Chess will never be solved, here's why

     Many Americans do indeed consider their own culture and nation as the finest in history and demand an unrealistic degree of respect from "lesser" peoples. The same seems to be true of nearly every other society. Perhaps we can learn from England, adopting the philosophy that "the wogs begin at Calais" and resigning ourselves to the necessity to "take up the white man's burden".

Holy damn this conversation is a year old

     Yes, and if keeps going until chess is truly "solved" it will outlive all of us.

If you were to stop talking about a strategy simply because others before you have talked about a strategy, it might help because it would make it simpler and easier to understand. The fact that others have obfuscated In the past is no good reason to obfuscate. Finding moves to counter an opponent's moves is not a strategy, except in the broadest sense in that it is the meta-strategy strategy of all turn based games. The alternative is to deliberately find bad moves or not to select good moves and to lose.

You really shouldn't come onto tygxc in the way that you do, by telling him that something has been pointed out to him already, because you don't adjust your thinking when you make mistakes that are pointed out to you. Continually calling the selection of moves a "strategy" isn't a major error. But it is obscurantist and is part of the reason why so many people are so confused.

A further criticism is that it is very much a matter of opinion whether algorithms can and should be used, to cut down the number of bad moves that are analysed. Without an algorithmic assessment of lines, there is no possible way to tell winning lines from drawing and losing lines. It all gets lost. I pointed that out several years ago. You don't listen, so you really shouldn't accuse others of not listening.

"Many Americans" must be exceptionally ignorant, then. You said it! But a failure to learn from the mistakes of others is a far worse error than making such mistakes in the first place, especially when we have the capability, in the light of present day knowledge, to see those mistakes made in the past for what they are. England has learned from those mistakes and moved on but the USA wants to repeat them all and more besides. You repeat the point about "taking up the white man's burden" but England has never been a particularly racist society. However, I was talking about free speech and hypocrisy!

The English learned from their civil war and from the Reformation. That's why its parliament's army and not the king's. Americans never experienced anything so bad as the Thirty Years War losing over a third of its population like the Germans did. Americans are naive. We haven't even assimilated the lessons of our own civil war. We're still fighting it.

That's a good point. As a nation, we learned that extremes tend to be bad. Cromwell was a reformer and many Parliamentarians were extremists. After 20 years of not being allowed to sing or dance people naturally became completely fed up of them and the monarchy was restored. But also we'd learned the lesson that a king by divine right is also an extreme viewpoint which can lead to despotism. Anything can lead to despotism if it's thought to exist by divine right. So the World, as in the English nation centred on London, turned back the clock somewhat, which led to a period of stability and happiness for a while.

Meanwhile, there were a lot of people who were intolerant of the restoration of the monarchy and all it entailed. They tended to be the people who went to America on the Mayflower, so they could continue to make the lives of others a misery. Of all the colonies founded, only one in America was based on tolerance. All the others were at least to some degree despotic and governed by bigots.

I have seen Him in the watchfires of a hundred circling camps; They have builded Him an altar in the eveneing dews and damps; I can read His righteous sentence by the dim and flaring lamps, His day is marching on.

A hundred circling camps! We didn't learn a thing. We are destined to repeat our own history.

"How much computation was done in the
proof? Roughly speaking, there are 10^7 positions
in the stored proof tree, each representing a
search of 10^7 positions (relatively small because
of the extensive disk operations). Hence, 10^14 is a
good ballpark estimate of the forward search
effort "- the cited article.  

Legal checkers positions:  5*10^20

I would say that's a lot less than 2/3, although feel free to correct me.

for some games, it's actually possible to 'solve' the game without considering any more than a couple positions.  Chomp is a great example (look it up on wiki).  the first player always has a winning position, but yet no winning moves are explained.

What an amazing view of history.

- Seat of military might...yes...seat of civilization and "center of the world"...no. 

- Discovered democracy and/or championed power for the people...no...missed that one by more than a millennium.  Cromwell was a local lesson, and not a new one, merely a reminder of a lesson already learned many times.  A pop-quiz question for the rest of the planet, usually involving the term roundheads.

- American colonists emigrating to form despotic nations rather than being persecuted for religious beliefs...no...nobody planned to have a revolution until good old George decided to turn the thumbscrews farther than he could handle.

"So the World, as in the English nation centred on London" is a viewpoint that fits oh-so-well with your worldview and your outlook on your own personal importance to the planet.  So not only are you head and shoulders above your fellow man by superior intellect and divine gift of paranormal ability...you trod upon the hills at the very seat of the ancient world like a Cumbrian Colossus pondering the plight of man...a demi-god, almost.  Maybe more like a Prometheus, since you bridle against authority, knowing better as you always do.  It's a wonder you have not woven Stonehenge into your narrative somehow...think of it...3,000BC and the mystery of the druids and yourself a conduit of power, guided by benevolent brilliance...but then, it's not in Cumbria.  Pity.

@7426

"worst moves according to an unreliable evaluator that you know to be entirely wrong sometimes (i.e. the move it thinks is best is a losing blunder)."
++ I have even quantified the error rate: 1 error in 10^5 positions for a 10^9 nodes/s engine calculating 17 s/move. Thus 1 case in 10^20 positions where the table base exact move is not among the top 4 moves of the 10^9 nodes/s engine running 17 s/move.
As only 10^17 positions are relevant to weakly solving Chess,
that means 0.001 error in the solution, i.e. not a single error at all.

The peer reviewed paper 'Games solved: Now and in the future' by Prof van den Herik states:
'it is often beneficial to incorporate knowledge-based methods in game-solving programs'

The peer-reviewed paper 'Acquisition of Chess Knowledge in AlphaZero' has 'knowledge' in its title. It leads to things we know, not things we guess, believe, or think.
It has only the Laws of Chess i.e. axioms as input and performs only boolean operations i.e. logic to acquire knowledge i.e. theorems.

This paper ranks the first moves in figures 5 and 31:
d4 > e4 > Nf3 > c4 > e3 > g3 > Nc3 > c3 > b3 > a3 >
h3 > d3 > a4 > f4 > b4 > Nh3 > h4 > Na3 > f3 > g4.

Once black has one path (there may exist several) to the 7-men endgame table base draw against the 4 best moves that oppose most to the draw,
then it is trivial to find such path to a draw or even a win for the 16 worst moves.
You could object it is not complete, but you cannot object it is not valid.

In general a chess engine cannot correctly evaluate a chess position,
only the 7-men endgame table base can.
However, some positions with > 7 men are clear wins or draws and need no further calculation.

1 e4 e5 2 Ba6? is an example of a clear loss for white. Stockfish says -8.1.
A full bishop up with all the rest being equal is more than a pawn up and is thus enough to win.
I have even demonstrated it is a forced checkmate in 72.

The final position of this game is an example of a clear draw with 12 men.
https://www.iccf.com/game?id=1164259 

wow!

When you have a nation where the dominant ethos is self-confidence and economic prosperity and yet such a high proportion of that population is ignorant, the result is the equivalent of a third world country with too much power. When you import masses of the poorest people from some of the most backward nations, leave them with their prejudices and don't try to educate them except in the basics, that is going to happen. I think that prejudices are so deep-rooted that it will take another 50 years to reach a level where people can participate usefully in things like elections but I also think there's evidence that the process is starting. When you have too much self-confidence, it's very destructive when things go wrong.

More like 200 years. The prejudices are a deep seated inherent problem. We also get wealthy diasporas. Like the Shah of Iran's emigres. The Armenians and on and on. Pretty much anybody wealthy enough to flee their situation bring us their money. The poor that come work very hard. They have a superhuman work ethic. It is the consequence of empire. The consequence of America's original sin.

@7436

"More like 200 years."
++ If done in a very dumb and stupid way weakly solving Chess takes 500 years.
If done in a smart and clever way using knowledge it takes 5 years.

Typically @tygxc can't tell the difference between solving chess and participating in an election.

I feel dumb just reading this entire forum lol

To observe a tiny hole in @tygxc's flaky thinking, there are quadrillions of positions where one side is a pawn up but it is not a win for that side to an (unavailable) oracle.

@tygxc believes that all you need to do is to find an unreliable evaluator to boost your confidence so that you can continue to believe they are all winning.

[Note that the same is true of other material differences],

@7440

"there are quadrillions of positions where one side is a pawn up but it is not a win for that side"
++ That is right. The rule is: 'Other things being equal, any material gain, no matter how small, means success'. 1 e4 e5 2 Ba6? is material gain and other things are equal.

"all you need to do is to find an unreliable evaluator"
++ No, on the contrary. In most positions the only way is to calculate until the 7-men endgame table base to get the exact evaluation draw / win / loss.
Some positions like 1 e4 e5 2 Ba6? are clear losses and need no further calculation.