👋
Chess will never be solved, here's why
Sort:
MARattigan wrote:
I agree that computers are pretty irrelevant apart from speeding things up a bit (a bit in this context). If chess ever gets a practical solution (it's been solved according to @tygxc's definition probably since it was first thought of) I think human intelligence will play a far larger rôle than Superficial Intelligence.
The difference between you and @Elroch and I is not about computers. It's about what would constitute a solution and the nature of any human thought that would arrive at one (though computers could be necessary to effect the arrival). In simple terms we don't count guessing as a solution.
That's no fun. If guessing isn't a solution then the title of this topic is right. Chess will never be solved. Why? Because guessing isn't a solution.
But I still say for all practical purposes it's already been solved.
@6581
"computers are pretty irrelevant" ++ Without computers it is not possible.
Only with 3 powerful computers during 5 years non stop is it feasible to weakly solve chess.
10^17 relevant positions is too huge a number for unaided humans.
"human intelligence will play a far larger rôle"
++ Yes, incorporation of chess knowledge is beneficial.
That is also why Sveshnikov asked first for good assistants and second for modern computers.
"what would constitute a solution and the nature of any human thought"
++ We have examples: Checkers, Losing Chess, Connect Four, Nine Men's Morris...
"we don't count guessing as a solution" ++ No, nobody does.
@6584
"for all practical purposes it's already been solved"
++ For all practical purpose chess is ultra-weakly solved: the game-theoretic value of the initial position is a draw.
We also have > 1000 perfect games with optimal play from both sides: ICCF WC Finals draws.
However chess is not yet weakly solved.
We know chess is a draw, but we do not yet know exactly how.
How to draw against 1 e4, 1 d4, 1 c4, 1 Nf3?
Here is again the proof that 1 e4 e5 2 Ba6? loses by force for white.
If white plays differently, then white loses quicker.
A win is easier to prove than a draw.
Even without such constructive proof it is and was known that 1 e4 e5 2 Ba6? loses: a full piece down with no compensation of any kind.
I gave above proof that 1 g4? loses by force for white, which is more surprising a result.
Dec 12, 2022
0
#4987
tygxc wrote:
@6581
"computers are pretty irrelevant" ++ Without computers it is not possible.
Only with 3 powerful computers during 5 years non stop is it feasible to weakly solve chess.
10^17 relevant positions is too huge a number for unaided humans.
Gauss managed quadratic reciprocity for ℵ₀ pairs of primes. That's more even than than the number you should have quoted. And they hadn't even invented computers.
In any case with your definition of "solved" it doesn't even need an abacus.
"human intelligence will play a far larger rôle"
++ Yes, incorporation of chess knowledge is beneficial.
That is also why Sveshnikov asked first for good assistants and second for modern computers.
I said "intelligence". Not the same as "stupidity" - look it up in Webster's.
"what would constitute a solution and the nature of any human thought"
++ We have examples: Checkers, Losing Chess, Connect Four, Nine Men's Morris...
"we don't count guessing as a solution" ++ No, nobody does.
What, guess or count their guesses as solutions? Some people give a very good imitation of both.
tygxc wrote:
@6584
"for all practical purposes it's already been solved"
++ For all practical purpose chess is ultra-weakly solved: the game-theoretic value of the initial position is a draw.
We also have > 1000 perfect games with optimal play from both sides: ICCF WC Finals draws.
However chess is not yet weakly solved.
We know chess is a draw, but we do not yet know exactly how.
How to draw against 1 e4, 1 d4, 1 c4, 1 Nf3?
We don't know chess is a draw. For example, I know it's a forced win for white. So what "we know" is really just a guess. And as Mar said, guessing isn't a solution.
When chess was invented it was intended to be a contest between two people. The better player wins, regardless of color choice. So what I mean when I say for all practical purposes it's been solved is that chess is no longer just a contest between two people. Now, computers will beat humans every time. So, the game of chess as it was intended has been solved. Computers always win.
But as far as which color wins, or doesn't win, with computers, we don't actually know. We guess. So as long as we are guessing, white wins by force.
tygxc wrote:
Here is again the proof that 1 e4 e5 2 Ba6? loses by force for white.
If white plays differently, then white loses quicker.
A win is easier to prove than a draw.
Even without such constructive proof it is and was known that 1 e4 e5 2 Ba6? loses: a full piece down with no compensation of any kind.
I gave above proof that 1 g4? loses by force for white, which is more surprising a result.
There's no proof using engine analysis. Engines are imperfect, as proven by the fact that they still get measurably better with every release. This has only been pointed out to you about a hundred times in this and other threads. That isn't even hyperbole.
Your whole premise to solving chess is to redefine the solution into your dead hero's statement. Backwards, as ever.
Dec 12, 2022
0
#4990
Optimissed wrote:
MARattigan wrote:
I agree that computers are pretty irrelevant apart from speeding things up a bit (a bit in this context). If chess ever gets a practical solution (it's been solved according to @tygxc's definition probably since it was first thought of) I think human intelligence will play a far larger rôle than Superficial Intelligence.
The difference between you and @Elroch and I is not about computers. It's about what would constitute a solution and the nature of any human thought that would arrive at one (though computers could be necessary to effect the arrival). In simple terms we don't count guessing as a solution.
The difference is between you and me and not between you and I. Anyway, it's your assertion that I'm guessing.
Imagine this. You are asked to add together 27 and 65. You can't do it because it involves a "carry over". Teacher gives you the right answer though, but you accuse her of guessing because she can't possibly work out such a difficult sum.
It's just the same, even though you may think it isn't. To people who understand chess, the answer is just as clear as 27 + 65.
The difference is between you and @Elroch and me, but point acknowledged.
The cases aren't the same. The teacher would be able to work out the sum, but you can't work out the position - otherwise youl'd have no problem with my challenge to demonstrate it against SF.
Dec 12, 2022
0
#4991
tygxc wrote:
Here is again the proof that 1 e4 e5 2 Ba6? loses by force for white.
If white plays differently, then white loses quicker.
A win is easier to prove than a draw.
Even without such constructive proof it is and was known that 1 e4 e5 2 Ba6? loses: a full piece down with no compensation of any kind.
I gave above proof that 1 g4? loses by force for white, which is more surprising a result.
Is that pathetic or is that pathetic?
It's possible - in fact likely - that the advantage of the first move might be enough to win by force. A hypothetical super power chess computer running at insane speeds might claim, prior to making its first move: "Mate in 2,212,598,303,505,004,977"
Dec 12, 2022
0
#4993
I nowhere said the position is not winning for Black. I don't know. Neither do you.
I did in fact try Black against SF a couple of times and both turned out to be losing for Black.
Dec 12, 2022
0
#4994
HurtU wrote:
It's possible - in fact likely - that the advantage of the first move might be enough to win by force. A hypothetical super power chess computer running at insane speeds might claim, prior to making its first move: "Mate in 2,212,598,303,505,004,977"
But not if the 50 move rule is in force.
Dec 12, 2022
0
#4995
Contrary to @Optimissed's claim, I am 100% certain that those who claim that 2. Ba6 loses are guessing. Of course those who are the subject of this knowledge may erroneously believe I am only guessing this, but they would be wrong to do so.
@6595
"A hypothetical super power chess computer running at insane speeds might claim, prior to making its first move: Mate in 2,212,598,303,505,004,977"
++ No, that is mathematically impossible. There are only 10^44 legal positions, most of them nonsensical. After a search of width w candidate moves with depth d moves we reach w^d positions, assuming we reduce w to avoid transpositions.
This gives the following maximum depths for various widths:
width depth
2 148
3 93
4 74
5 63
6 57
7 52
8 49
9 46
10 44
11 42
12 41
13 40
14 38
15 37
16 37
17 36
18 35
19 34
20 34
Any checkmate in more than 148 moves must consist of forced moves only and by both sides.
The hypothetical super power chess computer running at insane speeds can only state, prior to making its first move: 'I offer a draw, because it is a draw'.
Dec 13, 2022
0
#4999
I feel my post #6600 works equally well as a joke and as a statement of fact. It is only a shame that it was not post #6660.
I'm honestly impressed that this forum is still basically on topic.