I mean, you are a complete barm cake. A nut job. Just look at that post. Stream of consciousness but unhealthily paranoid too. You'd do better NOT to post your thoughts about others, if they're negative. Could get you into trouble.
'You' 'invited him over' to see that ?
Your namecalling?
Yea I'm following now so let's treat everyone the same and enjoy a discussion that is in place without anything else. I must admit its a high discussion so some good points here
Hi @StumpyBlitzer !
since I don't 'own' the thread - I can't 'invite you over' ... 
I believe there's only one person 'namecalling' in the forum.
And its not I.
But the term 'somebody' doesn't qualify as namecalling I believe.
It refers to no name - is not a 'pejorative'.
I've got it right too - that nobody here can outlaw English language pronouns.
Nobody here has that kind of power?
Well - staff and the ownership of the website can do what they want I guess.
But generally speaking - the staff are reasonable here.
English pronouns in the third person are allowed !
Yes. There had been no problems for a week. Everyone was getting on fine and then out of the blue, you started on your vendetta again. There was no call for it and it just seemed vindictive and I warned you I would report you if it continued. It did continue and I decided to contact the moderator personally, explaining that it had just started up apparently for no reason again, and that I'd told you what I thought of you. I think I wanted it to be seen that there are some circumstances where the chess.com rules are very disempowering. It's a cultural thing. In some cultures people play apparently by the rules and try to screw each other up. In mine, people tell each other exactly what they think & that's probably healthier. But here, my culture is dominated by one that's alien. That's ok because it's bound to happen a lot. You can call it cultural imperialism. I'm not trying to offend you and apologise if I am doing so but I would like you to understand it from perspectives other than your own.
Apparently a 'somebody' maintains he doesn't understand that its he who is doing the namecalling.
Another member here has informed us that that 'somebody' has been muted before by staff - and that the staff has had to explain to him several times over the years that others would not be actioned because they had not done anything to cause same.
You have never called my by name. You have consistently tried to use your "somebody" technique. You started it again yesterday for no reason. I find it extremely offensive and so did the others who left this thread because of your behaviour here. I was personally messaged and informed that you also do the same in clubs. I asked you to stop the behaviour I find offensive and you continued with it. After a few repetitions, I told you exactly what I thought of you. I meant it.
I now want to completely stop posting in this thread. You can have your customary last word, which you insist on in any case.
<<Another member here has informed us that that 'somebody' has been muted before by staff - and that the staff has had to explain to him several times over the years that others would not be actioned because they had not done anything to cause same.>>
I would wish to point out that this passage reflects your customary dishonesty and preference never to tell the truth if a lie would seem to suffice.
This was reported to staff several weeks ago:
Point is, I was talking about Ponz's thread and you didn't understand that. So there are often misunderstandings.
I knew exactly which thread you were talking about...thus the refutation of your characterization of it as "definitive" and the links to the threads Ponz's thread came from/drew from. That should be pretty clear.
So why pretend otherwise and attempt to obscure the discussion, other than the fact you are clearly a psychopath?
//////////////////////////////
After that was reported by staff (it was not directed at me) -
the offensive term (on page 23 of this forum) was edited by staff -
and apparently the member got a warning - as he was 'careful' for a while after that.
But he constantly 'tests' - over and over.
Should we expect that this behaviour will continue from him ?
He's blatantly calling another member 'clearly a psychopath'.
Is the point of his exercise to demonstrate that only he can do so?
Or to demonstrate that only he can report somebody?
Both? Far beyond that apparently.
My point: He can be reported.
But if members here prefer he be stopped - then inaction won't get that done.
When people go to a car race - do they go to see the race - or the accidents and the 'drama' ?
In internet chat rooms - many want the 'drama'. Its obvious.
Some drama is good. But what about lopsided 'power play' drama ?
Do we need that too?
Everybody in the forum can decide that for themselves.
I mean, you are a complete barm cake. A nut job. Just look at that post. Stream of consciousness but unhealthily paranoid too. You'd do better NOT to post your thoughts about others, if they're negative. Could get you into trouble.
Are you familiar with the old maxim "Practice what you preach"?
A good post by @mpaetz and with a good purpose -
but it had the effect of getting 'the other guy' 'started' again.
Unfortunately.
But now - we could get back to the originally intended forum subjects.
Earlier - I had posted the solution to a chess problem I had posted.
But that got 'buried'. So maybe I'll delete that solution and post it again at a better opportunity.
There's a great deal of scope within the forum topic.
So far - the most popular sub-topic has been 'strong versus weak' solving or what that could mean.
You are the only one here who is engaging in "power play" and, instead of prefering to discuss the topic, obviously attacking others *out of preference*. I'm far from the only one you do it to. I've recently discovered it happens to absolutely everyone who objects to your behaviour. If that isn't "power play", what is? You've just used another whole page to make yourself look really clever, on the ball and not the slightest bit foolish. Well done, as usual, for setting a good example.
You posted a puzzle a couple of days ago and I got the answer in a few seconds, so you promptly told me I'd either seen it before or that I put it in an engine. Then you immediately deleted the evidence. I've just realised that was when you started your campaign against the disreputable someone you disapprove of. Otherwise known as a tirade and campaign of attempted humiliation. If it was allowed, I'd call you some more very choice names in public and I think by far the majority of people would strongly approve of it!
Regarding the subject matter of the thread, I've patiently explained, a number of times, that finding a so-called weak solution is impossible without finding a semi-strong solution first. The only trouble with that is that the so-called authorities haven't defined a semi-strong solution, probably becuse they are not quite as bright as some members apparently imagine them to be. Since no-one is going to understand the reasoning until a suitable time has elapsed, when it can be claimed that it was known all along, I'm not accomplishing anything by trying to educate the ineducable.
from @tygxc in post #1523
"++ Can we please agree that the 10^44 is too high? Tromp exactly counted the number of possible chess positions. Then he randomly sampled 10000 of these. He found 543 of these legal. Thus he arrived at 10^44 legal positions. However all of his 543 randomly sampled positions found legal contain multiple excess underpromotions. Such positions can never occur in a reasonable game with reasonable moves and hence not in an ideal game with optimal moves."
Several remarkable things there:
1) First - that 'Tromp exactly counted' - but the figure is missing from the post.
2) Then Tromp 'randomly sampled' 10000 of '10∧44' positions ...
but out of those 10,000 positions ... 9,457 were 'illegal' ??
That's an illegal rate of over 95% !!!
Too bad the computer couldn't simply report the number of illegal and legal positions. Is that too - 'not feasible' ?
'arrived at' should be qualified ...
3) Every Single One of his 'legal' 543 positions had 'muliple underpromotions' ???
That sounds very very unlikely.
4) Idea: Tromp should have launched pre-projects to do what's easier first -
count up (with computer assistance) all positions without any 'have to be promotions/underpromotions' at all - along with the computer labelling each of those counted other positions as legal or illegal and whether they're checkmate or not.
So as to get three totals.
5) Note that if a position contains 'forcing' options towards stalemate or hopeless material draw ... that doesn't 'solve' the position - because the player with such options doesn't necessarily have to play for that ..
6) "Such positions can never occur in a reasonable game with reasonable moves and hence not in an ideal game with optimal moves."
'can never occur' there doesn't follow - therefore is not a valid premise - therefore the second part is invalid also ...
Key point could be:
What actual number of 'positions' would be truly solvable by today's supercomputers in a 'practical' period of time ?
There is no such 'exact number' - it would be a 'range' of numbers.
But its very difficult to define.
Because just one position - is enough.
For insolvability with today's computers.
The opening position. Also known as the Grand Position.
Isn't solved. It only takes that One
#1555
"1) First - that 'Tromp exactly counted' - but the figure is missing from the post."
++ The figure is 8726713169886222032347729969256422370854716254 positions
https://github.com/tromp/ChessPositionRanking
"2) Then Tromp 'randomly sampled' 10000 of '10∧44' positions ...
but out of those 10,000 positions ... 9,457 were 'illegal' ??
That's an illegal rate of over 95% !!!"
++ 94.62% are illegal.
"Too bad the computer couldn't simply report the number of illegal and legal positions. Is that too - 'not feasible' ?"
++ First Tromp generated positions and he allowed adjacent kings. Then he sampled. Then he determined legal or not of the samples first with a program, then manually. The Gourion paper avoids adjacent kings in the first place.
"'arrived at' should be qualified ..."
++ He took his number 8726713169886222032347729969256422370854716254 and then multiplied it by the 0.0538 he found by sampling to estimate the number of legal positions
"3) Every Single One of his 'legal' 543 positions had 'muliple underpromotions' ???
That sounds very very unlikely. "
++ It may sound unlikely, indeed all his 538 sampled positions he found legal contain multiple excess underpromotions. Most of his positions have 9 excess promotions. Excess promotions to R, B, N are just as likely as to Q in his count. Look at the 3 random samples in his paper. All of his 538 positions look like that.
"4) Idea: Tromp should have launched pre-projects to do what's easier first -
count up (with computer assistance) all positions without any 'have to be promotions/underpromotions' at all - along with the computer labelling each of those counted other positions as legal or illegal and whether they're checkmate or not. "
++ Do not tell Tromp what he should have done. I like the Gourion approach better: avoiding adjacent kings and avoiding excess promotions.
https://arxiv.org/pdf/2112.09386.pdf
"5) Note that if a position contains 'forcing' options towards stalemate or hopeless material draw ... that doesn't 'solve' the position - because the player with such options doesn't necessarily have to play for that .. "
++I claim that 1 e4 e5 2 Ba6? loses for white, without having to calculate until checkmate.
"6) "can never occur' there doesn't follow"
++ Yes, it does follow. Nobody right in his mind promotes a pawn to a rook, a bishop, or a knight without a compelling reason: to avoid stalemate, to promote to a knight with check, to promote to a knight to deliver smothered checkmate. Excess promotions happen, but rarely and usually to a queen i.e. no underpromotion. Underpromotions happen, but rarely and usually to a piece already captured i.e. no excess promotion. I presented a proof game that proves that the first position sampled by Tromp is legal indeed. That proof game has an accuracy near 0% and an average centipawn loss near 500. A reasonable game would have an accuracy around 50% and an average centipawn loss around 50. An ideal game with optimal moves as is needed to solve chess would have an accuracy close to 100% and an avarage centipawn loss near 0, where the difference from 100% and from 0 is not the fault of the game, but the fault of the engine evaluation judging the game.
"What actual number of 'positions' would be truly solvable by today's supercomputers in a 'practical' period of time ?"
++ The more positions, the more time. The more cloud engines, the less time.
I arrived at Sveshnikov's 5 years for weakly solving chess. That is open to debate.
Whoever should never have allowed adjacent Kings.
That should be put in at the beginning.
If you allow it - and then take a random sample and find 95% of positions illegal that's ridiculous.
"Nobody right in his mind promotes a pawn to a rook, a bishop, or a knight without a compelling reason:"
Correct - but there can be compelling reasons.
To understand that better - you have to have looked at some tactics problems.
Even in the openings there's sometimes underpromotion ...
like in the Albin countergambit there's an underpromotion line.
"Such positions can never occur in a reasonable game with reasonable moves"
is invalid.
And there are valid multiple underpromotion situations.
But more to the point - it would be better to do the project in simplified stages.
I would be curious to know how many of the 'tablebases' are really done.
The 7-piece skips castling and en passant ... however unlikely those might be - they're still possible with very few pieces on board.
They should only skip en passant when there's no pawns or only one pawn or its not possible anyway -
and castling should only be skipped if both Kings are off their original squares - or there's no rooks or they're off their squares ...
these things should not be hard to take care of with powerful computers available.
Maybe the 6-piece isn't properly done either.
Just simply allowing illegal positions with an expression like 13∧64 (the first upper bound) - isn't very impressive.
#1557
"Whoever should never have allowed adjacent Kings."
++ I would have done it otherwise, but that is how Tromp did it. His Haskell program is public, so you can modify it yourself to exclude adjacent kings from the beginning. Gourion excluded adjacent kings and same color bishop pairs from the beginning and I like that approach better.
"If you allow it - and then take a random sample and find 95% of positions illegal that's ridiculous."
++ That is how Tromp did it. You can modify his Haskell program if you want. Part of his automated legality check is to remove sampled positions with both kings in check and that also removes sampled positions with adjacent kings.
"Correct - but there can be compelling reasons."
++ That is right there can be compelling reasons for 1 underpromotion, even 1 excess underpromotion, but there are no compelling reasons for multiple excess underpromotions. Multiple excess underpromotions occur in all of Tromp's sampled legal positions. The vast majority of the positions counted by Tromp contain multiple excess underpromotions.
That is why I say the whole time that Tromp's number 10^44 is too high for the purpose of assessing the feasibility of solving chess.
"like in the Albin countergambit there's an underpromotion line."
++ I know, the Lasker trap, but that is an exception. It is a single excess underpromotion, not multiple excess underpromotions. The Lasker trap cannot happen if white plays optimal moves.
"And there are valid multiple underpromotion situations."
++ No , there are no valid multiple excess underpromotions. Show me one real grandmaster or engine game where multiple excess underpromotions occur and I eat my words.
"But more to the point - it would be better to do the project in simplified stages."
++ Exactly, that is why I propose to start with a single ECO code like C67. That would also settle the square root conjecture.
"I would be curious to know how many of the 'tablebases' are really done."
++ 7 men is completely done.
Work on 8 men is in progress. Many pawnless 8-men are already done.
"The 7-piece skips castling and en passant ... however unlikely those might be - they're still possible with very few pieces on board."
++ It makes no difference. En passant is possible in the solution, but there is no separate en passant flag in the position to save storage space.
Castling makes no sense. By the time a 7-men position is reached castling rights are lost in practice. It saves storage space not having to include 2*2 = 4 castling rights flags.
"these things should not be hard to take care of with powerful computers available."
++ That is right, castling and en passant can be included in table bases without any effort, but at the cost of more storage. They went through all the effort to compress the 7-men table base so it fits on a desktop computer by eliminating needless complications like an en passant flag and castling rights flags.
"Maybe the 6-piece isn't properly done either."
++ 6 and 7 pieces are done properly. They do however include illegal positions and insensible positions, so they could be compressed more.
So many things...
@tygxc, you seem quite intelligent, and that's why I still don't understand whether you don't understand some crucial points, or you pretend to not understand. I would like to understand clearly: do you think you have proven that chess would be solved in 5 years, given adequate and reasonable resources, or you believe it, but you think it's a well educated opinion/theory?
#1559
I firmly believe chess can be weakly solved in 5 years with present cloud engines and think it is a well educated opinion/theory. I had to make 3 assumptions that I consider plausible.
1) The most crucial one is the square root conjecture. It is borrowed from the peer reviewed paper on how checkers was solved and the square root is more than the 4th root that showed in the peer reviewed proof how losing chess was solved. Right now I see no way to arrive at a more plausible value than the square root. We will probably only know after it is done for 1 ECO code e.g. C67.
2) The number of sensible positions 10^36 also follows from some estimates. The Tromp number 10^44 is too high. The Gourion number 10^37 is both too low (no excess promotions) and too high (the sampled positions without excess promotions are found insensible). I tend to believe Tromp's conjecture that only 1 position in 10^6 can occur in a reasonable game.
3) The factor 10 by leaving out unnecessary ECO codes seems all right too. I gave the 19 ECO codes that can prove a draw against 1 e4 instead of the 200 possible B00 to C99.
#1559
I firmly believe chess can be solved in 5 years with present cloud engines and think it is a well educated opinion/theory. I had to make 3 assumptions that I consider plausible.
1) The most crucial one is the square root conjecture. It is borrowed from the peer reviewed paper on how checkers was solved and the square root is more than the 4th root that showed in the peer reviewed proof how losing chess was solved. Right now I see no way to arrive at a more plausible value than the square root. We will probably only know after it is done for 1 ECO code e.g. C67.
2) The number of sensible positions 10^36 also follows from some estimates. The Tromp number 10^44 is too high. The Gourion number 10^37 is both too low (no excess promotions) and too high (the sampled positions without excess promotions are found insensible). I tend to believe Tromp's conjecture that only 1 position in 10^6 can occur in a reasonable game.
3) The factor 10 by leaving out unnecessary ECO codes seems all right too. I gave the 19 ECO codes that can prove a draw against 1 e4 instead of the 200 possible B00 to C99.
Good to me as answer, thank you.
from @tygxc today
"His Haskell program is public, so you can modify it yourself to exclude adjacent kings from the beginning."
Now that looks like a good constructive post there !
"Correct - but there can be compelling reasons."
"++ That is right there can be compelling reasons for 1 underpromotion, even 1 excess underpromotion,"
the word 'excess' is being misused there ... it is far too subjective for the mathematical nature of the project and shouldn't be used.
People here are forgetting something. Solving chess is like solving soccer, or dancing.
The probability space is far too vast to become finitely graspable by any of our computers for the next millenia.
There's chaos theory, quantum mechanics, etc. required for these types of wicked problem-solving spaces. Human intuition is our closest counterpart, developed through iterative refinements in findnig the sharpest moves.
Therefore it remains more of an art of acuity.
Chess is not a problem
People here are forgetting something. Solving chess is like solving soccer, or dancing.
The probability space is far too vast to become finitely graspable by any of our computers for the next millenia.
There's chaos theory, quantum mechanics, etc. required for these types of wicked problem-solving spaces. Human intuition is our closest counterpart, developed through iterative refinements in findnig the sharpest moves.
Therefore it remains more of an art of acuity.
Chess is not a problem
I agree there's a lot of 'art' in chess.
Its often like a kind of 'portal' between art and science.
Yea I'm following now so let's treat everyone the same and enjoy a discussion that is in place without anything else. I must admit its a high discussion so some good points here