Will computers ever solve chess?

I never flip flopped. There are many plausible time-frames for when chess can be solved based on current knowledge. But they don't agree with your assertion that "chess can never be solved". I'll show you again a Venn diagram why 18 years is plausible, because it is within the known range of when chess might be solved:

surely if a match is "perfect" white may try to dictate the opening strategy but black has an equal chance to make it a black strategy. If perfect play both have equal chances of winning and it must therefore culminate in a draw. It is only the imperfection that allows a win. 

If I play a GM with a book opening I know by heart and so achieve to reach say move 12 with a balanced board. What is inevitable? I make a mistake or the GM sees an opening and so ultimately wins. If I play a perfect game then being perfect I cannot lose. If this is true for white it is also true for black. But if both play perfectly then neither can win so it must be a draw.

This line of thinking does not account for subtle tactical tricks (like triangulation or Zugzwang).  This is why it's so unclear as just a thought experiment: to be sure the result is a draw, we would need to know that every candidate line contains no such tricks

I find it interesting too that the current research is approaching the problem from both sides.  As endgame tablebases expand, the path from the starting position to any solved tablebase entry becomes shorter.

If you can prove that a theoretically drawn endgame can always be reached from the starting position, you've solved chess!

In a perfect game their would be no tactical trick in being a perfect game. I can play a perfect opening game, but any slip means I will lose. So perfection does what it says on the tin, it has no slip or trick, miscalculation or oversight. Playing computer v computer is not a perfect game either as it is really only programmer v programmer. A perfect game has no brilliancy and the small advantage for white in the opening even over the board can be nullified. 

https://www.quantamagazine.org/why-self-taught-artificial-intelligence-has-trouble-with-the-real-world-20180221/

As always, you go back to your straw man, and your rinky-dink diagrams.  Thanks for not adding a link  to a paper you don't even fathom.  You know damn well that I have never once said "chess will never be solved".  It won't be solved in our lifetimes, though.  I'll try to remind you of this for as many decades as Chess.com and this thread exist .

Meanwhile, you are hiding behind "there are many..." as usual.  I'm not talking about some nebulous many, I'm talking about your directly quoted numbers.

All I will say about ^ is that technological discoveries can be surprising heck maybe even mathematics will solve it before computer science.  Or maybe mathematics will prove that it cannot be solved even with computers!

A Mars landing in the next decade would qualify as "surprising".  Solving Chess is way, way, way, way beyond that difficulty.

My point is that maths has a way of turning mountains into a bunch of seemingly unrelated molehills.  Perhaps some abstract symmetry on the tree will be discovered that reduces the problem exponentially, for example.  Have they even invented abstract algebraic game theory yet?

As a mathematician, I can be very confident that chess lacks symmetries that would reduce the size of the computational problem much more. It is an extremely arbitrary game.

People lack imagination. They can't understand how big chess is. They're capped at "really big" and since computers are "really fast" it should be "really possible"

But that's now how scales work, and "really" is a feeling, not an amount.

Some problems are too complex to ever be solved, regardless of how fast a computer is.

 I am aware that 'really' is not a number but thanks anyway.  The undecidability thing is what I was getting at in #4505

Yeah, when people hear it, in hindsight they realize it should have been pretty obvious.

I think the actual number is something like 10^42.

So you can do the very simple math from there... that it's far beyond the technology humans are likely to ever achieve, even if we're around for 100 million years. Solving chess is very simple, we already know how, the problem is 10^42 means there's a lot of it to solve

I thought it was more on the order of 10^120.  Isn't ~10^80 the number of atoms in the observable universe?

I'd also prefer to receive replies that aren't so condescending, if that's okay with you.

IIRC 10^120 is games, and so there'd be a lot of duplicate positions.

10^42 (something around there) are (roughly) all the positions we'd need a win/loss/draw eval for to make our 32 man endgame tablebase.

So if we could store 1 position + its eval on the size of an atom, it would take something around the size of the sun just to store all the positions... so... good luck with that humanity.

Again, very easy to do, the problem is there are so many positions.

Yes that comparison rings a bell

Yeah, I've used it a few times over the years on this site.

It is indeed the number of positions that is more relevant. A tablebase has no need to store every possible ending: it merely needs to contain the evaluation of every move in every possible position (as it involves very little extra computation, tablebases using contain the number of moves to mate against the most obstinate defense for decisive moves). Certainly, a lot of positions can be omitted as not being relevant to a specific strategy, but what is needed is to reduce the logarithm of the number of positions a lot, and this is not so easy.

The Sun is no good for storing information, even if we could, because of its temperature! Finding as much cold matter would be tricky. I'm beginning to think chess is not worth the bother.  As a cautionary tale, here is Arthur C. Clarke's short short science fiction story published in Isaac Asimov's Science Fiction Magazine, First Issue, Vol 1, No. 1, Spring 1977

_______________________________________________________________________________________________________

Quarantine 

Earth's flaming debris still filled half the sky when the question filtered up to Central from the Curiosity Generator.

"Why was it necessary? Even though they were organic, they had reached Third Order Intelligence."

"We had no choice: five earlier units became hopelessly infected, when they made contact."

"Infected? How?"

The microseconds dragged slowly by, while Central tracked down the few fading memories that had leaked past the Censor Gate, when the heavily-buffered Reconnaissance Circuits had been ordered to self-destruct.

"They encountered a - problem - that could not be fully analyzed within the lifetime of the Universe. Though it involved only six operators, they became totally obsessed by it."

"How is that possible?"

"We do not know: we must never know. But if those six operators are ever re-discovered, all rational computing will end."

"How can they be recognized?"

"That also we do not know; only the names leaked through before the Censor Gate closed. Of course, they mean nothing."

"Nevertheless, I must have them."

The Censor voltage started to rise; but it did not trigger the Gate.

"Here they are: King, Queen, Bishop, Knight, Rook, Pawn."

How did you arrive at this conclusion?