< Earlier Kibitzing · PAGE 1 OF 2 ·
Later Kibitzing> 
Jul3016   RandomVisitor: 22...N8d7 or 22...Rc8 might be improvements 

Jul3016   Razgriz: Saw Rx7+ but the continuation got a bit moe complex for me 

Jul3016
  andrewjsacks: What kind of move is 22...Rxf3? 

Jul3016   plumbst: My guess was 20.Ng5 Kxh6 21.Nxe6, but 20...Bg8 seems to hold. In the game line, White only gets a pawn for the exchange but the powerful knight and epawn carry him to victory. This week seems to feature a lot of unclear sacrifices though, no instant wins here. 

Jul3016   RandomVisitor: <andrewjsacks><What kind of move is 22...Rxf3?>A move that is worth considering, along with: click for larger viewKomodo1064bit:
+0.30/37 22... Rc8 23.f4 N8d7 24.e5 Nf8 25.Nc5 Re8 26.Nxb7 g5 27.Nd6 Re6 28.f5 Rxe5 29.Nde4 Nd5 30.Bc4 Kg7 31.Bxd5 cxd5 32.Nxg5 Rc8 33.Kb1 Rc5 34.a3 Rc4 35.Nf3 Re3 36.Nd4 Rg3 37.f6+ Kg6 38.Nf5 Rh3 39.Nxd5 Re4 40.Ka2 Rh7 41.Nfe7+ Kg5 42.Rc1 Rf7 43.Rf1 Rxg4 44.Rf5+ Kh4 45.b4 Rg2+ 46.Kb3 Rg3+ 47.Kb2 +0.34/37 22...N8d7 23.Nxf8 Rxf8 24.Be2 Ne5 25.Rf1 g5 26.Nd1 Ng6 27.Ne3 Nf4 28.Bd1 Kg7 29.Rh1 Rd8 30.Kc2 Kf6 31.Rh7 Rd7 32.Rh8 Ne6 33.b3 Rd8 34.Rxd8 Nxd8 35.Nf5 Ke5 36.Kd3 Ne6 37.Ke3 Nf4 38.Nh6 Nc8 39.Be2 Ng2+ 40.Kf2 Nf4 41.Bf1 Nd6 42.Ke3 Ne6 43.Nf5 Nf4 

Jul3016
  wtpy: I liked ng5 kh6 20 ne6 Rf7 21 Bc4 g5 22Bb3 and while white is exchange down pieces are well placed and along with passed e pawn I think white is better but just got back from party so judgement possibly impaired. 

Jul3016   dnp: actually I surprised myself by solving today's puzzle. BTW I don't use programmes for solving. 

Jul3016   waustad: Oh my. I got several ply into the solution in seconds, after having nearly a liter of wine. Usually late in the week I find it a lot harder. 

Jul3016
  FSR: The sac is obvious, but it's not obvious that it wins. Indeed, as the analysis <RandomVisitor> posted shows, it doesn't. Another busted puzzle. 

Jul3016   stacase: Does getting the first four moves count? 

Jul3016
  Once: <FSR> and <Randomvisitor> I have to agree. Fritzie 15 is saying that with best play 20. Rxh7+ leads to a tiny white advantage of less than half a pawn. It seems to be the best move that white has available and it sets lots of traps for black. But it ought to have been a draw. Fritzie is saying that 25...Nf6 is when things start to go wrong. Instead 25...Ne5 would have been half a box of eggs  0.00. Just like <plumbst> I looked at 20. Ng5, but the sneaky 20...Bg8 causes problems for white. click for larger viewWhite's attack on h7 doesn't work and he now has to defend against Kxh6.
Black also has a threat of Rxf3 (if the Ng5 is forced to move). The move white would like to play is 21. Rh3 but that takes away the last escape square for the knight. 21...h6 would be embarrassing. So white is forced to look for a rook move that isn't Rh3, but everything apart from 21. Rxh7+ allows 21...h6 and the knight has to abandon the defence of the f3 pawn. All in all, 20. Ng5 looks tempting but doesn't quite work. It drops the f3 pawn when the passed e4 doesn't look anything like so scary. It looks like 20. Rxg7+ is white's best move but it should not have been winning with best play. 

Jul3016   The Kings Domain: Amateurish blunders by black which cost him dearly. 

Jul3016   jd4chess: Stockfish actually narrowly prefers 20 Rh4 and scores the game continuation level until 25..Nf6 which suddenly loses (Ne5 is better). 

Jul3016   morfishine: Every POTD, or every decisive game ever played can be busted by a modern engine at some point in the game 

Jul3016   RandomVisitor: After 22.Nxe6
click for larger viewKomodo1064bit:
<+0.19/45 22...Rc8 23.e5 N8d7> 24.f4 Nf8 25.Nc5 Re8 26.Nxb7 g5 27.Nd6 Re6 28.f5 Rxe5 29.Nde4 Nd5 30.Nxg5 Nh7 31.Ne6 Ne3 32.Re1 Nxg4 33.Be2 Ne3 34.Bd3 Kf7 35.Ne4 Nxf5 36.Nf4 Rae8 37.Bc4+ Ke7 38.Nd3 Kd8 39.Nxe5 Rxe5 40.Kd2 Nf8 41.b3 Kc7 42.Bd3 Ne6 43.Rf1 a5 44.Kc3 Nd6 45.Nxd6 Kxd6 46.Rf6 Ke7 47.Rh6 Re3 48.Kd2 Rg3 49.Bc4 +0.33/45 22...N8d7 23.Nxf8 Rxf8 24.Be2 Ne5 25.Rf1 g5 26.Nd1 Ng6 27.Ne3 Nf4 28.Bd1 Kg7 29.Kc2 Kf6 30.Rh1 Rd8 31.b3 Ne6 32.Nf5 Ke5 33.Rh7 Rd7 34.Rh8 Rd8 35.Rxd8 Nxd8 36.Kd3 Ne6 37.Ke3 Nf4 38.Nh6 Kf6 39.b4 Nd7 40.a3 c5 41.bxc5 Nxc5 42.Kd4 Nce6+ 43.Kc4 a6 44.Kc3 Ke5 45.Nf7+ Kf6 46.Nd6 b5 47.Nf5 Nd8 48.a4 +0.52/45 22...Rxf3 23.Nc7 Rf8 24.a4 a5 25.Nxa8 Nxa8 26.e5 Nc7 27.Bc4+ Kg7 28.e6 Re8 29.Ne4 Nd5 30.Nd6 Re7 31.Re1 Kf6 32.g5+ Kxg5 33.Nxb7 Kf6 34.Nxa5 Rxe6 35.Rf1+ Ke5 36.Rf7 Re8 37.Rg7 Kf6 38.Rb7 Ke6 39.b4 g5 40.b5 Kd6 41.Kd2 Rh8 42.Be2 Nf4 43.Bf3 cxb5 44.Rxb5 Nd7 45.Nc4+ Ke7 46.Rxg5 Ne6 47.Rg6 Rh4 48.Be2 Ndc5 49.a5 Nf4 50.Rg7+ Kf6 

Jul3016   YouRang: Saturday 20.?
click for larger view
I don't suppose that I was the only one to spot the "clever" exchange sac <20.Ng5> threatening both Nxe6 and Rxh7. I expected <20...Kxh6 21.Nxe6>
click for larger view
This attacks Rf8, and I threaten to mate after g5+ and moving by Bf1 out of the way of Rh1#. Black needs to attack my N with <21...Re8>, but then I have <22.g5+! Kh5 23.Ng7+> forking K and R.
click for larger view
This combination effectively wins a piece with and leaves black's king in a highly compromised position. It was so pretty, that I was sure that this must be the solution... ~~~~~
Checking with the computer, I learned that black had a much better option than 20...Kxh6, namely <20...Bg8!>, which moves the bishop out of take and guards Ph7.
click for larger view
Now white has no attack, and in fact black will win the backward Pf3. White has to withdraw the rook with <21.Rh2> (note that we can't protect the Pf3 with 21.Rh3? because h3 is the knight's only escape square after ...h6) <21...h6 22.Nh3 Rxf3> and black is better. Phhft. 

Jul3016   Patriot: 20.Rxh7+ Kxh7 21.Ng5+ Kg8 22.Nxe6 Rxf3 23.Nc7 is pretty easy to find. The counterplay isn't so easy but black has a lot of work to do to maintain a balance. 

Jul3016   YouRang: As for white's 20th move, my engine suggests that white has two reasonably playable moves: [1] 20.Rxh7 (as played)
[2] 20.Rh4
With <20.Rh4> black can take the pawn with <20...Rxf3> but then <21.Ng5!> forks R and Be6, leaving black with nothing better than <21...Rxc3+ 22.bxc3 Bg8>
click for larger view
And here the engine likes 23.a4! because the tempting 23...Nxa4 is met by 24.e5!, and having to stop this passed pawn creates some surprisingly serious trouble for black. But even if black resists 23...Nxa4, and plays the recommended <23...a5>, white is still better after <24.Rh2>. It seems that white can regroup to attack black's weak queenside and push the passed epawn. Following this line a ways, I see evals around +1.00 and rising, so arguably <20.Rh4> is just as good or better than 20.Rxh7. 

Jul3016   Patriot: <Once> On 20.Ng5 Bg8, I would try 21.Rxh7+ Bxh7 22.Ne6+ Kf6 23.Nxf8 Kxf8 24.Rd8+ and "if" 24...Ke7 25.Rh8 wins the trapped bishop. 24...Kg7 and black needs to unwriggle with something like ...Nbd7, ...a5, etc. But it's best to avoid this altogether. This shows the importance of considering all the defenses. And 20...Bg8 is very logical since it accomplishes two things: 1) Defends the bishop; 2) Defends the weak h7 square. Aside from those it renews the threat on h7 and leads to winning f3. 

Jul3016   Patriot: <Once> I guess he doesn't have to do a lot of unwriggling  ...Nbd7 is good enough since the rook is guarded by Nb6. 

Jul3016
  agb2002: The material is identical.
Black threatens Kxh6 and Rxf3.
The defenseless bishop on e6 and pawn on h7 and the possibility of creating a mating net suggest 20.Ng5: A) 20... Kxh6 21.Nxe6
A.1) 21... Rxf3 22.g5+ (22.Bg2 Rxc3+ 23.bxc3 g5 due to Rh1# 24.Nc7 +  [R+B vs 2N+P]) 22... Kh5 23.Be2 Kg4 24.Rg1+ and 25.Bxf3 +  [B vs P]. A.2) 21... Rf7 (or 21... Rc8) 22.Bd3 g5 23.e5 Rxf3 24.Rh1+ Rh3 25.Rxh3#. A.3) 21... Rf6 22.g5+ Kh5 23.gxf3 wins a piece at least. A.4) 21... Re8 22.g5+ Kh5 23.Ng7+ Kxg5 24.Nxe8 +  [B vs P]. B) 20... Bg8 21.Rh2 h6 22.Nh3 Rxf3 loses a pawn.

Another option is 20.Rxh7+:
A) 20... Kxh7 21.Ng5+
A.1) 21... Kg7 22.Nxe6+ Kf7 23.Nxf8 wins a pawn.
A.2) 21... Kg8 22.Nxe6
A.2.a) 22... Rxf3 23.Rd8+ Kh7 24.Ng5+ and 25.Nxf3 +  [B]. A.2.b) 22... Rf7 23.Rd8+ Kh7 24.Ng5+ Kg7 25.Nxf7 wins a pawn at least. A.2.c) 22... Rc8 23.Nc5 seems to provide enough compensation for the material. For example, 23... Rc7 24.Rd8+ Kf7 (24... Kg7 25.Ne6+ and 26.Nxc7 wins) 25.f4 Ke7 (25... N8d7 26.Rxa8 Nxa8 27.e6+ wins decisive material) 26.Rg8 Kf7 27.Rg7+ Kxg7 28.Ne6+ Kf7 29.Nxc7 and 30.Nxa8 with an extra pawn and the better ending. A.3) 21... Kh6 22.Nxe6 looks similar to A after 20.Ng5 with the difference of the pawn on h7. For example, 22... Rc8 23.Ba6 Kh7 (23... bxa6 24.Rh1#) 24.Bxb7 Re8 25.Ng5+ Kh6 26.Nf7+ Kg7 27.Nd6 wins two pawns at least. A.4) 21... Kh8 22.Nxe6 is similar to A.2.
B) 20... Kg8 21.Rxb7 Rxf3 22.Ng5 Rf6 23.e5 wins decisive material. 
I think I'd play 20.Rxh7+. 

Jul3016   RandomVisitor: On 20.Rh4 black has 20...N8d7
click for larger viewKomodo10.164bit:
<+0.09/36 20...N8d7 21.Be2 Rae8 22.Rh1 Rh8> 23.b3 h6 24.f4 Nc5 25.f5 Bc8 26.Nf2 a5 27.g5 h5 28.fxg6 Ref8 29.R1h2 Nbd7 30.Kd2 Ne5 31.Ke3 Kxg6 32.Rxh5 Rxh5 33.Bxh5+ Kxg5 34.Kd4 Ncd7 35.Be2 b6 36.Rg2+ Kf6 37.Ng4+ Nxg4 38.Bxg4 Rd8 39.Bxd7 Bxd7 40.e5+ Kf7 41.Rf2+ Ke7 42.Ne4 c5+ 43.Ke3 Rh8 44.Ng5 Be6 

Jul3016   Shamot: Why 23. Rd8+ was not played by white? It leads to material gain. Any move by black king to get him out of check leads to white knight forking the black king and rook. 

Jul3016   RandomVisitor: <Shamot>After 23.Rd8+ Kf7 black escapes with the better position click for larger viewKomodo10.164bit:
0.78/27 24.Rf8+ Kxe6 25.Rxf3 N8d7 26.Rh3 Ne5 27.Be2 a5 28.g5 Rf8 29.a3 Rf2 30.Rh8 Nbd7 31.Ra8 Nc5 32.Kd1 Ned3 33.Kc2 b5 34.Rxa5 Nf4 35.Kd1 Nxe4 36.Nxe4 Rxe2 37.Nc5+ Kf5 38.Ra6 Rxb2 39.Rxc6 Kxg5 40.a4 Kf5 2.05/27 24.Ng5+ Ke7 25.Nxf3 Kxd8 26.Ne5 g5 27.Nf7+ Ke7 28.Nxg5 N8d7 29.Nf3 Rf8 30.Be2 Nc4 31.Nd4 Nce5 32.Kd2 Rg8 33.Nf5+ Kf6 34.b4 Nxg4 35.Nd6 b6 36.b5 Nge5 37.bxc6 Nxc6 38.Bb5 Rg2+ 39.Ke3 Rg3+ 40.Kd2 Nde5 41.Nf5 Rg2+ 42.Ke3 Kg5 43.a3 Rh2 44.Ne2 Rh3+ 45.Neg3 a5 46.Kd2 2.65/27 24.Nc7 Rxf1+ 25.Kc2 Rf2+ 26.Kd1 Ke7 27.Rg8 Rf8 28.Rxg6 N8d7 29.Nxa8 Nxa8 30.Ne2 Ne5 31.Rg7+ Rf7 32.Rg5 Kd6 33.Ng3 Nc7 34.Rg8 Kc5 35.Nf5 Rd7+ 36.Ke2 Ne6 37.g5 Rd3 38.b3 Nf4+ 39.Ke1 b6 40.Kf2 Kb4 41.Rg7 a5 42.Re7 

Jul3016   RandomVisitor: 20.Rh4 still looks about equal
click for larger viewKomodo10.164bit:
<+0.03/41 20...N8d7 21.Be2 Rae8 22.Rh1 Rh8> 23.f4 Nc5 24.b3 Nxe4 25.Nxe4 Bd5 26.Nhg5 h6 27.Bd3 hxg5 28.Rxh8 Rxh8 29.Rxh8 Kxh8 30.Nxg5 Kg7 31.Kd2 Bg8 32.f5 gxf5 33.Bxf5 Kh6 34.Ne4 Nd5 35.Bd7 b6 36.g5+ Kg7 37.Nd6 Kg6 38.Bxc6 Ne7 39.Bg2 Kxg5 40.Ke3 Nf5+ 41.Nxf5 Kxf5 42.Kd4 Kf4 43.Bd5 Bh7 44.a3 Bg6 45.Bc4 a5 46.Kd5 b5 47.Be2 Bf7+ 


< Earlier Kibitzing · PAGE 1 OF 2 ·
Later Kibitzing> 