Attempts at Composition

Sort:
bangalore2
Remellion wrote:

1...Nc5 2. c8=Q Nxd7 3. Qxc3! (not 3. Qxd7? Ne4) is death for black. So there might be a problem there.

Yeah, that is a tougher variation. In retrospect, that should have been the actual puzzle!

Zigwurst

Wow, I was actually wrong, 2...Nxd7 doesn't work due to 3.Qxc3! and there is no fortress for Black.

Zigwurst

4.Qxe6

bangalore2

Oops...

bangalore2

Ok, I have got it. Black Knight on b3 to a3, White king to g7.

AKAL1

This is a geometrical oddity..sacking the queen for two knights doesn't work because the pawn promotes on the same color of the bishop!

bangalore2

No, it is just foresight!

Remellion

Another question: For the new version, 1...Nc4 2. d8=Q Ne4 3. c8=N. I have no idea what the outcome is here.

bangalore2

Amazing, Remellion! You certainly have something against my compositions! :) If anyone has the Lomonsonov tablebases, they can check! That would be much appreciated, thanks!

bangalore2

Inspired by Frankwho's thread, I will post a proof game in 4.0 moves.

AKAL1

Not too deep...

bangalore2

I know, I'm working on a deeper one.

Remellion

Unfortunately not unique, since 2...Bh6 and 3...g4 can be played in either order. I find PGs to be very hard to compose, even short ones. Getting a PG to be unique, let alone interesting, is very difficult.

Then again, I only compose very basic retros, so what do I know...

bangalore2

More than me ....😃

chaotic_iak
bangalore2 wrote:

Inspired by Frankwho's thread, I will post a proof game in 4.0 moves.

Thank you for the inspiration. (Post #75)

bangalore2

Give the last move, and solve the problem.

 
 
Sorry if this is easy. I'm new to this genre.
Remellion

The "last move" stipulation doesn't work. For the first it's -1...f7x~g6 or -1...Kg8xNh8; for the second it's -1...g7-g6 or -1...f7x~g6. 5 possibilities in either case (uncapture N, B, R or Q.)

Also for the second one 1. Bf7 h5 (1...h6 2. Bxg6) 2. Be8 Kh7 3. Bxg6+ wins too.

It's pretty hard to compose endgame studies, and pretty hard to do retros. And more than 4 times as hard to combine the 2 genres usually, since less material typically increases the retro possibilities, and more material means it isn't an endgame. Also retros tend to require extremely abnormal unit positioning relative to a sensible game, which makes it hard to compose forward play around it.

Lastly, I apologise for again pointing out a set of cooks.

bangalore2

Seeking the last move, and the best move for White.

bangalore2

Whoops! I noticed a cook...Place White pawns on b5 and c3, so, is this cooked too, Remellion?

Remellion

Best move for white: literally anything wins.

Last move: Cooked again. -1...Kg8-h8 -2. c4x~b5, or -1...Kg8xNh8 -2. Nf7-h8 or -2. c4x~b5.

Recommended reading for introduction to retros: http://www.chess.com/forum/view/fun-with-chess/retros-for-dummies. The ability to see retractions as easily as forward moves is a huge help in composing and solving.