well, just look at this thread and tell me what you think?
http://www.chess.com/forum/view/game-showcase/runnung-someone-out-of-time
Do we want that such thing happens, when this is a clear draw? I don't like it, and I would be among the first to vote for it to be changed, but we would have to listen to opinion from others as well.
@Dragec - I would prefer a draw in that game especially if there is still a lot of time on both clocks.
We will be able to call arbiters to the board once chess.com starts having LIVE tournaments.
i agree. looking into it...
i'd love some help from you guys here to help determine what ARE insuffient mating material situations.
so, if you run out of time and your opponent has [FILL IN THE BLANK], it is a draw, not a loss.
K
K+N
K+B
K+N+N
what else?
also, does it matter what YOU have when you run out of time?
here are some references:
http://www.fide.com/fide/handbook?id=124&view=article
http://www.chessbase.com/newsdetail.asp?newsid=4893
"Law 6.10 of the FIDE Laws of Chess states that: "If a player does not complete the prescribed number of moves in the allotted time, the game is lost by the player. However, the game is drawn, if the position is such that the opponent cannot checkmate the player's king by any possible series of legal moves, even with the most unskilled counterplay."
for those of you looking for EXTRA reading material, please check: http://www.amazon.com/Insufficient-Mating-Material-Rowena-Cherry/dp/0505527111
-------------------------------------
let me know your thoughts!
K + B vs. K + Q and K + N vs. K + Q should be a draw when the side with the queen times out.
K + B vs. K + R should be a draw when the rook times out.
In all three situations, it is impossible to construct a helpmate where the side with the queen or rook can't block the check or capture the opposing piece.
It does matter what you have when you time out, not just what your opponent has.
Notably K+N vs K+P would be a loss when the player with K+P times out because a checkmate is possible using those pieces despite K+N generally being considered insufficient material.
Sometimes, although very rarely, having less material can be advantageous when on the verge of timing out.
i'd love some help from you guys here to help determine what ARE insuffient mating material situations.
so, if you run out of time and your opponent has [FILL IN THE BLANK], it is a draw, not a loss.
K
K+N
K+B
K+N+N
what else?
also, does it matter what YOU have when you run out of time?
here are some references:
http://www.fide.com/fide/handbook?id=124&view=article
http://www.chessbase.com/newsdetail.asp?newsid=4893
"Law 6.10 of the FIDE Laws of Chess states that: "If a player does not complete the prescribed number of moves in the allotted time, the game is lost by the player. However, the game is drawn, if the position is such that the opponent cannot checkmate the player's king by any possible series of legal moves, even with the most unskilled counterplay."
for those of you looking for EXTRA reading material, please check: http://www.amazon.com/Insufficient-Mating-Material-Rowena-Cherry/dp/0505527111
-------------------------------------
let me know your thoughts!
I think this could open a can of worms! If one player has a King and Knight, that does not mean that checkmate is impossible (unless the opponent has a bare King, obviously). As long as FIDE rule 6.10 is adhered to as closely as possible, then everything should be okay. After all, I have to admit that Ringwraith2021's example should probably have been a draw. But if rule 6.10 isn't followed closely enough, then a deluge of complaints could follow, as players protest that draws were handed out when they should have been wins, etc.
K + B vs. K + Q and K + N vs. K + Q should be a draw when the side with the queen times out.
K + B vs. K + R should be a draw when the rook times out.
In all three situations, it is impossible to construct a helpmate where the side with the queen or rook can't block the check or capture the opposing piece.
No it's not:
K+Q and K+R can always defend against K+B helpmate (via interposition).
K+B and K+N cannot, so in those cases it would be a draw.
Similarly, K+Q can always defend against a K+N helpmate (via capture).
K+B, K+N, and in this case K+R cannot, so in these cases it would be a draw.
I think one challenge is that the rule reads "any possible series of legal moves" which complicates the calculation quite a bit. For example:
K+N vs K+P should only be a draw when K+P times out if the pawn is not a rook pawn.
K+B vs K+B should only be a draw if the bishops are the same colour.
etc.
K + N vs. K + P would not be a draw when the pawn times out, since the pawn can be underpromoted to a knight, allowing a helpmate.
The problem with this is that it may be possible to construct positions with more pieces where helpmate cannot be achieved. Here is a bit of a silly example, just to show that it may be difficult to create a list of every possible position where one side cannot win by any series of legal moves.
I fully agree with TheGrobe that the opponent's material matters, of course.
For automatic evaluation purposes, only the FIDE rule concerning "no win by any sequence of legal moves" can be applied. As far as I know now, and as far as I have seen in the examples named in this thread, this is limited to:
if you run out of time and your opponent has [FILL IN THE BLANK], it is a draw, not a loss. :
* K vs anything
* K + N vs K
* K + B vs K
* K + B vs K + Q
* K + B vs K + R
* K + N vs K + Q
* if, and only if, the automatic software is able to safely identify fortress positions in which no legal mate is possible (see post #38 for an example), then also these fortress positions should be included in this list.
However, if you run out of time and your opponent has
K + N + N vs K
it is a loss according to present FIDE rules, because a helpmate is possible here.
All other cases, especially cases such as evaluating a game as a draw according to the quick play finish rules, or adjudicating a correspondence chess game as a draw, would have to be done by a human arbiter.
An automated system should not get the job of aborting a game as a draw when a material distribution such as K+N+N vs K is reached. For example, there are beautiful studylike positions in which there is a forced mate with this material. Of course, as soon as there is a human arbiter involved, a much wider range of positions can be adjudicated as draws, especially in correspondence chess time controls.
K + N vs. K + P would not be a draw when the pawn times out, since the pawn can be underpromoted to a knight, allowing a helpmate.
...
Yes, good point. I stand corrected.
my head hurts. can't imagine the staff's heads with this one.
so, out of curiousity, can a chess engine be set up to determine this for situations that are not clear? or perhaps a dispute button after the game which will have the game reviewed (by person or engine) when it seems justified by the non-timed out person?
In fact, there are very few positions which cannot make a checkmate with worst play for one side.
Oh, and what does FIDE say about completely blocked positions where there are many pawns, but it is impossible to break through (for example only kings left).
ozzie, just read above. The "completely blocked" positions are just as drawn as soon as they fulfil the criterion that there is no sequence of legal moves that leads to mate.
We have given already a diagram example in this thread, and I have called such positions "fortress positions such as in post #38" in the context of this thread.
Thank you for the clarification.
Yup, just looked at #38.
For the record, I don't know how important it is to get all of those things right. Just get the hardcoded ones right and leave the rest as a feature request. Bigger fish to fry.
Ah, ok. :) I think it is a good suggestion.