@5547
"Because I said so"
++ Indeed, I said so before. I repeat:
Chess will never be solved, here's why
Sort:
Sep 29, 2022
0
#4263
tygxc wrote:
@5546
"A reference to a user with a 400 rating is not so useful."
++ There is no threshold of rating or college degrees.
I didn't say there was, I just pointed out the reference is not so useful (unless you're hoping the user you're responding to won't notice).
"positions with 7 men or less only because the facts can be checked in those positions"
++ Yes, some positions of 7 men may be useful to check facts, e.g. endgames KRPP vs. KRP.
"they're valid for 32 men they should be also valid for 5 or 7 or 26"
++ Yes, from 32 to 8. You can use 7 too for verification.
So why don't you, as requested, do just that?
"earlier KNNKP" ++ Not relevant. A draw is claimed in 7 men, so 5 men is never reached.
Where does it say that in the FIDE handbook?
@5544
"What do you need the supercomputers for?"
++ They do the bulk of the work: 3 engines, 5 years, 24/7.
The 3 good assistants work 5 years, 40 h/week to launch the calculations and to occasionally terminate them in case of a clear draw or loss so as to save engine time.
@5551
"Where does it say that in the FIDE handbook?"
++ All positions with 7 men or less are already strongly solved (apart from castling rights, which in practice are lost in a 7-men position.)
The over 1000 perfect games with optimal play from both sides we have from the ICCF world championship finals never go to 7 men: they claim an endgame teblebase draw.
Weakly solving Chess does the same.
Weakly solving Checkers did the same.
Sep 29, 2022
0
#4266
tygxc wrote:
@5544
"What do you need the supercomputers for?"
++ They do the bulk of the work: 3 engines, 5 years, 24/7.
The 3 good assistants work 5 years, 40 h/week to launch the calculations and to occasionally terminate them in case of a clear draw or loss so as to save engine time.
The whole of the work, using your definition of "weakly solve" (or "strongly solve") can be comfortably done by one human in less than an hour. Again, why the supercomputers?
Incidentally no show yet for your calculation of the theoretical result and error rates in my games here. Are you still working on it?
Sep 29, 2022
0
#4267
Good news: both of the forum bugs recently introduced (the missing delete button and the broken link button functionality) have been fixed in the last 12 minutes!
@5543
"Induction from a tiny sample of imperfect examples"
++ Induction from over 1000 perfect games with optimal play from both sides: ICCF WC Finals draws. Also from millions of human and engine games. Also from common sense.
It is possible to artificially construct a game that reaches a 7-men position with castling right,
but it is sure that it is not with optimal play from both sides.
Sep 29, 2022
0
#4269
tygxc wrote:
@5543
"Induction from a tiny sample of imperfect examples"
++ Induction from over 1000 perfect games with optimal play from both sides: ICCF WC Finals draws. Also from millions of human and engine games. Also from common sense.
It is possible to artificially construct a game that reaches a 7-men position with castling right,
but it is sure that it is not with optimal play from both sides.
Induction from a tiny sample of of imperfect examples that don't necessarily use the same rules and are in many or most cases terminated by an inductive imperfect evaluation even.
Incidentally no show yet for your calculation of the theoretical result and error rates in my games here. Are you still working on it?
If you do that we can all concentrate on the topic instead of discussing your proposals.
Sep 29, 2022
0
#4270
tygxc wrote:
@5543
"Induction from a tiny sample of imperfect examples"
++ Induction from a pathetically tiny sample of over 1000 believed to be by one imprecise human, but in no sense proven to be perfect games with believed to be by one imprecise human, but in no sense proven to be optimal play from both sides: ICCF WC Finals draws.
Also from millions of human and engine games most of which have even weaker claim to perfection (although, to be pedantic, there is no weaker claim than no claim at all). Also from common sense based on human induction from a tiny sample of games all of which are not provably perfect.
tygxc wrote:
@5547
"Because I said so"
++ Indeed, I said so before. I repeat:
This is not helpful. I threw in the "Because I said so" as an analogy, and THAT is what you choose to respond to? How about the part where you say "I know, because I know, because every serious chess player knows, because Grand Masters know." That's how you know? My point, which you clearly missed, was that it's not enough to say "I know" without some supporting evidence. Maybe you feel you've already provided that, but if so, you could have cited it as a reference. But your statement was, prima facie, weak.
Sep 29, 2022
0
#4272
tygxc wrote:
"earlier KNNKP" ++ Not relevant. A draw is claimed in 7 men, so 5 men is never reached. [line inserted to set context]
"Where does it say that in the FIDE handbook?"
++ All positions with 7 men or less are already strongly solved (apart from castling rights, which in practice are lost in a 7-men position.)
No. All positions with 7 men or less are not already strongly solved apart from castling rights; you just have a learning problem.
The over 1000 perfect games with optimal play from both sides we have from the ICCF world championship finals never go to 7 men: they claim an endgame teblebase draw.
Weakly solving Chess does the same.
Weakly solving Checkers did the same.
You didn't answer my question.
Incidentally still no show yet for your calculation of the theoretical result and error rates in my games here. Are you still working on it?
Once you've done that we can stop discussing your proposal.
It doesn't work.
Does the fact that it's not appeared mean you've already done it and reached the same conclusion, but you still post because you like to troll?
I didn't assume "we cannot know that". I just needed more from tygxc than just "I know because everyone knows". He may very well know, and everyone may very well know, but I was looking for some evidence from him, since he was the one making the claim.
Sep 29, 2022
0
#4275
tygxc wrote:
@5544
"What do you need the supercomputers for?"
++ They do the bulk of the work: 3 engines, 5 years, 24/7.
The 3 good assistants work 5 years, 40 h/week to launch the calculations and to occasionally terminate them in case of a clear draw or loss so as to save engine time.
You obviously missed the point.
I can post you a solution if we accept the definition of solution you are quoting. (I nearly said, "the definition of solution you are using", but of course that would be wrong.)
So again; why do you need the supercomputers?
Incidentally still no show yet for your calculation of the theoretical result and error rates in my games here. Are you still working on it?
Once you've done that we can stop discussing your proposal.
It doesn't work.
Does the fact that it's not appeared mean you've already done it and reached the same conclusion, but you still post because you like to troll?
Sep 29, 2022
0
#4276
tygxc wrote:
...
++ Yes, some positions of 7 men may be useful to check facts, e.g. endgames KRPP vs. KRP.
...
Coincidentally, I have a batch of KRPP vs. KRP games running at the moment.
They take some time to complete because I give SF15 up to 34 minutes per move, but I'll post them as soon as possible.
Will you then check if your "calculations" work on that set?
Then we can stop discussing your proposals.
@5561
'Other things being equal, any material gain, no matter how small, means success,' - Capablanca
++ 'Any material gain' is easy to see: 1 e4 e5 2 Ba6? loses a bishop.
'Other things being equal' is more difficult to judge, that is why the good assistants should be ICCF (grand)masters. In this case it is easy: white has no compensation of any kind.
Thus white is lost. The provided analysis only confirms that: checkmate in 82.
@5567
"I have a batch of KRPP vs. KRP games running at the moment"
++ Interesting. Let us see.
Sep 29, 2022
0
#4279
tygxc wrote:
@5561
'Other things being equal, any material gain, no matter how small, means success,' - Capablanca
++ 'Any material gain' is easy to see: 1 e4 e5 2 Ba6? loses a bishop.
'Other things being equal' is more difficult to judge, that is why the good assistants should be ICCF (grand)masters. In this case it is easy: white has no compensation of any kind.
Thus white is lost. The provided analysis only confirms that: checkmate in 82. Don't you mean 2 (for White)?
Any of the outright beginners in the chess craze at my local that I mentioned earlier could have told you that.
Their judgement has about the same chance of correctness.
@5540
I do not work on positions of 7 men or less. Those have been strongly solved by the 7-men endgame table base. Do not tell me about castling rights. In practice when a 7-men endgame position is reached, castling rights are lost.
I've just realised this was meant to be a response to my post.
He did refer to the number of the post but erroneously added an @ before it.
Unfortunately, the link function is one of the things that has just been broken by chess.com in their latest random vandalism on the code (the post delete button also vanished). Now the link is (uselessly) to the page, rather than to the post.