Nov-04-20 J H Donner vs H Ree, 1971 
| |
| Predrag3141: <After 33 Rxf7+, can Black survive with 33 ... Kg8> Black can go down the exchange but it's a tough game to win with an exposed king. The exact lines vary depending on where Black's king goes in response to checks along the 7th and 8th ranks, so I'll describe it instead of |
|
|
|
Oct-31-20 V Arbakov vs Milov, 1994
| |
| Predrag3141: I could see to the end better after 23 ... g5 24 fxg5 Be5 -- clearly unacceptable to White. Instead, White has to allow Black to take over the dark squares with gxf4, etc. Stockfish finds a clever defense though: 23 ... g5 24 Nd7 Rf7 25 fxg5 Rxd7, returning the piece. The attack |
|
|
|
Oct-30-20 Y Neishtadt vs Gipslis, 1955
| |
| Predrag3141: I expected 20 Rh8+, which Stockfish rates just over +10. I couldn't see it to the end, but Black gives back a rook and the attack continues. The game continuation, 20 Qh5, is over +16, soe either way is fine. |
|
|
|
Oct-29-20 Oll vs Psakhis, 1987
| |
| Predrag3141: The puzzle of the day starting before Black's 22nd move has to involve ... Bxh2+ and ... Rxb3, but it took a while to work out when to play each. <Rxb3 first> 22 ... Rxb3 23 Rxb3 Bxh2+ 24 Kh1 Qh4 25 g3 looks advantageous to Black but unclear because White has banked an ... |
|
|
|
Oct-28-20 N Gaprindashvili vs A Pomar Salamanca, 1973
| |
| Predrag3141: The error in puzzle positions seems to occur in one place where a human enters the wrong move number, from which the server generates a matching (but incorrect) caption and URL. Here as in a recent case where the move number was after the end of the game, the caption and URL ... |
|
|
|
Oct-27-20 E Bakhmatov vs Tseitlin, 2007
| |
| Predrag3141: Stockfish (in a 6-second run) considers 29 Qxf3 (as played) to be best, even though it allows 29 Qxf3 Qa7. Allowing the bishop to remain on the board is even worse because ... Be4+ looms. Since White should accept the bishop sacrifice, the winning combination really starts a ... |
|
|
|
Oct-22-20 I Findlay vs L Day, 1980
| |
| Predrag3141: <The non-obvious 26 ... Qf8 seems to leave most options open> Yes, 26 ... Qf8 looks good. |
|
|
|
Oct-20-20 I Cheparinov vs O Salgado Lopez, 2003
| |
| Predrag3141: 22 ... cxb4 would have made the b4-pawn a target. 22 ... c4 handed White a modest advantage but a bigger practical advantage: he can place his bishop on e3 and (if Black is not careful), queen on d5, and await tactical opportunities. Once the queenside was frozen in place, Black |
|
|
|
Oct-19-20 C Zeng vs Nguyen Huynh Minh Huy, 2013
| |
| Predrag3141: At first I thought the solution to the puzzle was 25 Qxd7, winning two pieces. Surprisingly though, it only wins back the one piece White is down, due to the weakness of White's back rank. And 25 Bxh7+ wins a whole queen. |
|
|
|
Oct-17-20 Dubov vs R Svane, 2019
| |
| Predrag3141: The hardest move to see in the puzzle position before move 31 was 34 Qg2. I was planning 34 Qc6+, but that is only equal according to Stockfish. The point of 34 Qg2 was to force Black to move his c5 bishop, opening the c-file to allow 37 Qc1+. Since my plan led to equality, no ... |
|
|
|
| indicates a reply to the comment. |
|
Kibitzing
