Chess being solved is a given
No.
Chess being solved is a given, considering quantum computers
No.
quantum computers will be commercially available soon.
No.
I've only read the first sentence, but off to a bad start.
Chess being solved is a given
No.
Chess being solved is a given, considering quantum computers
No.
quantum computers will be commercially available soon.
No.
I've only read the first sentence, but off to a bad start.
Stockfish has probably quite close to "solving" chess already, and it has little impact on how most people (newbies) play the game.
Yes
Agreed. If chess is ever solved, the solution will have no practical value for ordinary chess players like us. It will have only limited impact even at the Super-GM level.
Agreed. If chess is ever solved, the solution will have no practical value for ordinary chess players like us. It will have only limited impact even at the Super-GM level.
Yeah, functionally speaking, we already have an oracle. Chess engines are massively stronger... and how do professionals (and even many amateurs) use that information? We don't choose lines that are best, we choose lines that are difficult for our human opponents. We choose lines that lead to positions we understand.
Solving chess would change nothing.
why do so many people worry about it in the forum?
Us forumers have too much time on our hands. And when you have a lot of time, and nothing to do with it... well, you get philosophers, if you're smart, and as for the rest of us...
What does it mean to solve chess? Is that when you find an irrefutable line that makes it pointless to play? FIDE would just make a rule against it. Chess.com would just ban you for discussing it let alone playing it. So no, I guess it doesn't matter.
The point is that such a line is worthless to you unless you can remember the refutation to every possible alternative move that your opponent might play on each and every move from the beginning to the end of your line.
So if your "winning line" begins with 1. c4 then you have to memorize the proper winning answer to every legally possible Black reply, not just on move one but every possible legal alternative on every single move right out to the eventual mate.
Billions of billions of billions of billions of lines. Literally.
The "solution" to chess is almost worthless if it won't fit into a human mind.
What does it mean to solve chess? Is that when you find an irrefutable line that makes it pointless to play? FIDE would just make a rule against it. Chess.com would just ban you for discussing it let alone playing it. So no, I guess it doesn't matter.
Look at an endgame tablebase. It'll be something like 10 moves win, 10 moves draw, 10 moves lose.
A solution to chess would have something like half of the legal moves draw on every turn... and if the average drawn game is 120 ply (which is 60 moves) and if, conservatively, there are only 4 "good" moves in each position on average, that's 4^120 moves... or about 10^72... transpositions would reduce this number, but anyway, that's in the ballpark of number of atoms in the observable universe.
"one irrefutable line" lol
@1
"I have yet to memorize a single opening, let alone learn the entire game."
++ Chess grandmasters know about 10,000 games by heart. If he knows by heart 10,000 perfect games with optimal play from both sides, then he has a substantial advantage.
@13
"that's 4^120 moves... or about 10^72"
++ People keep repeating this nonsense.
There are only 10^44 legal chess positions, the vast majority of them making no sense at all.
To weakly solve chess requires 10^17 relevant positions, that is about 10^15 games.
Memorising 10^4 games is a tiny portion, but may give a substantial edge.
"that's 4^120 moves... or about 10^72"
++ People keep repeating this nonsense.
There are only 10^44 legal chess positions
Yes, I can do the very simple (two term, high school level) calculation myself and arrive at a number of positions fewer than 10^72 (unlike you who has to quote something). But you'll notice I wasn't talking about number of positions, I was talking about nodes on a game tree. This is relevant because when humans memorize openings they do it in sequences of moves, not as individual positions.
To weakly solve chess requires 10^17 relevant positions
Repeating this does not make it true.
Just like Reddit, the downside of the vote feature is that people who don't know anything vote... OP has 5 upvotes and gets 3 things wrong in the first sentence... ridiculous.
-
@16
"I wasn't talking about number of positions, I was talking about nodes on a game tree."
++ Me too. Node = position + history + evaluation.
"when humans memorize openings they do it in sequences of moves"
++ No, you do not memorize 1 e4 e5 2 Nf3 Nc6 differently from 1 e4 Nc6 2 Nf3 e5
You do not memorize 1 e4 e6 2 d4 d5 3 exd5 exd5 4 Nf3 Nf6 differently
from 1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 d3 Nf6 6 d4 d5
"To weakly solve chess requires 10^17 relevant positions
Repeating this does not make it true."
++ There are only 10^44 legal positions https://github.com/tromp/ChessPositionRanking
So you can never arrive at more than 10^44 positions.
The vast majority of these legal positions make no sense,
as the 3 displayed random samples show: 3 or more bishops/rooks on both sides.
A better estimate is thus 10^37 positions https://arxiv.org/pdf/2112.09386.pdf
but a random sample of 10,000 such positions show none to be able to result from optimal play from both sides either. On the other side positions with 3 or 4 queens are known to occur in perfect games with optimal play from both sides in ICCF WC Finals so multiply by 10.
That leaves 10^37 * 10 / 10,000 = 10^34 positions.
So you can never arrive at more than 10^34 positions with reasonable play from both sides.
Now to weakly solve chess you need only 1 black strategy to draw against all white moves.
E.g. in the initial position you do not need 20 black replies to each of the 20 white opening moves, but 1 black reply for each of the 20 white opening moves.
Thus 20 * 1 = 20 instead of 20 * 20 = 400 possibilities, i.e. 20 is the square root of 400.
In this sense Sqrt (10^34) = 10^17 positions are relevant to weakly solving chess.
If you could memorize the 10^17 relevant positions, you could always play perfectly.
No human can memorize 10^17 positions, but 10,000 games i.e. 10^6 positions is feasible.
"I wasn't talking about number of positions, I was talking about nodes on a game tree."
++ Me too. Node = position + history + evaluation.
"when humans memorize openings they do it in sequences of moves"
++ No, you do not memorize 1 e4 e5 2 Nf3 Nc6 differently from 1 e4 Nc6 2 Nf3 e5
You do not memorize 1 e4 e6 2 d4 d5 3 exd5 exd5 4 Nf3 Nf6 differently
from 1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 d3 Nf6 6 d4 d5
"To weakly solve chess requires 10^17 relevant positions
Repeating this does not make it true."
++ There are only 10^44 legal positions https://github.com/tromp/ChessPositionRanking
So you can never arrive at more than 10^44 positions.
The vast majority of these legal positions make no sense, as the 3 displayed random samples show: 3 or more bishops/rooks on both sides.
A better estimate is thus 10^37 positions https://arxiv.org/pdf/2112.09386.pdf
but a random sample of 10,000 such positions show none to be able to result from optimal play from both sides either. On the other side positions with 3 or 4 queens are known to occur in perfect games with optimal play from both sides in ICCF WC Finals so multiply by 10.
That leaves 10^37 * 10 / 10,000 = 10^34 positions.
So you can never arrive at more than 10^34 positions with reasonable play from both sides.
Now to weakly solve chess you need only 1 black strategy to draw against all white moves.
E.g. in the initial position you do not need 20 black replies to each of the 20 white opening moves, but 1 black reply for each of the 20 white opening moves.
Thus 20 * 1 = 20 instead of 20 * 20 = 400 possibilities, i.e. 20 is the square root of 400.
In this sense Sqrt (10^34) = 10^17 positions are relevant to weakly solving chess.
If you could memorize the 10^17 relevant positions, you could always play perfectly.
No human can memorize 10^17 positions, but 10,000 games i.e. 10^6 positions is feasible.
You may find this link (below) interesting. In it you'll find that several people have been explaining the various ways in which you're wrong for 6 months or so.
https://www.chess.com/forum/view/general/chess-will-never-be-solved-heres-why?page=341
@19
"several people have been explaining the various ways in which you're wrong for 6 months or so"
++ No, those several people are wrong in various ways and I try to explain that to them.
Some are too lazy to read, some are too stupid to understand.
@19
"several people have been explaining the various ways in which you're wrong for 6 months or so"
++ No, those several people are wrong in various ways and I try to explain that to them.
Some are too lazy to read, some are too stupid to understand.
I think it would help your case if you were able to convince at least 1 person. When the situation is either everyone in the world is crazy or you're crazy then your odds aren't so good.
@21
"I think it would help your case if you were able to convince at least 1 person."
++ I am no debater, communicator, or wordsmith,
but I might be the only scientist on that thread.
"either everyone in the world is crazy or you're crazy"
++ Like when Galileo was forced to swear the Earth does not revolve around the Sun...
++ Like when Galileo was forced to swear the Earth does not revolve around the Sun...
Yes, those cases are so rare in history that we can afford to remember one so long ago as Galileo. A very good example of the exception proving the rule.
This is not a case of opposing epistemologies. The Church derives knowledge from tradition, the people arguing against you derive knowledge from the same modern sources as you do.
++ I am no debater, communicator, or wordsmith,
but I might be the only scientist on that thread.
There are several people with advanced degrees arguing against you.
Scientist... one annoying aspect is how you seem to have worked backwards. You started with a 15 year old quote about solving chess, and massaged your calculations until they fit Sveshnikov's baseless statement, and you did so seemingly without realizing that modern computers (super or otherwise) are 50 to 100 times faster than they were in 2017.
Chess being solved is a given, considering quantum computers will be commercially available soon. Well, solved as the ultimate "best move" to each possible move will be found. However, why do so many people worry about it in the forum?
I have yet to memorize a single opening, let alone learn the entire game. For the average person or even high-level grandmasters, chess being solved does not matter. Stockfish has probably come quite close to "solving" chess already, and it has little impact on how most people (newbies) play the game.
Why the fuss?