So a few more months of bickering, what's the consensus / argument about these days?
I haven't seen ponz in a while.
He hangs out on the thread he controls, which is why he split it off in the first place. But yes, I'm worried that things have finally caught up to him :/.
So a few more months of bickering, what's the consensus / argument about these days?
I haven't seen ponz in a while.
I really have no idea. Mainly fanciful ideas trying to bring the number of possible positions down to a number that is still probably beyond any practical possibility and/or shading the idea of solved into something that doesn't mean searching the whole game tree because of the prohibitive numbers.
Really the most reasonable argument I can come up for it being solved would be to believe in the existence of an omniscient higher being and at that point why not. It's a good way around the practical issues.
#1267
This is how checkers was solved.
http://library.msri.org/books/Book29/files/schaeffer.pdf
It was not necessary to search the whole game tree. See figure 1.
True or false? Chess will never be solved! why?
True. Because chess is the game of the gods/goddesses!
The normal eight year old can solve tic-tac-toe. Checkers is no more complicated, except the math is deeper.
Euclid was one of the smartest people ever to live. He got it so right, his books were used for teaching until modern history. To me this elevates him above other great minds of his period, most of whom made errors by faults in their thinking. (Although, having said that, I am not sure I can find fault with Eratosthenes, another hero of mine).
Euclid understood something that Optimissed appears to have forgotten - that proving intuitively "obvious" things is worthwhile.>>
As I pointed out, you hopeless great mutt, to demonstrate methods of constructing formal proofs. As usual, you're saying the same thing and pretending it's different. You and Rattigan should get together more often.
No, not only that. Also because "obvious" things are not always true, as I explained with a crucial example.
I used "obvious" to mean "obviously true". Look, you're a statistician. Please stop imagining that "statistician" means "brighter than Optimissed", which you most certainly are not! 
There may be some whom you impress but you don't impress me and many others when you try so hard to be on top.
Chess will be a solved is what knewbs say who will never become good at chess and always suck.
This is a good example of this thread being pointless because most of the time I can't figure out what definitions of "solved" people are working with. I have no idea of how to define it to have this post make any sense to me.
It's defined like this:
I mean, I know what it means, but even given that there is a pretty wide range of what it means. Strong, weak, etc. Someone suggested that searching the game tree wasn't necessary to solve a game and I'm trying to wrap my head around that one. This thread is the chess version of arguing whether golf is a sport.
Searching the entire game tree isn't the way to do it. The way to do it is to invent a new generation of algorithms that work with chess situations that are in the process of becoming unbalanced. Searching the entire game tree would only be necessary if chess is a forced win because, then, it would be necessary to find that win. If chess is a forced draw with best play, that can potentially be discovered by analysing unbalancing situations, starting off with positions that are much simpler than the starting game position.
I found your copy in a used book store. Your notes were flawed. I threw the book away.
Concise, encompassing, and directly to the point. Thank you Ziryab. 
In practice, however, chess cannot be solved because it’s beyond the capacity of any human mathematicians and beyond the capabilities of computers. There are 10120 potential variations of games in chess and around 10^43 different potential positions on the board. To fully solve chess, every single one of these must be compared against the other.
@#1277 You seem to have got the hang of this subject, anyway. I'd say that was exactly correct. Have a cigar! 
I found your copy in a used book store. Your notes were flawed. I threw the book away.
That's a coincidence. I think I found it when it turned up in a book bank in Wigan, Lancashire, a couple of years ago. I assumed it was the author's notes in the margin and sold it for £10,000. I didn't agree with the bit where it said that there's a sort of domino effect from the first move, where white starts setting up the centre and black plots to overthrow it.
At first, I thought it was Hegelian dialectic, anotated by Karl Marx, who, of course, lived in London. Marx, as we know, turned Hegel on his head so where, for Hegel, the ultimate synthesis of thesis (white) and antithesis (black) is the Nation State, for Marx it became the Commissariat and the ultimate triumph of the Workers (black) against the Nation State. As I read further though, it turned out to be Nimzo, describing the ultimate triumph of black as white's bastions are one by one isolated, weakened and overrun.
I decided he was wrong, though, because he wasn't treating the game as a composite entity and nor was he affording white the same chance for self-protection that he afforded black. Seemed to think white had to continue to attack, instead of transposing to a complex ending, slightly favourable to white.
#1277
"around 10^43 different potential positions on the board.
To fully solve chess, every single one of these must be compared against the other."
++ To strongly solve chess each of the 10^44 legal positions needs to be considered.
To weakly solve chess only 10^17 legal, sensible, reachable, and relevant positions need to be considered.
Losing Chess has been weakly solved considering only 10^9 positions.
Checkers has been weakly solved considering 10^7 * 10^7 = 10^14 positions.
Ultra-weakly solved means that the game-theoretic value of the initial position has been determined
Weakly solved means that for the initial position a strategy has been determined to achieve the game-theoretic value against any opposition, and
strongly solved is being used for a game for which such a strategy has been determined for all legal positions.
Chess may never be solved for one simple reason, we have to make this super super calculator AI. If many chess program AIs are already stronger than humans, why buy a new program with a more powerful AI when the existing ones will already be more or less unbeatable for humans player at the maximum level ?
Maybe they are superior but not yet unbeatable for a human which means that there will still be others but when this is done there will be no point in making new ones and in the end the game will not be solved because there will not be a super, super calculator having calculated everything.
#1281
"we have to make this super super calculator" ++ As Sveshnikov said: modern computers with good assistants can weakly solve chess in 5 years. That is without new hardware or software.
"why buy a new program with a more powerful AI when the existing ones will already be more or less unbeatable for humans player at the maximum level ?"
++ Why buy a faster car when the existing cars already are much faster than humans.
There will always be stronger engines that make even less errors.
So a few more months of bickering, what's the consensus / argument about these days?
I haven't seen ponz in a while.