Reported to staff
BUG in new puzzle scoring; equally good win near the end penalized as full MISS
Sort:
1 day ago
0
#3
The actual puzzles didn't change in the update. They were all re-rated but they're the same puzzles and that one should be removed since there's more than one solution.
1 day ago
0
#4
C'est chiant car j'avais 2500 en ( résoudre des problème) et récamment alors que je n'ai rien touché sa ma mis 1600
Count me in the minority - I'm glad for the recent rescaling of puzzle ratings. I'm curious what has changed in the scoring; e.g., I gather the +5 "minimum win" went away, which is fine. And maybe it's okay that some "near solutions" no longer get a positive score. BUT ...
in solving the following, I got the first 5 moves right, after which I chose
6. Kf1 instead of the preferred Kh1.
https://www.chess.com/puzzles/problem/2395660
On shallow analysis, Kh1 was slightly (~0.1 points) better; but when Stockfish went deeper, Kf1 actually came out narrowly on top (captured in image). In any case, it's the difference between +4.3 and +4.2. Before, you would have called the puzzle over after the first five moves, since afterwards there are multiple ways to win equally effectively. Instead, I lost 9 rating points, for a best 6th move! This cannot be how you want the system to work -- something around the determination of how long the winning moves must match, got broken badly with the recent changes!