The Greatest Forgotten Chess Puzzle 

Lesson to learn - if you have one pawn ready to become a queen, don’t wait for 2 more.

That basically combines every weird rule in chess into one puzzle, ie - double pawn first move - En passant - Castling - Promotion and under promotion

The real puzzle is how the king and the other rook didn't move while three pawns made their way into the second rank

I love the thing that the bishop at c8 is just chilling and watching his king and pawns fight for the king's life

These puzzles where you have to randomly castle are really wierd. How could I have known, this was still an option?

"Imma castle with no pawns on the 2nd rank, in the end game, on move 30-something." Utter madness lol.

The castle move is hard to spot because it is not even clear that casteling is still alowed, in a position this late in the game the king or rook most probably moved at some point (and end up at starting squares) making casteling ileagal.

A Google search says that this problem was composed by Yochanan Afek in 2011. I don't think that RU-vid will allow me to post a link, but try searching for the ARVES chess endgames study association.

Actually this puzzle shows the so-called "Valladao-Task": an en-passant capture, castling and pawn promotions in the same chess puzzle :)

Great find! This is the so-called Valladão Task, a problem showing promotion, en passant capture and castling. The underpromotion to a knight is a bonus. Btw., I have not found this problem in the known problem databases. Probably someone else will have more luck.

Just to clarify: it's a convention in chess problems that castling is valid unless you can prove it isn't. The convention is made that way round, because, otherwise, problems wouldn't be able to involve castling, because you can't *prove* that castling is valid. In some problems it looks as if black may castle, but something about the position means you can prove that black has lost the right to castle, so white may do something which works, but would fail against castling.

Ah. Hidden castling. I hate those, since you can‘t know if the move is legal.

The purpose of the bishop on c8 is to prevent quicker checkmates that include Rf5

Study by Yochanan Afek from 2011 (a famous israelien chess study composer and also IM as a chess player). He received 3rd commendation for this Valladao theme study (but in his carreer he often got first prizes in many tournaments).

Great composition! It has all of the little tricks that make a puzzle interesting.

I started with 0-0 as the 1st move, and I still do not see why it does not win. White threatens mate in 3 (g4+, Rh7+, Bg5X), and it seems that after 1....Kg6 2.g4 we are back to the propsded solution (B:g4 does not help). So the interesting defense is 1....f3 2. R:f3+ Kg6 3. Rg3+ Kf5 (3.... Kg6 4. Bg5+ and mates) 4. Rg5+! Kf4 (4....K:f6 5.R5-g7+ Kf5 6. g4+ Kf4 7. Rg-f7+ Bf5 8.R:f5X) 5.f7! letting black to promote 2 queens, but it looks like they can't help, like 5.... c1-Q+ 6.Kh2 b1-Q 7. g3+ Kf3 8. f8-Q+ Bf5 9: R:f5+ with mate, or 5....b1-Q+ 6.Kh2 Qg1+ 7.K:g1a1-Q+ 8.Kh2 Qh8 9.Rg8 and wins. So it looks like the solution is not really unique. Or did I miss something?

Love these types of videos!

Thank you for the amazing video!

I knew that as soon as I saw the white king and rook's position in the thumbnail, that there was going to be a hidden castling move here

Hard to plan a move if castling is unknown to be available this late in a game.

Thought I was clever for seeing the pawn to g4 forcing en passant but I missed the castle check and never would have gotten the rest of the puzzle lol. Great but very complex

Your lessons are the best on RU-vid.

en passant is the most obscure move in chess, yet it makes this puzzle so much harder...

The castle was the first thing I spotted, it would've been my first move if I were working through this puzzle myself

En passant, castling, and underpromotion! Love it!

You didn’t forget the “en passant” rule. Pawn to G4. F4 takes G4 by moving to G3. I was very surprised when I learned that rule.

Tip for doing puzzles: A check is usually the answer

Chess Vibes uploading chess puzzle videos speaks for itself.

1:37 En Passant:

Nelson really found a good study

I feel like the castling was a bit of a cheat, as it seems unlikely that castling would still be possible this late in a game.

Pretty good puzzle, like that it uses enpassant, which I have never been able to use in an actual game XD

Pretty sure I found the solution from the picture Push pawn, opponent queens , you move king, opponent takes rook, you queen mate, but they will probably defend the promotion with the queen

At 8:11 the checkmate could still be delayed If the black king move to Kh4 the white queen can try to move Qh6 for check Black queen can blocked Qh5 but the bishop for white will checkmated at Bf5#

In this puzzle, It is very good exam that king safety is the most important. Despite has three extra pawn nearly promote, black lose. Fascinating!

As @TeaInTheMorning4929 pointed out earlier, it is not clear that castling is possible. I have seen puzzles where the king and a rook were on their starting squares with solutions which hinged on proving the fact that castling was not possible. Accordingly, does it not make sense to provide a proof game for the given position to show that this position (with castling allowed) is feasible?

Superb puzzle as usual 🎉

What's the problem with castling first and then playing g4?

Anyone else thinking that there is no way that king and rook hasnt move this whole game to even think that castling was an option

The black bishop could take that pawn, then in line of sight to take out the rook, so the rook would have to move, unless you want to sacrifice your rook, or try to keep black's king in check.

This 76 year old remembers my favorite puzzle where both kings were almost totally locked into their respective back two ranks and the drawing side had to “ match” squares with the side trying to win…. I loved that long lost puzzle.

The Greatest Forgotten Chess Puzzle speaks for itself

Underpromoting to a knight is subtle - and brilliant!

Since it initially looked impossible, I realized that there had to be a castling move to gain tempo while protecting the king. :-)

The purple is wrong 6:20 we can just promote the pawn by pushing it

End game teaching excellent

I swear I've seen this one too but couldn't tell you where sadly

Move the white peon two spaces up, you check the king, so has to move from f5 to g6, move tower in front of king g7, only place left to move for the king is h6. Peon that moved two spaces finished the deal

At 6:21 I’d like to explore the black pawn moving to G1 and becoming queen instead of taking the bishop. I’m sure it’s clear but my mind can’t work that far ahead.

2:14 so unless it’s the only move you have you can’t en passant out of check.

I finally got one of these puzzles right!!!

1:32 my guess is which should castle to get the rook in like to attack any potential queens. 1:36 en passant... don't need to move the king. Moving the king is a blunder here

Gonna remember this when i get to this position

Black's bishop: 🍿

I knew it should become a knight at first glance!

NGL this is the first time I've seen An pasant in action. I've only just heard of it

I like this dude because he every possible move

I actually thought about Qc1 usntead of Bc1 to trade and maybe bishop & rook vs bishop is winning since they're on opposite squares? I guess not though

Yeah, I've seen parts of it, like first three moves I did it just by looking at the picture, then I got confused somewhat.

Love the cadtling idea but its hard to know if it isnt stated that the king has never moved

Did not see this one. Got distracted by the pawns and tried to figure out how to stop them...Lol

Man that was insane and interesting

if you played bg6 at 5:20 black would be forced to take the bishop with the queen and afterwards rxg5 if you make a queen it doesnt help because i will play f8=q+ and i think that is loosing for black. and if bg6 qxg6 rxg6 and kxg6 i think h4+ and maybe h4 is not needed because we can promote to a queen ultimately drawing the position

This puzzle goes crazy. Good explanation though!

If castling is available in what's approaching an endgame, it should be stated up front.

The king side castle is dirty. Because, we don't know if it's still available.

Thing is, if I was white in that position, I would’ve already assumed that I couldn’t castle.

i saw mate in 3 and thought you lost your edge, then on croissant

i didn't know castling was an option.

7:24 , black moves queen to F2, forcing the trade and ultimately wins? Edit: It leads to situation where you have to choose between eating bishop or a pawn with the king, if you eat bishop you will lose and if you eat pawn it leads to draw.

There is a much easier solution, Rock to f5 while bishop is in g5 and king in h5 is the key, you don't need to castle in that position.

OMG THAT'S THE FIRST TIME I SAW EVERY BEST MOVE ON A TRICK PUZZLE ON THIS CHANNEL * _ *

You can move 2 to the white paida Or soldier

g4+ was the pretty obvious 1st move (only move, w/o much thinking actually.) 0-0 was a bit tricky, as we didn't know if it was still an option at such a late point in the game. Once 0-0 was allowed, rest was also pretty obvious, as all moves are more or less forced.

Wow there are so many rules I didn't even know about

hmm, there seems to be an out for black throught promoting pawn C1 into knight and check on the king

I saw correctly until castles part which seemed funny at such an endgame scenario

Castling in end games are crazy 💀

I too completely missed the castle. It might help if you saw the game from the beginning, but OTOH you might get distracted by other possible moves.

7:24 has Qc2 forcing queen trade...but still winnable with bishop and rook....

Very nice puzzle.

That's rather cheesy since it's hard to imagine the king has never moved with that kind of carnage on the board.

At 3:00 how do we know if rook or the king didn't move and we can castle?

But there is no condition in this problem to determine whether either the rook or king for white has moved already. We don't know whether we can castle.

Wasnt aware the king hasnt move to allow castling.

i knew you would castle

What I don't like about the castling situation in a puzzle is you're assuming up to that point in the game the king hasn't been in check already, either of those 2 pieces has moved or castled already. It's a late game situation, so chances are one of those 3 situations probably has happened.

Black and play a draw position at 7:24mins in. Queen to f2. it forces a trade of queens. the line goes white queen f2, pawn f2, king f2, bishop f3.

6:12 if you look carefully there's that black bishop that can take down the white pawn then threatening the white rook forcing it to move it's position where it's unguarded by the queen. Then black pawn to a1 forcing the rook to take it down and black getting another queen using the b2 black pawn to take down the rook and getting promoted then forking the rook and check matting the white king without yet getting the pawn promoted to a queen forcing the king to move then black queen taking down the rook and threatening the pawn from promoting

6:11 (pause and think) A2 Pawn promotes to queen. if you promote your g7 pawn to queen or anything else then black can win because you cant put him in check without saving your king, he can move his queen to a7 delivering a check and if you block with your rook to F2 he can eat the rook with the queen and checkmate you, if you move your king to h1 its checkmate with the bishop to B7 (black wins by checkmate) also if white blocks then queen check with his bishop to E3 then black can simply take the bishop and still checkmate white

It was unclear that castling was available because it wasn't stated whether King and Rook had moved.

If I'm not even registering the moves where he's like "You might be tempted to do this..." does that mean I'm getting better at these or do I just have a different problem now? lol

Yea, thought about this for like 15 minutes and never considered castling was still legal

Great ending!

I’m getting better at these. I actually called most of these moves

Thanks!

on 6:34 what if the queen check the king and a follow up with their bishop?

Would have had this the first time if I knew that White had Castling available. Might be my fault but I would have assumed castling was not possible this late in the game.

I said I didn't learn any chess besides the basic rules but this time I found the castling. It is a sudden realize "aren't both of them at their original position?" Yep castling should be considered available unless otherwise specified. But the "en passant checkmate in 1" episode you proposed before is really confusing. How do I know in that puzzle this en passant is available? Or in other words, since the puzzle is always based on real games shouldn't the previous opponent move be stated in the puzzle as well?

This is the second time you've done this in a puzzle I've seen. When a game is THIS far along, how are we supposed to know if castling is still possible? Should it just be assumed because they're in the starting positions?