Chess will never be solved, here's why

Sort:
MARattigan
Optimissed wrote:

... There's nothing to suggest that chess is anything other than a draw ...

Nor anything of substance to suggest it's anything other than a win.

playerafar

Chess is not 'solved'.
Therefore other things about chess aren't known.
People react in different ways to whatever being unknown.
Even ignoring the context sometimes.
Having a 'policy' towards the unknown rather than accepting it as such.
Related to this: Cognitive disonnance. And its cousin - cognition bias.

playerafar

From Wiki:
"Game theory is the study of mathematical models of strategic interactions.[1] It has applications in many fields of social science, and is used extensively in economics, logic, systems science and computer science.[2] Initially, game theory addressed two-person zero-sum games, in which a participant's gains or losses are exactly balanced by the losses and gains of the other participant. In the 1950s, it was extended to the study of non zero-sum games, and was eventually applied to a wide range of behavioral relations. It is now an umbrella term for the science of rational decision making in humans, animals, and computers. "
'Rational'.
There would seem to be a big overlap with psychology.
Controversy in game theory? Would seem to be obviously so.
Set theory? Again - yes.
Semantics of the word 'luck'.
Do such semantics matter? Yes because the word is used a lot.
But its only a word. Meant to serve us. Not us to serve the word.
Doesn't always work out that way.

Elroch

Another key extension to game theory is having more than 2 players. A trivial result is that every non-zero sum game with N players is equivalent to a zero sum game with N+1 players.

But the majority of games with more than 2 players are awkward to deal with unless co-operation is banned, because they can be transformed into games with 2 players in multiple ways - the ways in which the players can team up.

N-player games (Britannica)

DiogenesDue
Optimissed wrote:

RISK! was an excellent example!

One informal rule we used (around 1973-4) is that co-operation of players is banned if both are not genuinely trying to play to win. It meant that in practice, if a RISK! player combined with another and was obviously not attempting a winning strategy but simply a strategy of preventing the best player from winning, it would be questionable as to whether the other players would allow them to play in future.

Risk is an awful game. Couldn't be more luck based and still pretending to be about world conflict. Try something better, like Axis and Allies, or even Diplomacy, though that game is skewed towards England and Italy is kind of screwed.

MARattigan

Shame. Sounds like you blew it because of your inflated opinion of yourself. Not surprising though.

RyanZ_MD
ParanoidAndroid88 wrote:

Funny how wrong the first couple comments in this post are. Guy saying cloud computing could solve chess if someone just paid right now. LMAO.

well, if we get good enough computers, maybe. but probably another few millions years of technology

playerafar
RyanZ_MD wrote:
ParanoidAndroid88 wrote:

Funny how wrong the first couple comments in this post are. Guy saying cloud computing could solve chess if someone just paid right now. LMAO.

well, if we get good enough computers, maybe. but probably another few millions years of technology

There's another idea though. And that's computing power increasing exponentially.
Could make a big dent in the trillions of trillions of years it would otherwise take to tablebase all of chess.
More progress is being made with that present day.
The so-called Quantum computers.

DiogenesDue
Optimissed wrote:

We didn't have many good alternatives in those days to games like Minesweeper or Risk. I liked Risk at the time and even designed another verson of it and made the board for it, with an extra continent. Not sure I can remember ... maybe it was South Asia or something. I don't think I've played Risk since 1974 but at the time it was very enjoyable. A game might last from 11 pm til 2:30.

I don't think it was luck based. Very strategic and also based on probabilities, which you could work out.

They deal your starting countries from a deck of cards and the battle results are determined by dice rolling. Enough said.

Back then in the "more than two players" strategic board game arena you had Diplomacy, for starters. Feudal, which would appeal to chess players, was also around. In the 70s more started coming out.

Then there were full-on wargames with the hexagon maps, etc., but those are little much for most people. I still have Afrika Corp but I never really played it, too much setup and too many hours for what you get out of it.

DiogenesDue
Optimissed wrote:

You would have to understand how to play the game in order to understand that it's actually a difficult game, rather like chess. You can't go on superficial evaluations. I played diplomacy exactly once and didn't enjoy it I think because it's so artificial. Later I enjoyed civilisation and played Civilisation II but when Civilisation III came out I thought it was completely ruined and reverted to playing Civilisation I, which is by far the best game of the civilisation family. Again, the sequels were top heavy, clumsy, artificial and too slow.

Lol, I used to play online Risk with developer friends. It got boring winning most of the time without much really happening, then occasionally losing to ridiculous dice runs (like army of 3 holding against 18, etc.). Risk is dirt easy. There's barely any mechanics to it at all, and no real strategic decisions to be made, just statistics and luck with a bare minimum of logistics. Wait for overwhelming force to attack, or take a gamble...so many choices!

Are you going to pretend that it takes brainpower to decide to attack Kamchatka with your 5 to 1 advantage, or that it is a strategy to "sneakily" spread out your army and amass using "just-in-time" tactics, or that deciding to take a smaller continent for the bonus and barricade the ways in are robust gameplay?

I haven't played in decades and there's so little strategy involved I just covered pretty much all of it.

Are you sure you didn't like Diplomacy because there's no luck and it relies on your ability to form trust and negotiate?

DiogenesDue
Optimissed wrote:

The only shoot-em-up I played was Wolfenstein. I could get through all the levels without losing a single life. The same reaction speed that made me a good table football player. I miss table football.

Load up Doom or Quake and I would be happy to test your reflexes at dodging rockets. Blowing you up over and over sounds like a lot of fun. Wolfenstein 3D (the original one) is pretty slow.

I have met John Carmack, and while he will always have a soft spot in his heart for Wolfenstein, Doom and Quake were quantum leaps better in terms of "twitch" gameplay and real reaction times.

Thee_Ghostess_Lola

Some of the top "world Risk players" often mentioned include: Matchstickx, Jimmie The Frosty, Robt Rustbolt 2, Bradex - these names frequently appear at the top of leaderboards on the Hasbro Risk platform, with Matchstickx often listed as the number one player based on skill points.

these players are just lucky ..boy do u have ALOT to learn. lol !!

DiogenesDue
Thee_Ghostess_Lola wrote:

Some of the top "world Risk players" often mentioned include: Matchstickx, Jimmie The Frosty, Robt Rustbolt 2, Bradex - these names frequently appear at the top of leaderboards on the Hasbro Risk platform, with Matchstickx often listed as the number one player based on skill points.

these players are just lucky ..boy do u have ALOT to learn. lol !!

Learn to read, instead. I said Risk was as luck-based a game as you could have while still trying to pretend it's a game that models strategic conflicts...and it is. Nevertheless, the ridiculously named players you mentioned can lose to a mediocre player if the dice roll the wrong way or if, say, the worse player gets dealt all of South America randomly. Whereas in chess a 2700 hundred player will not lose an even odds game to a 1500 player in your lifetime..

Why do you think that the top Poker players stopped winning the WSOP as soon as the number of players exploded?

DiogenesDue
Optimissed wrote:

Never played it online. I agree it would become dull if that (the game itself) was the only object but the tossing of the dice online is what would have spoiled it because back then, synthesised or pseudo random number production was very poor. Nowadays it's better but truly random numbers have to have a real life source. That means that some real life, random and measurable source must be used to produce truly random numbers. Of course, some imagine that everything in the universe is predetermined but that's an outlandish assumption. The universe would have to conspire to conceal a deterministic function producing simulated randomness; and for what? As ever, the premise of genuine randomness naturally existing is the correct, scientifically based assumption, since it reduces the unknown variables to one.

Seeding the RNG with the system clock and various other methods took care of the bulk of the random number issues back in the 80s and 90s. In 2010 or so dedicated hardware with even better seeding took over. OSes also provide their own random seeding using various chaotic interactions between hardware interrupts, etc. You needn't worry about getting the same "random" series of number more than once, that was really only in the early years when PCs were not supposed to be used for gaming (according to IBM, anyway).

DiogenesDue
Optimissed wrote:

It improved things but it was still completely obvious to those who could remember complex patterns that it was pseudo0random, with none of the kind of surprising outcomes one gets in real randomness.

You really don't need to worry about it, especially nowadays. No Man's Sky has 15 quintillion procedurally generated planets, not one of them the same, for example.

Anytime after 1990, I would definitely say that any perception of seeing the same string of numbers would be a case of confirmation bias, and seeding based on electronic noise and thermal fluctuations makes it a practical impossibility today.

shadowtanuki
DiogenesDue wrote:
Optimissed wrote:

We didn't have many good alternatives in those days to games like Minesweeper or Risk. I liked Risk at the time and even designed another verson of it and made the board for it, with an extra continent. Not sure I can remember ... maybe it was South Asia or something. I don't think I've played Risk since 1974 but at the time it was very enjoyable. A game might last from 11 pm til 2:30.

I don't think it was luck based. Very strategic and also based on probabilities, which you could work out.

They deal your starting countries from a deck of cards and the battle results are determined by dice rolling. Enough said.

Back then in the "more than two players" strategic board game arena you had Diplomacy, for starters. Feudal, which would appeal to chess players, was also around. In the 70s more started coming out.

Then there were full-on wargames with the hexagon maps, etc., but those are little much for most people. I still have Afrika Corp but I never really played it, too much setup and too many hours for what you get out of it.

When more than two people play Risk, according to the original rules, players take turns claiming territories one at a time. They aren't randomly dealt out. When two people play, the territories are randomly dealt out, but the players take turns deploying their armies into those territories two at a time, and they place one unit of the neutral army, so there is always an element of strategy in the game's starting conditions for any number of players.

That being said, because of the relatively few armies that you start out with, and the resultant decimation of your forces that can take place on a single unfortunate assault, luck would seem to play quite a role, especially at the beginning. The first player has an advantage in the chance to reduce the number of armies the next player gets in reinforcements, and the last player in the round may begin with a considerably weakened force. The game is strategically pretty straight forward, but a lot will be determined by the luck of the first few attacks. More players, all trying to win, would do a lot to reduce the influence of luck on the outcome of the game, though. Would it be enough to give a player a chance to come back and win from a serious initial disadvantage? Probably not, but then again total victory doesn't have to be the player's goal. For many players, the hope of a partial victory would be enough of a motivation to keep playing.

Grand_Mister

I don't understand the point of this thread. What is this all about? Is the OP stating that it is already solved? I am too lazy to read all that again.

For an answer, chess can never be solved for the reason I mentioned in my previous post.

playerafar
Grand_Mister wrote:

I don't understand the point of this thread. What is this all about? Is the OP stating that it is already solved? I am too lazy to read all that again.

For an answer, chess can never be solved for the reason I mentioned in my previous post.

With many things - there are multiple points. Not just one.
'the' point is therefore often a misinterpretation.
In the case of this this thread - the word 'solved' has different meanings to different people - depending on what meaning or meanings they elect to assign.
Even if a common meaning were assigned - such as 'solved' meaning that the entire game and all its trillions of trillions of possible positions have been 'tablebased' - then opinions would still vary as to whether that will ever be done or not or can ever be done or not.
Since 'solved' has many meanings and all over the world every day millions of people try to 'solve' their chess games by trying to win them and since this is a chess website - well if you factor in those realities then the 'points' of this thread become more visible and understandable.
But even with that logic and realism available - the 'points' of any thread are subject to further interpretation.
People decide on their own what the 'points' of a thread are.
Some assign a single point that 'doesn't work' and they get muted or banned by the website.

playerafar

Related to this: some may try to insist that chess 'isn't solved' or 'won't be solved' or 'will be solved' or 'is solved' for one simple reason. Because they say so.
Chess is not solved. But that's not about any poster or posters in the thread.
It simply isn't. That's using a definition involving something called tablebasing.
As basic as the earth going around the sun plus spinning on its own axis and being round too.
What is the current situation regarding tablebase 'solving' of chess?
Most 7 piece and fewer pieces situations have been tablebased by computers.
'Solved'.
Martin is very good at finding exceptions.
Although castling and other rules have not been factored into 7-piece and fewer.
Current projects with 8 pieces are struggling.
The math suggests that given the increase in difficulty as more pieces are added - it could take trillions of trillions of years to tablebase all positions.
But for now - 7 piece has been 'cooked'. Its available 'to go'.
But if you want castling sprinkled on it that's not available.
And 8 piece is not on the menu yet, It will be. In our lifetimes probably. To most of us.

DiogenesDue
shadowtanuki wrote:

When more than two people play Risk, according to the original rules, players take turns claiming territories one at a time. They aren't randomly dealt out. When two people play, the territories are randomly dealt out, but the players take turns deploying their armies into those territories two at a time, and they place one unit of the neutral army, so there is always an element of strategy in the game's starting conditions for any number of players.

That being said, because of the relatively few armies that you start out with, and the resultant decimation of your forces that can take place on a single unfortunate assault, luck would seem to play quite a role, especially at the beginning. The first player has an advantage in the chance to reduce the number of armies the next player gets in reinforcements, and the last player in the round may begin with a considerably weakened force. The game is strategically pretty straight forward, but a lot will be determined by the luck of the first few attacks. More players, all trying to win, would do a lot to reduce the influence of luck on the outcome of the game, though. Would it be enough to give a player a chance to come back and win from a serious initial disadvantage? Probably not, but then again total victory doesn't have to be the player's goal. For many players, the hope of a partial victory would be enough of a motivation to keep playing.

That's a pretty good assessment. Yes, the first real engagement very often determines far too much for the individual player. I should say, the first and the last engagements...if someone spends most of the game amassing an army that loses a battle where they outnumber by more than 2 to 1, the game is generally over for them. It is better for players to avoid each other and try to consolidate, but as soon as someone corners a continent, the clock starts ticking and that needs to be broken up. Extra armies every turn is like card advantage in MTG, the longer it sits there, the worse it gets for the other player(s). That's why Indonesia/Australia is the best continent to grab early game. One way in and out, and if you take it you just amass armies at the choke point, preferably Siam so you also have a buffer and prevent any shenanigans with Asia...not that it's not kind of good for you if someone tries to take Asia, which is bad idea with more than 2 players.

Like Diplomacy, everyone will turn on you immediately if you get that strong. Except Diplomacy has basically the same fun gameplay, minus the luck. The problem with Diplomacy is that England has somewhat of an advantage as the game is set up. Turkey and Russia also, but not as much. Italy and Germany are screwed, one being all bottled up and the other being open to attack in 3 directions from the get-go. Italy is forced to be a naval power to have any chances, which is a laugh, historically speaking.

Gee, I wonder how a game designed in the US ended up slightly skewed that way? Anyway, the other issue is you might have to put a moratorium on negotiation time between summer and winter orders, because some people take way too long negotiating. Usually the ones that end up with only 4 countries and slowly wither away. But Diplomacy has no luck in its design, and the moves are simultaneous, so no first move advantage either.

Haven't played either game in probably 25-30 years, though.