I am pretty sure a lot of QvQ endings would go on forever without an exchange if one player chose that as an objective. They get drawn by one of the other drawing rules.
True all I am saying is that in a drawn KQKQ ending where one player refuses to agree a draw the other can force a draw before the 50 move rule would be relevant anyway.
Mostly yeah, it's easy to either force the king to go to the same squares, in a circle, or in a skewer.
Gotta go, C Ya
I am pretty sure a lot of QvQ endings would go on forever without an exchange if one player chose that as an objective. They get drawn by one of the other drawing rules.
True all I am saying is that in a drawn KQKQ ending where one player refuses to agree a draw the other can force a draw before the 50 move rule would be relevant anyway.
Mostly yeah, it's easy to either force the king to go to the same squares, in a circle, or in a skewer.
Or self stalemate. I would guess it couldn't be extended beyond 20 moves.
Yes.
EASY
You appear to be playing both sides (which does tend to make it easy).
Your post #661 appeared to be saying there are positions where one side can't force a draw quicker than he could claim under the 50 move rule. If you don't think this is one such, can you suggest a position yourself, or have you changed your mind?
No, I was trying to play black as if he was trying to avoid perpetual.
In a Q v Q endgame, the king will eithergo back and forth or continue in a circle, so it's a guaranteed repetition
I am pretty sure a lot of QvQ endings would go on forever without an exchange if one player chose that as an objective. They get drawn by one of the other drawing rules.
True all I am saying is that in a drawn KQKQ ending where one player refuses to agree a draw the other can force a draw before the 50 move rule would be relevant anyway.
Mostly yeah, it's easy to either force the king to go to the same squares, in a circle, or in a skewer.
Or self stalemate. I would guess it couldn't be extended beyond 20 moves.
Logically, if one side is checking every move, the other player can only move the king, which makes the possibility of being able to achieve a repetition of position a high likelihood. (The only way the defending player can move their queen is to interpose it, and that permits the exchange that they are trying - for some unknown reason! - to avoid).
I am pretty sure a lot of QvQ endings would go on forever without an exchange if one player chose that as an objective. They get drawn by one of the other drawing rules.
True all I am saying is that in a drawn KQKQ ending where one player refuses to agree a draw the other can force a draw before the 50 move rule would be relevant anyway.
Mostly yeah, it's easy to either force the king to go to the same squares, in a circle, or in a skewer.
Or self stalemate. I would guess it couldn't be extended beyond 20 moves.
Logically, if one side is checking every move, the other player can only move the king, which makes the possibility of being able to achieve a repetition of position a high likelihood. (The only way the defending player can move their queen is to interpose it, and that permits the exchange that they are trying - for some unknown reason! - to avoid).
The reason for playing on is presumably in the hope of an error from one's opponent. That was presumably the reason Nikolic didn't agree a draw here:
http://www.chessgames.com/perl/chessgame?gid=1268705.
(His hopes were justified; Arsovic several times slipped into a losing position, even though the initial and final positions in the KRBKB endgame were theoretical draws.) That is presumably also the reason why most people don't offer a draw at the start of the game (though it's unclear whether this is in fact a theoretical draw). Stockfish at any rate won't accept a draw offer in KQKQ and attempts to avoid exchanging queens or repeating positions.
The reasons you give certainly point to a forced draw under FIDE's pre 1st. July rules, but the question is whether the draw can always be forced within the 50 move rule and, in relation to my question to EndgameStudy, within 30 moves. Given that a position can be repeated twice without incurring a draw it is not clear from EndgameStudy's example that this is the case. He doesn't, for example, consider 2...Kg8.
If black plays g8, then q just checks on the back rank again, leading to repetition, and I don't move 3 times because its implied that if the position was already repeated, it will be repeated again. Removing the repetition rule was NONSENSE by FIDE and I still don't get the logic behind it.
If black plays g8, then q just checks on the back rank again, leading to repetition, and I don't move 3 times because its implied that if the position was already repeated, it will be repeated again. Removing the repetition rule was NONSENSE by FIDE and I still don't get the logic behind it.
But the repetitions add up. Do they add up to less than 30 in the worst case? E.g. 2...K.g8 3.Qe8+ Kg7 4.Qe7+ Kh8 5.Qe8+ Kh7 6.Qe7+ Kh8 7.Qe8+ Kh7 8.Qe7+ Kg6 and we have the position for your move 2 but it's taken 8 moves with more of the same to come. I don't say these are necessarily your responses to 2...Kg8 just that you don't consider it.
Totally agree with your views on FIDE's removal of the repetition rule.
I don't know, why were we interested in 30 again?
I don't know, why were we interested in 30 again?
I thought your post was in response to my suggestion that you try stretch the endgame to 30 moves. Apart from that it has no particular significance (apart from being less than 50). My original contention was that either side in a drawn KQKQ position can force a draw before the 50 move rule becomes relevant.
I estimate its less than 50. About 30-40 moves from That position, as the queen has the king trapped on the side the whole time to avoid skewers.
I estimate its less than 50. About 30-40 moves from That position, as the queen has the king trapped on the side the whole time to avoid skewers.
I'm now not so sure, but it's an interesting question. Certainly there are many quick draws, but I'm not at all sure after looking at it a bit more whether all or even the majority of positions can be forced to a draw within 50 moves against correct play.
It should be analysable given a spare weekend.
U did suggest at one point, having specific rules for different types of endgames. That would be better. If the rule was 100, no less, for all endgames, instead of 50, I wouldn't be making a big deal about it at all.
Yes I also thought of that. And then becuase there are so simple I also would not mind to give some easy endgames less then 50 moves. For example Mate with the Queen 20 moves, Mate with the Rook 25 moves, two Bishops 25 moves, Bishop and Knight 40 moves , Two Knights agianst a pawn 130 moves etc etc.
The problem with this is that the endgames are not as simple as you suggest. KBNK and KNNKP in particular are mostly not winnable OTB no matter who you are unless you have already analysed or learned the endings at least to some extent. See e.g.
http://www.chessgames.com/perl/chessgame?gid=1295765
or Ushenina-Girya post #97.
The rules have take account of all strengths of player. A beginner who reaches KBBK will probably be able to work out a mate, but may well not manage a 19 move DTM position in 25 moves OTB against accurate defence.
But why should such levels of accuracy be called for in these endgames? Nobody thinks you need to be so accurate in KQRBBNNPPPPPPPPKQRBBNNPPPPPPPP.
Trust me I can checkmate you with K B N against K just on my increment in about a move or 36. Indeed I did practice it.
And I practice KBNK against the Nalimov EGTB and at about the same speed. And around 90% of the time I make it in the optimal number of moves (the worst case is 33 moves by the way).
My point is that a lot of people haven't had that much practice, and if players like Epishin and Ushenina can fail to make it at all it would seem only reasonable to allow people 50 moves at least. Chess isn't a test of how much people have studied an endgame, it's about whether you can beat your opponent. Why should an extra requirement for accuracy be placed on just basic endgames?
25 moves for KBBK would leave only 6 moves leeway in the worst positions. That is an endgame where a player could work it out OTB without ever having looked at it, but with only 6 moves to spare? For any level of player?
And you're saying you would allow 40 moves for the second diagram in my previous post but only 20 for the first. Is that right?
I don t know exactly what you mean but in the rook endgame you showed I would play Rb2 then the king to a5 then to b6 play Rc1 and then checkmate on c8. The most simple I think. The Bishop plus Knight endgame is a checkmate in only one move as everone can see. ( Bf3 # ) and the queen endgame I would play Kc2 or K b3 to open the diagnonal of the queen then play Qf6 ( or Qg7 ) and close the net and check mate him. It is all not so difficult I think.
But about for example mating material there are only 4 direct cases. Q plus K vs K , R plus K verus K , Two Bishops and K versus K , and Bishop and Knight plus K versus K . Thus I think this is for a chess player not too hard to practice and master it without any problems. I do n know why chess players could not asked for to practice a bit on endgames. For me is the same a a tennis player practice on his serives. For me it goes also not about this very simple endgames that are to master by the most of us but for the harder endgames. Like two bishops against a knight. Two knight agianst a pawn. some of them are not even possible to checkmate them in 50 moves. So I advocate to give chessplayers more moves for very hard endgames and certian for endgames that are not possible to checkmate within the 50 moves. I made only the comparation with very easy endgames that differ so much in the number of moves you needed for a win then the very hard ones. That I think it is not fair to give this endgames also only 50 moves. At the other hand I think that if someone is not able to give a checkmate in for example queen verus rook and pawn and making no progress at all. It must be possible to claim a draw. Also when it is winning in for example 50 moves and this person has complety no clue or what so ever how to win this only by if the opponent is blundering or passes the clock. That in such case it makes no sence to give it 100 although it is hard if some one make no progress or what so ever. But for example the two Knights against a pawn are not possible to mate in 50 moves if you understands it or not. In such cases I think must be more moves given. In endgames that are hard but theoretical draw likle Bishop and rook versus rook , 50 moves is fair to show that you are able to defend it. This is my opinion. I only said that in comparing a queens mate or rook mate is so easy that I would not mind if they gives this even less moves. I surport to give difficult but winning positions more moves.
But in a rook and bishop vs rook endame, he can end up being mated in 2 moves, but as long as 50 moves go by, he gets away with it. He can ignore mates in 5, because there's 2 or 3 moves before 50. It should be 100 for all endgames regardless of progress. It's not that big a deal. Not 50, 75, 100. The progress rule is great, but 100 moves should be allowed for all endgames to start with, then implement the progress rule.
The point I was making is that having k move rules for generic endgames with different k for different generic endgames doesn't make sense. When you reach an endgame in a game of chess you don't play the generic endgame, you play only a specific position in that generic endgame. A generic endgame may contain winning positions that require a high value for k while another generic endgame only positions that require a lower value for k. But that doesn't mean these values should be in effect when you reach specific positions in those generic endgames, because each position in the endgame has it's own k. If there are positions in a generic endgame that could be won under a k move rule against accurate defence then there are also positions in the same generic endgame that could be won under a 1 move rule.
In the examples I gave, the generic endgame KQK needs at least a 10 move rule if all theoretically won positions in it are to be winnable against accurate defence. The KBNK generic endgame requires at least a 33 move rule. But in the specific positions I gave an 8 move rule would be sufficient in the KQK position (though not after Kc2 or Kb3) and a 1 move rule sufficient in the KBNK position. It doesn't really make sense to have a 20 move rule in effect for the first and a 40 move rule in effect for the second.
I know this is the approach adopted by FIDE in the past (i.e. assigning different k move rules to different generic endgames) when it became clear that there were positions that couldn't be won under a 50 move rule against accurate defence, but FIDE don't ever seem to have got their act together regarding this rule.
The other point I wanted to make was that the rules of chess should be usable by any strength of player. If a player who has just learned the moves gets into some won KBBK positions it may well take him more than 19 moves to achieve mate. If he's a little dim he could well take more than the 25 moves you suggest, but this is no reason to allow his opponent to claim a draw at that stage. After all his opponent must be a little dimmer to have arrived on the wrong side of the endgame. The point of chess is to arrive at the result by making the moves.
As for your suggestion that players be asked to practice basic endgames, it's really up to them what they want to spend their time on. If you don't survive the opening and middle game you don't get to play an endgame, so it would seem to be least important to study the endgame. In Topolov-Karpov 2000 (http://www.chessgames.com/perl/chessgame?gid=1295765) it was painfully obvious that neither player had ever spent any time looking at KNNKP, but could you say they would have achieved more if they had? I think Topolov was ranked world number 1 at the time and Karpov was world champion for about a decade not long before.
You didn't say when I asked before. Do you think my no progress rule discussed in various previous posts would answer the problems you discuss? If the number of moves before a no progress claim could be made were set to 20n or 30n, where n is the number of pieces on the board additional to the kings, it would at least give a reasonable chance of limiting KRK and KQK quicker than under the 50 move rule (see the KRK example in #636) while allowing any theoretically won position for which a reliable EGTB has been published to be won with sufficiently accurate play.
But in a rook and bishop vs rook endame, he can end up being mated in 2 moves, but as long as 50 moves go by, he gets away with it. He can ignore mates in 5, because there's 2 or 3 moves before 50. It should be 100 for all endgames regardless of progress. It's not that big a deal. Not 50, 75, 100. The progress rule is great, but 100 moves should be allowed for all endgames to start with, then implement the progress rule.
That would mean that if you have a winning position you need only progress towards mate by one move in each 100 move progress period, which makes it relatively easy for the winning side, but it would also mean you couldn't claim a draw in a drawn KBKN position until you've made 100 moves if you're playing against Stockfish or someone equally intransigent.
How did you arrive at 100? It really is an awful lot of moves if nobody's getting anywhere.
100 is just a good arbitrary number. 3 digits, ya know. A move limit should be in the 3 digits. 100 isn't that much more than 50. The progress rule could then be implemented under whatever move retsrictions thereafter, like 100 free moves where u can do whatever u want, THEN the progress restrictions. See what I'm saying. Not progress every 100 moves, but 100 moves, and then the whole progress tracking begins.
I am pretty sure a lot of QvQ endings would go on forever without an exchange if one player chose that as an objective. They get drawn by one of the other drawing rules.
True. All I am saying is that in a drawn KQKQ ending where one player refuses to agree a draw the other can force a draw before the 50 move rule would be relevant anyway.