| Sep-20-15 | | abuzic: It must be 21...Rxc3 22.bxc3 Qxa3 (threat ...Nc1+) 23.Qb1 Rc8 and ...Rxc3 |
|
| Sep-20-15 | | diagonalley: ummmm.... yes... well... seems i'm lost for words... anyone for tennis? |
|
Sep-20-15
 | | Once: Somehow this doesn't feel like a puzzle. The idea of Rxc3 and Qxa3 seems like a useful exchange sacrifice on general principles. We get our queen in amongst the white pieces and we create a passed pawn. We need to be wary of white playing Rxg7, but we can always check the white king to f1 and then force the exchange of queens. This is either a very tough sunday or a very easy one, depending on how much brute force calculation we think we need. |
|
Sep-20-15
 | | Alex Schindler: ... I guess the "insane" designation means I was supposed to think harder about whether rxg7 bothered me? This was a pretty intuitive exchange sac, like once said. |
|
Sep-20-15
 | | agb2002: Black has a knight for a bishop.
White threatens 22.Rxg7.
The pawn on b2 is overburdened with the defense of the knight and the a-pawn. This suggests 21... Rxc3 22.bxc3 Qxa3: A) 23.Rxg7 Qb2+ 24.Kf1 Qc1+ 25.Kg2 (25.Ke2 Qd2+ 26.Kf1 Qxd3+ even loses more material) 25... Qxg1+ 26.Kxg1 Kxg7 - + [2N vs B]. B) 23.Qb1 Rc8 seems to win a second pawn for the exchange. For example, 24.Qc2 Na1 25.Qd2 Rxc3 26.Rh1 Rc8 27.Rhg1 Nb3 28.Qd1 Nc1+ wins the bishop. C) 23.Bb1 Qb2+ 24.Kd3 Nc1+ wins the queen.
D) 23.Kf1 Qc1+ 24.Kg2 Qxc3 25.Bf5 Nc1 with two pawns for the exchange and the slightly annoying threat Ne2. I can't dedicate more time to this puzzle today. |
|
| Sep-20-15 | | dfcx: guessed the first two moves, black gets a knight and a pawn for the rook and will win another pawn soon. The passed pawns on the queen side will win the game for black. |
|
| Sep-20-15 | | lostgalaxy: Puzzle stocks running out lol... |
|
| Sep-20-15 | | The Kings Domain: Nice puzzle, got this one. It's all about breaking white's defence. Strange opening play by the women, what with the double pawns and white bypassing castling thinking she had a good chance at black's king. Good game. It should be a game of the day. |
|
| Sep-20-15 | | morfishine: I settled on <21...Rxc3> and enjoyed the roller-coaster next 16 moves |
|
| Sep-20-15 | | Cheapo by the Dozen: I saw the appealing exchange sacrifice, didn't see a quick, clear-cut win after that, and stopped working on the puzzle. It's not at all obvious to me that Black has a win if White simply defends with 23 Qb1. Indeed, it's not immediately obvious how Black gets a second pawn for the exchange. If White keeps her queen and king near c3 to defend, and eventually swings a rook or two to the queenside, what does Black do? |
|
| Sep-20-15 | | karik: As a premium member I agree. |
|
Sep-20-15
 | | An Englishman: Good Morning: Examined 21...Nh4 first before settling upon the game continuation. White's 24.Rg7 did surprise me, as I thought the game would continue 24.Bf5,Qc3 and if 25.Rg7??,Qc4+. |
|
| Sep-20-15 | | devere: Nice exchange sac, but the only thing "insane" was White's defense. |
|
| Sep-20-15 | | whiteshark: A bit unusual to see ♕+♘ invading decisively on an unoccupied queenside |
|
| Sep-20-15 | | patzer2: In an OTB game, I would have played 21...Rg8 = in today's Sunday puzzle position. Though I calculated 21..Nxh5? 22. Rxh5 Nxd4+ 23. Kf1  going bad for Black, I didn't consider the sound positional exchange sacrifice <21...Rxc3!> 22. bxc3 23. Qxa3  , which favors Black by clearing the way for the advance and promotion of one of Black's Queenside pawns. White's 24. Rxg7?, allowing 24...Nxd3 (-5.83 @ 21 depth), appears to make the win easy for Black. White can complicate and set a trap with 24. Bb5!? a6 25. Rxg7 (diagram below). 
click for larger viewHere (diagram above) Black wins with 25...Qb2! (-2.95 @ 22 depth, Deep Fritz 14) Not 25...axb5? 26. Rxf7! Rg8 27. Rxf6! Rxg1+ 28. Kxg1 Kg7 29. Rf5 = (0.00 @ 28 depth, Deep Fritz 14) when White can force a draw by perpetual with the two active Rooks. P.S.: Early in the opening I prefer 4. g3 as in Ding Liren vs Karjakin, 2015. |
|
| Sep-21-15 | | AnotherNN: For Black I had 37...Nxf2 and if 38.KxN, then 38...Rf5+. |
|
Kibitzing
sac: 21...Rxc3