Chess will never be solved, here's why
Sort:
Botwin74
Sep 4, 2022
0
#3661
?
Botwin74 wrote:
?
It means that if someone's idea is entirely off the mark, there is no point in discussing all the specific parameters with them at a detailed level.
Sep 4, 2022
0
#3663
btickler wrote:
MARattigan wrote:
And the rest. He's talking about competition rules.
When someone proposes to make a car engine out of cheese, I don't feel the need to tell them their piston timing is off.
Not a matter of piston timing. The number of game states (which is what I refer to as positions and what SF works with, so presumably what @tygxc should be starting with) in KBNvK under competition rules is probably comparable with Tromp's number of basic rules positions for the whole game (though it's difficult to say). Not a minor adjustment by any stretch.
MARattigan wrote:
btickler wrote:
MARattigan wrote:
And the rest. He's talking about competition rules.
When someone proposes to make a car engine out of cheese, I don't feel the need to tell them their piston timing is off.
Not a matter of piston timing. The number of game states (which is what I refer to as positions and what SF works with, so presumably what @tygxc should be starting with) in KBNvK under competition rules is probably comparable with Tromp's number of basic rules positions for the whole game (though it's difficult to say). Not a minor adjustment by any stretch.
I don't think it's a minor adjustment. I chose that analogy to be expedient. It covers the point I was making, but is not 100% applicable.
@4618
"The number of chess games is finite" ++ Yes, but too large: between 10^29241 and 10^34082, that is why positions are better than games: 10^44 legal of which 10^17 relevant.
"I had a book with every possible game of chess that could be played
(it might have 10^100 pages or even 10^1000 pages)"
++ At 1 page per game that would be between 10^29241 and 10^34082 pages.
"I would consider that solving chess."
++ No, a 32-men table base with all 10^44 legal positions would strongly solve chess.
Chess can be weakly solved with 10^17 relevant positions.
@4621
"Weakly solved means that for the initial position a strategy has been determined
to achieve the game-theoretic value against any opposition"
"Where do "reasonable" moves come into it?"
++ any opposition indicates an act of opposing, i.e. white opposes to black achieving the game-theoretic value of a draw.
1 e4 e5 2 Nf3 Nc6 3 Ng1 does not oppose, it helps.
@4622
"I just don't believe that we can use GMs, engines, or a combination of the two to definitively decide which moves are unreasonable."
++ Yes we can. I am no GM and I can definitively decide that 1 e4 e5 2 Ba6, 1 e4 Nf6 2 Qh5, 1 e4 d5 2 Qg4 etc. are unreasonable: they do not oppose to black achieving a draw.
It definitely needs a combination of (ICCF) (grand)masters and engines.
The human assistants launch the calculations, preferably from 26-men positions and they terminate pointless calculations adjudicating a draw or a win when there is no doubt.
Thus the humans save time for the engine calculations.
"Top human players and the best engines have proved themselves capable of making mistakes in evaluation." ++ That is right, therefore solving chess cannot rely on engine evaluation.
"Deliberately letting imperfect entities eliminate a great number of positions from consideration in the search for a solution will create reasonable doubt as to the validity of the conclusion."
++ No, if any doubt at all, then calculate.
If no doubt at all, then eliminate so as to save pointless calculation.
"If we do not eliminate so many lines from examination the size of the task and the amount of time and resources needed to complete it, will grow."
++ Yes, of course, but using chess knowledge to eliminate unreasonable lines from calculation is beneficial to shrink the size of the task and the time and resources needed to complete it.
bollander wrote:
My god, the fact that this DUMB discussion is still running blows my mind
What are you doing here then? Run along.
bollander wrote:
And here comes the brainless ant that keeps pushing the discussion commenting.
Go hug the other lunatic and discuss this idiocy in your DMs
Go hug the other lunatic and discuss this idiocy in your DMs
It seems like a fruitless course of action to complain about a topic you are voluntarily reading.
"Why am I watching this commercial?! I'm so mad right now..."
"Ummm..."
Sep 5, 2022
0
#3670
A little test for @tygxc:
How many states (necessary generalisation of position) are there in chess with the 50 move rule?
Can you accurately analyse chess with the 50 move rule while ignoring the 50 move rule?
@4637
The 50 moves rule plays no role at all. It is just a practical rule to avoid competition games to drag on for days or weeks as one side tries to win e.g. KRB vs. KR.
Just weakly solve chess disregarding the 50 moves rule.
The solution with the 50-move rule is the same.
In none of the perfect games we have from the ICCF WC did the 50 moves rule play any role.
Sep 5, 2022
0
#3672
Optimissed wrote: To repeat, Game Theory consists of the application of the theory of games to real life situations,
No. The clue is in the word "theory". Strangely, you seem to confuse "the fact that game theory can be applied" with "its nature".
GT is inapplicable to chess.
GT is inapplicable to chess.
An odd nonsense! For example, the existence of an optimal strategy for each player is a theorem of game theory that applies to a class of games including chess. This theorem applies to chess and implies there exists an optimal strategy for each player in chess.
General theory and application.
<ultra-weakly solved means that the game-theoretic value of the initial position has been determined>
That's fair enough. Chess is [very probably] a draw with best play from both sides.
That's fair enough. Chess is [very probably] a draw with best play from both sides.
The "very probably" is important. No expert would say that chess has been ultra-weakly solved, because it hasn't. Some other games have.
it hasn't been determined, by deductive reasoning.
Now you're getting it!
"Ultra-weakly" very clumsy, so you'd do better to call it the "game-theoretic value" or the "game-theoretic result". Not "The ultra-weak solution is that it's a draw", because it's completely unnecessary to use such jargon.
Chess ain't ultra-weakly solved. This is a meaningful fact about the state of knowledge.
<weakly solved means that for the initial position a strategy has been determined to achieve the game-theoretic value against any opposition>
You're already in trouble. That's the problem with jargon. You can always use it to sell ideas or confuse the opposition. Here, you've evidently confused yourself.
<weakly solved means that for the initial position a strategy has been determined to achieve the game-theoretic value against any opposition>
You're already in trouble. That's the problem with jargon. You can always use it to sell ideas or confuse the opposition. Here, you've evidently confused yourself.
Well, he may have confused you, but he is 100% correct, providing no basis for the claim of personal confusion.
Looking at the layout of the solution from top-down, with the game theoretic value being the top of the tree (effectively "the user interface"), that cannot be determined without understanding the stategies that may be used to achieve a draw in all circumstances {"against any opposition"). This is why you're much better off speaking in English and not Jargonese. Forget the idea of "stategy", because in any case it gives the wrong idea, since any full solution of chess is tactical only.
Looking at the layout of the solution from top-down, with the game theoretic value being the top of the tree (effectively "the user interface"), that cannot be determined without understanding the stategies that may be used to achieve a draw in all circumstances {"against any opposition"). This is why you're much better off speaking in English and not Jargonese. Forget the idea of "stategy", because in any case it gives the wrong idea, since any full solution of chess is tactical only.
Here you confuse yourself by conflating two uses of the word "strategy". It is always very important to remember that concepts are what are discussed. When a word refers to two different concepts this does not conflate the concepts. Here "strategy" refers to a complete procedure, like the fully analysed solution of a chess problem. In practical chess or in the Art of War, strategy means something very much more vague.
You need to accept that the only "strategy" is the exploration of all tactical lines which may be relevant, not forgetting to allow considerable leeway for examining surprising lines, which may otherwise be missed but may lead to a conflicting result.
Dead right.
Deciding which lines may be relevant is an incredibly complex task. Your belief that Stockfish can achieve it, with the assistance and guidance of three GMs, is incredibly naive, if taken seriously. If those trying to argue against you deserve criticism, it's for allowing your ideas to achieve any credibility at all. Your ideas are wrong. There's no doubt about it. It isn't up for argument and you'd do best to realise that, no matter how belatedly.
Deciding which lines may be relevant is an incredibly complex task. Your belief that Stockfish can achieve it, with the assistance and guidance of three GMs, is incredibly naive, if taken seriously. If those trying to argue against you deserve criticism, it's for allowing your ideas to achieve any credibility at all. Your ideas are wrong. There's no doubt about it. It isn't up for argument and you'd do best to realise that, no matter how belatedly.
Yes.
Taking on board what I'm suggesting is the only way you will be able to communicate effectively and, more importantly, to start to get your thoughts in order. There isn't anyone else, commenting on this thread, who seems capable of pointing out the problems you're causing for yourself, by using terms which you don't seem to properly understand, even in the context in which you're trying to use them."
Taking on board what I'm suggesting is the only way you will be able to communicate effectively and, more importantly, to start to get your thoughts in order. There isn't anyone else, commenting on this thread, who seems capable of pointing out the problems you're causing for yourself, by using terms which you don't seem to properly understand, even in the context in which you're trying to use them."
There is a sizeable subset of participants who are 100% ok with the definitions, though.
Sep 5, 2022
0
#3673
tygxc wrote:
@4637
The 50 moves rule plays no role at all.
Only if the result of a game plays no role at all (i.e. you change the rules so a draw and a win as the same value. After which there is no game).
The 50 move rule changes the optimal value of many positions in games. You should be able to understand that.
So, have another go.
@4640
No, not at all.
> 99% of ICCF draws are perfect games. In none is the 50-moves invoked.
You can safely ignore the 50-moves rule.
The solution to chess is the same as with the 50-moves rule.
Likewise the stalemate rule plays no role either.
As AlphaZero has shown, even if stalemate is made a win, chess still remains a draw.
The 50-moves rule and the stalemate rule change the optimal value of some positions,
but such positions are not reached in a perfect game.
Sep 6, 2022
0
#3676
tygxc wrote:
Has chess been solved? No
Can chess be solved? Yes, it takes 5 years on cloud engines.
Will chess be solved? Maybe, it depends on somebody paying 5 million $ for the cloud engines and the human assistants during 5 years.
Have humans walked on Mars? No
Can humans walk on Mars? Yes
Will humans walk on Mars? Maybe, it depends on somebody paying billions of $ to build and launch a spacecraft.
Hi, can you read my name backwards and change your diapers?
drutyllemscxgyt wrote:
Hi, can you read my name backwards and change your diapers?
Reported. Enjoy your vacation.
@4642
"I proclaim that your proclamation is false."
++ I prove, you proclaim.
tygxc wrote:
@4642
"I proclaim that your proclamation is false."
++ I prove, you proclaim.
You have yet to prove anything.
@4646
"You have yet to prove anything."
++ I have proven many things so far.
Some do not read or do not understand.
Sveshnikov proclaimed:
"Give me five years, good assistants and the latest computers
- I will bring all openings to technical endgames and "close" chess."
I have proven him right.
In 5 years there are 157788000 seconds. The latest computers calculate a billion positions per second. So 3 such computers - or 3000 desktops - can calculate all 10^17 (hundred million billion) relevant chess positions in 5 years to weakly solve chess.
@Elroch proclaimed the 50-moves rules plays some role in weakly solving chess.
I have proven him wrong.
Using statistics and probability - which he should understand - and looking at a sufficiently large tournament with sufficient level and a sufficient number of independent players e.g. the 30th ICCF World Championship finals I have proven that several 1000 of such games are perfect games with no error. In none of those perfect games and in none of the imperfect games either was the 50-moves rule invoked. Thus the 50-move rule plays no role in weakly solving chess.
The weak solution of chess without the 50-moves rule is the same as with the 50-moves rule.