It is currently Sat Nov 30, 2024 10:18 pm

All times are UTC - 6 hours [ DST ]




Post new topic Reply to topic  [ 19 posts ] 
Author Message
PostPosted: Sun Jan 29, 2017 3:17 pm 
Offline
Member
User avatar

Joined: Sep 12, 2015
Posts: 87
Identity: girl
so my first question is, given infinite time to consider each of its moves, could a regular modern-day computer brute-force a solution to chess?

My next question depends on the answer to the first one:

if no:

If you created the most efficient possible computer using all of the material in the universe, would that computer be able to brute-force a solution to chess?

if yes:

how long would it take?


Like this post
Top
 Profile  
 
PostPosted: Sun Jan 29, 2017 3:22 pm 
Offline
Member
User avatar

Joined: Sep 12, 2015
Posts: 87
Identity: girl
hi door (dooooooooooor)


Like this post
Top
 Profile  
 
PostPosted: Sun Jan 29, 2017 5:01 pm 
Offline
Member
User avatar

Joined: Oct 30, 2013
Posts: 7305
Location: England
What exactly do you mean by a 'solution' to chess? I assume you mean a strategy by which it could win every game? What is the problem you are trying to brute force here?

_________________
Welcome! I'm Garren and I'll be your designated villain for the evening.


Like this post
Top
 Profile  
 
PostPosted: Sun Jan 29, 2017 5:11 pm 
Offline
Member
User avatar

Joined: Sep 12, 2015
Posts: 87
Identity: girl
Yeah, perfect play. Win or draw with 100% certainty on black.

Bruteforce in this case would mean that the decision to make each move comes from manually mapping out every possible chain of responses to any possible move, and not by applying some complicated mathmematical solution, should one exist.


Like this post
Top
 Profile  
 
PostPosted: Sun Jan 29, 2017 9:33 pm 
Offline
Conqueror of Eldangard
User avatar

Joined: Sep 25, 2013
Posts: 14140
Location: Kamloops, BC
Identity: Male
People have been trying to do it for a while now. That would indicate that at least, nobody's found out it's impossible. I think I may have heard once that it is in fact possible.
Current computers, even very big ones need to take a stupidly long time to solve the problem. Theoretically, even a very weak computer could do it given enough time.

_________________
Cato wrote:
CotW is a method for ranking cards in increasing order of printability.

*"To YMTC it up" means to design cards that have value mostly from a design perspective. i.e. you would put them in a case under glass in your living room and visitors could remark upon the wonderful design principles, with nobody ever worring if the cards are annoying/pointless/confusing in actual play

TPrizesW
TPortfolioW


Like this post
Top
 Profile  
 
PostPosted: Mon Jan 30, 2017 12:33 am 
Offline
Member
User avatar

Joined: Oct 30, 2013
Posts: 7305
Location: England
Apple wrote:
Brute force in this case would mean that the decision to make each move comes from manually mapping out every possible chain of responses to any possible move, and not by applying some complicated mathmematical solution, should one exist.

I think your definition might be a little off here - if only because when talking about chess your proposed brute force method and the 'complicated mathmatical solution' are the same thing. At high level chess is a puzzle game - it's less about knowing every possible move and more about knowing what your opponent is doing and working to disrupt their plan then it is about winning (especially when you are playing on black).

To answer your question though I'd say yes it's possible so long as both 'win' and 'draw' are acceptable outcomes of your solution - playing on the black especially their are simply some games where knowing every possible moves still means the best you can hope for is a draw. If you just want 'win' however then no it isn't possible simply because of how the game works.

_________________
Welcome! I'm Garren and I'll be your designated villain for the evening.


Like this post
Top
 Profile  
 
PostPosted: Mon Jan 30, 2017 6:55 am 
Offline
Member
User avatar

Joined: Nov 08, 2013
Posts: 620
Location: Zaragoza, Spain
There is a finite number of pieces and a finite number of squares, so there's also a finite number of game states and moves. Therefore a brute force solution is possible given an enormous and completely unrealistic amount of time and space. More potent computers won't make it feasible.


Like this post
Top
 Profile  
 
PostPosted: Mon Jan 30, 2017 9:37 pm 
Offline
Conqueror of Eldangard
User avatar

Joined: Sep 25, 2013
Posts: 14140
Location: Kamloops, BC
Identity: Male
I've heard two definitions of "solving" chess.
One is a specific series of moves, which if taken in order will always result in victory or tie regardless of what the opponent does. I'm not sure that this exists.
The other is a strategy- "a map" of all possible chess moves cut just cut out the bits that lead to your loss. What you're left with is a flowchart to chess victory/tie. If opponent takes action X, do response Y. Follow the strategy guide and you can't lose.

_________________
Cato wrote:
CotW is a method for ranking cards in increasing order of printability.

*"To YMTC it up" means to design cards that have value mostly from a design perspective. i.e. you would put them in a case under glass in your living room and visitors could remark upon the wonderful design principles, with nobody ever worring if the cards are annoying/pointless/confusing in actual play

TPrizesW
TPortfolioW


Like this post
Top
 Profile  
 
PostPosted: Mon Jan 30, 2017 10:39 pm 
Offline
Member
User avatar

Joined: Sep 23, 2013
Posts: 2354
What's more interesting to me is that there are computers that have used more complicated methods to beat some of the best Chess and Shogi players.

_________________
Ambiguity, a role-playing contest
List of Chopped Champions

Previous seasons of Chopped:


Like this post
Top
 Profile  
 
PostPosted: Thu Feb 02, 2017 1:34 pm 
Offline
Member
User avatar

Joined: Sep 12, 2015
Posts: 87
Identity: girl
the lower bound on the number of possible games in go is something like 10^800, which is way way more than the number of planck volumes in the visible universe. Considering that a computer exists inside our universe, even if it took up all the space in the visible universe and was built in the most efficient possible way, its possible that it couldn't bruteforce a solution to go, because it needs more space to represent the games than it has available to it in the universe.

if even a crappy home computer can eventually solve chess completely with 8GB of ram and maybe 1TB of memory otherwise, then physical space shouldn't be a limiting factor in the ability to physically find a solution, and time versus processing speed will be the only factors, in which case the question as to whether or not it is physically possible is basically "if a computer had access to all the space in the universe to calculate at the fastest physically possible speeds, would it be able to find a solution before entropy prevented it from physically being able to manipulate information in order to calculate?"


Like this post
Top
 Profile  
 
PostPosted: Fri Feb 03, 2017 12:25 am 
Offline
Conqueror of Eldangard
User avatar

Joined: Sep 25, 2013
Posts: 14140
Location: Kamloops, BC
Identity: Male
Yeah, I kind of assumed infinite storage space and time as the parameters. In practical terms, that's not possible.
By "Brute forcing" are we assuming that the computer has to try every possible board position? Because it seems like you should be able to cut out many board states as being blatantly unwinnable and many game patterns as meaninglessly indistinct from one another. I mean, you and the opponent could just move your pieces back and forth for an infinite arbitrarily long time and produce infinitely many sequences of moves. If a computer could detect that loop ahead of time, it could cut off that whole avenue of investigation. Or is that taken into account in your figure of 10^800 possible games? And are we certain there isn't a way to bring that figure down further?

_________________
Cato wrote:
CotW is a method for ranking cards in increasing order of printability.

*"To YMTC it up" means to design cards that have value mostly from a design perspective. i.e. you would put them in a case under glass in your living room and visitors could remark upon the wonderful design principles, with nobody ever worring if the cards are annoying/pointless/confusing in actual play

TPrizesW
TPortfolioW


Like this post
Top
 Profile  
 
PostPosted: Fri Feb 03, 2017 4:15 am 
Offline
Member
User avatar

Joined: Sep 12, 2015
Posts: 87
Identity: girl
there are generally rules in games like chess that prevent the game from continuing forever, like the 50 move rule or threefold repetition.

I think that if a computer used logic to determine that there wasn't any point in reading certain moves, then that wouldn't qualify as a brute-force solution.

the 10^800 figure was for go and not chess, which I imagine is not brute forceable in our universe, although I don't know enough about computers or algorithms to actually crunch the numbers. The most common number I've found for the lower bound on the number of possible games of chess is 10^120.


Like this post
Top
 Profile  
 
PostPosted: Fri Feb 03, 2017 2:52 pm 
Offline
Member
User avatar

Joined: Sep 23, 2013
Posts: 8248
Identity: Spambot
Preferred Pronoun Set: 0, 1
Don't brute force it, Monte Carlo it. For each first move, have the computer play a thousand games where each player makes arbitrary but legal moves. Then, choose whichever move had the highest winrate, and repeat the process for black's first move, and so on and so forth.

_________________
Any resemblance to actual persons, living or dead, is purely coincidental.


Like this post
Top
 Profile  
 
PostPosted: Sun Feb 05, 2017 12:48 am 
Offline
Member
User avatar

Joined: Sep 12, 2015
Posts: 87
Identity: girl
door (dooooooooooor) is my favourite name of yours and enigmocracy is my least favourite although it is still an alright name


Like this post
Top
 Profile  
 
PostPosted: Tue Feb 14, 2017 1:46 pm 
Offline
Member
User avatar

Joined: Sep 25, 2013
Posts: 3084
It is theoretically possible, that given an impossibly large computer that's impossibly power running for an impossibly long period of time, you could brute-force chess. Of course, when dealing with literal infinites, even the slowest computer in existence would eventually be capable of it if it existed for a period of time longer than the computation takes place across.

_________________
Quote:
"If you refuse to use rock, you will never beat scissors." — Galef, Dakka Dakka Forums


Like this post
Top
 Profile  
 
PostPosted: Tue Feb 14, 2017 3:06 pm 
Offline
Conqueror of Eldangard
User avatar

Joined: Sep 25, 2013
Posts: 14140
Location: Kamloops, BC
Identity: Male
I guess you could rely on luck too. Maybe the first random attempt at a solution works.

_________________
Cato wrote:
CotW is a method for ranking cards in increasing order of printability.

*"To YMTC it up" means to design cards that have value mostly from a design perspective. i.e. you would put them in a case under glass in your living room and visitors could remark upon the wonderful design principles, with nobody ever worring if the cards are annoying/pointless/confusing in actual play

TPrizesW
TPortfolioW


Like this post
Top
 Profile  
 
PostPosted: Wed Feb 28, 2024 2:58 am 
Offline
Member

Joined: Feb 28, 2024
Posts: 1
Brute force is a method of solving a problem by trying all possible options until finding the best one. In chess, brute-force would mean searching all possible moves and outcomes to find the optimal strategy.

However, brute force is not physically possible for chess, because there are too many possible chess positions and games to search.


Like this post
Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 19 posts ] 

All times are UTC - 6 hours [ DST ]


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to:  
Powered by phpBB® Forum Software © phpBB Group