correspondence chess in the age of NNUE

[MikeB]

FWIW, I have never seen a Crazyhouse chess game end in a draw.

[lkaufman]

FWIW, I have never seen a Crazyhouse chess game end in a draw.

According to old talkchess posts, in high level engine play Crazyhouse gets about 4% draws, which isn't bad, but the real problem is that of the other 96%, White won 85% to 11%. So it's basically a handicap game. With some opening restrictions it could be made fairly even though. But it's almost as much a shogi variant as it is a chess variant.

[Ferdy]

I suppose the actual wording should read something like: If a move is played that repeats the position for the third time, either player may call it a repetition, with the repeating side getting 1/4 point, the other side 3/4 point. For engine vs. engine play I suppose it would be automatically scored as such. The main point isn't to avoid repetitions played to avoid a no draws by agreement rule, it is that testing has shown that prohibiiting (or penalising) repetitions reduces draws far more than the stalemate + bare king rules do. There are many drawn endings where if you aren't allowed to repeat, you will lose (many even material King and pawn endings would be in this category). I don't claim that this rule avoids all draws or even most draws; I do claim that it roughly doubles the percentage of decisive games when the level of play is high enough that that percentage is a fairly small number. If you want to eliminate all or even most draws we would need more rules.

Hi Larry.
I assume that you are aware, that this rule will change the character of the game massivly. Not only that KPvKP are effected, but also many other totally equally balanced endings. Even KRvKR and KQvKQ can only try to escape by the 50-move-rule. Chances tobe adopted by ICCF? In my oppinion close to zero. I seen to much reluctantness in accepting rule that affects the game characteristics, even ones, that didn't smell such "unfair" in many situations.
Cheers,

In one of my simulations I saw KRvKR ended in repetition. This is ugly separating them by 3/4-1/4, but perhaps another rule could be added like in this type of position just award 1/2-1/2.

[Desperado]

There's probably nothing that can be done to remove the overwhelming influence of computers in correspondence chess at this point, even with modifications to the game big and small. Implementing most rules changes into Stockfish would take a decent programmer all of minutes to hours, and then they could train a very, very strong NNUE on consumer hardware in a day or two. Bugfixes, retraining, and refinement would be carried out over the course of a match as necessary.

Anyway, my favorite chess variant has always been where you start with a coin on an empty square (most often e4, but you can pick any square for variety), and then the coin duplicates the motion of both players' pieces with every move. (For example, 1. Nf3 would move the coin from e4 to d6. Castling is a king move.) The catch is that if the coin wouldn't land on an empty square, you can't make the move!

Having no moves counts as a loss, and checks and checkmates work the same way they do in normal chess, even if the coin restriction would technically prevent the attacking pieces from capturing the king. (You can just imagine that any move that would capture a king is allowed to ignore the coin.)

You are right that there's little that can be done about the influence of computers in correspondence chess, but the topic here is draw reduction, which is a completely different issue. You can either change the start position (i.e. chess960 or mandated openings), or change the rules. For me the only interesting rule changes are those that reduce draws with the least effects in most positions, keeping it as close as possible to normal chess. If you want a very different chess, shogi is the best option in my opinion.

Sorry if I just jumped into the discussion.
Maybe the following has already been said a dozen times.

For me, the problem does not look so difficult. Why not play time handicaps. This is illustrated when an engine plays against itself in short time, for example 2 minutes against 5 minutes. The results change significantly. The principle should be transferable to correspondence chess. I am convinced that an imbalance will be created, which in the long run will favor the stronger players and will ultimately be expressed in the rating.

Ultimately, the game remains the same in terms of rules and basic requirements.
Handicaps in the context of time can, in my opinion, express the playing strength much better than using material handicaps.

Regards

[Ferdy]

There's probably nothing that can be done to remove the overwhelming influence of computers in correspondence chess at this point, even with modifications to the game big and small. Implementing most rules changes into Stockfish would take a decent programmer all of minutes to hours, and then they could train a very, very strong NNUE on consumer hardware in a day or two. Bugfixes, retraining, and refinement would be carried out over the course of a match as necessary.

Anyway, my favorite chess variant has always been where you start with a coin on an empty square (most often e4, but you can pick any square for variety), and then the coin duplicates the motion of both players' pieces with every move. (For example, 1. Nf3 would move the coin from e4 to d6. Castling is a king move.) The catch is that if the coin wouldn't land on an empty square, you can't make the move!

Having no moves counts as a loss, and checks and checkmates work the same way they do in normal chess, even if the coin restriction would technically prevent the attacking pieces from capturing the king. (You can just imagine that any move that would capture a king is allowed to ignore the coin.)

You are right that there's little that can be done about the influence of computers in correspondence chess, but the topic here is draw reduction, which is a completely different issue. You can either change the start position (i.e. chess960 or mandated openings), or change the rules. For me the only interesting rule changes are those that reduce draws with the least effects in most positions, keeping it as close as possible to normal chess. If you want a very different chess, shogi is the best option in my opinion.

On shogi, disallow repetition with check could be a good improvement for chess correspondence. This would offer more challenges to players than just go for easy draw because your opponent has a higher rating.

[Ferdy]

An example game from iccf with draw result, why the players ended the game?

I am ok with draw result but it should be well fought.

[pgn][Event "WC30/final"]
[Site "ICCF"]
[Date "2017.06.20"]
[Round "?"]
[White "?"]
[Black "?"]
[Result "1/2-1/2"]
[BlackElo "2540"]
[ECO "D02"]
[WhiteElo "2666"]

1.d4 { [%ccsnt 2017.07.02,23:31] } d5 { [%ccsnt 2017.07.03,14:41] } 2.Nf3 { [%ccsnt 2017.07.04,13:02] } Nf6 { [%ccsnt 2017.07.05,19:46] } 3.Bf4 { [%ccsnt 2017.07.07,18:46] } c5 { [%ccsnt 2017.07.13,08:37] } 4.e3 { [%ccsnt 2017.07.14,07:52] } Nc6 { [%ccsnt 2017.07.14,18:55] } 5.Nbd2 { [%ccsnt 2017.07.16,16:49] } Qb6 { [%ccsnt 2017.07.18,15:30] } 6.dxc5 { [%ccsnt 2017.07.19,13:25] } Qxb2 { [%ccsnt 2017.07.25,09:38] } 7.Rb1 { [%ccsnt 2017.07.26,07:30] } Qc3 { [%ccsnt 2017.08.01,10:19] } 8.Bb5 { [%ccsnt 2017.08.02,10:07] } e6 { [%ccsnt 2017.08.03,00:09] } 9.O-O { [%ccsnt 2017.08.19,21:08] } Be7 { [%ccsnt 2017.08.23,22:51] } 10.Ne5 { [%ccsnt 2017.08.24,19:35] } Bd7 { [%ccsnt 2017.08.24,22:11] } 11.Nxd7 { [%ccsnt 2017.08.27,19:27] } Nxd7 { [%ccsnt 2017.08.28,20:48] } 12.e4 { [%ccsnt 2017.08.29,17:07] } O-O { [%ccsnt 2017.08.29,22:21] } 13.exd5 { [%ccsnt 2017.09.01,15:24] } exd5 { [%ccsnt 2017.09.07,22:32] } 14.Ne4 { [%ccsnt 2017.09.19,00:36] } dxe4 { [%ccsnt 2017.10.19,06:52] } 15.Qxd7 { [%ccsnt 2017.10.22,01:34] } Bxc5 { [%ccsnt 2017.10.22,23:27] } 16.Qxb7 { [%ccsnt 2017.10.23,20:22] } Rac8 { [%ccsnt 2017.10.30,21:03] } 17.a4 { [%ccsnt 2017.11.08,23:41] } Bb6 { [%ccsnt 2017.11.21,13:18] } 18.Rbd1 { [%ccsnt 2017.12.05,23:20] } e3 { [%ccsnt 2017.12.07,06:34] } 19.fxe3 { [%ccsnt 2017.12.20,11:59] } Bxe3+ { [%ccsnt 2017.12.20,12:45] } 20.Bxe3 { [%ccsnt 2018.01.03,12:19] } Qxe3+ { [%ccsnt 2018.01.03,14:50] } 21.Kh1 { [%ccsnt 2018.01.04,13:51] } Ne5 { [%ccsnt 2018.01.04,14:01] } 22.Rfe1 { [%ccsnt 2018.01.09,13:56] } Rcd8 { [%ccsnt 2018.01.09,18:37] } 1/2-1/2[/pgn]

[Thomas A. Anderson]

I suppose the actual wording should read something like: If a move is played that repeats the position for the third time, either player may call it a repetition, with the repeating side getting 1/4 point, the other side 3/4 point. For engine vs. engine play I suppose it would be automatically scored as such. The main point isn't to avoid repetitions played to avoid a no draws by agreement rule, it is that testing has shown that prohibiiting (or penalising) repetitions reduces draws far more than the stalemate + bare king rules do. There are many drawn endings where if you aren't allowed to repeat, you will lose (many even material King and pawn endings would be in this category). I don't claim that this rule avoids all draws or even most draws; I do claim that it roughly doubles the percentage of decisive games when the level of play is high enough that that percentage is a fairly small number. If you want to eliminate all or even most draws we would need more rules.

Hi Larry.
I assume that you are aware, that this rule will change the character of the game massivly. Not only that KPvKP are effected, but also many other totally equally balanced endings. Even KRvKR and KQvKQ can only try to escape by the 50-move-rule. Chances tobe adopted by ICCF? In my oppinion close to zero. I seen to much reluctantness in accepting rule that affects the game characteristics, even ones, that didn't smell such "unfair" in many situations.
Cheers,

Usually, in fully "equal" endings KRvKR and KQvKQ, the game will ultimately be drawn by 50 move rule instead of by repetition, no big change here since the players can agree to a draw without waiting for 50 move rule. But there will be many equal material endgames including those two where one side has an advantage but not enough to win, yet enough to force the opponent to repeat or lose, so those endgames will get 3/4 of the point[/size]. I consider this a good thing, but of course not everone will agree. The only alternative to this or similar rule changes that would make correspondence chess playable again at the top level is to mandate bad openings or a bad startposition (with paired games), as in TCEC. I leave it to the chess world to decide which is the better option, or whether Correspondence chess should fade away.

Maybe I get you wrong, but In contrast to you, I can not imagine of any 4-piece endgame position (KRvKR, KQvKQ, etc) where one side has an "advantage" that "deserves" 3/4 of the point. The effect of this rule (3-fold-position-repetition isn't 0.5 points) would be that any of today's dead drawn endgame must be reevaluated, if one side can escape with a 50-move-0.5 before it is forced to step in the 3-fold-position-repetition-trap and only get 0.25. Personally I I would probably being one of the players benefit most from this rule, because I am pushing my opponents into exactly that type of position you define as "favourable", but the majority of drawn endgames are simply equal in a manner you can not argue, that hitting the repetition first has to be judged as being the weak side. In his latest paper Arno was trying to define that kind of positions that has to be considered as favourable. Very hard, if not impossible, task and not by chance his latest proposals is that we need a jury that decide what player deserves more than the half point in a drawn game. While I really like this idea, I doubt it will become reality

[jefk]

Uri Blass,

about engine opening moves, guess you're right,
with Nnue apparently you don't need an opening book anymore.

Just had a quick look to confirm, while the Nnue engine moves
for Black in the opening phase are not always perfect, they
seem good enough for a draw (and probably are good enough
after enough analysis time, whether its 1 hr or a bit more).

So indeed something has to be done about the current rules...

And when the drawing margin would be reduced with the new Kaufman/Nickel
rules (*), indeed allowing less time for White eg. one day average
(instead of three days as 'normal' may be an option.

(*) As for the Iccf arbiters etc. i don't see why they wouldn't
be open for suggestions (eg. a new letter from LK/AN to them
because of the current 'nnue' situation where we have almost
achieved 'perfect' play (with perfect meaning good enough
(for Black) to maintain a draw.

[Ozymandias]

I can not imagine of any 4-piece endgame position (KRvKR, KQvKQ, etc) where one side has an "advantage" that "deserves" 3/4 of the point. The effect of this rule (3-fold-position-repetition isn't 0.5 points) would be that any of today's dead drawn endgame must be reevaluated, if one side can escape with a 50-move-0.5 before it is forced to step in the 3-fold-position-repetition-trap and only get 0.25.

Hi Neo,

The goal of the rule is to transform some draws into something other, by design. To deserve half a point you have to do better and see further than your opponent, so that you don't fall into those draw looking endgames (or positions) were you're actually getting just 1/4 of a point, and maybe anticipate so much than you actually get 3/4 of a point, and not just half. This has always how chess has been played, but we're focusing on something other than wins now, because those are becoming so scarce than the difference in playing strength no loger gets reflected on the scoreboard.

Hope that helps

[lkaufman]

I suppose the actual wording should read something like: If a move is played that repeats the position for the third time, either player may call it a repetition, with the repeating side getting 1/4 point, the other side 3/4 point. For engine vs. engine play I suppose it would be automatically scored as such. The main point isn't to avoid repetitions played to avoid a no draws by agreement rule, it is that testing has shown that prohibiiting (or penalising) repetitions reduces draws far more than the stalemate + bare king rules do. There are many drawn endings where if you aren't allowed to repeat, you will lose (many even material King and pawn endings would be in this category). I don't claim that this rule avoids all draws or even most draws; I do claim that it roughly doubles the percentage of decisive games when the level of play is high enough that that percentage is a fairly small number. If you want to eliminate all or even most draws we would need more rules.

Hi Larry.
I assume that you are aware, that this rule will change the character of the game massivly. Not only that KPvKP are effected, but also many other totally equally balanced endings. Even KRvKR and KQvKQ can only try to escape by the 50-move-rule. Chances tobe adopted by ICCF? In my oppinion close to zero. I seen to much reluctantness in accepting rule that affects the game characteristics, even ones, that didn't smell such "unfair" in many situations.
Cheers,

Usually, in fully "equal" endings KRvKR and KQvKQ, the game will ultimately be drawn by 50 move rule instead of by repetition, no big change here since the players can agree to a draw without waiting for 50 move rule. But there will be many equal material endgames including those two where one side has an advantage but not enough to win, yet enough to force the opponent to repeat or lose, so those endgames will get 3/4 of the point[/size]. I consider this a good thing, but of course not everone will agree. The only alternative to this or similar rule changes that would make correspondence chess playable again at the top level is to mandate bad openings or a bad startposition (with paired games), as in TCEC. I leave it to the chess world to decide which is the better option, or whether Correspondence chess should fade away.

Maybe I get you wrong, but In contrast to you, I can not imagine of any 4-piece endgame position (KRvKR, KQvKQ, etc) where one side has an "advantage" that "deserves" 3/4 of the point. The effect of this rule (3-fold-position-repetition isn't 0.5 points) would be that any of today's dead drawn endgame must be reevaluated, if one side can escape with a 50-move-0.5 before it is forced to step in the 3-fold-position-repetition-trap and only get 0.25. Personally I I would probably being one of the players benefit most from this rule, because I am pushing my opponents into exactly that type of position you define as "favourable", but the majority of drawn endgames are simply equal in a manner you can not argue, that hitting the repetition first has to be judged as being the weak side. In his latest paper Arno was trying to define that kind of positions that has to be considered as favourable. Very hard, if not impossible, task and not by chance his latest proposals is that we need a jury that decide what player deserves more than the half point in a drawn game. While I really like this idea, I doubt it will become reality

I'm not certain, but I believe that it would be only in rare cases that a KRvKR or KQvKQ ending would unavoidably end in a repetition. In the vast majority of cases, such endings will normally end in repetition, but if the repeater gets just a quarter point most will end with the fifty move rule. Exceptions will probably mostly be cases where the need to repeat will be clear from the reaching of this endgame, so the inferior side can attempt to avoid it if possible. But there is no denying that it changes the game somewhat, it's just the smallest change I know of which will have a dramatic effect on the draw rate.