Will computers ever solve chess?

Sort:
Avatar of Alltheusernamestaken

The computational power of quantum computers will solve chess. Im guessing in less than 30 years

 

Avatar of sirHandberg
Agreed
Avatar of lfPatriotGames
Alltheusernamestaken wrote:
lfPatriotGames wrote:
Alltheusernamestaken wrote:
ptd570 wrote:

Will there ever be a computer strong enough to solve chess to the point where white uses its half tempo advantage to always beat black no matter what moves black plays (in otherwords the same computer can never win with black even after a thousand random games against itself)

 

I beleive one day there will be a computer so strong and so big that it will solve chess completely but perhaps that is 50 or 100 years off, its possible to solve it but we may never see it even in a 100 years

Black will always be alway to draw/stealmate with perfect play

As far as you dont know, that is.

It's obiovus as you can draw in a losing position, so with perfect play white would keep his edge but it wouldnt be enought to win

It's obvious some losing positions can result in a draw. But it's also obvious some losing positions lose, no matter what. As for perfect play for an entire game, no one knows what that is or has ever seen it. As was pointed out before many times, if there is a result with "perfect play" this topic would not exist, chess would already be solved. So as far as you absolutely dont know, black will always be able to draw with perfect play. 

Think of it this way, white starts with an advantage and with our primitive computers and limited knowledge about chess it will not be for many decades, possibly centuries before whites advantage is turned into a forced win every single game. So as far as I don't know, white will eventually always win with perfect play.

Avatar of ponz111

It is a matter of opinion and evidence if chess is a draw or a win for one side [with perfect play]. So far the players who know most about chess--the grandmasters and supergrandmasters believe chess is a draw with perfect play by both sides. I only know of one grandmaster who MIGHT disagee with this and he was promoting a book.

It is also a matter of opinion and evidence has a perfect game been played? I believe dozens [or more] such games have already been played.

 Also this forum would still be here regardless as some nongrandmasters do not know or believe chess is a draw with perfect play.

Avatar of phillidor5949
staples13 wrote:

Perhaps a more interesting question. Do you think there is any position after both players have made 1 move where one side is lost with pefect play?

The two worst first moves according to Droidfish 1.72 using Stockfish Nine (9):

Depth = 38.

  • 19. [38] -0.29: 1.f3 e5 2.e3 Nc6 3.c4 Nf6 4.Nh3 d5 5.cxd5 Nxd5 6.Nf2 Be6 7.Bb5 Qd7 8.Qa4 Be7 9.Bxc6 bxc6 10.0-0 0-0 11.Na3 f6 12.Rd1 f5 13.d3 f4 14.exf4 exf4 15.Nc4 Bf6 16.Re1 Nb6 17.Qc2 Nxc4 18.dxc4 Rab8 19.Qe4 Rfe8 20.Qxf4 Qf7 21.Be3 Bxb2 22.Qxf7+ Kxf7 23.Rab1 Bxc4 24.Bxa7 Rb7 25.Rxe8 Kxe8;
  • 20. [38] -0.57: 1.g4 d5 2.e3 Nc6 3.d4 e5 4.dxe5 Nxe5 5.h3 h5 6.g5 c6 7.e4 Bb4+ 8.Nc3 Qe7 9.Be3 Bxc3+ 10.bxc3 dxe4 11.Qd4 f6 12.Qxe4 fxg5 13.Qb4 Nf6 14.Qxe7+ Kxe7 15.Bxg5 Kf7 16.Bf4 Re8 17.0-0-0 Ne4 18.Bxe5 Rxe5 19.Bd3 Nxc3 20.Nf3 Nxd1 21.Rxd1 Ke6 22.Bc4+ Kf5 23.Nxe5 Kxe5;

According to Stockfish PA_GTB 010713 64 SSE4.2 g (first 11 lines at depth 40, remainder at depth 39):

 

  • 19. =/+ (-0.47): 1.f3 e5 2.e3 d5 3.d4 exd4 4.exd4 Bd6 5.Be3 Ne7 6.Nc3 0-0 7.Qd2 c6 8.0-0-0 Nd7 9.Kb1 b5 10.Bd3 Nb6 11.Bf4 a5 12.Bxd6 Qxd6 13.Re1 Re8 14.g4 b4 15.Nce2 Nc4 16.Qc1 f5 17.Nh3
  • 20. -/+ (-0.86): 1.g4 e5 2.Bg2 h5 3.gxh5 d5 4.d4 exd4 5.Qxd4 Rxh5 6.Bf3 Nc6 7.Qd1 Re5 8.Bf4 Rf5 9.e3 Nf6 10.Nc3 d4 11.Nce2 Nd5 12.Nxd4 Nxd4 13.Qxd4 Nxf4 14.Qxd8+ Kxd8 15.exf4 Rxf4 16.0-0-0+ Bd6"

 

Cite: http://oeco.hopto.org/mediawiki/index.php/Initial_Position

Both suggest 1.g4 as clearly worst with possibly 1...d5 (or maybe 1...e5) as Black's most promising reply.

Further analysis of Black's replies perhaps using a variety of engines is needed to better answer this question.

 

Avatar of vesna10

idonno

Avatar of vinny-kumar

There are many  draws in world class chess engine matches, one tempo advantage doesn't lead to victory, in fact one tempo  advantage can backfire in many endgame positions. Intentions behind mischievous questions to discredit chess should be exposed.

Avatar of phillidor5949
staples13 wrote:

Perhaps a more interesting question. Do you think there is any position after both players have made 1 move where one side is lost with pefect play?

Or, inverse of #8495 (worst White, best Black), what are Black's most suicidal replies after White's strongest first moves? ( [38] 0.40: 1.d4 ; [38] 0.40: 1.c4 ; [38] 0.34: 1.e4 ; [38] 0.32: 1.Nf3 ; [38] 0.30: 1.e3 ) (best White, worst Black)

Avatar of mybd

I think chess will be "solved" by computers sometime.

Avatar of troy7915

To 8466: Knowing for certain and believing are two different things. The former implies proof, the latter doesn’t.

 

  Two, playing the game with the assumption that White is slightly better before his first move, or that the starting position is more or less equal, has nothing to do with the actual fact of the starting position being equal or not.

  Simply put, there isn’t enough factual evidence (implying best moves, objectively, not based on past, limited experience suggesting either way), so only a fool would claim to know the correct evaluation of the starting position, based on the way they (others—GMs—or himself) have been conditioned to play the game.

 

  Then again, this is coming from a player who is not aware of the psychological factor present at over-the-board games. Perhaps he needs more reading time...if he lacks keen observation, that is.

Avatar of cami1o2999
Who would win between Skynet and the Matrix?
Avatar of pm12345
cami1o2999 wrote:
Who would win between Skynet and the Matrix?

Matrix! Skynet is like the beta version of Matrix.

Avatar of KZNinjago
It depends , they don’t really play unless they aren’t backed up with human knowledge
Avatar of Alltheusernamestaken
lfPatriotGames wrote:
Alltheusernamestaken wrote:
lfPatriotGames wrote:
Alltheusernamestaken wrote:
ptd570 wrote:

Will there ever be a computer strong enough to solve chess to the point where white uses its half tempo advantage to always beat black no matter what moves black plays (in otherwords the same computer can never win with black even after a thousand random games against itself)

 

I beleive one day there will be a computer so strong and so big that it will solve chess completely but perhaps that is 50 or 100 years off, its possible to solve it but we may never see it even in a 100 years

Black will always be alway to draw/stealmate with perfect play

As far as you dont know, that is.

It's obiovus as you can draw in a losing position, so with perfect play white would keep his edge but it wouldnt be enought to win

It's obvious some losing positions can result in a draw. But it's also obvious some losing positions lose, no matter what. As for perfect play for an entire game, no one knows what that is or has ever seen it. As was pointed out before many times, if there is a result with "perfect play" this topic would not exist, chess would already be solved. So as far as you absolutely dont know, black will always be able to draw with perfect play. 

Think of it this way, white starts with an advantage and with our primitive computers and limited knowledge about chess it will not be for many decades, possibly centuries before whites advantage is turned into a forced win every single game. So as far as I don't know, white will eventually always win with perfect play.

You never reach those positins with a very very small advantage, black would always draw

Avatar of Alltheusernamestaken
lfPatriotGames wrote:
Alltheusernamestaken wrote:
lfPatriotGames wrote:
Alltheusernamestaken wrote:
ptd570 wrote:

Will there ever be a computer strong enough to solve chess to the point where white uses its half tempo advantage to always beat black no matter what moves black plays (in otherwords the same computer can never win with black even after a thousand random games against itself)

 

I beleive one day there will be a computer so strong and so big that it will solve chess completely but perhaps that is 50 or 100 years off, its possible to solve it but we may never see it even in a 100 years

Black will always be alway to draw/stealmate with perfect play

As far as you dont know, that is.

It's obiovus as you can draw in a losing position, so with perfect play white would keep his edge but it wouldnt be enought to win

It's obvious some losing positions can result in a draw. But it's also obvious some losing positions lose, no matter what. As for perfect play for an entire game, no one knows what that is or has ever seen it. As was pointed out before many times, if there is a result with "perfect play" this topic would not exist, chess would already be solved. So as far as you absolutely dont know, black will always be able to draw with perfect play. 

Think of it this way, white starts with an advantage and with our primitive computers and limited knowledge about chess it will not be for many decades, possibly centuries before whites advantage is turned into a forced win every single game. So as far as I don't know, white will eventually always win with perfect play.

If a lost positions can't be drawn them it means white has a good advantage and with perfect play white never gets such an advantage

Avatar of I3otOlmo
Is this the longest running thread i chess.com history?
Avatar of josephyossi

no

 

Avatar of josephyossi

destinys thread was over 400000

Avatar of josephyossi

or 40000

Avatar of pawn8888

I think that if you could convert chess into numbers it would just be a matter of doing the whole game in a sort of a plus and minus formula. Since it reverts to numbers means it has to be solved or solvable, since the numbers will reach an end eventually.