Chess will never be solved, here's why

@13242

"50 move info in the hashed positions"
++ None of the ICCF WC Finals draws end by the 50-moves rule.
An average ICCF WC Finals draw lasts 39 moves.
Those games are 99.992% certain to be perfect games with optimal play by both sides.

That is until they prove kings gambit is a draw or not

",oh but know one is foolish enough to do that "

Well stockfish drew occasionally against the kings Gambit they might not have winning chances but actually PROVE it

@13222

"@tygxc defines "position" as FEN-{move number,ply count} and then assumes his positions correspond 1-1 with nodes in the competition rules game tree and also in the ICCF game tree."
++ No.
Diagram = location of men on the board

Position = diagram + side to move + castling rights + en passant flag = FEN without move # or ply count. That is not my definition but that of 9.2.3

Node = position + history (takes care of 3-fold repetition and 50-moves rule) and provisional, heuristic evaluation, e.g. +0.33. That is not my definition but here.

generally 1 diagram = 2 positions, except when a king is in check
1 position = 0.5 node, except in up/down symmetrical positions

after Nxa6 white is a piece down.
But a bishop down doesn't always lose.
Chess theory says a rook down without compensation loses - apparently based on the idea that the rook and king can mate without other assistance.
But in a situation of exchanging down - a bishop up - black will not be able to mate with King and bishop versus King.
And - can black always force other pieces to be exchanged?
No.
White still has six pieces to mate black's King with and still has eight pawns for that too. 8 promotable pawns.
------------------------
We don't know whether white can't force a draw or not.
But do we 'know' that white can't force a win from there?
Technically - no.
Because the 'game tree' with 31 pieces still on the board is just too labyrinthine.
-----------------------
endgame tablebases only go up to seven pieces.
But is there any reason that computer projects would not be finding many 'forced win' and 'forced draw' positions from all the other situations?
Yes - if there are too many pieces on the board.
Because then you've got 'game tree' instead of 'retrograde tablebase analysis'.
But maybe various projects have found zillions of 'forced results' when pieces are fewer than 17.

so by definition not a solution of any sorts.

you either have complete certainty or you have to treat each case as worst possible outcome. thats how a mathematical proof works.

@13210

"if you replace 35 with 3"
++ Let us look at a relevant position.

This occured in 11 ICCF WC Finals draws, 9 times with this move order and twice by transposition.
It also occured in > 100,000 human master games.
It is thus a relevant position.
Legal 6th white moves ranked:

1. Be3: 5 ICCF WC Finals draws
2. f3: 3 ICCF WC Finals draws
3. Bg5: 1 ICCF WC Finals draw
4. Nb3: 1 ICCF WC Finals draw
5. Bd3: 1 ICCF WC Finals draw
6. Be2: >10,000 human master games
7. h3: > 8000 human master games
8. Bc4: >7000 human master games
9. f4: > 4000 human master games
10. g3: > 3000 human master games
11. a4: > 1000 human master games
12. Rg1: > 400 human master games
13. Qd3: > 100 human master games
14. a3: > 100 human master games
15. h4: > 100 human master games
16. b3: > 10 human master games
17. Nf3: played in a few human master games
18. Qd2: played in a human master game
19. g4: played in human master games, loses a pawn, but gets an open file and 2 tempi as compensation
20. e5: played in 1 human master game, loses a pawn, no compensation
21. b4: can be dismissed by logic
22. Na4: can be dismissed by logic
23. Nde2: can be dismissed by logic
24. Rb1: can be dismissed by logic
25. Nf5: loses a tempo
26. Nce2: loses a pawn
27. Nd5: loses a pawn
28. Nb1: loses a pawn
29. Ncb5: loses a knight
30. Ndb5: loses a knight
31. Nc6: loses a knight
32. Ne6: loses a knight
33. Bb5: loses a bishop
34. Qg4: loses the queen
35. Qh5: loses the queen

@13227

"neither are ICCF players. In fact they're even lower rated than me"
++ In the ongoing ICCF WC Finals, Turgut and Chytilek are over the board IM and Terreaux is FM.

none of those moves are addressed with any mathematical rigor, try again.

tygxc why havent you addressed the fact that I had mathematicians personally verify that your arguments are completely delusional?

@13240

"whether the position after 1.e4 e5 is a White win, a draw or a Black win"
++ It is a draw. 21 ICCF WC Finals draws, 99.992% certain to be perfect games with optimal play by both sides prove it.

Retrograde tablebase analysis offers practical potential to thoroughly solve many (but not all) positions not just with 8 pieces on board - but with more than that.
Because the computers can 'handle' the greatly reduced number of possible positions as compared to 'game tree from the front'.
But 'game tree forward' analysis from 32 pieces offers no practical potential to 'thoroughly solve'. There's too much game tree.
That's why even e4 e5 Ba6 isn't 'solved'.
But for effective 'gameplaying' purposes - Ba6 there can be 'pruned' though.
Gameplaying isn't 'solving'.

ah yes, proof by high probability, accepted in which mathematical journals???

tygxc why arent you addressing the fact that I personally had mathematicians verify that your arguments are delusional?

tygxc doesn't want to get it that high probability isn't proof.
He has disdain for mathematical objectivity.
Its interesting?
By his 'logic' what would be some common unproven things that would then be 'proven' in his view?
Future events?
A weather forecaster knows its raining hard three miles east of him and the prevailing winds are blowing the rainstorm west towards him.
With those facts in place and him then assigning 99% chance of rain where he is - does that mean he's now 'proven' it?
In tygxc's world - apparently that's a Yes.
But in reality its a No. Its not proven.
----------------------------
In what kind of situation would tygxc be 'right'?
A movie set is about to have a shooting scene.
A real gun is put on a table but its unloaded although having a 'clip' attached.
Before filming:
All the actors and movie staff file by the table and examine the gun - removing the clip and making sure its empty - racking the chamber and making sure that's empty and looking down the barrel and making sure its clear.
Is it now 'proven' that the gun is unloaded? In a 'tygxc' way?
I would say Yes.
Some could say 'No. That's not solved. Everyone on the set could be drugged and it doesn't register on them that the gun is chambered with a cartridge.'

extremely low probability death risk problems have always been a fascinating problem for me.

tygxc i see you read my questions, so answer them.

Proof by high probability is accepted in which math journals? you claimed a high probability was a proof. so Im asking which math journals will accept that proof.

Why arent you addressing the fact that I went and talked to mathematicians and they personally verified what ive been saying all along, that your arguments are delusional?

I got reprimanded for wasting their time with arguments as bad as yours, so how about you make it worth my time and address it?

I'm in the same situation. I suspect that the only source explaining what is done is delving into the details much. User documentation doesn't seem to have a high priority.

And which mathematical journals accept ludicrous red telephone figures for probabilities anyway?

True. This is why I contrasted with Tromp's work, where probabilities are based on perfectly random samples from well-defined sets. Like so much of what @tygxc says, there is no valid reasoning behind it, merely the reasoning of a practical chess player trying to make a choice of move in an uncertain position.

@tygxc, you need to answer my question about you using "logic". It seems there is not a single example of you actually using logic, based on your failure to reply thus far.

Also, you need to admit that your trust in the "perfection" of games played by an engine that cannot play accurately with 6 pieces on the board is based on a religious reverence for the evaluation function that is the sole basis for ignoring moves that could logically refute analysis.

[Note: given his inability to address any questions put to him, I see @tygxc has taken to manically down voting all posts by everyone else! If only he could do logic as well as click].

He seems to have answered your question. There's a -1 next to the down arrow.