lichess.org/analysis/standard/8/3P3k/n2K3p/2p3n1/1b4N1/2p1p1P1/8/3B4_w_-_-
Perhaps its just my settings
Perhaps its just my settings
what is the answer?
Unable means not given enought time for stockfish. I saw same post in talkchess that Stockfish 10 can solve in 1 hour.
A basic rating for bug free chess program is 2000 rating. Stockfish has improved algorithm to prioritize search in checkmate, piece capture etc which is probably 95% important moves but least priority in Zugzwang and giving away pieces.
That being said, if positions include the latter 5%, older Stockfish programs or basic chess program can probably solve in 5 -10 minutes.
A basic rating for bug free chess program is 2000 rating. Stockfish has improved algorithm to prioritize search in checkmate, piece capture etc which is probably 95% important moves but least priority in Zugzwang and giving away pieces.
That being said, if positions include the latter 5%, older Stockfish programs or basic chess program can probably solve in 5 -10 minutes.
Also the supposed solution to this is not 100% forcing so Stockfish dismissed it
@MightyMouse27 1.Nf6+ Kg7 2.Nh5+ Kg6. From there Stockfisch understands it.
@Morozov ahh thats a pretty good puzzle
Probably pruning going wrong.
You can even go one move deeper. After 2. Nh5+, Stockfish shows at a depth of 40 that the king has to go to g6 (evaluation: -7.3) or it's forced mate. It doesn't seem to include the answer 3. Bc2+ in its calculations at all. Only when that move is actually played, it does start to calculate it deeper and of course sees very quickly that the move wins.
I don't know how sophisticated SF's pruning mechanisms are. Or rather, how much they are influenced by silly hardcoded human rules. Bc2 certainly violates some general principles at face value (hangs material, removes the last possible obstacle for the e3-pawn) and doesn't immediately lead to mate, coupled with a relatively broad search tree. After Bc2+ however, SF only needs a depth of 31 to determine that White is better, so it certainly was not impossible to see the same thing at depth 40 before.
You can even go one move deeper. After 2. Nh5+, Stockfish shows at a depth of 40 that the king has to go to g6 (evaluation: -7.3) or it's forced mate. It doesn't seem to include the answer 3. Bc2+ in its calculations at all. Only when that move is actually played, it does start to calculate it deeper and of course sees very quickly that the move wins.
I don't know how sophisticated SF's pruning mechanisms are. Or rather, how much they are influenced by silly hardcoded human rules. Bc2 certainly violates some general principles at face value (hangs material, removes the last possible obstacle for the e3-pawn) and doesn't immediately lead to mate, coupled with a relatively broad search tree. After Bc2+ however, SF only needs a depth of 31 to determine that White is better, so it certainly was not impossible to see the same thing at depth 40 before.
Tal‘s pet study. I have seen it in every forum at least 10 times...
I am sure I seen this puzzle one time, does anyone know how is it called?
Knights of Round Table is the name of the title in chessable's endgame course.
This topic has been archived and can no longer be replied to.