Mates that are difficult for engines

No worries, engines can't do most of these puzzles either, so you are in good company .

i think qinshu was using the ohio meme

and for endgame enthusiast

what does realistic mean?

material imbalance is wacky, that rook on b7 makes zero sense, and those doubled pawns too which are blocked by their own rook somehow?

What exactly is the "Ohio meme"?

u will know when it will happen to you

ie never

0.1% whine so loud here

try to understand chess.coms pov instead of shouting "unfair" at everything

Even if whatever you are crying about did happen to me I wouldn't spam other people's threads with my own personal grievances.

I mean I can’t post puzzles

Pieces can get blocked up in endgames in more weird ways than you might think. The piece count is realistic and the position doesn't looke overly grotesque (artificially constructed) like some of the other ones.

SELFMATE

You're a diamond member, so you should be able to post puzzles unless you mean you don't know how to post puzzles.

Mate in 21, some engines MIGHT be able to find this one, but the vast majority won't:

The last position looks remarkably similar to a 10-mover I corrected 3 years ago. The toughest repair job in my life! The solutions are quite different though.

#10

The interesting thing about this one is that it was meant to be a mate in 27, however I found a shorter solution.

Here is a link to this problem: Mate in 27 problem

Nice underpromotion in the one you corrected to secure the mate in 10, otherwise it would be a mate in 11.

There are probably a million solutions for #11. So many close calls when I analyzed it during creation. The pawns f2/f3/h3/h4 are all just cookstoppers added to the original which had like a billion solutions (or so it seemed to me)

Here is a mate in 40 that was originally intended as a mate in 49, however I found a shorter mate, it utilizes cyclical Queen checks in order to lose a tempo and repeats itself until Black is forced into zugzwang:

Here is a link to the original Mate in 49, unfortunately it doesn't show the solution.

That one is actually very realistic.

here i agree

Qh1 a bit egregious but i can see h1=Q as having happened

Here is a Mate in 34 that utilizes an interesting "reverse" staircase maneuver to get the mate in 34 that engines have trouble with:

Here's a new blog that looks at a couple of short problems SF has trouble solving. One is the M5 by Kohnlein that @drdos7 found and as promised I gave you a mention in the blog!

Kling combination: a self-stalemate theme that perplexes Stockfish

Thanks Peter, I'll take a look at it. I've already looked at some of your other blogs and found them to be of a high quality.