Will computers ever solve chess?

Sort:
troy7915
s23bog wrote:
troy7915 escribió:
ponz111 wrote:

overconfidence but that is not the opposite of confidence.

 

  Correct. What IS the opposite of confidence?

Insecurity.

 

  Correct.

ponz111
lfPatriotGames wrote:
s23bog wrote:

"Solve" means determine if the game is a forced win for either side, or a draw, with best play.

Wouldn't the terms "solve" and "best play" be the same thing? It seems like it's unnecessary to say solve means figuring out the game with best play because there is no other way to figure it out. It seems unlikely chess will be solved with ordinary play, good play, or even great play. The moment best play ever happens for an entire game, chess will be solved at that same moment.

 

If what you say is correct--then chess has already been "solved".

lfPatriotGames
ponz111 wrote:
lfPatriotGames wrote:
s23bog wrote:

"Solve" means determine if the game is a forced win for either side, or a draw, with best play.

Wouldn't the terms "solve" and "best play" be the same thing? It seems like it's unnecessary to say solve means figuring out the game with best play because there is no other way to figure it out. It seems unlikely chess will be solved with ordinary play, good play, or even great play. The moment best play ever happens for an entire game, chess will be solved at that same moment.

 

If what you say is correct--then chess has already been "solved".

If chess has already been solved then "best play" has already happened. I dont see how thats possible since all the time computers and people come up with new and better ideas about how to play chess. What we think is "best play" usually gets improved on a few years later. It seems like both computers and people are getting better. Which to me means best play hasn't happened yet. Which means chess hasn't been solved yet. I just think they will both happen simultaneously, if they ever happen at all.

ponz111

It is the overwhelming opinion of the strongest players that chess is a draw with best play by both sides.

If, in fact, chess is a draw then there would be millions of possible "perfect games". There would be enough leeway for making a perfect game. [by this i mean a game where neither side made an error which would change the theoretical result of the game.]

If, however, chess is a win for one side or the other when neither player made a mistake--then it would seem that there would be a much smaller number of "perfect games" [as compared to the number of perfect games if, in fact, chess is a draw]

It is not at all hard to make the best possible move.  I have shown several diagrams to prove this. Even when i was 8 years old--i know i made some perfect moves. When we learn how to win with king and rook vs a lone king--we are making perfect moves.  There are often dozens of different sequences leading to a forced win when we have king and rook vs a lone king. So, just about every  chess player reading this--has played some perfect moves!? Smile  

ponz111

It is not necessary to solve chess to play a perfect game. A perfect game can possibly happen by chance.

Or if a player has a high rating [say 2500+]  and is playing someone with an approximately equal rating--they may happen to play a perfect game without even realizing it.

Or, if the game is very short, it is quite possible to play a perfect game.

Almost all of us have played a perfect endgame.  Almost all of us have played maybe hundreds of perfect endgames. 

Perfect games happen and are not so very rare. Now proving you have played a perfect game is not so easy. I could try but there are too many players who will not accept a perfect game has been played until chess is "solved" [which will not happen in our lifetimes, if at all...]   

ponz111
s23bog wrote:

That might be a worthwhile statistic to monitor.  How many positions have been "solved"?  Solving the game is down the road.

Of course the game of chess  has not been solved. I never said or implied the game of chess has been solved.

However there have been millions/maybe billions of positions which have been solved. Way too many to monitor!

I have played thousands of positions which i have solved.

lfPatriotGames
ponz111 wrote:
lfPatriotGames wrote:
s23bog wrote:

"Solve" means determine if the game is a forced win for either side, or a draw, with best play.

Wouldn't the terms "solve" and "best play" be the same thing? It seems like it's unnecessary to say solve means figuring out the game with best play because there is no other way to figure it out. It seems unlikely chess will be solved with ordinary play, good play, or even great play. The moment best play ever happens for an entire game, chess will be solved at that same moment.

 

If what you say is correct--then chess has already been "solved".

So you are saying, or implying, that chess has already been solved. Because if what I say is correct, that "best play" can only happen to solve chess, are you saying best play has already happened for an entire game? 

I dont think it has. I think that many millions of chess positions or puzzles have been solved because best play exists in those. But I dont think best play has ever happened in an entire game of chess because both computers and people are always getting better. Something better than best would have to exist for that to happen in your scenario. I agree amazing or great play has happened for an entire game, but not best. I dont think there has ever been a game where every move was perfect. If it ever did happen, no one ever knew it happened.

ScootaChess

The question is why would anyone want to? 

ponz111
lfPatriotGames wrote:
ponz111 wrote:
lfPatriotGames wrote:
s23bog wrote:

"Solve" means determine if the game is a forced win for either side, or a draw, with best play.

Wouldn't the terms "solve" and "best play" be the same thing? It seems like it's unnecessary to say solve means figuring out the game with best play because there is no other way to figure it out. It seems unlikely chess will be solved with ordinary play, good play, or even great play. The moment best play ever happens for an entire game, chess will be solved at that same moment.

 

If what you say is correct--then chess has already been "solved".

So you are saying, or implying, that chess has already been solved. NO! i am not saying or implying that chess has been solved. I have repeatedly said that chess has NOT been solved and probably will not be solved for a long time [if ever] 

 

Because if what I say is correct, that "best play" can only happen to solve chess, are you saying best play has already happened for an entire game? The problem is what you are saying is incorrect.

I dont think it has. There is a difference in saying that chess has been solved and in saying a perfect game has been played. The two are not the same thing. I have repeatedly mentioned that a perfect game can be played quite by accident without ever solving chess. I have also mentioned other ways that a perfect game can be played without ever solving chess.

 

I think that many millions of chess positions or puzzles have been solved because best play exists in those. This is one thing we agree on.

But I dont think best play has ever happened in an entire game of chess because both computers and people are always getting better. Here is where we disagree. As mentioned a perfect game can be played quite by accident. Chess players do not have to be the best chess player on the planet to play a perfect game.

If chess is a draw with best play [because the White advantage is not enough to force a win] then there probably have been thousands of perfect games already played. I believe chess is a draw with perfect play by both sides and I am sure thousands of perfect games have already been played. 

 

Something better than best would have to exist for that to happen in your scenario. What senario?  I believe [as do the vast majority of the strongest players] that chess is a draw with perfect play by both side. By "perfect play" i mean a game where neither side makes an  error/mistake which would change the theoretical result of the game. Example--if we give the hypothetical that chess is a draw with perfect play--that means the game of chess will always be drawn if neither side makes a mistake which would change the theoretical result of the game.


 

I agree amazing or great play has happened for an entire game, but not best. If a game is played where neither side makes an error which would change the theoretical result of the game--then a perfect game has been played and this has happened thousands of times.

I dont think there has ever been a game where every move was perfect. If it ever did happen, no one ever knew it happened. Disagree, I think there have been thousands of such games.  And most very strong players know that games with no errors by either side have been played and they also know that all such games have ended in draws.

HalmaDori

wink.png

ponz111

It is also my opinion that the strongest GMs with a 2700 or above rating have a very good understanding of chess. Yes, the best chess engines can beat the 2700+ players but this is mainly because the chess engines can calculate so much faster.  If the current top human players had sufficient time to calculate -- they would come close to holding their own against the engines.

The main reason you see the very top players losing a game is simply that they have to play under severe time limits and thus make errors.

However with sufficient time to calculate [this will never be allowed to happen] you would see even more draws between top players than you see today...

[because, after all, chess is a draw when played under the best conditions Laughing]

troy7915
ponz111 wrote:
s23bog wrote:

That might be a worthwhile statistic to monitor.  How many positions have been "solved"?  Solving the game is down the road.

Of course the game of chess  has not been solved. I never said or implied the game of chess has been solved.

 

 

 

  Then you have no idea what a perfect move is, and so what perfect games are.

ponz111
troy7915 wrote:
ponz111 wrote:
s23bog wrote:

That might be a worthwhile statistic to monitor.  How many positions have been "solved"?  Solving the game is down the road.

Of course the game of chess  has not been solved. I never said or implied the game of chess has been solved.

 

 

 

  Then you have no idea what a perfect move is, and so what perfect games are.

You are quite wrong.  A perfect game is a game played without any error or errors which would change the theoretical result of the game.

A perfect move is a move [or moves]  which are best for the position. A perfect move would lead to a draw or a win, depending on the position.

Here, I will give you an example of a perfect move--White to play and make a "perfect move".



lfPatriotGames
ponz111 wrote:
troy7915 wrote:
ponz111 wrote:
s23bog wrote:

That might be a worthwhile statistic to monitor.  How many positions have been "solved"?  Solving the game is down the road.

Of course the game of chess  has not been solved. I never said or implied the game of chess has been solved.

 

 

 

  Then you have no idea what a perfect move is, and so what perfect games are.

You are quite wrong.  A perfect game is a game played without any error or errors which would change the theoretical result of the game.

A perfect move is a move [or moves]  which are best for the position. A perfect move would lead to a draw or a win, depending on the position.

Here, I will give you an example of a perfect move--White to play and make a "perfect move".

 



You are right, that is a perfect move.  Perfect and best are the same thing. No other move or moves would win in one move. But Troy is also right when he says you have no idea what a perfect game is. Nobody does. A perfect game means a game where no mistakes are made. I dont see what "theoretical" has to do with anything. Pefect or "best play" would be a game where the very first move is THE best move possible. The only way to know that is to know what the best continuation is. It seems to me that if that ever happened, then chess would be solved. I dont see how perfect play for an entire game, best play for an entire game, and solved game can be different from one another.

Anyone who thinks a perfect game has been achieved would have no problem easily beating the worlds best humans and the worlds best computers because BOTH play chess far less than perfectly. If they did play perfectly, they would never increase in rating or ability. 

Icecream4crow

I don't think they don't have to solve it-it's  like tic tac toe, best play on both sides results in a draw-

troy7915
ponz111 wrote:
troy7915 wrote:
ponz111 wrote:
s23bog wrote:

That might be a worthwhile statistic to monitor.  How many positions have been "solved"?  Solving the game is down the road.

Of course the game of chess  has not been solved. I never said or implied the game of chess has been solved.

 

 

 

  Then you have no idea what a perfect move is, and so what perfect games are.

You are quite wrong.  A perfect game is a game played without any error or errors which would change the theoretical result of the game.

A perfect move is a move [or moves]  which are best for the position. A perfect move would lead to a draw or a win, depending on the position.

Here, I will give you an example of a perfect move--White to play and make a "perfect move".



 

  You. misunderstood again: the goal was to show you that you have no idea what a perfect game is, using the fact that you have no idea what perfect moves are. That means, from the beginning of the game. Not from some random position, where many imperfect moves have already been played. But from the very beginning. And so it is.

 

  I have eliminated mating moves long time ago, and having a good memory I’m not going back to do it again. 

ponz111

PatriotGames

Where  you are going wrong is to assume that there is only one first move which leads to a perfect game. Most of the first moves lead to a perfect game.

In most positions there are numerous best/perfect moves. Moves that will lead to the theoretical result.

You also go wrong in  thinking that someone who thinks a perfect game has been achieved would have no problem beating the best humans and the world's best computers.  One thing wrong with that assumption is that with best play or perfect play--chess is a draw--not a win.

Another thing wrong with that assumption is that one does not have to be the very best player in the world to know that a perfect game can be and has been played.

There are often several moves in any given position  which will lead to  the perfect/theoretical result.  Here i will give one such position where there are several moves which will lead to the best result--and it is a draw-not a win... 



ponz111

All of the most played first moves lead to a draw. The include 1. e4

1. d4  1. c4   1. Nf3   1. e3   1. g3  [there are more...]

troy7915

The problem is that ‘chess is a draw, with best play’ is not a fact, stated from the present viewpoint. It may or may not be, but at this point it’s just a guess.

lfPatriotGames
ponz111 wrote:

PatriotGames

Where  you are going wrong is to assume that there is only one first move which leads to a perfect game. Most of the first moves lead to a perfect game.

In most positions there are numerous best/perfect moves. Moves that will lead to the theoretical result.

You also go wrong in  thinking that someone who thinks a perfect game has been achieved would have no problem beating the best humans and the world's best computers.  One thing wrong with that assumption is that with best play or perfect play--chess is a draw--not a win.

Another thing wrong with that assumption is that one does not have to be the very best player in the world to know that a perfect game can be and has been played.

There are often several moves in any given position  which will lead to  the perfect/theoretical result.  Here i will give one such position where there are several moves which will lead to the best result--and it is a draw-not a win... 

 



I might be wrong in assuming there is only one first move which leads to a perfect game. But I might also be right. You seem to think it must be one way, and not possibly the other even though neither of us know for sure.

I've already agreed, and I'm sure pretty much every chess player agrees, that when it comes to endgame positions or even some middle game positions there can in fact be only one perfect or best move. But to solve chess I think we would have to start from the most important position, the beginning. 

And still, if someone believes perfection has already been achieved, all that person would have to do to easily beat the worlds best humans and computers is duplicate that perfect game. Because both humans and computers play far below perfection, they lose quite often which is far less than perfect. Perfect beats anything less than perfect.

Assuming chess is a draw with perfect play makes absolutely no sense not only because it's an assumption, but also because common sense tells us knowledge and ability keeps improving when it comes to playing chess. So what might seem like perfection one year is proven to be less than perfection the next year. As long as there is improvement, perfection (or best play) has not happened yet.