lichess.org
Donate

Why not use the comp as an arbiter in puzzles?

I'm probably not the only one who thinks that some puzzles are a bit unfair. I mean, some moves are rejected even if they are clearly winning. They are just considered "not the best".

Why not set the rules of the puzzles as something like "win from this position against the computer at highest level" ?
Because the point in training puzzles is to find the best possible sequence of moves to take..that would help you to become a better player :)
I would argue that against the machine, if you don't find the best possible sequence of moves, you'll probably lose. And if you win anyway, then you probably found a pretty damn good sequence of moves still.
But there are many puzzles where you already are loads of material up and are just trying to find the quickest checkmate sequence. These puzzles become pointless.
But finding the quickest sequence can mean the difference between a win and a loss in timed games.
Seems the only way there could be a conflict is if there is more than one mating sequence with the same number of moves. I don't remember encountering such a puzzle, but that doesn't mean they don't exist.
With many puzzles I had the feeling that there arre more than one moves which are completely equivalent. I did not check it with the engine ( because there is no possibility I believe to copy the fen and it is a pain to set it up manually ) but I'm almost certain.

This is contrary to the essence of puzzles which assumes that there should be only one clear outstanding best move in every position. If in a position more than one move is equivalently good, the puzzle should end there.

This topic has been archived and can no longer be replied to.