Chess will never be solved, here's why
Sort:
Jul 16, 2024
0
#10863
How much money are Elroch and Dio and others paying Optimissed for Optimissed to look as stupid as possible in front of them to make Elroch and Dio and others look extra-good?
Optimissed is doing a 'good job' of making himself look stupid and insipid.
Maybe he should get a raise in pay. After all its his life work.
Why is any of it worth reading?
The responses are - because they're from people using their intelligence while Optimissed continues to think and troll with his rather congested nose.
A similiar reason to how the responses to tygxc are worth reading.
tygxc puts up his 'illogic softballs' and others hit them out of the park.
Constantly.
Jul 16, 2024
0
#10864
tygxc wrote:
@14022
"why doesn't the engine know its a draw?"
++ That is another reason why the human is necessary in addition to the computers.
Sveshnikov asked first for good assistants and only then for modern computers to weakly solve chess in 5 years.
That is tygxc's pathetic defense to strong engines not being able to recognize an obvious draw.
Which means he is conceding that point.
While trying to pretend that 'adding a human' would neutralize that issue.
It doesn't.
He'll just refuse to acknowledge the implications of the computer imperfection.
Crass and gross and intense imperfection of computer chess software and of hardware too.
The obvious failures of the engines completely refute illusions and elusions and delusions of 'optimal play'.
------------------------------------
Its apparently not bothered with much if at all - playing older engines against much more advanced present day engines. Not part of 'business'.
Do they even bother keeping all the old software?
How do you prove that engines of ten years from now given 20 days per move wouldn't beat today's engines then getting 5 days per move at that time?
--------------------------------------------------
Phony arbitration around '5 days per move' or 'strong engines' isn't the same as 'resisting Solipsism'.
Attempts to equate such false equivalencies should be avoided - in a truly intellectual discussion of these subjects.
but it appears to be tygxc's 'job' to make the false equvalencies so that they constantly get refuted. Year in year out. With Optimissed constantly and desperately wanting 'a piece of that action'.
tygxc should be O's boss and decide on his 'raise in pay'.
In other words 'the money' should by paid to tygxc by whoever. Not to O.
@14106
"it is a sample of 3" ++ All positions counted by Tromp look like that.
Most have 9 promotions to pieces not previously captured.
99.95% have 3 to 16 promotions to pieces not previously captured.
"I was for a few months the highest rated daily chess player on chess.com."
++ chess.com daily is a joke. One may think for a day and the opponent for a minute.
One may play one game and the opponent 100 games simultaneously.
"No engines, no tablebases" ++ Impossible to verify.
"very unusual things happen occasionally" ++ Chess is a very logical game.
"As many master games show."
++ Show one real master game with an underpromotion to a piece not previously captured.
Not a Nakamure trolling blitz game with 6 knights, not the Lasker trap.
"If that happens once in less than 10^7 games, that might happen 10^10 times in 10^17 games" ++ But a rare exception should not change the total count.
"in an enormous sample"
++ It is no random event, there is an iner logic resulting from the Laws of Chess.
A pawn needs time to reach the back rank, and in that time pieces get exchanged,
therefore promotion to a piece not previously captured is rare.
An average Tromp position has 9 promotions to pieces not previously captured.
The most reasonable piece to promote to is a queen.
Promotion to a knight is rare, but happens sometimes for the unique properties of the knight. Promotion to a rook or even bishop is even more rare,
and only happens to avoid a draw by stalemate.
Underpromotion to a piece not previously captured is the product of two rare events that does not occur in master games, not because the sample is tiny, but because of the inner mechanics of Chess.
That is why 10^38 is more reasonable than 10^44.
Even the 10^37 is too many: here is a legal position as counted in the 10^37, but which cannot be reached by reasonable play, let alone optimal play by both sides:
Jul 17, 2024
0
#10866
ah tygxc i see you are not only moving the goalposts of your own arguments while pretending you werent, I see you are also continuing with your red herring fallacy by ignoring the core failures of your delusion, that have been pointed out to you many times.
a) your total positional estimation method ignores the vast amount of positions that need to be made EVEN ASSUMING ZERO PROMOTIONS, because you falsely assume that unoptimal moves are not to be considered. you cant figure out which moves are optimal if you dont have unoptimal moves in calculations. plus, a weak solution demands you calculate for unoptimal moves from at least one side. you claim otherwise, but its a shame that every mathematician and game solution paper disagrees with you. of course, you dont read the game solution papers you cite, so you wouldnt know that.
b) you falsely rule out positions before their game tree is calculated (or invariant proved) as "solved" positions based on personal intuition and rules of thumb.
c) you then leave those "solved" positions out of your calculation total, which by definition invalidates your calculations as a solution.
d) draws made by parties not proven to be perfect mean nothing. I remember someone asked you "at which number of draws did it switch from being unproven to proven perfect" and you refused to answer. typical. probably because you know you couldnt.
theres more, but i decided to let you off easy.
Jul 17, 2024
0
#10867
tygxc wrote:
@14106
"it is a sample of 3" ++ All positions counted by Tromp look like that.
Most have 9 promotions to pieces not previously captured.
99.95% have 3 to 16 promotions to pieces not previously captured.
"I was for a few months the highest rated daily chess player on chess.com."
++ chess.com daily is a joke. One may think for a day and the opponent for a minute.
Top level daily chess is the highest quality of chess on chess.com.
According to your criterion all chess.com chess where people think for less than a minute per move is a joke, which accounts for all the chess you have played on chess.com. It's just that in your games both players are playing badly rather than in a weak daily chess game where only one player takes a minute a move.
The rating that people achieve at daily chess depends both on their ability and how much effort they put in. I had to put in a lot of effort, to be frank! While a long way short of perfection, I did achieve a good standard. I do the same in a more relaxed way in vote chess (with contributions from other players, some of whom are OTB masters). (You can look at our discussions if you like).
You can be sure that everyone with high ratings (top 50 say) at daily chess on chess.com spends more time on their moves than at standard time controls. And many of those players have FIDE FM, IM and GM titles, so you should not be surprised that the quality of play at the top is extremely good. (For top OTB players it would be so even if they spent one minute a move. I certainly had to put in a lot more effort, making up for a lack of master level OTB skills with explicit analysis in the editor).
One may play one game and the opponent 100 games simultaneously.
And their ratings reflect this. You won't find any players playing a lot of games at the top - even against weak players. Often I played just 2 games (due to the format of World League matches where I played board 1 for England, and the like), and some others do the same.
"No engines, no tablebases" ++ Impossible to verify.
Are you totally ignorant about chess.com's cheat detection, the best in the chess world?
There was a period long ago (2010) where I lost many games in a row due to inadequate cheat detection (players got banned too slowly in daily chess - later they were banned). More recently, it has improved greatly, and the reason for my more recent peak rating being so high was two opponents getting banned for cheating during my games. My rating became more representative since then due to losing points against some of the very strongest players (a mixture of draws and losses against players over 2500). It was nice to beat an FM twice before I decided to save several hours a week by giving it up!
"very unusual things happen occasionally" ++ Chess is a very logical game.
"As many master games show."
++ Show one real master game with an underpromotion to a piece not previously captured.
Not a Nakamure trolling blitz game with 6 knights, not the Lasker trap.
Show me a true Scotsman!
And perhaps you have forgotten that the 10^7 sample of master games - weak players, 1000 points or more from optimal! - compares to more than 10^20 times as many games with perfect play that are relevant to a weak solution. Tiny samples of low quality data only convince those with inadequate understanding.
It is surprising that you retain a complete lack of comprehension that you miss the extremes completely in a small sample.
"If that happens once in less than 10^7 games, that might happen 10^10 times in 10^17 games" ++ But a rare exception should not change the total count.
This is complete nonsense. You are arguing for making an entire class of legal moves illegal for the purpose of making solution simpler. You have no reason to believe one of those moves is not crucial to the solution of chess.
"in an enormous sample"
++ It is no random event, there is an iner logic resulting from the Laws of Chess.
A pawn needs time to reach the back rank, and in that time pieces get exchanged,
Some do, but this is vague generalisation from mediocrity again.
therefore promotion to a piece not previously captured is rare.
Yes, RARE. Rare is NOT NEVER.
Almost uinimaginably rare events occur billions of times in large sample spaces. You are using common intuition where it is inappropriate.
To quantify this, if for every promotion to a piece not previously captured, another occurs 1/1000 of the time (or whatever) - a model that deliberately underestimates substantially - you might find 2 occurring once in a million game, 3 once in a billion games (already outside our empirical dataset), and 10 occurring once in 10^30 games. You have no genuine basis for saying this does not happen.
An average Tromp position has 9 promotions to pieces not previously captured.
See above. The problem is that you don't know which positions don't matter. That is the nature of solving a problem - you know almost everything does not matter, but you need to solve the problem to find out WHAT doesn't matter.
The most reasonable piece to promote to is a queen.
Generalisation from mediocrity again. What you should say is that when there is a promotion, there is a high probability (but less than 0.99) that the best piece to promote to is a queen (promotions to knight comprise 1.8% of the master database). This formulation is more enlightening, as it reveals the very significant subset of underpromotions.
Promotion to a knight is rare,
1.8% in the chessbase dataset.
but happens sometimes for the unique properties of the knight. Promotion to a rook or even bishop is even more rare,
c. 0.1%, ignoring examples where it doesn't matter
and only happens to avoid a draw by stalemate.
Yes.
Underpromotion to a piece not previously captured is the product of two rare events that does not occur in master games,
It occurs even in the tiny sample of 10^7 master games.
not because the sample is tiny, but because of the inner mechanics of Chess.
Nonsense proclamation from a position of ignorance.
That is why 10^38 is more reasonable than 10^44.
Repetition of nonsense does not make it pass peer review.
Even the 10^37 is too many: here is a legal position as counted in the 10^37, but which cannot be reached by reasonable play, let alone optimal play by both sides:
Go through the entire 10^44 and eliminate all such positions by hand, then come back.
The serious point here is that it is WAY more difficult to eliminate such positions than finding a weak solution (the real definition, not a bodged one), and finding a weak solution is way beyond computational practicality.
There are some number os possible positions than the number of atoms in the universe, it'll take multiple thousand years for a couple thousand supercomputers.
hmm
Jul 17, 2024
0
#10870
idk13243 wrote:
There are some number os possible positions than the number of atoms in the universe,
No. It's much less.
it'll take multiple thousand years for a couple thousand supercomputers.
No, much more than this.
hmm
Jul 17, 2024
0
#10872
Optimissed wrote:
Whisky Tasting.
After doing some work today and Christine wanting me to make her a g&t when she got back from a long walk, I ..... You can buy a bottle of Glen Ord online for £220 from a reputed supplier or maybe get it for around £100 elsewhere. Added it to my likes list and will buy another for maybe up to £60 if one comes around.
you're trying to flex IRL stuff in an online chess forum
do you realize how pathetic you look?
also, you completely ignored how you are objectively the worst debator on this forum despite claims you have made
again, a debater is only good if they can convince people. can you give any evidence of anyone in this forum that has been convinced?
@14136
"all chess.com chess where people think for less than a minute per move is a joke" ++ Yes.
"I played just 2 games" ++ proving my point
"ignorant about chess.com's cheat detection" ++ ICCF allows it because they cannot patrol it.
"the 10^7 sample of master games"
++ Chess games are no random events, they follow logic, especially at higher levels.
"You are arguing for making an entire class of legal moves illegal" ++ No, I am arguing that the game stays the same if underpromotions to pieces not previously captured were illegal.
"reason to believe one of those moves is not crucial to the solution of chess"
++ Just logic.
By the time a pawn reaches the last rank there usually is one knight, bishop, rook traded.
Underpromoting is kind of a sacrifice. It is justified in rare circumstances.
The combination of underpromotion and promotion to a piece not previously captured does not happen in master games.
"rare events occur billions of times in large sample spaces" ++ A chess game is no random event. The stronger the players, the less randomness and the more logic.
"Promotion to a knight is rare, 1.8%" ++ OK
"Promotion to a rook or even bishop is even more rare" c. 0.1% ++ OK
++ So that makes 2% of promotions. Not each game has promotions.
"It occurs even in the tiny sample of 10^7 master games."
++ Please show one with an underpromotion to a piece not previously captured.
"Get it peer-reviewed." ++ 3*10^37 is peer reviewed.
"Go through the entire 10^44 and eliminate all such positions by hand, then come back."
++ Tromp did not check all 10^44 for legality, but 2*10^6 which you would call a tiny sample.
"it is WAY more difficult to eliminate such positions than finding a weak solution"
++ I am not talking about eliminating positions,
but at estimating the number of positions relevant to weakly solving Chess, i.e. 10^17
"finding a weak solution is way beyond computational practicality"
++ The 17 ICCF WC Finalists with their engines are doing it now.
The 112 draws out of 112 games are at least part of the weak solution of chess:
redundant and thus fail safe, but not yet complete.
The 112 draws represent 10^17 positions considered.
hmm
@14137
"There are some number os possible positions than the number of atoms in the universe" ++ No
"it'll take multiple thousand years for a couple thousand supercomputers"
++ No. The 17*2 severs 90*10^6 positions/s of the 17 ICCF WC finalists are weakly solving Chess in 2 years. Sveshnikov predicted 5 years.
As for strongly solving Chess that can be expected by 2100.
Jul 17, 2024
0
#10877
tygxc wrote:
@14136
"all chess.com chess where people think for less than a minute per move is a joke" ++ Yes.
"I played just 2 games" ++ proving my point
Not at all. It's about targeting quality not quantity. Ironically, you yourself rely on a tiny sample (106 games was it?) of relatively high (but very uncertain) quality for making false conclusions!
12 of the top 20 players have FIDE titles, despite not many playing the daily format. Only 2 of these are playing over 6 games (one GM and one IM) with most playing fewer. Got to be impressed by the two who are playing more!
Playing a full ICCF tournament (before it became 100% cheating) would have been almost a full-time job. Even playing 2 games used a few hours a week for me. My aim was solely to achieve the highest standard of play legally possible. My final rating was 2398 after 160 games, which would be #20 in the leaderboard as of today.
Note the phenomenon that longer time controls have a narrower Elo distribution because of the lower randomness in results compared to all live chess. This also extends to centaur chess on ICCF.
"ignorant about chess.com's cheat detection" ++ ICCF allows it because they cannot patrol it.
ICCF made a crucial, weak decision back in the dark ages when engines were weak and they had no knowledge or resources for cheat detection, and have never gone back on this. They stopped being correspondence chess and became centaur chess, a very different thing.
Chess.com had the advantage of being able to invest many millions of dollars in cheat detection, because it is relevant to commercial success. They recognized most people wanted to play chess, not jockey engines.
"the 10^7 sample of master games"
++ Chess games are no random events, they follow logic, especially at higher levels.
You don't understand the notion of a sample of variation.
"You are arguing for making an entire class of legal moves illegal" ++ No, I am arguing that the game stays the same if underpromotions to pieces not previously captured were illegal.
A claim which definitely can't be justified. You are appalling at reasoning logically.
"reason to believe one of those moves is not crucial to the solution of chess"
++ Just logic.
By the time a pawn reaches the last rank there usually is one knight, bishop, rook traded.
Underpromoting is kind of a sacrifice. It is justified in rare circumstances.
Yes, 1.8% of the time in the chessbase database.
The combination of underpromotion and promotion to a piece not previously captured does not happen in master games.
Yes, it does.
"rare events occur billions of times in large sample spaces" ++ A chess game is no random event. The stronger the players, the less randomness and the more logic.
"Promotion to a knight is rare, 1.8%" ++ OK
"Promotion to a rook or even bishop is even more rare" c. 0.1% ++ OK
++ So that makes 2% of promotions. Not each game has promotions.
"It occurs even in the tiny sample of 10^7 master games."
++ Please show one with an underpromotion to a piece not previously captured.
"Get it peer-reviewed." ++ 3*10^37 is peer reviewed.
Not the number, the falsehood that you can ignore other positions in solving chess. You are talking about solving a different game.
Unfortunately, underpromotion to rook or bishop is also necessary for proving chess is a draw. The reason is that by ignoring this, a strategy would blunder into any position where underpromotion was necessary for the the other side to win. You should be capable of understanding how disastrous this is.
I will point out one interesting fact that you have missed - you CAN ignore positions that have extra rooks and bishops for BOTH sides. The reason is that for a drawing strategy, only the opponent of the strategy is trying to win. The proponent can play suboptimally in the case where an underpromotion is necessary to avoid a draw, because a draw is his objective.
BUT, you can't exclude multiple additional rooks and bishops for one side. So don't ignore these positions. Remember that point about what happens in a sample of 10^30.
"Go through the entire 10^44 and eliminate all such positions by hand, then come back."
++ Tromp did not check all 10^44 for legality, but 2*10^6 which you would call a tiny sample.
Correct. Let me introduce you to the very basics of statistical estimation by pointing out that he was estimating an average (and doing it correctly in a valid way). This is unaffected by extremely rare extremes.
How many times do I need to remind you how different an AVERAGE and the EXISTENCE of one example are? 10^17 times?
"it is WAY more difficult to eliminate such positions than finding a weak solution"
++ I am not talking about eliminating positions,
but at estimating the number of positions relevant to weakly solving Chess, i.e. 10^17
Nah, you only need one. The first position.
Yeah, that's sarcasm.
[snipped boring confirmation that you don't know what a weak solution is]
Jul 17, 2024
0
#10878
tygxc wrote:
@14105
"how absurd you (a weak player) think three positions selected by Tromp are"
++ He did not select, he showed 3 random samples.
If you cannot see these 3 positions are absurd, then you are a very weak player.
I even proved none of those 3 can result from optimal play by both sides.
...
Liar. You didn't prove a single one of the whole sample.
And it's not relevant because you're planning to rely on SF play which is not optimal.
Jul 17, 2024
0
#10880
@MARattigan, do you accept my point that we can ignore all positions where both sides have extra rooks or bishops?
With the way positions for black and white have dual use in a solution (i.e. a legal move that reaches a position that has already occurred in the analysis tree for the other side need not be analysed further) we need only deal with positions where "side B" has <=2 bishops and <=2 rooks.
That should save a useful amount on the Tromp number. Or at least what would be a useful number if a solution was close to feasible.
I wonder how much it saves?
Wow I just went back to do some work and while I was working I was also thinking. You know I've seen quite a few people on here call you a sadist. I don't mean to your face but if they're talking about you. But you aren't that. You want to know what I think? I think you're a masochist. Seriously. I know that sado-masochism is supposed to go together but you're so bad at the stuff you try to do and yet you keep on with it apparently forever and claim you don't care what others think of you and how you're the site police, informal branch etc. Definitely crazy but in what way?
Think I may have understood you at last. A mini breakthrough but sleep time now. I promise not to hurt you! G'night, Christine's waiting 4 me.
I don't claim to be the site police. That's you. I don't claim to be sadistic or masochistic. That's you. I don't claim to be inherently smarter than the posters I engage with. That's you. I don't claim to know people's maladies from continents away. That's you.
You may be noticing a pattern here...
You have claimed "breakthroughs" many times. What you seem to mean is that you are contorting your narrative in a new way because the old way is not working. This seems to happen to you often.