@playerafar
If you're trying to weakly solve, the three-fold repetition rule can be ignored, but not the 50 move rule. (Did you look at the posts I mentioned and http://galen.metapath.org/egtb50/ ?)
Agreed draws would have to be discounted.
If you are trying to solve (sans "weakly"), the three fold repetition rule can't be ignored. See post #219 here https://www.chess.com/forum/view/general/chess-will-never-be-solved-heres-why?page=11. That would just be Nightmare on Elm Street.
#343
The 3-fold repetition rule is essential theoretically as well as in practice.
Theoretically the 3-fold repetition rule makes chess finite and thus solvable.
Most draws in practice are 3-fold repetition.
In other games like Go and Stratego repetition is forbidden: i.e. if you repeat then you lose.
#334
Yes, all positions of 7 men or less are known exactly.
However, castling is not allowed.
Correct. We haven't solved 7 man chess yet.
That would almost certainly be possible given 5 years on a supercomputer.
#345
The endgame tablebases do not include castling.
Rightly so, it would only confuse the isssue.
After a position is down to 7 men castling rights are usually gone.
@playerafar
If you're trying to weakly solve, the three-fold repetition rule can be ignored, but not the 50 move rule. (Did you look at the posts I mentioned and http://galen.metapath.org/egtb50/ ?)
Agreed draws would have to be discounted.
If you are trying to solve (sans "weakly"), the three fold repetition rule can't be ignored. See post #219 here https://www.chess.com/forum/view/general/chess-will-never-be-solved-heres-why?page=11. That would just be Nightmare on Elm Street.
@MARattigan I think we're going to disagree on these points.
Or maybe - it would be about 'weakly' semantics. Would that be worth our time?
I believe the 50 move rule and the three-fold rule can both be ignored much as we'd ignore the situations on the clocks.
They can all be seen as 'additional' and pertaining to games rather than positions.
"Hey I can draw by reaching fifty moves with no captures/pawn motion !"
Is there any situation where the 50 couldn't be ignored ?
If positions exist where the win is impossible because it would take more than 50 ... (I don't know whether there are such positions) -
then they could be 'noted'.
More practical would be to have an 'unclear' category.
Where the computer spits out positions where a 'solution' isn't readily available - and they're instantly assigned to other computers specialized for that.
Then the other computers could further categorize.
Point - you could have all positions 'accounted for' as opposed to 'solved'.
As an intermediate step.
Further: the project of 'solving' all positions regardless of 50 and three-fold and clocks and could the position have legally got there - is itself an 'intermediate step' rather than a 'weak step'.
En passant positions possible is 'easy' its just a matter of doubling the numbers ... or more if more than one en passant is theoretically possible.
Yes - that is possible when considering positions rather than games.
Castling is tougher though.
#345
The endgame tablebases do not include castling.
Rightly so, it would only confuse the isssue.
After a position is down to 7 men castling rights are usually gone.
Does that mean the 7 men 'solved' work is already invalid?
Beginning to look like it. 
#343
The 3-fold repetition rule is essential theoretically as well as in practice.
Theoretically the 3-fold repetition rule makes chess finite and thus solvable.
Try to get rid of that rectifier. Read this link I already posted https://en.wikipedia.org/wiki/L_game. An infinite game is not necessarily unsolvable.
The three-fold repetition rule would make chess a nightmare to solve, but is not relevant for a weak solution.
Most draws in practice are 3-fold repetition.
Nonsense. Most draws in practice are agreed draws (and under competition rules would more often become draws under the 50 move or dead positions if played on). You had it as stalemates a couple of posts back.
In other games like Go and Stratego repetition is forbidden: i.e. if you repeat then you lose.
@MARattigan is correct ...
most draws are Not 3 fold repetition.
I would say Lone Kings and Kings + one minor piece and a and hpawn draws and K+P drawn because of opposition (but not with 6th rank King)
or attacking King not in front of pawn ..
and add in various equal and close enough to equal material situations and then all such draws greatly outnumber all other draws combined ...
Strange thread...most people seem to have no idea what "solving" a game means.
Solving a game does not mean being dominant over all other humans...otherwise chess was "solved" by Morphy, even though we can find many suboptimal moves in his play now.
Solving also doesn't mean calculating every permutation.
What it means is a mathematically best strategy has been shown -- either an unstoppable strategy that always wins for one player, or a strategy that forces a draw (where an unstoppable can be proved to not exist).
At this time, the game of chess has not been solved, but there is no reason why it is impossible, or intractable.
I think Go will be solved before chess, and when that happens we can suspect chess is coming soon. Go has many more permutations than chess, so is sometimes described as a more complex game, but I think, given only one kind of "piece" and "move" it looks a better candidate for finding an unbeatable strategy.
finally! someone gets it. yes!
Chess could be thoroughly 'solved' - given enough time.
But 'enough' could be 50 million years or more.
And 'solved' could mean what whoever wants it to mean.
#349
When they agree on a draw, it is because a 3 fold repetition or a stalemate or a dead position, or a 50 moves draw is inevitable. Of these 4 the 3-fold repetition is the most common and the 50 moves rule is the least common.
Recent example:
https://en.chessbase.com/post/tata-steel-chess-2022-r10
I know your L-game where there is a cunning solution. Also antichess is solved. However for chess it is essential that it ends in a finite number of moves and the 3-fold rule ensures a finite albeit huge number.
A single GM tournament isn't a good representative sample regarding draws.
And GM's may often agree to a draw because they know better than some other players if its headed to 'dead'.
I've seen tournament games where an endgame draw was agreed to because the pawns were symmetrical to each other.
But if players of a lesser caliber had continued - one might 'easily' go wrong and lose even in such positions.
The Kings get in close and somebody does something technically wrong.
The lower the level of play the more it happens.
So there's another obstacle ...
the 'ability' of either contestant to 'go wrong' in the position.
How are the computers going to resolve that ?
By working backwards and arguing a draw can be forced?
But would players find it?
Players in tournaments and other rated play are not allowed computer assistance !
I think chess is safe from computer 'doom'.
Wow - those pawns are going to get a Workout !
You could have a chess athletic game.
Players have to run around with heavily weighted pieces - with the clocks running - on a very large board.
But - could it compete with tennis ?
Strangely I don't think such a contest ever evolved.
@playerafar
If you're trying to weakly solve, the three-fold repetition rule can be ignored, but not the 50 move rule. (Did you look at the posts I mentioned and http://galen.metapath.org/egtb50/ ?)
Agreed draws would have to be discounted.
If you are trying to solve (sans "weakly"), the three fold repetition rule can't be ignored. See post #219 here https://www.chess.com/forum/view/general/chess-will-never-be-solved-heres-why?page=11. That would just be Nightmare on Elm Street.
@MARattigan I think we're going to disagree on these points.
Or maybe - it would be about 'weakly' semantics. Would that be worth our time?
I believe the 50 move rule and the three-fold rule can both be ignored much as we'd ignore the situations on the clocks.
They can all be seen as 'additional' and pertaining to games rather than positions.
"Hey I can draw by reaching fifty moves with no captures/pawn motion !"
Is there any situation where the 50 couldn't be ignored ?
If positions exist where the win is impossible because it would take more than 50 ... (I don't know whether there are such positions) -
then they could be 'noted'.
More practical would be to have an 'unclear' category.
Where the computer spits out positions where a 'solution' isn't readily available - and they're instantly assigned to other computers specialized for that.
Then the other computers could further categorize.
Point - you could have all positions 'accounted for' as opposed to 'solved'.
As an intermediate step.
But we don't need to continue in disagreement.
The link I gave gives many positions that are wins under basic rules but not with the 50 move rule in effect. If you go into the https://syzygy-tables.info/ site you can set up positions with up to 7 men on the board and it will show the percentage of positions that are "frustrated wins" - that is positions that can be won under basic rules but can't be won under competition rules with the 50 move rule in effect. The figure assumes the ply count is 0 under competition rules - obviously the figure would rise with the ply count.
But use at least 5 men - that's when the frustrated wins first appear. There are many positions with ply count > 0 with fewer than 5 men that can't be won under competition rules but could be won under basic rules, but these would not be relevant for a weak solution of the competition rules game.
Further: the project of 'solving' all positions regardless of 50 and three-fold and clocks and could the position have legally got there - is itself an 'intermediate step' rather than a 'weak step'.
I think we're talking at cross purposes. For "solving" and "weakly solving" see https://en.wikipedia.org/wiki/Solved_game.
En passant positions possible is 'easy' its just a matter of doubling the numbers ... or more if more than one en passant is theoretically possible.
Castling is tougher though.
Heck, all the pieces would get a good workout, not just the pawns. Blitz on a board like that? It'd be hilarious.
#349
When they agree on a draw, it is because a 3 fold repetition or a stalemate or a dead position, or a 50 moves draw is inevitable. Of these 4 the 3-fold repetition is the most common and the 50 moves rule is the least common.
Recent example:
https://en.chessbase.com/post/tata-steel-chess-2022-r10
I know your L-game where there is a cunning solution. Also antichess is solved. However for chess it is essential that it ends in a finite number of moves and the 3-fold rule ensures a finite albeit huge number.
I don't know where you get your statistics from. How many of the agreed draws in the ICCF tournament you linked to were headed for a draw under the 3-fold repetition rule? But it's academic anyway, for the reason I gave in #327 and also because many could be headed for take your pick.
The L-game is not a cunning solution. It took me two days, but that was only because I was working during the day.
It's certainly not essential for chess. For a weak solution (of either chess) positions that include even a single repetition can be ignored.
"The link I gave gives many positions that are wins under basic rules but not with the 50 move rule in effect."
We don't have to call it 'disagreement'.
We could call it qualification.
The computer looks to see if a mate can be forced or a draw can be forced.
It works backwards from simpler or clearer positions already analyzed - to do so.
By ignoring how many moves left (which pertains to 'game' rather than position) the whole project becomes more feasible.
We could argue forever about 'solved' and 'weakly' - but - I don't have forever.
As to what the Finale of the project might look like ...
could it look something like this?:
For all 20 white first moves in chess - a system of supercomputers has now assigned classifications of 'forced win' for whatever side as the case might be - or 'option to force a draw' in some cases - but then further classifying as to if that option is passed up - what does it mean about the possible black reply moves in terms of options to force draw or force win that white might have ...
Is that what 'solved' is supposed to look like?
Quite daunting. That's with 50 mover and three fold and clocks thrown out and castling and en passant in because every opening move was arrived at working all the way back from all possible endings.
It might leave some illegals still in.
They'd 'immigrate'.
@tygxc
Agreement to draw doesn't always happen when one of the other four situations are 'inevitable'.
Players often agree to a draw anyway.
Plus I think the three-fold can be ignored too. Like the 50-mover.
Would somebody want to factor in the clock situations too?
Somebody's 'down' and the computer has to worry about that too?
Lol !
Again - a position's a position.
The most common draw situation is probably 'dead' position.
But even that takes different forms.
Two kings plus a minor piece.
You can't even set up mate !
On this site - computers arbitrate those as draws.
But King and rook versus King and rook ... No.
Even King and knight versus King and knight ...
a mate position is possible. And bishops of opposites.
Much more difficult for the computer is when there's more material than those - but its a draw because 'neither side can make progress' including by deterrent.
That's going to slow down El Computro !