vs KQRBPPP is a win. But people have gotten used to this in online chess, since it has always been this way.

Lichess can change this, however, and if possible, I think they should. It would give Lichess an even greater advantage over the other chess servers, being the one that finally uses the proper chess rules.
As to #32, wouldn't the 90 move theoretical checkmate fall under the 50 move rule anyway?

As to the others, I'd recommend discussing it with thibault if you want a rule change. He listens to reason, and you're more likely to hear his reasoning.

Given the bad sportsmanship here: http://en.lichess.org/UGPawo3K/black#150 my position has softened somewhat. I don't think all theoretical wins should be automatic wins, but that position certainly should be.
In the 2 knights vs pawn checkmate, the defending side usually can't avoid making some pawn move within the first 50 moves (or otherwise get mated in time). I haven't checked the given example though. ;)
For the 2 knights against pawn, because white has a pawn the player with 2 knights can use zugzwang to force the pawn to move to prevent the 50-move draw.

Incidentally, if yu follow the top tablebase line for the particualr position given, at one point you reach a position with white: pawn h6, king g7 and black: knights h7 and c6 and king e7, where white to move cannot capture the knight on h7 due to exactly the same one-knight corner checkmate as in the 'bad sportmanship' game http://en.lichess.org/UGPawo3K/black#150

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