People solve it, the rating goes down. People fail, the rating goes up.
I can only think of one method that is better, but it involves Harry Potter and his magic wand.
Tactics trainer puzzles engine derived ?
Sort:
Ubik42 wrote:
People solve it, the rating goes down. People fail, the rating goes up.
I can only think of one method that is better, but it involves Harry Potter and his magic wand.
Perhaps I pass the human puzzles and fail the engine derived puzzles...lol
I was asking myself, other than the rating system and my particular skill set, what could cause me to feel like some puzzles of a given rating range are so easy and others are so impossible to see the right answer for...?
8 hours ago · Quote · #32
Jpatrick
Jeez. looks like ambiguity due to horizon effect.
nameno1had
Jeez. looks like ambiguity due to horizon effect.
It is funny you should say that, I have often questioned how many of the tactics that ended up here are somehow generated by an engine or engine analysis. I don't think we should have to be asked to solve those, but it is too bad their isn't a good way to tell.
Perhaps I pass the human puzzles and fail the engine derived puzzles...lol
I was asking myself, other than the rating system and my particular skill set, what could cause me to feel like some puzzles of a given rating range are so easy and others are so impossible to see the right answer for...?
I think maybe your comment has enlightened me.
Also, I am now questioning again a better way to rate puzzles, based on multi pv engine analysis of a position, as opposed to how well people fair doing them.
Too many people could get lucky, others could have seen the same puzzle 5 times, and still yet others could be using a bot to solve them. I'd rather use a more objective way to rate them, based on something more static