Chess, a solved game in the future?

Factual Questions
41

[QUOTE=Malacandra]
Sorry, here I understood him to be saying something about checkers, and then saying he didn’t know the first thing about the game. My reading comprehension skills must be not too sharp today, or something.
[/QUOTE]

Ah, I see.

No, I don’t think he said anything in particular about checkers–he was just relying on a very general notion, applicable to all games, that if draws are impossible, then one player must lose. That means any program written to play that game is able to lose–since if it played against itself, it would lose against itself. (As well as winning against itself, of course.)

-FrL-

42

[QUOTE=Trunk]
Isn’t it possible that before any moves are made, white is in zungzwang?
[/QUOTE]

(zugzwang). It’s possible, but in the context of this discussion, that comes close to begging the question. It may be that if chess is solvable, black always wins an error-free game - since there is no do-nothing move; even an opening Knight move changes that Knight’s parity - but we aren’t even close to establishing that.

Okay. Taking the thirty seconds to discover whether checkers can be drawn would have been time well spent, IMO. :slight_smile:

43

[QUOTE=Trunk]
Isn’t it possible that before any moves are made, white is in zungzwang?
[/QUOTE]
Epistemically, yes. And it would mean, of course, that Black wins.

…

Wouldn’t it be a hoot to see two grandmasters sit a board. The clocks are started. The player with white looks over the board, makes a notation on his score sheet, and tips over his king.

:smiley:

44

[QUOTE=Liberal]
Epistemically, yes. And it would mean, of course, that Black wins.

…

Wouldn’t it be a hoot to see two grandmasters sit a board. The clocks are started. The player with white looks over the board, makes a notation on his score sheet, and tips over his king.

:smiley:
[/QUOTE]

Effectively that has been done, though not because chess has been solved. In a fit of pique, Fischer refused to turn up for his second match game against Spassky in 1972. So you could argue that the score sheet for that game reads:

  1. Resigns

…a record that should hold up. :smiley:

45

[QUOTE=Liberal]
Wouldn’t it be a hoot to see two grandmasters sit a board. The clocks are started. The player with white looks over the board, makes a notation on his score sheet, and tips over his king.

:smiley:
[/QUOTE]

Futurama had a bit like this once; two robots sit down at a chessboard, one looks over the starting alignment, says “Mate in 135 moves”, and the other says “I resign”.

46

[QUOTE=Liberal]
According to Wikipedia, the game tree complexity of chess is 10[sup]123[/sup]. If computers can do no better than brute force examination of sequential positions, they ain’t ever going there.
[/QUOTE]

Recently there was available on the web a panoramic representation of sizes/quantities of everything in the universe from 10[sup]-15[/sup] to 10[sup]15[/sup].

Writing a program for exploring all the alleged possible moves in chess being 10[sup]123[/sup] may require more time and computing capacity than currently available.

47

This question makes me wonder if there are other “games” or “problems” that were once predicted to never be solvable that were eventually solved.

I know about the size and time requirements of 6 piece end game solutions, and that business about more moves than atoms in the known universe, etc etc.

But, I can also conceive of almost limitlessly fast computers, and virtually limitless amounts of quickly accessible storage.

My intuition breaks down when trying to attack the one with the other. The old immovable object and unstoppable force conundrum.

I doubt that chess will ever be solved. Certainly not in my lifetime. But that’s far different than the claim that it will never be solved. Maybe it’s just way more likely that the Sun blows up before we make the necessary advances.

48

What about the more popular variants of Chess, such as Chinese and Korean Chess? Are they in the same state of partially-solved as FIDE rules Chess?

49

[QUOTE=Trunk]
Isn’t it possible that before any moves are made, white is in zungzwang?
[/QUOTE]

No.

By definition, to be in zugswang, any move you make must weaken your position.
This definitely does not apply to the opening, where it is easy to improve your position.

Zugswangs are most likely to occur in the ending, where one side is defending.
The simplest would be White King d6, White pawn e6, Black King d8. White to move draws, Black to move loses.

50

[QUOTE=Malacandra]
Effectively that has been done, though not because chess has been solved. In a fit of pique, Fischer refused to turn up for his second match game against Spassky in 1972. So you could argue that the score sheet for that game reads:

  1. Resigns

…a record that should hold up. :smiley:
[/QUOTE]

[boring nerd voice ON]
Actually Fischer didn’t resign - he defaulted.
There was a game in an English tournament (Miles - Reuben if memory serves) which went 1. Draw agreed. :eek:
There was quite a furore about whether this was legal (or moral).
[boring nerd voice OFF]

51

[QUOTE=CJJ*]
Futurama had a bit like this once; two robots sit down at a chessboard, one looks over the starting alignment, says “Mate in 135 moves”, and the other says “I resign”.
[/QUOTE]

Astonishingly you can actually get a situation where two players with access to tablebases are playing and reach a specific position with King, Rook and Bishop v King and 2 Knights.
Player one says “I win in 223 moves” and player 2 agrees. :eek:

(Source: John Nunn’s ‘Secrets of Pawnless Endings’)

52

[QUOTE=glee]
No.

By definition, to be in zugswang, any move you make must weaken your position.
This definitely does not apply to the opening, where it is easy to improve your position.

Zugswangs are most likely to occur in the ending, where one side is defending.
The simplest would be White King d6, White pawn e6, Black King d8. White to move draws, Black to move loses.
[/QUOTE]

I’m not sure about the technicalities of chess terminology, but in the theoretical possibility that black is able to force a win in chess, then wouldn’t the opening be zugzwang for white, in that, if white were able to forfeit its first move, it would be effectively switching the roles of the two players, and thus switching from a forcible loss to a guaranteed win?

Specifically, it would be a case of “reciprocal zugzwang”, I gather, where the first player to move (and thus to essentially assume the role of white, whatever his actual color) hands to the other player the ability to force a win.

53

[QUOTE=Xema]
There’s an interesting - and troubling - corollary to the fact that chess computers now reliably beat humans: The time is now in view when human-human chess matches can’t be reliably fair.

This is because technology for undetectable communication is proceeding apace. All you need is an assistant who uses a good computer program to analyze your game and has a means of passing the recommended moves to you. The players at the board will become proxies for the machines behind the scene.
[/QUOTE]

See Toiletgate ; Indian chess cheat; US chess cheat.

But metal detectors should help…

54

[QUOTE=glee]
No.

By definition, to be in zugswang, any move you make must weaken your position.
This definitely does not apply to the opening, where it is easy to improve your position.

[/quote]

But if it turns out that a complete analysis of chess shows that black can force a win, then wouldn’t every first move from white count, strictly speaking, as weakening his position?

Maybe we should make the notion of “weakening one’s position” more precise. What do you mean by that phrase in your definition of zugzwang?

(I always thought of zugzwang as being the following situation: If it were possible to pass, that would be your best move. On that understanding, white certainly would be in zugzwang if black can actually force a win from the opening position.)

-FrL-

ETA I see Indistinguishable has a post already with substantially similar content to mine. Just acknowledging that.

(Actually, Indistinguishable, that has happened to me a lot since you came around. It’s kind of weird actually.)

55

[QUOTE=Indistinguishable]
I’m not sure about the technicalities of chess terminology, but in the theoretical possibility that black is able to force a win in chess, then wouldn’t the opening be zugzwang for white, in that, if white were able to forfeit its first move, it would be effective switching the colors of the two players, and thus switching from a forcible loss to a guaranteed win?

Specifically, it would be a case of “reciprocal zugzwang”, I gather, where the first player to move (and thus to essentially assume the role of white, whatever his actual color) hands to the other player the ability to force a win.
[/QUOTE]

Your theoretical idea would conceivably work in games like Nim (where the last player to take a piece away loses). Effectively the starting position is lost, provided the second player knows how to play perfectly.

However all the evidence for chess is that moving first as White is a clear advantage.
I think a survey of thousands of GM games gave a 55% advantage to White.
Also in single quickplay game tie-break situations for example, White is given more time, but Black has draw odds (i.e. he wins if the game is drawn).
As I said, since it’s easy to improve your position in the first few moves, I’m certain the possiblity of ultimate zugswang will not apply.

56

[QUOTE=glee]
Your theoretical idea would conceivably work in games like Nim (where the last player to take a piece away loses). Effectively the starting position is lost, provided the second player knows how to play perfectly.

However all the evidence for chess is that moving first as White is a clear advantage.
I think a survey of thousands of GM games gave a 55% advantage to White.
[/QUOTE]

Ah, this explains the confusion. The person you were orginally responding to, who first mentioned zugzwang, was playing jokingly with the off possibility that it turns out black has a forced win rather than white. We all know it appears white has the advantage, but it’s just possible this could turn out ultimately not to be the case.

-FrL-

57

[QUOTE=Indistinguishable]
I’m not sure about the technicalities of chess terminology, but in the theoretical possibility that black is able to force a win in chess, then wouldn’t the opening be zugzwang for white, in that, if white were able to forfeit its first move, it would be effectively switching the roles of the two players, and thus switching from a forcible loss to a guaranteed win?
[/QUOTE]

Not necessarily true.

While I doubt it would turn out to be true, wouldn’t it be possible that the slight difference in starting position between white and black, the transposed king and queen, would be significant?

58

[QUOTE=Frylock]
(I always thought of zugzwang as being the following situation: If it were possible to pass, that would be your best move. On that understanding, white certainly would be in zugzwang if black can actually force a win from the opening position.)
[/QUOTE]
Some writers restrict zugzwang to situations where both players would prefer to pass, and refer to the situation where only one player wishes to pass as a squeeze.

59

[QUOTE=Frylock]
But if it turns out that a complete analysis of chess shows that black can force a win, then wouldn’t every first move from white count, strictly speaking, as weakening his position?

[/quote]

That’s true.
However all the evidence points to White having a clear advantage, and I would guess the likeliest result of solving chess would be 90% draw / 9.9% White wins / 0.1% Black wins.

[QUOTE=Frylock]

Maybe we should make the notion of “weakening one’s position” more precise. What do you mean by that phrase in your definition of zugzwang?

(I always thought of zugzwang as being the following situation: If it were possible to pass, that would be your best move. On that understanding, white certainly would be in zugzwang if black can actually force a win from the opening position.)
-FrL-
[/QUOTE]

Being in zugswang means any move weakens your position fatally, so another way to say it is that your best move is to pass (which of course is not legal in chess!).

I gave one example already:

White King d6, White pawn e6, Black King d8. White to move draws, Black to move loses.
(1. e7+ Ke8 2. Ke6 stalemate; 1. … Ke8 2. e7 Kf7 3. Kd7 and promotes.)

Here’s another:

White King e4, White pawns c4,g4, Black King e6, Black pawns c5,g5. White to move draws, Black to move loses.
(1. Kd3* Ke5 2. Ke3 draw; 1. … Kd6 2. Kf5 wins.)

*not 1. Ke3? Ke5 winning.

You can generate such positions with more material, but it is rare, so it just wouldn’t apply to the starting position.

60

[QUOTE=Garfield226]
Not necessarily true.

While I doubt it would turn out to be true, wouldn’t it be possible that the slight difference in starting position between white and black, the transposed king and queen, would be significant?
[/QUOTE]

It’s hard to see what difference that makes.
I already gave statistics based on thousands of GM games…