Chess will never be solved, here's why

or 94 or 83 mins of chess with @ifemo

#2035

"You think that Stockfish makes no more than one losing error in every 100,000 moves"
++ I infer that by extrapolation from the peer reviewed Alpha Zero paper. I know the published error rate at 1 s/move and 1 min/move and extrapolate to 1 h/move and 60 h/move and thus arrive at 1 error of the top 1 move per 10^5 positions. Thus I conclude that including the top 2, top 3, top 4 moves should lower this to 1 error per 10^20 positions, which is more than the expected search space of 10^17 positions.

yeah

#2052

"I chose the endgame because of the number of zugzwangs."
"the majority of @tygxc's processing time would eventually be spent in such positions."
++ Zugzwang is typical for endgames and thus happens more after the table base is hit. There are a few famous Nimzovich games with middle game Zugzwang, but those are rare exceptions.
As said earlier the errors are because of the failure of the evaluation function. Winning the pawn in KNN vs. KP is drawing.
"Three trillion moves - now that would be long."
++ There is less evidence that a 3 trillion move checkmate exists than that the Loch Ness monster, the Spaghetti Monster, or Martians exist. From GM and ICCF games I rather tend to believe that a forced 3-fold repetition or table base draw can be reached in 50 moves.

By including three more candidate moves -
errors thus get divided by 1000 trillion ??
looks like another Leap !  

"error" might be poorly defined too.
Does a supercomputer really 'know what its doing' in chess ?
Its a paradox.
People would say yes - because it beats the world's best human players.
But its also no - because the computer is only following formulaic instructions from human programmers.
In tactical situations - supercomputers would be great.  Fantastic.
But what about 'positional' ?
Insisting the computer - in picking 'best move' only makes a 'mistake' once every 100,000 moves ?   In relatively 'positional positions' ?
Doesn't make sense.
There may be several 'best' moves in such positions.
And several 'second rate' moves.  
The computer is assigning numerical differences to evaluate positional moves according to obscure minor technical points which may have little to do with 'best move'.  

Edited moderator SB 

No advertising 

And I repeat - finding 'best move' in unsolved positions - is dubious at best.
The opening position is unsolved.
Computers cannot solve it at millions of 'nodes per second' ...
the positions generated from it are also unsolved. - also cannot be 'solved' at whatever 'nodes per second'.  And so on.

Arguing that 'best move' is found accurately in 99,999 out of 100,000 unsolved positions is ridiculous.
That might work in some short already-solved tactics puzzles ....
But if a position isn't solved - it can also be argued that much of the time if not most of the time - the best move just isn't known.
Nor the best four moves.
Especially since most unsolved positions - continue to be Just That !
Unsolved !  

"Circular argument" 
"oh the computer finds best move in an unsolved position - so therefore since it 'found that move' it follows that the position is therefore now 'solved"  (in positions where the best move can't be known because it isn't solved)

#2036
"If White wins in any of these lines, it would require no less time, because the engine should check White's strategy against all the reasonable alternatives for Black (you say that), not just one."
++ That does not happen: white does not win in any of these lines. If you just let Stockfish autoplay at reasonable time even on your desktop then you end with... a draw. That is empirical evidence, not proof. So the setup can assume it does not happen. In the unlikely case it would happen, the calculation must stop and the good assistants must step in and investigate what is happening.

#2060

"By including three more candidate moves - errors thus get divided by 1000 trillion ??"
For 1 candidate move:   1 error in 10^5 positions.
For 2 candidate moves: 1 error in 10^5 * 10^5 = 10^10 positions.
For 3 candidate moves: 1 error in 10^5 * 10^10 = 10^15 positions.
For 4 candidate moves: 1 error in 10^5 * 10^15 = 10^20 positions.

'"error" might be poorly defined too.'
++ error = move that turns a drawn position into a lost position

"because the computer is only following formulaic instructions from human programmers."
++ No, AlphaZero got no other instructions but the rules of the game

"In tactical situations - supercomputers would be great.  Fantastic."But what about 'positional' ?"
++ Positional = tactical at more depth

"There may be several 'best' moves in such positions. And several 'second rate' moves."
++ There are only two kinds of moves: good moves and errors.
Good move = move that turns a drawn position into another drawn position
Error = move that turns a drawn position into a lost position

"The computer is assigning numerical differences to evaluate positional moves according to obscure minor technical points which may have little to do with 'best move'."
++ The computer essentially counts the material in pawn units with some tweeks. This is a good approximation for the middle game. It is totally flawed in endgames, like KNN vs. KP.

#2068
weakly solved means that for the initial position a strategy has been determined to achieve the game-theoretic value against any opposition
https://web.archive.org/web/20170912011410/https://pdfs.semanticscholar.org/8296/bc0ab855841088b31190c9f2923951853d7b.pdf 

Your mum will never be solved

No it doesn't.

The White strategy

         Play 1.Qa7+

achieves the game theoretic value against any opposition. Is that a weak solution of that initial position?

I already pointed that out. Do you read the posts addressed to you?

If you're proposing to find to find a weak solution of "chess" it would probably be a good idea to find out first what "weak solution" means.

Edit: Fixed position.

As I already said, I was just following your reasoning:

Did you change your mind? Fine, it's a draw, then. But that does not answer the question I asked the last time (and thrice before):

@MARattigan, an objection can be made to your last post, but I will leave it to @tygxc 😁

Yes, will fix.

Extrapolation is not necessarily at all reliable. There is not even a unique way to do it, especially from (absurdly) two data points! (Transform the values using a wide range of monotone functions to get pretty much any answer you want).

But neither would it be reliable with 100 data points unless you know the form of the relationship. You don't.

Thanks for clarifying that those numbers of yours must be ignored.

Fascinating. 

How did they work out the error rate? Have they beaten you to it?

(And were the games played under basic rules with the addition of a double repetition rule. If not that would have a significant effect on the error rate.)

When one side is winning, a strategy is only required for that side to have a weak solution. That is the case here, so there is no need for a strategy for white.

A weak solution requires a black strategy that will reach a win against any white defense.

I know that and you know that, but going by his definition @tygxc doesn't.

It should read:

weakly solved means that for the initial position a strategy has been determined for a player with a best game-theoretic value to achieve the game-theoretic value against any opposition

But since you don't know the game-theoretic value until you have a real solution, that's not very useful.

Better:

weakly solved means that for the initial position either a timely strategy has been determined for one player that achieves a win against any opposition, or a timely strategy has been determined for both players that avoids a loss against any opposition. 

Edit: Added the word "timely" to the definition in the last paragraph. Otherwise there is the obvious solution of just trying out all possible continuations up to a maximum length of the number of possible positions. 

#2079
I used the definition by Prof. van den Herik and I gave a reference to his paper.
Personally I prefer a simpler, more practical definition but then I get reproached of using my own definitions instead of the generally accepted ones, which is what you do here.
You even find fault with his generally accepted definition.
I am glad I am not the only one being critiqued.
You should take your criticism to prof. van den Herik this time.