I just want this thread to be closed and deleted.
Locked, fine. Deleted, no. Unlike the climate-change-huge-hoax-3 thread, this was not started as a trolling thread, there are plenty of good posts buried within the dreck from Tygxc and Optimissed.
I just want this thread to be closed and deleted.
Locked, fine. Deleted, no. Unlike the climate-change-huge-hoax-3 thread, this was not started as a trolling thread, there are plenty of good posts buried within the dreck from Tygxc and Optimissed.
the climate change ehox thread should be closed the only person even disagreeing is ign or whatever his name was
@10743
"10^17 positions will be reached in a mere 56 moves. Top engine games using a tablebase typically take longer than this to resolve." ++ But with plenty of forced moves and plenty of transpositions. ICCF WC Finals games end in average 38 moves.
Chess is full of transpositions. As is known d! grows faster than w^d.
Anyway 10^38 = 3^(2*40)
Some games go longer than 40 moves, but with forced moves and with many transpositions.
Finite numbers are all TINY compared to almost every other finite number (but not as tiny).
YOU made 2 mistakes. First, infinite is spelled wrong and there are only 1 infinity
This is very wrong, and I can give you a real life practical example.
The number off Odd numbers are infinite, the number of Even numbers are infinite, and they are the same sice of infinite. However the number of all the odd numbers + all the Even numbers are also the same infinite. This is called countable infinity. Ie start by making two lists. 1 (Odd) then two (Even) and you Will see that these two lists both are equaly Long.
However, make two new lists. On one list all the natural numbers (1, 2, 3) etc. On the other writa all the rational numbers. (1, 1.1, 1.01, 1.001) ypu Will now have to cont til infinity before you reach 2. So one list Will be longer than the other.
The practical eexcample comes from computer science. A computer can always (if given enought time) solve any problem that takes a countable infinity amount of time to solve. However, a computer can never (Even with infinite amount of time) solve a problem that takes a uncountable infinity amount of time to solve.
Hi ! @mrhjornevik
Are you suggesting that all infinities are equal?
That there can't be greater or lesser infinities?
If so its an interesting idea. Or maybe you're not suggesting that.
If its argued that one infinity has a two to one correspondence with another and therefore must be bigger one could just then argue that there's always 'more' of the second one to 'catch up' with the first one.
-----------------------------------------------
I've seen at least one person claim that infinity is only a term in mathematics and doesn't exist in the real world.
But I believe that's invalid.
You may as well say '5' doesn't exist in the real world ...
but it does since it describes and refers to much that is real ...
unlike 'reverse time travel' which does not refer to anything real but can still make money in movies!
Some would argue 'No! its a concept that exists in minds so it therefore exists!'
Existence versus non-existence not A or B though.
Objects exist. The 'real'.
Objects have descriptors. Referring to the 'real'.
But entities that don't exist also have descriptors. The 'ethereal'.
----------------------
Nobody will ever prove that the universe is finite. Nor that its infinite.
Since nobody can disprove physical infinity then a claim that it doesn't exist is invalid and so is a claim that its descriptor can only be abstract where such descriptor refers to reality.
@10736
"even if chess is solved, no one would have to worry cuz there is no way someone even Magnus could memorize that much moves"
++ That is right, but someone who had opportunity to study the solution and draw conclusions from it would have a huge advantage over somebody who has not.
Nobody can memorize the 7-men endgame table bases, but somebody who has studied these and has drawn conclusions from them has an idea of which endgames are drawn and which are won and knows in general terms how to win or how to draw, that is a huge advantage over somebody who has not.
@10736
"even if chess is solved, no one would have to worry cuz there is no way someone even Magnus could memorize that much moves"
++ That is right, but someone who had opportunity to study the solution and draw conclusions from it would have a huge advantage over somebody who has not.
Nobody can memorize the 7-men endgame table bases, but somebody who has studied these and has drawn conclusions from them has an idea of which endgames are drawn and which are won and knows in general terms how to win or how to draw, that is a huge advantage over somebody who has not.
Could also be used somehow to assist chess coaches.
The issue of 'solving chess' refers to investment of time.
Most of humanity does not invest in chess.
Maybe most people know how the pieces move and what checkmate and stalemate are but probably not much beyond that.
Have most people played at least one game of chess?
My guess. No. But its possible.
Point: I believe most people would see chess as a colossal waste of time.
No I am not one of those people.
-----------------------------------------------
But I think chess is a game of leisure for persons with a privilege or circumstances of leisure.
------------------------------------------
I believe that students in military academies often play or study chess.
'Marshalling one's forces'
'Art of war'
'Science of Conflict'.
-------------------------------
In movies I'm guessing its more the bad guy who is keen on chess.
And the rich powerful bad guy who has an expensive chess set and board in his luxurious living rooom.
Why? The author wants the audience to dislike the bad guy.
Sometimes - criminals in movies are the 'good guys' or protagonists and play chess.
McQueen in at least two of his movies for example.
Chess is seen as a 'fringe activity' by most of society I believe.
-------------------------------------------
Point:
Would there be an irony in chess players seeing a project to solve chess as a 'waste of time and money'?
The computer projects to 'solve chess' might well be adjuncts to developing better chess software and other software.
Its like Toyota or Honda building faster racing cars.
'Cutting edge of technology'.
Or like an adjunct of the space program (hey maybe adjunct is the wrong word and somebody'll get excited) with that same space program providing weather satellites and GPS/communications satellites for cellphones.
@10743
"10^17 positions will be reached in a mere 56 moves. Top engine games using a tablebase typically take longer than this to resolve." ++ But with plenty of forced moves and plenty of transpositions. ICCF WC Finals games end in average 38 moves.
When two humans agree a draw with about 16 pieces left. It is safe to say the draw is not proven, just believed.
Chess is full of transpositions. As is known d! grows faster than w^d.
It's full of positions too. over 10^44 of them with basic chess rules. There are already over 3 x 10^9 positions after white's fourth move.
Anyway 10^38 = 3^(2*40)
Some games go longer than 40 moves, but with forced moves and with many transpositions.
If waving your hands was a valid way to prove anything, you might have a chance of proving something. To suggest that there are LESS than 2 plausible moves as a geometric average (which fully takes into account "forced" moves, which are almost all not forced in the correct sense) is just clutching at straws.
But the main point is that your entire attitude is irrelevant to solving chess. You simply CAN'T ignore most of the legal opponent moves based on a zero ply evaluation. Any of those moves could be best.
Not me: I stated a hard fact. You argued with my quote containing only the 112 words YOU wrote.
As anyone else can verify.
You don't need four: one move per move includes the best move, with certainty. The problem is knowing which move that is!
Any heuristic method of choosing 4 moves based on zero ply analysis is going to miss the best move occasionally. It's worth remembering that this is not a matter of getting an evaluation for each move with some search depth: it is getting an evaluation with zero search depth. It's what happens at the leaf nodes of an analysis.
@10833
"To suggest that there are LESS than 2 plausible moves as a geometric average (which fully takes into account "forced" moves, which are almost all not forced in the correct sense) is just clutching at straws." ++ I say there are average 3 non transposing moves, because Chess has so many transpositions. 10^38 = 3^(2*40). Chess has 10^38 legal positions without underpromotions to pieces not previously captured and an average ICCF WC finals game lasts 38 moves. Of course there are positions with more than 3 legal moves: 20 in the initial position.
There are however many many transpositions.
"Any of those moves could be best." ++ No. 1 e4 e5 2 Ba6? cannot be best. 1 a4 could be as good as 1 e4, fit to draw, but it is unthinkable that 1 e4 draws and 1 a4 wins for white.
@10838
"4 ^80 =1.4615 e + 48" ++ There are only 10^44 legal positions, the vast majority of which are nonsense because of multiple underpromotions from both sides. There are only 10^38 legal positions without underpromotions to pieces not previously captured.
Inspection of a sample of 10,000 such positions shows none can result from optimal play by both sides, that leaves 10^38 / 10,000 = 10^34
"If there's a square root involved to cover for repeated positions"
++ No, the 10^44 or 10^38 covers for repeated positions.
The square root is for weakly solving instead of strongly solving: N white moves * 1 black reply instead of N white moves and N black replies.
That leaves Sqrt (10^34) = 10^17
@10843
"If there were a forced win for white,
it would probably be found after a first move such as 1. a4."
++ That defies all logic.
'Chess is a very logical game' - Capablanca
'From the outset two moves, 1.e4 or 1.d4, open up lines for the Queen and a Bishop.
Therefore, theoretically one of these two moves must be the best,
as no other first move accomplishes so much' - Capablanca
All this was corroborated by AlphaZero, with no other human input but the Laws of Chess.
See Figure 31. If we see the Laws of Chess as axioms, then we can see the resulting knowledge as theorems derived from the axioms by boolean logic operations.
This is also the expert opinion of the 17 ICCF World Championship finalists:
they played 1 d4, 1 e4, and 1 Nf3 and none 1 a4 for good reason so.
@10846
"assuming a 40 move game, taking 4 alternative candidate moves per ply as a lower estimate and therefore getting 4 ^80"
++ No, there are only 10^44 legal positions. 4^80 is wrong.
"need to be analysed for what you seem to call a weak solution"
++ No weakly solving Chess requires Sqrt (10^37*10/10,000) = 10^17 positions.
10^44 positions would be to strongly solve Chess to a 32-men table base.
It cannot logically be wrong, since we know that games can last for 80 half plies and we know that there are potentially 4 candidate moves per ply. Therefore 4^80 is reasonable and actually very low for an estimate of the number of different moves that become possible if the game potentially branches four ways every half move for 40 moves.
This is wrong because it's too low but I just wanted to see where you were with it. What are your qualifications in maths, tygxc? I don't really have any maths qualifications, except for an English "A level" in maths and I did pass the first year of a mechanical Engineering degree, which involves a great deal of maths, with very good marks at Liverpool University in 1970 but I dropped out and went abroad. I also passed the first year of a computing degree in 1992 but switched to philosophy. In each case, there were 80 plus people on the course and in each case my overall marks were well within the top 10%. That doesn't mean I'm good at maths although I was an arithmetic prodigy when I was about 9. It just means I know my way around the subject. So what are your qualifications?
We dont know if ur lying or not
@10849
"we know that games can last for 80 half plies and we know that there are potentially 4 candidate moves per ply"
++ There are only 10^44 legal chess positions:
worst possible white moves, worst possible black replies, stupid underpromotions gallore.
"Therefore 4^80 is reasonable" ++ No 4^80 > 10^44 so it is WRONG.
"This is wrong" ++ Yes.
"because it's too low" ++ Because it is too high, > 10^44.
You cannot end up with more positions than there are legal positions.
"What are your qualifications in maths" ++ More than any here.
@10853
"there is no forced win for either side in chess" ++ Correct.
"it's a draw given good play by both sides" ++ Correct.
"1. e4 is the worst place to look since we know that 1. e4 e5 is certainly drawish"
++ The 17 ICCF World Championship Finalists think differently.
They mostly opened 1 d4, but second was 1 e4, then 1 Nf3.
"It is therefore logical to look in precisely such positions as occur after 1. a4"
++ No. Read what Capablanca wrote. Look what AlphaZero found with no human input but the Laws of Chess. Look what the ICCF World Championship finalists play.
"non-existent forced win for white"
++ Non existent indeed, but 1 e4 and 1 d4 make it hardest for black to draw.
"not seeing the truth in what I'm telling you means that you don't think all that well"
++ I can say the same to you.
"consider something and to reject it after well-placed thought" ++ That is what I do.
And yet I've made more useful posts than anyone else here and you have yet to attempt to comment on the actual thread content, which is evidently beyond you to even attempt.
Odd how your mind works, innit. Every post you write like that further reduces your credibility, though you've been well within the bounds of incredibility for years.
You are probably wrong on that assertion 50 times over at this point in the thread. Which is why I don't often continue to post about actually solving chess in this trainwreck of a thread trying to pass for a discussion.
If you or Tygxc actually come up with some new wrinkle to be debunked, I might weigh in. Otherwise all your arguments are falling into the same 3 categories I laid out yesterday. You are nothing if not a creature of habit and predictable rigidity.
@10857
"10 ^17, which seems too low" ++ It may seem so, but after some well-placed thought you might understand. The 10^37 comes from An upper bound for the number of chess diagrams without promotion.
No promotions to pieces not previously taken is a bit too restrictive, that is where the *10 comes from to include positions with 3 or 4 queens.
The / 10,000 is pruning: it stems from inspecting a random sample of 10,000 positions without promotions to pieces not previously captured and finding none can result from optimal play by both sides.
The Sqrt is for weakly solving: only 1 black reply for N white moves, i.e. 1 * N = N positions instead of N black replies to N white moves, i.e. N * N = N² positions.
"the arguments of Svesnikov" ++ I admit that at first I was surprised by GM Sveshnikov's claim to weakly solve chess in 5 years if given good assistants and modern computers.
After considering it and after well-placed thought I found he was right.
Now the ICCF World Championship Finals confirm this: 106 draws out of 106 games.
I just want this thread to be closed and deleted.