If I come across the Herik book I'll read it but my instinct is that he's probably more interested in self-publicity than in honesty. I know what games theory is and I'm aware it's statistical in nature, because it consists of analysing real life situations rather than games, by TREATING them as games.
no its that tygxc as usual is completely misunderstanding and deliberately misconstruing what the article says lol.
@9195
"the game state assumes a perfect evaluation" ++ Yes.
"which you don't have" ++ I have the 7-men endgame table bases with their perfect evaluation.
White tried to win, black tried to draw. They reach a 7-men endgame table base draw or a prior 3-fold repetition. Black succeeded, white failed. All black's moves are justified in retrospect.
im still baffled at how you can justify this to yourself.
they reach a draw, so their play was perfect!?!?!?
by that logic, all my games that were draws had no errors on black's end.
absolute buffoonery.
im noticing how you cant address my other points tygxc. probably because they arent convenient to your.... idk your fantasy? thats all that i can think of to describe your position on chess's solvability. a fantasy. or an elaborate troll
not a single article you cite is taken in the correct context
egs: a gm claiming that computers will bridge the gap between mid game and end game theory -> you claiming hes talking about solving chess completely?!?!?! random chess computer databases of high level computer play. --> perfect chess play!??! (your logic is literally "according to this game engine, this SAME game engine made the best move")
not a single calculation you make is without basic errors
egs: your calculation of move speed accuracy is off by a factor of 100 million, because you double counted nodes as moves. read the article you cited, its one move per minute at a supposed 90+% accuracy, with 100 million nodes, not 100 million nodes each making one move per minute each at 90%+ accuracy.
Not a single mathematical concept is used correctly.
egs: you claiming that chess errors follows poisson distribution despite it only following one of the four properties needed, you claiming differences between ultra weak and weakly solved despite your "proofs" failing to actually take those differences into consideration. your complete lack of understanding how parity and zugzwang make strategy stealing impossible.
Chess be solved before 2040. 32.men EGTB isn't needed. 20 men EGTB and Expanding Opening Theory will meet in the middle. Some opening position is a clear win for one side.
Ofc with best play by both colors will end in.a draw.
@9211
"chess is a draw with perfect play" ++ Yes. If chess were not a draw, then all 104 finished ICCF WC finals games would contain only an odd number of errors (?): 1, 3, 5, 7... and none an even number of errors (?): 0, 2, 4, 6.... That is absurd. Thus Chess is a draw.
"this is perfect play because it ended with a draw"
++ No, this is perfect play because all 104 finished games between 17 strong entities ICCF (grand)master + engine at 5 days average per move ended in draws with not even one single decisive game. It is absurd that there would only be games with an even number of errors (?): 0, 2, 4, 6... and none with an odd number of errors (?): 1, 3, 5... Thus these must be games with 0 error (?), hence perfect play.
""chess is a draw with perfect play" ++ Yes. If chess were not a draw, then all 104 finished ICCF WC finals games would contain only an odd number of errors (?): 1, 3, 5, 7... and none an even number of errors (?): 0, 2, 4, 6.... That is absurd. Thus Chess is a draw."
your argument is literally "oh its absurd for there to be an odd number of errors"
thats an appeal to stone fallacy.
"++ No, this is perfect play because all 104 finished games between 17 strong entities ICCF (grand)master + engine at 5 days average per move ended in draws with not even one single decisive game"
thats literally exactly the same logic. they were draws, you claim the engines are perfect because they ended with draws.
there's nothing special about the engines.
saying that its absurd doesnt make it false lmfao.
yeah, a troller wouldnt make such a stupid line of logic. they would actually put THOUGHT into their claims.
The problem is that you need some method to determine if there were errors made for your method to have any validity. We know that GMs opinions of what is "best" evolves overtime, and we know that engines constantly improve, today's engines outperform yesterday's. Yet these are the flawed "authorities" you use to decide which lines are irrelevant.
its funny, im trying to think of a fallacy that you havent made, tygxc. (excluding wierd informal ones).
The problem is that you need some method to determine if there were errors made for your method to have any validity. We know that GMs opinions of what is "best" evolves overtime, and we know that engines constantly improve, today's engines outperform yesterday's. Yet these are the flawed "authorities" you use to decide which lines are irrelevant.
yep, but according to tygxc, because those engines resulted in draws, its a proof!
@9218
"The problem is that you need some method to determine if there were errors made"
++ If Chess were not a draw, then the 100% ICCF WC Finals' draws must contain an odd number of errors (?): 1, 3, 5... and none would contain an even number of errors (?): 0, 2, 4, 6...
That is absurd. Thus Chess must be a draw with an even number of errors (?) 0, 2, 4...
Likewise it is absurd, that the 100% drawn ICCF WC Finals' games would contain even numbers of errors (?): 0, 2, 4, 6, and none would contain an odd number of errors: 1, 3, 5, 7...
That is absurd.
Thus the 104 ICCF WC Finals draws must contain 0 error: optimal play.
"we know that engines constantly improve" ++ Yes, they make fewer and fewer errors over time.
"today's engines outperform yesterday's" ++ Yes, they make fewer errors.
"these are the flawed authorities you use to decide which lines are irrelevant."
++ No, the authority is the 7-men endgame table base. When a game ends in a 7 men endgame table base or a prior 3-fold repetition, that justifies all black moves in retrospect as fit to draw. It does not justify all white moves as unfit to win, but 104 attempts by 17 ICCF (grand)masters with engines at 5 days / move failed to win even a single game.
The above logical argument about the number of errors (?) proves that Chess is a draw and that all 104 games are perfect with optimal play by both sides.
chess can and will be solved, just needs more computing power and storage
@9218
"The problem is that you need some method to determine if there were errors made"
++ If Chess were not a draw, then the 100% ICCF WC Finals' draws must contain an odd number of errors (?): 1, 3, 5... and none would contain an even number of errors (?): 0, 2, 4, 6...
That is [very unlikely]. Thus Chess [is likely] a draw
Likewise it is absurd, that the 100% drawn ICCF WC Finals' games would contain even numbers of errors (?): 0, 2, 4, 6, and none would contain an odd number of errors: 1, 3, 5, 7...
That is absurd.
Thus the 104 ICCF WC Finals draws must contain 0 error: optimal play.
"we know that engines constantly improve" ++ Yes, they make fewer and fewer errors over time.
"today's engines outperform yesterday's" ++ Yes, they make fewer errors.
"these are the flawed authorities you use to decide which lines are irrelevant."
++ No, the authority is the 7-men endgame table base. When a game ends in a 7 men endgame table base or a prior 3-fold repetition, that justifies all black moves in retrospect as fit to draw.
This is ridiculous. The problem is doing the calculation to get to table base positions. This cannot be thoroughly done with any feasible computer. The table bases only offer a modest reduction in the size of the problem.
You cannot disagree that there have been decisive games between top players in recent years. These results PROVE the imperfection of those players (this is a correct use of the word prove - if two agents play a game with the the same rules twice and don't get the same result, at least one of them is probably imperfect.
It does not justify all white moves as unfit to win, but 104 attempts by 17 ICCF (grand)masters with engines at 5 days / move failed to win even a single game.
The above logical argument about the number of errors (?) proves [wrong] that Chess is a draw and that all 104 games are perfect with optimal play by both sides.
No, it is evidence that it is LIKELY.
Again, your lack of competence at dealing with uncertainty, combined with apparent lack of knowledge of the meaning of the word proof - a mathematical term - is leading you astray.
Is it possible for a fair coin to come up heads a thousand times?
Does getting 1000 heads by tossing a coin PROVE it is unfair?
Answer these questions correctly and you will see why your statements were wrong. [Hint: it's YES and NO].
Stats nerds that are wrong(bt kinda right ) vs chess nerds that are wrong(but kinda aright )
this feels like kramnik ....
I think tygxc would be a good opponent for kramnik he doesn't provide any proof so kramnik wouldn't be able to "disprove" what he is saying
@9225
"The problem is doing the calculation to get to table base positions."
++ It does not matter how the move sequences from the initial position to the 7-men endgame table base draw or a prior 3-fold repetition were generated: with with 1 or more computers,
by praying to Caissa, by dark magic to summon the spirit of Capablanca...
If a sequence ends in a 7-men endgame table base draw or a prior 3-fold repetition,
then that justifies all black moves as fit to draw, but not all white moves as unfit to win.
"The table bases only offer a modest reduction in the size of the problem." ++ Yes. But modern computers plus human ICCF (grand)masters can now bridge the gap between the initial position and the 7-men endgame table base, exactly as GM Sveshnikov predicted.
"there have been decisive games between top players in recent years."
human (grand)master + engine > engine > human grandmaster
In the 2024 Toronto Candidates' tournament there were average 1.1 errors per game.
In the ICCF World Championship finals there have been decisive games: every year fewer and now none. They now have reached perfection. If some alien intelligence would have access to a 32-men table base, then it could not win a single 5 day/move correspondence game against an ICCF grandmaster with engine as now competing in the ICCF World Championship Finals.
"if two agents play a game with the the same rules twice and don't get the same result"
++ 17 agents (ICCF (grand)masters with engines at average 5 days / move) played 104 games with the same rules and got 104 draws. Not a single decisive game.
@9233
"deductive proof isn't available"
++ Even that. The white advantage is 1 tempo. 3 tempi = 1 pawn. 1 pawn is needed to win.
The white advantage is not enough to win. Chess is a draw.
@9225
"The problem is doing the calculation to get to table base positions."
++ It does not matter how the move sequences from the initial position to the 7-men endgame table base draw or a prior 3-fold repetition were generated: with with 1 or more computers,
by praying to Caissa, by dark magic to summon the spirit of Capablanca...
If a sequence ends in a 7-men endgame table base draw or a prior 3-fold repetition,
then that justifies all black moves as fit to draw, but not all white moves as unfit to win.
"The table bases only offer a modest reduction in the size of the problem." ++ Yes. But modern computers plus human ICCF (grand)masters can now bridge the gap between the initial position and the 7-men endgame table base, exactly as GM Sveshnikov predicted.
"there have been decisive games between top players in recent years."
human (grand)master + engine > engine > human grandmaster
In the 2024 Toronto Candidates' tournament there were average 1.1 errors per game.
In the ICCF World Championship finals there have been decisive games: every year fewer and now none. They now have reached perfection. If some alien intelligence would have access to a 32-men table base, then it could not win a single 5 day/move correspondence game against an ICCF grandmaster with engine as now competing in the ICCF World Championship Finals.
"if two agents play a game with the the same rules twice and don't get the same result"
++ 17 agents (ICCF (grand)masters with engines at average 5 days / move) played 104 games with the same rules and got 104 draws. Not a single decisive game.
You consistently leap to conclusions without even being aware of the uncertainty that is obvious to someone with basic familiarity with such things.
How detectable do you think one blunder in 1000 games would be in your sample? After engaging your brain and reassessing do you think you are in a position to conclude from your sample that the error rate is zero rather than one per 1000 games?
The problem with interpreting inductive evidence as a proof (as you have done) is that 1000 can be replaced by 10^6 or 10^9 with the same point being made about the uncertainty of inductive reasoning.
If you choose to ignore such a simple point, you are clearly committed to being foolish.
@9233
"deductive proof isn't available"
++ Even that. The white advantage is 1 tempo. 3 tempi = 1 pawn. 1 pawn is needed to win.
The white advantage is not enough to win. Chess is a draw.
yeah this confirms you are stupid.
your logic is literally "1 tempo isnt enough to win because 1 tempo isnt enough to win"
you dont get to make up the value of the starting position LMFAO
@9195
"the game state assumes a perfect evaluation" ++ Yes.
"which you don't have" ++ I have the 7-men endgame table bases with their perfect evaluation.
White tried to win, black tried to draw. They reach a 7-men endgame table base draw or a prior 3-fold repetition. Black succeeded, white failed. All black's moves are justified in retrospect.
"What you do have is an imperfect evaluation which is very useful but not 100% reliable."
++ The perfect evaluation comes from reaching the 7-men endgame table base, or a prior 3-fold repetition.
"there have been losses for agents just like those participating now"
++ Yes, but each year fewer and fewer. Now 100% draws.
dude how are you so stupid to see how circular your logic is.
"chess is a draw with perfect play"
"this is perfect play because it ended with a draw"