Chess will never be solved, here's why

Characteristic of many chessplayers: they want conversations to be verbal chess games with winners and losers. But they project that obsession though. Common with obsessions.
Its a kind of Bobby Fischerism. 'the idea of chess is to deflate the ego of the opponent'.
Bobby's obsessions with that idea spilled over into other aspects of his life.
Unfortunately for him.
Part of his paranoia and probably led to his early death.
Fischer and Morphy being the most clear examples of mental illness among the best players in the world - in their time.
------------------
VCY (has good intentions) caught the projection.
I just hope 'O' and team don't succeed in baiting VCY into something.
'O' is known for baiting people and then reporting them.
-------------
and 'D' just tried to divert from his '100% tactical' ...
Does anybody care?
When people watch a fictional movie do they really care about what happens in the fiction?
No. But they have preferences.
A forum could be regarded as a kind of 'movie'. But posters are in the movie.

This thread is getting too excessive to follow throughly but if I get the gist of what you're saying correctly, I agree. Both "positional" and "tactical" are man made generalizations and extra vague for a game like chess. Computers just calculate the same way regardless of whether it exploits specific tactics or positional weaknesses. Weaknesses in pawn structure can leave open squares for tactics/mating nets/easier threats on uncoordinated pieces, and tactics can be used to pry open positions and make them more vulnerable. It's all intertwined. I would say that it leans more tactical, since to "win" you need to checkmate which means you need to win enough material and/or tactically exploit the king, but the actual execution of a winning sequence is going to be tactical at some extent, whether it's a winning endgame, material advantage, and/or mating net.

This is why computers often make bizarre looking moves like king walks instead of castling..etc, because they don't fall for the "emotional" generalizations of "king being cozy in the corner"..etc. They calculate deep and realize that walking the king to d2 or even e3 ends up having a subtle advantage 30 moves later in the lines. This is exactly why I am so impressed with how human chess players with all these flawed generalized concepts, manage to program such computers that seemingly do the opposite.

We know computers aren't playing "positional chess" against each other because both are calculating million of positions and thousands of lines dozens of moves deep each. They aren't going by surface-level things like how nicely the pawns are configured or total piece development. If you watch gothamchess computer match recaps, you'll see games it decides to mysteriously push an hour pawn to h5 or h6 in the middle of some opening sideline, and then 40+ moves later, it turns out that the pawn ends up being just close enough that it makes the position winnable after a ton of complicated maneuvering and that ends up being the reason it randomly chose a sideline of pushing an h pawn in the opening. The real purposes of seemingly random moves are realized dozens of moves later, which isn't how "positional" chess works.

Now somebody else is adding a new invalid 'wrinkle' and saying he 'agrees' with the other guy while actually not agreeing - with the other guy trying to insist chess is 100% tactical and 'positional' is fiction but with the 'somebody' suggesting they're both fictional.
As for 'we know' its actually 'somebody believes' that computers aren't playing positional chess.
Sure they are. Including when they play d4 or c4 or Nf3 on their first move.
When the computers play e4 though - maybe's there's a case that's not as weak for that move being tactical. An argument that it makes it less effective for black to respond with the very valualbe and important Nf6 at that point since white would then have the tactical pawn push immediately - and immediately thereby upsetting black's knight at its key post.
-----------------------------------
Computers do play positional chess and they're good at it.
But they don't go at it as humans do? Humans use heuristics.
Apparently computers use heuristics also. Different heuristics.
Those who maintain objectivity will keep in mind that chess isn't solved.
And will therefore realize that computers don't play perfectly.
----------------
Also - hyping on Kasparov's loss to Deep Blue as 'breaking his mind' is silly.
Obviously it upset him. Doesn't mean it broke him. 
It didn't.

Regarding Kasparov's loss to Deep Blue - Kasparov continued to be the top player in the world for another 8 years after that. Obviously Kassparov wasn't 'broken'.
And Kasparov's peak rating of 2851 has only been exceeded by Carlsen who has now held the top FIDE rating in the world since July 2011 with no interruptions.
(will somebody now get very excited and shout 'No! Computers have acchieved ratings over 3500!!)
In advance - I'm obviously talking about human players' (I'll refrain from adding 'you dope!' since that would tend to lower me towards the level of such persons)
--------------------------
Carlsen's peak rating is 2882. Only 15 players have achieved 2800 or higher.
Do I 'swear' by these numbers? No.
But some people try to maintain the earth is flat though. (or their 'correspondents' in chess trying to spam counterparts to flat earthism about chess)
What was also remarkable was that I first asked chatgpt for the info - and I quickly saw it was messing up. But like Copilot - chatgpt couldn't correct itself on it.
So I simply copied the session and posted it to Grok and asked Grok to correct the errors - which it found quickly and kind of ruthlessly.
I intentionally didn't mention it was a chatgpt session but Grok recognized it was chatgpt anyway and said so.
--------------
Summary: obviously Deep Blue did not 'break' Kasparov.

Deep blue wasn't fully developed to its max capability when it lost to Kasparov the first time. The programs now could beat all the top GMs working together as a team. Don't know if any of you watch star trek but that's why the scene where Deanna Troi beats Data at 3d chess has got to be one of the most laughable mistakes in the entire franchise. She or any of the other crew members wouldn't be able to beat a pocket smartphone let alone a 24th century supercomputer, in 3D..

'When it lost to Kasparov the first time' which was and is irrelevant.
Nor was there relevance to science fiction.
EE having phobias about 'mistakes' in science fictions that never pretended to be valid in the first place?
------------------------
Translation of EE post:
EE conceding he and D are wrong with their fiction about the realities of tactical and positional play and EE conceding that Deep Blue did not 'break' Kasparov who continued to be the top rated grandmaster of all time until Carlsen replaced him in that role several years later.
By 31 points.

@Optuimissed, your argument against pedantry was based on the implicit assumption that if someone uses the word "position" it will be understood as "game state". If not, then it was simply nonsense, because it is game states that matter and if you try to use positions (normal meaning) instead, reasoning fails.

Naive chess players (I am not suggesting any are here ) consider a position to be an arrangement of pieces on a chess board. Those with a bit more knowledge incorporate the player to move, full information on castling rights, and any en passant possibility. The reason is that all this information is needed for a basic chess game state, which determines all legal continuations and results with the basic chess ruleset.

The standard position representation notation for chess incorporates one more piece of information, the ply count since an irreversible move, as this is cheap and suffices for all n-move draw determinations.

This Forsyth notation and list of attributes describes the game state for the version of chess with an n-move (usually n=50) drawing rule. and no repetition rule. If this version of chess was specified, the information represented in Forsyth notation is both an (enhanced) position and a game state.

With the game of chess as we play it, with a 50-move rule and a repetition rule, Forsyth notation does not suffice to describe the game state. As in general it is very cumbersome to list all positions reached since an irreversible move has occurred, the full score of the game up to a certain move is used as the game state. However, I don't think anyone considers the score of a game to be a "position". That would be an non-standard use of the word. It does determine the game state. Notably, multiple partial game scores can determine the same game state, because only what is possible in the future matters. Eg no move variation - eg transposition - before the last irreversible move affects the game state.

In summary, it's appropriate first to specify the precise game (i.e. ruleset) you are discussing - "chess" is too vague a term for reasons that should be clear - and then to use the right game state definition for that game. If the word position is used, it is necessary to specify the definition because of the lack of standardisation mentioned above. (There is not universal agrreement whether it includes (1) who is to move (2) castling rights) (3) e.p. rights (4) information for a repetition rule).

As a last point let me emphasise the key general truth that what is of interest is (hopefully well-defined) concepts, and what we use to refer to such concepts are words and phrases. Forgetting the distinction is a source of much confusion (such as thinking that if there is a word like "position" or "chess", you can assume it always means the same thing regardless of the context). For example, above I have emboldened three phrases, referring to different games, all sometimes referred to as "chess".

hêlllo

este es el mejor juego que habia visto en mi vida. Le doy de puntos

This thread has seen much worse days in terms the LLMs doing most of the talking.

However Grok seems to be correct in a sense that the definitions of terms "perfect information" and "game state" fit chess just fine and precisely describe what they were intended to..

Again speaking for myself, I wouldnt mind discussing the terms and their definitions if you find chess has properties they dont fit well... or if you have another alternative definition in mind for the terms that would have a better use case than the existing ones, but you havent provided such.. So your protest for applying game theory to chess doesnt have that much basis yet.

You're posting Grok crap, not FACTS, but you can't tell the difference.

Is this a fact?

chess has an astronomical number of possible positions (estimated at 10^43 to 10^120, depending on how you count legal games).

It's quoted an obsolete lower bound and Shannon's number, which refers to an estimated number of games of a certain type. Just as you can't distinguish between a Stockfish evaluated node a game state and a game theoretic value, Gronk can't distinguish between games and positions (let alone say what it means by the latter).

You tell us in your own words what, "depending on how you count legal games", is supposed to mean in relation to the figures it's trying to quote.

If "position" is to be relevant in the context of solving, then it must be taken as "game state" and the figures are dependent on the version of chess. (What is ambiguous about how you count legal games for each version?) Grok appears not to have realised there is more than one version of chess each with different figures. (Have you? From the questions you ask it you don't seem to.) 

And again:

the game’s vast complexity (with an estimated 10^120 possible games)

Totally clueless. Under FIDE basic rules the number of possible games is infinite. Under competition rules a lower bound of 10^29241 is estimated here. Both just a bit different from 10^120.

Just like the team of GM analyst assistants at championship matches.

And a finite length of games. Former world champion Max Euve incorrectly claimed in his Mathematics PHD dissertation that an infinitely long game of chess is theoretically possible. Finite board, finite number of pieces, repetition rules..etc, not sure what I am missing but simple logic, all chess games ended eventually no matter how long you try and extend them out.

What you're missing is that the 50M and 3R draws do not have to be claimed and FIDE Removed the 50M and 3R rules from the basic game in any case in 2017.

Under competition rules game length is necessarily finite by the 75M/5R rules.

More Grok crap.

What are "standard rules"?

Under FIDE basic rules and pre 2017 competition rules there is no limit to the possible length.

Under FIDE competition rules the limit is exactly 8848.5 moves.

Under chess.com rules the limit is exactly 5898.5 moves.

You would.

Er, no it doesn't. Not if it's calculating the longest possible game.

In fact it doesn't really assume anything anywhere about chess players except they play according to the rules of the version they're playing.