True or false? Chess will never be solved! why?

Analyzing games of engines played against themselves at shallow depths will not get us any closer to determining whether chess is solvable or not.

Engines at large also have nothing to do with solving chess, because solving chess involves presenting a proof and engines are just practical tools for practical purposes and the way they play proves nothing ( it only reflects what humans think about the value of positions, since evaluation functions are programmed into the engines by humans - human guesses coupled with shallow imperfect searches mean nothing in terms of solving chess ).

1. In many cases it won't be a big enough advantage to win, but in some it will, because there are so many positions where a certain move looks like the best move but it has strong refutation that is very difficult to see (too difficult to see in fact until the game has gotten closer to that position). The discussion was about whether it will ALWAYS be a draw with best play, and my argument shows that's not the case given the assumptions.

2. No, I don't think it means the game is fundamentally flawed. You seem to believe that it SHOULD be a draw with best play, and I think it probably IS a draw with perfect play, but the best engines of today and the best human chess players are so far away from perfect play that whether it is ultimately a draw or not doesn't have much relevance for us.

Anyway, thanks for the discussion. Am off to bed for now.

I hate to bring this up again, but in the above position the highest rated engines want to play either Nc3 or Nd2 ( either of those moves blows their win ).

Why?

Because humans taught them that the knight is not good on the edge of the board...

If you systematically create all positions involving a legal collection of 6 chessmen ( or less ), then start from positions which are mate, stalemate or draw by insufficient material ( terminal positions with definite value ), look at all your legal moves which lead to those positions and give them value, then look at all opponent moves which enable you to make those moves, and give them value, and continue this as far you can reach positions that you have not seen before, you do this with every position that has definite value, then with this reverse tree search you can build a 6 men tablebase. If you have a 6 men tablebase then you are guaranteed to make the optimal move in any legal position involving 6 chessmen.

Engines do not work this way: they perform partial searches from given positions down to a certain depth and not down to terminal positions and instead of the value of terminal positions they rely on evaluation functions. Engines are designed in a way so that they can come up with a reasonable move in any position after a short think. This is very different from tablebases which think for ages, but then know everything about a certain limited set of positions.

I won't call an engine's second choice finding a winning move. If I have to rely on an engine I have no other rational option but to play its first choice. If I take this to the extreme and set the multipv level high enough so that every legal move gets analyzed then I can claim that the engine finds the best move in every position because it is among its choices, because all legal moves are among its choices.

You do. And everything in human life is personal.

The only thing that bothers me is whether there is an error in my reasoning.

My goal with writing to a forum is to learn something so I'm glad if errors in my reasoning are pointed out.

Here is a piece of advice: do not ask a question if you fear the answer.

True enough. I meant that you cannot tree-search 500 plies without storing the tablebases (which act as transposition tables). But since those are of a not-so-big size it means the tree-search could reverse to TB generation when  the number of pieces goes below 6, so my point is wrong.

Let me change that to: no combination blah blah can tree-search 500 plies from the starting position. That's enough to bust MM's claim.

False, Lasker solved it already. He just didn't give us the database. Do you remember ,"with perfect play from both sides, the result is a draw"? How did he new? LOL

Well, I have an elegant proof that chess is a win for Black. I just don't have the place to write it in the margin.

@irontiger, you copied my favourite joke.

So there you have it.

                    In post 1021, we have a win in 261 moves or plys.  Its the same with nuclear war and religion. If we kill, kill, kill, kill, kill, kill,kill, kill, kill, kill,kill, kill, kill, kill, kill, kill, kill, EVENTUALLY,    we get a win

@watcha: I have a question about your statement that you can find all legal positions with 6 chessmen when there are 6 chessmen on the board. I was thinking about positions with 6 chessmen that can only occur legally from a position with 7 chessmen. A good possibility for such a position is an undiscovered check using the en passant rule. Belongs a position with 6 chessmen, that can only be created from a 7 chessmen position to a 6 chessmen tablebase or to a 7 chessmen tablebase? Was it included in your statement?

I think it belongs to the 6 men tablebase. The origin of the positions in the 6 men tablebase does not matter as far as they are legal. It should contain all legal positions involving 6 chessmen no matter how those positions were reached.

I ran this position... it instantly found Kd3-kd4... though isn't finding the mate in 261 very quickly for obvious reasons.

But as previously said... how else do you propose it was found if not by an engine?

It was found by a tablebase, instantly (or not instantly, but in the time it takes to find indexed rows in a database, so effectively instantly for human purposes).  You need to read up on the difference between an engine and a tablebase.  There are no chess calculations/evaluations in real time using a tablebase, the calculations are already baked in.  That's the whole point.

I have tried to create a 6 chessmen position that can only come from a 7 chessmen position through en passant capture. I hope I got it right:

That is exactly what I meant indeed. :-)

Here is MIC from Galaxy XV5-R.4 (somewhere far in the milkyway): stop thinking about it, chess is already solved here 1000 years ago, silly humans!

              Math;  silentboy mis-spelt the last word.  new know   How did Lasker know, ( in a day with no computers) that at best, chess is a draw.            And Mic15, wearn't you pissed a few minutes ago about reading your wife's diary??  You seem to have recovered pretty well in a different thread.