Solving chess? With no BS. (moderated)

Sort:
DiogenesDue
Redgreenorangeyellow wrote:

Is three fold repetition also an acceptable way for a game to end? Coupled with the absence of the 50 move rule, both players could simply keep moving their piece(s) continually without making any progress. 

Well, yes, but for the for purposes of solving chess via tablebases that is irrelevant, because they work backwards from mate. 

When I said agreed draws cannot be evidence, I did not mean draws that can be claimed via perpetual check or by repetition ("look, it's a draw, I will check you forever...want to play it out?").  Also, a pawnlocked position is a draw, along the same lines as insufficient material, because checkmate has become impossible.

DiogenesDue
BadBishopJones3 wrote:

Or perhaps the opposite can be proven. For every example where Black mimics White an advantage is eventually found.

This won't work in chess, specifically because of check.  White can always force black to break the "mimic" pattern, to black's detriment.  Chess is actually extremely well "designed" (by evolution as much as anything)...which is why the stalemate and en passant threads are so annoying wink.png.  

"Look, if you understood the game, you wouldn't even ask..."

There are many games created that cannot answer this simple obstacle "what if the opponent just copies everything?" and they never go very far as a result.

blueemu
btickler wrote:

Also, a pawnlocked position is a draw, along the same lines as insufficient material, because checkmate has become impossible.

Correct. Rule 9.6

9.6 The game is drawn when a position is reached from which a checkmate cannot occur by
any possible series of legal moves. This immediately ends the game, provided that the
move producing this position was legal.

DiogenesDue
blueemu wrote:
btickler wrote:

Also, a pawnlocked position is a draw, along the same lines as insufficient material, because checkmate has become impossible.

Correct. Rule 9.6

9.6 The game is drawn when a position is reached from which a checkmate cannot occur by
any possible series of legal moves. This immediately ends the game, provided that the
move producing this position was legal.

Also I want to clarify that when I say "pawnlocked" here I mean what I would call a true/hard pawnlock, not a position where there's a break in the pawn wall but it's too disadvantageous to go through it, which I would call a "soft pawnlock".

One of the first votechess games I played here was a public game that I joined late, and my team was just in the process of having blundered and losing the exchange the upcoming move.  I saw that all the pawns were still on the board and that the position was favorable, and I pushed them to close the position and force a draw, in a way that the opposing team would not immediately realize what were going for.  One of my most satisfying games of Votechess, despite not winning.

Edit:  Found the game, from 2013...https://www.chess.com/votechess/game/37038

MARattigan
btickler wrote:
Redgreenorangeyellow wrote:

Is three fold repetition also an acceptable way for a game to end? Coupled with the absence of the 50 move rule, both players could simply keep moving their piece(s) continually without making any progress. 

Well, yes, I guess for purposes of solving chess that does have to be in the basic rules, although you could probably buy a cheap chess set and their rules book would not cover 3-fold repetition.  When I said agreed draws cannot be evidence, I did not mean draws that can be claimed via perpetual check or by repetition ("look, it's a draw, I will check you forever...want to play it out?").  Also, a pawnlocked position is a draw, along the same lines as insufficient material, because checkmate has become impossible.

I don't think it does need to be in the rules for the purposes of solving chess and it's not in FIDE's basic rules.

To determine the correct outcome of a game of chess you need to know only that there is a forced win for one side or other or no such forced win. If there is a forced win with even a single repetition there is necessarily a shorter forced win with no repetitions. The process for generating Nalimov tablebases takes no account of the rule and given enough time would result in a solution of chess. The draws by repetition will be shown to be draws by not appearing in the final version..

Steven-ODonoghue
blueemu wrote:
btickler wrote:

Also, a pawnlocked position is a draw, along the same lines as insufficient material, because checkmate has become impossible.

Correct. Rule 9.6

9.6 The game is drawn when a position is reached from which a checkmate cannot occur by
any possible series of legal moves. This immediately ends the game, provided that the
move producing this position was legal.

Which would then mean that in a position like this, white cannot legally play Kxa8 because the game is over. And in a rated blitz event, he could theoretically lose the game by playing Kxa8

 

MARattigan
Steven-ODonoghue wrote:
...

Which would then mean that in a position like this, white cannot legally play Kxa8 because the game is over. And in a rated blitz event, he could theoretically lose the game by playing Kxa8

 

Think that could do with some working on.

 

Ixneilosophye

I am of the opinion chess has been solved insofar as it ever will be. We know white or black may win with a modicum of human subjectivism; moving first is not an advantage but a matter of procedure. If we want some computer's opinion (or analysis) we take the fun right out of the game and certainly any human element. 

 

I do think chess is a perfect game, especially after one takes a chill pill. Oh. Hey. 

MARattigan
blueemu wrote:

...

Correct. Rule 9.6

9.6 The game is drawn when a position is reached from which a checkmate cannot occur by
any possible series of legal moves. This immediately ends the game, provided that the
move producing this position was legal.

Actually 5.2.2 these days - it's in the basic rules. In fact it was since its introduction. The basic rules had a forward reference to 9.6. 

DiogenesDue
MARattigan wrote:

I don't think it does need to be in the rules for the purposes of solving chess and it's not in FIDE's basic rules.

To determine the correct outcome of a game of chess you need to know only that there is a forced win for one side or other or no such forced win. If there is a forced win with even a single repetition there is necessarily a shorter forced win with no repetitions. The process for generating Nalimov tablebases takes no account of the rule and given enough time would result in a solution of chess. The draws by repetition will be shown to be draws by not appearing in the final version..

Well, there's immediately forced, and there's positionally "forced".  There has to be some mechanism to discard or move on from positions where the best logical (but not forced) move for both sides is to repeat if the 50 move rule is not being followed (which it isn't, in many tablebases).

MARattigan

If the best move for both sides is to repeat that means the position is (positionally) drawn. It will not appear in any Nalimov table. (Nalimov ignores the 50 move rule.)

DiogenesDue
MARattigan wrote:

If the best move for both sides is to repeat that means the position is (positionally) drawn. It will not appear in any Nalimov table. (Nalimov ignores the 50 move rule.)

Well, okay, I see what you are saying, and yes, it won't really "appear" in the tablebase because the tablebase is constructed backwards from checkmate, but its omission is effectively showing it to be a forced draw in the sense that if you go to a website using the tablebase and plug in that PGN it will tell you *something* wink.png.

JackRoach

There is no way chess will be solved... (opinion: but 99.9999999999999999% likely to be true.)

DiogenesDue

Someone mentioned to me that the link to the Shannon PDF was broken...fixed now, if anyone wanted to read it and could not get there.

Redgreenorangeyellow
btickler wrote:
Redgreenorangeyellow wrote:

Is three fold repetition also an acceptable way for a game to end? Coupled with the absence of the 50 move rule, both players could simply keep moving their piece(s) continually without making any progress. 

Well, yes, I guess for purposes of solving chess that does have to be in the basic rules, although you could probably buy a cheap chess set and their rules book would not cover 3-fold repetition.  When I said agreed draws cannot be evidence, I did not mean draws that can be claimed via perpetual check or by repetition ("look, it's a draw, I will check you forever...want to play it out?").  Also, a pawnlocked position is a draw, along the same lines as insufficient material, because checkmate has become impossible.

I mean maybe you should include standard tournament rules for this. There are far too many possible games without the 50 move rules. 

DiogenesDue
Redgreenorangeyellow wrote:

I mean maybe you should include standard tournament rules for this. There are far too many possible games without the 50 move rules. 

There are far too many games either way happy.png.  So it's kind of a moot point.  Why not have a tablebase that can tell you "mate in 127"?

StormCentre3
btickler wrote:
BadBishopJones3 wrote:

Or perhaps the opposite can be proven. For every example where Black mimics White an advantage is eventually found.

This won't work in chess, specifically because of check.  White can always force black to break the "mimic" pattern, to black's detriment.  Chess is actually extremely well "designed" (by evolution as much as anything)...which is why the stalemate and en passant threads are so annoying .  

"Look, if you understood the game, you wouldn't even ask..."

There are many games created that cannot answer this simple obstacle "what if the opponent just copies everything?" and they never go very far as a result.

Ahhh.... 

but btickler - you are guilty of the very thing you are trying to avoid !!

That of establishing intuition as being true. 
My point is it needs to be  proven. This would be but a 1st step. 
Take for example- The Collatz Conjecture. What is intuitively known to be true - remains unsolved. 

It looks like a simple, innocuous question, but that’s what makes it special. Why is such a basic question so hard to answer? It serves as a benchmark for our understanding; once we solve it, then we can proceed to much more complicated matters.

 

To begin the quest of solving chess ... simple matters require attention. 
If these simpler conjectures can not be proven- then I’d suggest proving a chess result to be a futile endeavor.

 

DiogenesDue
BadBishopJones3 wrote:

Ahhh.... 

but btickler - you are guilty of the very thing you are trying to avoid !!

That of establishing intuition as being true. 
My point is it needs to be  proven. This would be but a 1st step. 
Take for example- The Collatz Conjecture. What is intuitively known to be true - remains unsolved. 

It looks like a simple, innocuous question, but that’s what makes it special. Why is such a basic question so hard to answer? It serves as a benchmark for our understanding; once we solve it, then we can proceed to much more complicated matters.

 

To begin the quest of solving chess ... simple matters require attention. 
If these simpler conjectures can not be proven- then I’d suggest proving a chess result to be a futile endeavor.

You can't mimic white's game as black in a chess game.  It fails at the first check by white.  Not sure what there is to prove there?  I consider it proven with a moment's reflection (and that reflection for me was at the age of 7, beginning chess players learn this lesson very quickly).  Someone just posted a Seirawan video that covers how he also "discovered" it's a failing strategy while starting out in chess.

MARattigan

3.QxQ is also hard to mimic.

Ixneilosophye
JackRoach wrote:

There is no way chess will be solved... (opinion: but 99.9999999999999999% likely to be true.)

There are many responses here. Some are responses and others are RESPONSES. I will not go out on a limb and say which one this MAYBE. 

 

My opinion is that chess is, in fact, an ungame.