Forums

Chess.com blundered ??

Sort:
Gump_forest

this is a nice endgame study but chess.com blundered here..there are 2 solutions to this puzzle which shouldnt happen in a chess.com puzzle...

here they gave my move wrong but upon max analysis it turns out to be winning 

so where should i complain to get my points refunded lol

xjcl

Puzzle number / link?

xjcl

Looks like Kd4? h4! forces Ke5 back so it can't be the quickest way to win (Kd4? h4! Kxc4?? g5!)

chanelno5x

Answers to the puzzles must be in the fewest # of moves. There is another thread that explains this if you'd like the link.

magipi
chanelno5x wrote:

Answers to the puzzles must be in the fewest # of moves. There is another thread that explains this if you'd like the link.

This is not true. If a puzzle has 2 winning solutions, the puzzle is broken and should be reported.

A puzzle should have one solution that's clearly winning, and all other moves are clearly not winning.

Que-n

I'm used to blunders

Arisktotle

@magipi is right of course. Chess.com makes some exceptions for very short checkmate lines (like 2 or 3 moves) forcing you to find those. Longer alternatives are avoided which explains why chess.com abruptly aborts solutions before the solver is faced with the challenge to choose between winning moves. Often in very awkward positions from the solver viewpoint!

chanelno5x
magipi wrote:
chanelno5x wrote:

Answers to the puzzles must be in the fewest # of moves. There is another thread that explains this if you'd like the link.

This is not true. If a puzzle has 2 winning solutions, the puzzle is broken and should be reported.

A puzzle should have one solution that's clearly winning, and all other moves are clearly not winning.

"Broken" puzzles can be reported. Here's more info about puzzles. I missed one detail. The last comment is by Martin and explains the criteria for a puzzle with a mistake:
https://www.chess.com/forum/view/more-puzzles/i-found-a-mistake-in-puzzles-55751790

Mazetoskylo

How many solutions a puzzle can have? Apparently just one- here is a trivial example.

It seems that this one has two solutions, 1.Ka6 and 1.Kc6.

But in reality there is only one solution, as after 1.Kc6 Ka7 white does have to repeat the initial position to win the game.

Clearly not the case with the position in #1, where both 1...Kf4 and 1...Kd4 win following different paths.

Noo0bb
Mazetoskylo wrote:

How many solutions a puzzle can have? Apparently just one- here is a trivial example.

It seems that this one has two solutions, 1.Ka6 and 1.Kc6.

But in reality there is only one solution, as after 1.Kc6 Ka7 white does have to repeat the initial position to win the game.

Clearly not the case with the position in #1, where both 1...Kf4 and 1...Kd4 win following different paths.

this isnt a puzzle its a solved endgame

Asua-san

Аа

Elmolm
Thanks for the review Mazetoskylo
nguyendeptrai123ew2

thanks

magipi
NF-K wrote:

I analyze most of my games with a higher powered analysis and the accuracy is always different than chess.com game review...

"A higher powered analysis"?

magipi
NF-K wrote:

Sorry I meant to say engine bro

That doesn't make it any more clear. Engines don't show any kind of "accuracy" normally. I have no idea what you're talking about.

magipi

Is chessis a chess.com affiliate or something? If not, then there's no reason for their "accuracy" to be the same thing as chess.com's "accuracy".

Even if they are the same thing, there's a different depth at which they analyze, and that's why the numbers are different.

Crusaders1096

h4