Will computers ever solve chess?

Current engines are limited by their evaluation algorithms. They can only calculate to the best that their coding allows.

Alpha Zero highlighted some of the flaws in Stockfish 8's programming—SF8 thought it was doing fine in positions where AZ was actually winning.

Stockfish is a monster that uses centipawn values to weigh its lines. But the chess-playing AIs of the future might very well determine that centipawns are actually an inaccurate way to approach chess.

There is no doubt that they are. I can set up a legal position where Stockfish thinks it is 30 pawns up at 100 ply depth, and yet it is a dead draw. Chess is a game where the only significant thing is the final score, and the only reasonable proxy for that is an estimate of the final score (there may be nuances that differentiate between positions with the same expected score but different win probabilities, but these only matter if you are interested in something other than scoring highly).

chess.com announced today they have acquired Komodo with a new version adding Monte Carlo search and probabilistic assessments of this type, like a hybrid of AlphaZero or  LeelaZero and a conventional chess engine.

AZ already seemed to hint at that: it approached the position with a clear emphasis on mobility, rather than pawn/piece values.

The reason for this was that its experience found that the combination of many positional factors was a better way of estimating the result than being excessively focused on material. A couple of the published games illustrate this stunningly.

You cannot detect perfect moves in the beginning of the game. Says you.

Since you cannot do that, the idea of finding perfect games is quite silly, because you start with imperfect moves until the position is simple enough to let you find best moves.  What proof do you have that i start with incorrect moves??

But because you started with imperfect moves the resultant lines/ games are not perfect. This would be true if i started with imperfectt moves but what proof do you have that i start with imperfect moves?

Obviously. You keep failing to address this anomaly in your thinking. you failed to address most of what i said post #6385.

The burden of proof belongs to you. I’m not the one making statements. I simply said: I don’t know, nobody knows. If you think you know, you have to prove it. In this case, you have to prove  that the beginning moves are perfect. And unless chess is solved, you cannot do that. 

By the way, you didn’t succeed in loading any games, on both occasions.

maybe this will show up?

White to move, find the best moves. [there is only one set of best moves in this problem]

s23, just not lying about what Ponz and I have said!!! These People even do not apologize after it was exceptionally clear to everyone with a brain , due to multiple repetitions of us,  that they have wronged us.

When you say "I don't know, nobody knows." That is a statement.

I dont see 

It IS true that a relative truth is truth. 

I would guess it is WAY less than 5% of GMs who think chess is not a draw with best play on both sides. Probably less than 1%.

In my whole life i have only encountered 1 GM who thought White was a win with best play and he was promoting a book which said that.

 After the book was printed he was beat rather badly using his ideas.

I think his book promotion got to his head. However, even in his book, he could give no line where White had a win from the first move. 

When a grandmaster says chess is a draw, they usually mean that games played between two humans at grandmaster-level will most often (but not always) end in a draw. You can read minds to twist what he says? If he says chess is a draw--he means chess is a draw.

But I'm not aware of any grandmaster who has said unequivocally that chess has been mathematically proven to be a draw. Of course not! grandmasters are not stupid. Just about everybody knows that chess has not been mathematically proven to be a draw. I don't say that chess has been mathematically prove to be a draw either. 

Human play and perfect play are two different things. Very usually you are correct. But you are wrong in assuming a human cannot play a perfect game.

I have not seen anyone make a claim (other than speculation) that one type of game has the same result as the other. I do not even know what this means?  Do you mean nobody has made the claim that duplicate bridge has the same result as checkers?  Your sentence is so ambiguous as to not make much sense?  Maybe you could clarify what you mean?  

  There you go again: after playing 50 non-best moves, you are introducing an ending where a best move is possible to spot. I’ve already described this situation: it proves zilch.

  There you go again: after playing 50 non-best moves, you are introducing an ending where a best move is possible to spot. I’ve already described this situation: it proves zilch. you are making another mistake in  your assumption--this position DID NOT come from 50 non-best moves!