Oct1012   The Last Straw: White could have played 25.e6!, which wins material after 25...♕xc2 26.exf7+ ♔h8 27.fxe8♕ ♖xe8 28.♖xc2. 

Jan2316
  Penguincw: Hmm, looks like a pretty easy Saturday, which is good, as this week for WednesdayFriday, I didn't even come close. I saw 25.Nf6+, and if 25...exf6, 26.Qxf5, a pin, black is in trouble. My first instinct was 25.Nd6, but that just drops the white queen. Not sure if there's much a difference between taking the queen first vs. taking on e8. Perhaps trap based, because if black plays 27...Qxc2, 28.Rxc2, and the rook is saved. Unusual puzzle in which the material remains balanced at the end (at least from move 26 to 28). 

Jan2316   RandomVisitor: After 25...Ne6
click for larger viewKomodo9.364bit:
<+1.33/36 26.Nf6+ Kf8 27.Nxe8 Rxe8> 28.Qxf5 gxf5 29.Ra5 Nxd4 30.Rd3 Rd8 31.Kf2 a6 32.Rc5 b6 33.Rcc3 Ke7 34.b4 Ke6 35.Ra3 Kxe5 36.Rxa6 Rb8 37.Ra7 c5 38.Rxf7 Ra8 39.Rd2 g5 40.h3 Ra4 41.bxc5 bxc5 42.Re7+ Kf6 43.Rd7 Nb5 44.Rc2 Nd4 45.Rb2 Ke5 46.Rd2 f4 47.Rd8 Ra5 48.Rh8 Ra3 49.Rc8 Ra5 50.Rd8 

Jan2316
  al wazir: I missed 34. e6+, which of course was the whole point. I have to wonder how much Abramovic foresaw of the complications in the lines that weren't played. After 30...Qf5+ white could have gone wrong in several ways, e.g.: A) 31. Ke2 Qe4+ 32. Ke1 (32. Ke2 Qxg2+) Qe4+ 33. Kd1 Qb1+, etc. White cannot escape the checks except by interposing his ♖. B) 31. Kg3 g5! 32. Rh8+ Kd7 33. Rxa8 Qf4+ 34. Kh3 Qh4#. 

Jan2316
  patzer2: Got the first move 26. Ne6+! as my try at this Saturday puzzle, but was too quick on the trigger after 26...Kg8 in going for the second best move 27. Qxf5 (+1.18 @ 25 depth, Deep Fritz 15). I looked briefly at the stronger 27. Nxe8 to and visualized 27. Nxe8 Nxc5 28. Qxc5+ Kxe8 29. Rh3 but was uncertain about the possibilities after 29...Qb1+ 30. Kf2 . I didn't realize the White King could so easily escape the Black Queen's desperation checks. For example, after 30...Qxb2+ 31. Kg3 Deep Fritz 15 indicates it's mateintwelve. P.S.: Black's first decisive error was 24...Qf5??, when White could have won on the spot with 25. e6! (diagram below) click for larger view25...Qxc2 26. exf7+ (+4.78 @ 20 depth, Deep Fritz 15). Instead 24...Qd7 = (0.24 @ 20 depth) appears to hold comfortably for Black. After White overlooked the win 25. e6! and played 25. Ne4?, Black overlooked another opportunity to hold with 25...Qh4! (0.74 @ 20 depth, Deep Fritz 15). Overlooking these opportunities, Black's notsoobvious final error 25...Ne6? allowed the strong 26. Nf6+! to solution to today's Saturday puzzle. 

Jan2316   faulty: What is wrong with 34.Qd6+? 

Jan2316   Cheapo by the Dozen: I didn't try for very long, and went awry for a stupid reason: I somehow imagined that White's queen wasn't defended by the rook ... 

Jan2316
  gofer: I imagine we all found <26 Nf6+>. But this is not the move that matters... <26 Nf6+ ...>
26 ... gxf6?
27 Qxf5
26 ... Kh8??
27 Qxf5 gxf5
28 Rh3#
<26 ... Kf8>
The moves that really matters is one move later. The game line leads
to disaster, but black doesn't have to choose that path... <27 Nxe8 Qxc2>
<28 Rxc2 Kxe8>
click for larger viewWhite is "winning", but the rooks have no open lines to attack
I think this will be a difficult win, possibly a very difficult
win. What do our silicon friends say? 

Jan2316   Nick46: I hazarded a correct guess at 26.Nf6 but the rest was fog. 

Jan2316
  agb2002: The material is identical.
Black threatens 26... Nxc5 and 26... Nxd4.
The rook on g3 xrays the black king. This suggests, 26.Nf6+: A) 26... gxf6 27.Qxf5 wins decisive material.
B) 26... Kh8 27.Qxf5 wins (27... gxf5 28.Rh3#).
C) 26... Kf8 27.Nxe8
C.1) 26... Nxc5 27.Qxc5+ Kxe8 (27... Kg8 28.Rf3 Qb1+ 29.Rf1 followed by Nd6 with an extra piece and there's no perpetual) 28.Rh3 C.1.a) 28... Rd8 29.Rh8+ Kd7 30.Qd6+ Kc8 31.Q(R)xd8#. C.1.b) 28... Kd7 29.Qd6+ Kc(e)8 30.Rh8#.
C.1.c) 28... Qb1+ 29.Kf2 Qxb2+ 30.Ke3 looks winning.
C.2) 26... Nxd4 27.Qxf5 Nxf5 28.Nd6 (28.Rh3 Rxe8 only wins the exchange) 28... Nxg3 29.hxg3 +  [N vs P]. C.3) 26... K(R)xe8 27.Qxf5 gxf5 28.Rc4 +  [R vs N]. 

Jan2316
  morfishine: <26.Nf6+> 

Jan2316
  thegoodanarchist: Wow, I got it. Well at least I got the tactical idea to fork king and rook an the knight is immune. Did not calculate all the way to the end. Anyone who did, congratulations you are probably at least FIDE Master strength. 

Jan2316   Patriot: <agb2002> <The rook on g3 xrays the black king. This suggests, 26.Nf6+> Exactly! I always look for these xrays because they are so common. We are often told to study tactical "patterns". Obviously the positions we study probably will never come up in our games so the key is understanding what makes each tactic possible in order to look for those in our games. This understanding will help develop a better/faster tactical eye. Once a potential tactic is spotted one must calculate what may happen. 26.Nf6+ is therefore initially safe since 26...gxf6 27.Qxf5 wins the queen. 26...Kh7 and 27.Rh3+ forces 27...Qxh3 28.gxh3 gxf6 or 28...Nxc5 are possible continuations. Best is certainly 27.Qxf5 with a clear win. I thought 26...Kf8 was the move to beat. It not only leaves the knight hanging but has some potential to equalize and cause one to question the whole line. Re8 is threatened and so is Rc5. 27.Nxe8, after other considerations, seemed the only try to at least win material. 27...Qxc2 28.Rxc2 Kxe8 simply drops the exchange for black. So to equalize, 27...Nxc5 28.Qxc5+ Kxe8 29.Rh3 looks very menacing. I pretty much stopped there thinking "this must be it" but obviously there is more in 29...Qb1+. 

Jan2316   Patriot: Oops, sorry. 26...Kh7 is not possible in the second paragraph  it should say 26...Kh8. 

Jan2316
  patzer2: <gofer> I too was curious about the best way to proceed after 26.Nf6+ Kf8 27.Nxe8 Qxc2 28. Rxc2 Kxe8 (diagram below) click for larger viewSo I played it out in infinite analysis mode with Deep Fritz at 20 depth per move with the following result: 29. Rh3! Kg8 (29... Nxd4 30. Rd2 c5 31. b4 Ke7 32. bxc5 Ne6 33. c6 bxc6 34. Rb3 Nc5 35. Ra3 Ra8 36. Rd6) 30. Rc4 Rd8 31. Kf2
f6 32. Ra3 a6 33. Rb3 Rd7 34. Ke3 Kf7 35. h4 Ke7 36. Rb6 Nc7 37. Ke4 Nd5 38.
Rb3 Ke6 39. g3 Rc7 40. Rc1 Ne7 41. g4 Rd7 42. Rg1 Nd5 43. Rf1 Rf7 44. h5 gxh5
45. gxh5 f5+ 46. Kf3 c5 47. dxc5 Kxe5 48. Rd3 Nf6 49. b4 Rc7 50. Re1+ Ne4 51.
Rd8 Rf7 52. Re8+ Kd5 53. Kf4 Kc4 54. a3 Nf6 55. R8e6 Nd5+ 56. Kf3 f4 57. c6 Rc7 58. R1e4+ Kb3 59. cxb7 Rxb7 60. Rxa6 Nc3 61. Re5 Kc4 62. b5 Na2 63. Rc6+ Kb3 64. Kxf4 Ka4 65. Ra6+ Kb3 66. b6 Nc3 67. a4 Kb4 68. a5 Nd1 69. Ra7 Rb8 70. Rxg7 (+16.83 @ 20 depth, Deep Fritz 15 x 64). P.S.: Of course the strongest continuation for Black is <26.Nf6+ Kf8 27.Nxe8 Rxe8> given in <Random Visitor>'s 36 depth analysis above with Komodo 9.364 bit. 

Jan2316   Patriot: <agb2002> Take a look at <Random Visitor>s post and as <patzer2> noted. 26.Nf6+ Kf8 27.Nxe8 Rxe8 28.Qxf5 gxf5 29.Ra5. I think the problem with 29.Rc4 is 29...b5 30.Rxc6 Nxd4 lighting up the board like a Christmas tree. After this, black is winning the exchange back and with counterplay. 

Jan2316   mel gibson: Classic pin of the g6 pawn & a discovered queen check using the knight. A nice puzzle. 

Jan2416
  agb2002: <Patriot: <agb2002> Take a look at <Random Visitor>s post and as <patzer2> noted.> Thank you! I should have calculated a little further. 
