Chess will never be solved, here's why

What sparked the idea to think about this in your mind?

@7722

"What sparked the idea to think about this in your mind?"

The statement by GM Sveshnikov sparked the idea to think about this in my mind.

'Give me five years, good assistants and the latest computers
- I will bring all openings to technical endgames'

At first I found it surprising, but after checking facts and figures I found he was right.
Of the 10^44 legal positions only 10^17 are relevant to weakly solving Chess.
3 cloud engines of 10^9 nodes/s guided by 3 grandmasters can do that in 5 years.

Good God, wtf 600 plus new posts, you know any interesting groups i can debate with?

no, get out while u still can

FIGHT FOR YOUR LIFE YEAHHHHHHH

ILL HIT A KNOCKOUT PUNCH WITH DIS HEARTFELT SONG

shredding guitar solo

@7730

"No one is working on this simple project." ++ I do not know, maybe somebody is.

"And no one ever will" ++ Depends on funding: 3 million $ to rent 3 cloud engines and hire 3 (ICCF) (grand)masters during 5 years.

"you could only look at 7 moves or 14 plies"
++ No, calculate all the way until the 7-men endgame table base. 

"946,707,780,000,000,000 = 3 cloud engines of 10^9 NPS x 5 years."
++ 10^9 nodes/s/engine * 3 engines * 3600 s/h * 24 h/d * 365.25 d/a * 5 a = 4.73364e+17
That is more than the 10^17 positions relevant to weakly solving Chess.

I was nonplussed, as you obviously also were, by @tygxc's, "It improves on the 'at least' and 'for both players' with the wording 'against any opposition'", given that the phrases replaced have precious little to do with the replacement. 

I agree the phrase "against any opposition" would be superfluous in Allis' definition, because, "a strategy has been determined to obtain ... for both players ... ", can only sensibly be taken to mean that.

If it were omitted in Prof. van den Herik's version it would leave open the possiblility that a solution could be a strategy to be followed jointly by the players. That is because he fails to say who the strategy is for. If the game-theoretic value of the starting position were "draw", the strategy, "White offers a draw then Black accepts" would count as a valid solution in that case. That wouldn't be a reasonable interpretation as it actually reads because a joint strategy could not be "against any opposition".

On the other hand, including the phrase "against any opposition" in Prof. van den Herik's version doesn't completely cure the problem. He still doesn't say who the solution is for.

If the initial position happens to be a win for White (in some zero sum version of chess) and all White's winning moves result in positions that are selfmates for Black, then the strategy for Black of playing the appropriate selfmate arrives at the game-theoretic value and would be a solution according to Prof. van den Herik's definition. But Black doesn't have to follow the strategy and no strategy is determined for White in that case so it wouldn't be a solution according to anyone else.

No harm in including the phrase "against any opposition" either way. We can laugh even louder when @tygxc can't understand it even when it's spelt out. I included it in my version, which I think is clearer than either on that point.

I agree that "at least" added something (essential) to the meaning, but there is no reference in either of the definitions to optimal play, so the rest of your sentence is misguided.

According to Prof. van den Herik's definitions a strategy for a player that achieves better than the game-theoretic value against some opposition and the game-theoretic value for the rest would not count as a solution (or part of a solution).

Indeed it is not clear that any version of chess, even amended to give a definite yield, actually has a solution according to Prof. van den Herik's definitions. 

If the initial position happens to be a draw (in some zero sum version of chess) and also a selfmate for White and all White's first moves that draw result in positions that are selfmates for Black, then it's clear that no strategy for either player could achieve the game-theoretic value against any opposition.

Illegal positions legitimately occur all the time in games of chess if the FIDE definition of "illegal position" is accepted.

3.10.3
A position is illegal when it cannot have been reached by any series of legal moves.

A position in which the pieces are in the places shown below cannot have been reached by a series of legal moves, because all pieces in the initial position occupy squares and any legal move from such a position results in a position in which the same is true. Such a position is therefore illegal according to FIDE's definition, but can legitimately occur.

 Here, by "position" I mean simply "situation occurring", and by "chess position" I mean "situation occurring in a game of chess", which is the sense, I believe, in which it is used in the FIDE laws.

What you can say is that positions in which the pieces are in places they couldn't legitimately be during a chess game are not chess positions, but I do already take note of that, indeed I've argued the point at length in an "illegal position contest" thread.

When it comes to the definitions in question we cannot take into account art. 3.10.3 because the definitions are intended to apply also to games other than chess, so the FIDE laws are irrelevant. In combinatorial game theory the term "position" would normally mean the initial position and the positions after actions referred to in the rules are taken, so in that case I would agree that the epithet "legal" is superfluous.

It's inclusion does no harm and serves to emphasise that the positions referred to are not restricted to those arrived at by perfect play from the initial position.

I think you are completely wrong to think it's just as well to do away with the reference to "reasonable resources". That is obviously what the thread is about.

If the FIDE laws were amended to make it a zero sum game and dispense with imponderables such as clocks and arbiters in the competition rules game, we already have solutions under the definitions that omit that proviso and have had since the game was invented. I sketched a solution in #7717. I used the word "timely" to express the same idea in my version.

@tygxc is correct in calling the phrase vague, but omits to mention that Allis expands on that point (presumably because it would emphasise that his preferred definition means he wouldn't need his 3 supercomputers, his 7 maids with 7 mops or any sponsorship).

It has to be acknowledged that the term still remains vague with Allis' further comments, as does my term "timely".

The fact is that there are different meanings to "solution" depending on the resources available, so a definition of "solved" effectively needs as a parameter the available resources, one of which is the time available to apply the strategy or strategies the solution provides in the situation for which it is to be used.

A solution of checkers that takes an average of two minutes to produce a recommended move is not a solution of 3 min. bullet checkers and the 7 man Syzygy tables are not a solution of 7 man chess if you have no internet connection and only a ZX80.  

You gotta remember, chess has puzzles in it but is not a puzzle in itself, isn't the goal just to play and have fun?

Chess will be solved eventually.

What is the name of the chess font in this picture? I like it

This is almost as true as the original post

According to the trolls here:
Prof. van den Herik does not know about game theory,
GM Sveshnikov did not know about chess analysis,
and tygxc does not know about either...

Checking in. How is everyone doing? Has everyone calmed down? 

Suggesting that Prof. van den Herik had not had his morning coffee when he included your quoted definitions in his paper is hardly saying he doesn't know about game theory. He is in fact eminent in that sphere.

Similarly suggesting that Sveshnikov had taken something stronger than coffee when he made the remarks you quote does not detract from his abilities in analysing chess at a practical level. He was reputed to have a considerable talent for that.

We can agree on your last point.  

https://www.ctan.org/pkg/skaknew 

I mostly use exactly the same for figurines (white for White's moves and black for Black's.)

yes, i hope so only one way to find out

Thank you very much