well, I do not know if the position is a black win with perfect play, or simply draw, no one, engine or human, knows that, and no one will know that, engine or human, for another 100 years to come.Dann Corbit wrote:I saw a man who jumped out of a window when his building was on fire and he broke his leg.
Therefore, never jump out of the burning building.
Having 6 man tablebase files will solve more problems more quickly than not having them.
As for an engine making mistakes, yes. Stockfish does make mistakes. But it makes less mistakes than any other engine which is why it wins more games than any other engine.
It also makes less mistakes than any human on the planet, including Carlsen.
If a really strong engine failed to solve a problem, it probably means we did not give it enough time. Even bad pruning decisions will eventually be overcome by sufficient depth, because there is no algorithm that prunes down to zero (unless it is a pure loss or unless the program has a serious bug).
in order to be certain what a difficult position is, one must find all the best moves for some 100 plies, and that makes an awful lot; currently, top engines will find only about every 3rd best move.
non skipping even a single best move in a 100-ply search is completely impossible, in any position that is a bit more difficult.
all I know is that 90% of all lines are won for black or black has large advantage, 300+ cps, so my guess is black has the much better chances.
however, it is still possible, with perfect play there is a draw, I do not know and have no way of knowing it, neither SF and Komodo do, no matter the TC and tbs used.
so, for the time being, I will believe the position is won for black, and you might want to believe whatever you like.
