@DVRazor No, KRNvsK with touching Ks is actually a win, even if it might look like a total draw. See the last chapter in this study:
en.lichess.org/study/JYSYPQrNWhat you said about comments is right indeed. I'll have to adjust that behavior.
@ProgramFOX, oh wow, that's complicated, would never come to it myself. Thanks!
True, but it's actually quite rare, since if both sides have pawns remaining, any sane person would trade both the rook and the knight for the pawns and win easily.
Nice site ProgramFOX! I'll add atomic puzzles when I have the time to make/find good ones :)
Stupid question, but is there any way to see the puzzles that you've done already? I think I found an error in the last one I played for atomic (multiple solutions) but I can't actually send a message about it since I can't access the puzzle anymore. Still, very nice page! Thanks a ton!
Also, if you're including RNK vs K, you might want to include RBK vs K. Neither endgame will actually ever occur in practice, but if you're going to have one, you might as well have the other.
@Gannet There isn't such a page yet, but my database tells me that you refer to
chessvariants.training/Puzzle/90Thanks for the RBKvsK note, I'll add that soon!
Ach. I overlooked 1. Kc3 Ke5 ... Qb2.
To fix, shift everything left a file.
@Illion I think I really need to implement puzzle editing, but meanwhile, would you be fine if I set that puzzle state to rejected while you re-create it?
Yep sure. I've also sent a bunch more your way to vet - just standard openings/endings, nothing elaborate.