Uri Blass wrote:I think that you are wrong to assume that there is no stalemate
The following game that is legal in normal chess is also legal in dark chess
and the final position is clearly a stalemate in the strong meaning(no pseudolegal moves for black that allow capturing the black king).
One has to decide who wins in those conditions. I have always thought that the player with no moves should lose (which is what happens in checkers and many other games), and the whole stalemate rule is a bit strange.
So if it is decided that you lose when you have no moves, that seems very reasonable to me, but it should be included as part of the rules.
hgm wrote:I don't see myself writing an engine for Dark Chess anywhere in the near future, but that would not have to stop me from providing support for it in WinBoard, of course.
Wow. You're ambitious
Is anyone actually considering writing a Dark Chess engine? I can't even imagine how one would go about writing an engine for a variant wherein you don't have complete information... If I were of a mind to do something way off the beaten path, I'd probably try a multi-move variant like Marseillais Chess (http://www.chessvariants.org/multimove. ... seill.html). At least I know it's possible since Zillions-of-Game does it...
Greg Strong wrote:Is anyone actually considering writing a Dark Chess engine? I can't even imagine how one would go about writing an engine for a variant wherein you don't have complete information...
I can imagine some options, but they are not very attractive. One would have to treat this a bit like poker. Probably some Monte Carlo method can be used to both try to "read" where the opponent pieces might be and to search with uncertainty, but I don't know if you would get to 2000 ELO with this approach.
Then again, perhaps humans are not all that great at playing this game either, so it might be enough.
It seems to me that you could probably get a 'decent' playing engine by simply playing conservatively (so as not to leave a king capture or en prise piece) whilst maximising vision. It might even be an interesting way to have a low-strength engine that doesn't play like a complete idiot for play against a full-vision human (especially if the human isn't spending the game trying to take advantage of what the engine doesn't know but instead is just playing chess).
Uri Blass wrote:I think that you are wrong to assume that there is no stalemate
The following game that is legal in normal chess is also legal in dark chess
and the final position is clearly a stalemate in the strong meaning(no pseudolegal moves for black that allow capturing the black king).
Yes i know you are right i have spotted this back then when i was playing in Brainking, but i forgot about that.
The rules i have posted here have as a skeleton the text from Braiking.com's rules that i modified them slightly to avoid ambiguities and to make them better. But i forgot to remove the stalemate thing.
I have said that to Fencer(the owner of the site) before 5-6 years but he replied this is no issue to be fixed since these kind of games will never happen, but still the rules remain ambiguous with the stalemate mentioned in them.
Thanks for the useful remind.
After his son's birth they've asked him:
"Is it a boy or girl?"
YES! He replied.....
hgm wrote:I don't see myself writing an engine for Dark Chess anywhere in the near future, but that would not have to stop me from providing support for it in WinBoard, of course.
Wow. You're ambitious
Is anyone actually considering writing a Dark Chess engine? I can't even imagine how one would go about writing an engine for a variant wherein you don't have complete information...
This guy (whoever he is) has created a kind of "semi-engine" for Dark Chess. It can not play at all, it's some kind of helper engine, but the pattern of what it does is obvious and very powerful.
I'm speaking about Dark Chess Helper where it tries(and succeeds perfectly as i see) to find with deductive logic where the pieces stand in a game from one side's perspective.
So one has to include the logic of this "semi-engine"/helper and some kind of search to it and have a full Dark Chess engine.
After his son's birth they've asked him:
"Is it a boy or girl?"
YES! He replied.....
After each moves, both players get an update of the board, right? The program at the link does not seem to do that. It is not only that you cannot ponder effectively: you really get information by it you otherwise would not get. E.g. I play e2e4, and on the opponent's move e4 disappears from view (as it was unprotected). If Pe4 had been allowed to 'look around' after my own turn, I would now know that what captured it is not a Pawn. Otherwise it could be anything.
The latest beta ( http://hgm.nubati.net/WinBoard-4.5.beta.zip ) now has fully darkened display (in human-engine mode): no highlights, no animation, no move history, no engine PVs, no move printed above the board. I guess this is about as it should be. No changes on the back-end yet, though. But that means you can use a normal Chess engine, as long as it says that it plays "variant dark".
Animation for multi-moves is now also OK; it animates the first leg, without leaving debris in the display. (This is how castling has always been implemented as well.) The debris was actually was a bug that had reverted.
