|
< Earlier Kibitzing · PAGE 3 OF 3 ·
Later Kibitzing> |
| Dec-09-10 | | knight knight: After a few mins I've found 38...Bg4 attacking the rook: ai) 39. Ra1/b1/c1/e1/f1 Bxh3 40. gxh3 Qxh3+ 41. Kg1/Qh2 Qxd3 two pawns ahead aii) 39. Rg1 Bxh3 40. gxh3 Qxh3+ 41. Qh2 Qxd3 two pawns ahead b) 39. Be2 Bxh3 40. gxh3 Qxh3+ 41. Kg1 Qxc3 two pawns ahead c) 39. Ne2 Bxh3 40. gxh3 Qxh3+ 41. Kg1 Qe3+ 42. Kf1/h1 Nxe4 three pawns for the piece with exposed white king (43. Qxe5?? drops the queen after the appropriate knight check, mind you white could play 43. Bxb5 instead... then maybe 43...Nc3 44. Nxc3 Rxc3 black still has an attack, bit unclear tho) d) 39. Rd2 (the most annoying defence) Bxh3 40. gxh3 Qxh3+ 41. Kg1 well here I'm stuck Hang on, maybe 38...Bg4 is too fancy, perhaps just 38...Bxh3 39. gxh3 Qxh3+: a) 40. Qh2 Qxh2+ 41. Kxh2 Rxc3 two pawns ahead
b) 40. Kg1 Qg4+ 41. Kh2 Rxc3 42. Qxc3 Qxd1 two pawns ahead c) 40. Kg1 Qg4+ 41. Qg2 Qxg2+ 42. Kxg2 Rxc3 two pawns ahead Best defence might be 38...Bxh3 39. Kg1 Bg4 one pawn ahead. If 38...Bxh3 39. B/Nxb5?? Bd7+ 40. Kg1 Nxb5 41. N/Bxb5 Bxb5 42. Qxe5 black has a piece for a pawn, mind you white's pawns look a little dangerous. Let's see what happened...
PS: Any Mosconi Cup fans? I've got tickets for tonite yay!!! :-) |
|
Dec-09-10
 | | Sastre: <gmalino: I think 38...Rxc3 is essential because if you start with the bish white brings his Queen to the Kingside, at least drawing> 38...Rxc3 39.Qxc3 Bxh3 40.Qe1 Qxe1+ 41.Rxe1  . If 38...Bxh3 39.gxh3 Qxh3+ 40.Qh2, 40...Qxh2+ 41.Kxh2 Rxc3  . |
|
| Dec-09-10 | | knight knight: Missed 39...Bxg2! Nevermind... |
|
| Dec-09-10 | | gmalino: <Sastre> Ok, good argument, missed it completely!
Thx for opening my eyes... |
|
Dec-09-10
 | | agb2002: The material is even.
White would probably focus on the waek pawn on b5.
The white castle is not adequately protected because the white pieces operate on the other side of the board. Another detail is that the sole defender of the white rook is already under attack. Therefore, 38... Bxh3, exposing the white king and allowing a queen fork against the king and the rook: A) 39.gxh3 Qxh3+ 40.Kg1 (40.Qh2 Qxh2+ 41.Kxh2 Rxc3 - + [N+2P vs B]) Rxc3 (40... Qg3+ 41.Kf1 Rxc3 42.Qxc3 Nxe4 43.Qc8+ Kg7 44.Bxe4) A.1) 41.Qxc3 Qg4+ (41... Nxe4 42.Qe1 Ng3 43.d6)
A.1.a) 42.Kf2 Qxd1 43.Qc5 (43.Qc6 Qd2+ 44.Kg1 Qxb4 - + [N+3P vs B]) Qd2+ 44.Kg1 (44.Be2 Nxe4+ and 45... Nxc5 - +) Qxd3 45.Qxd6 Qxe4 46.Qb8+ Kg8 47.Qxb5 Qd4+ 48.Kg2 e4 49.Qc6 (49.Qc5 Qxc5 50.bxc5 Kf8 - +) e3 50.d6 e2 - +. A.1.b) 42.Kh2 Qxd1 is similar to A.1.a.
A.1.c) 42.Kf(h)1 Qxd1+ followed by 43... Qg4 - + [N+2P vs B]. A.2) 41.Qe2 Nc4 with the threat 42... Ne3 43.Ra1 Rxd3. A.3) 41.Qd2 Qg4+ 42.Kf1 (42.Kh1 Nxe4 43.Bxe4 Rh3+) Ra3 - + [N+2P vs B]. B) 39.Kg1 Bxg2 40.Kxg2 (40.Qxg2 Rxc3 - +) Rxc3 - + (41.Qxc3 Qg4+ is similar to A.1). |
|
Dec-09-10
 | | Phony Benoni: <MHassan: 2r3k1/3b1p2/3n2p1/1p1Pp2p/1P2P2q/2NB3P/1Q
4P1/3R3K b - -
...do not know why the diagram at end of move 44 of my analysis is not produced just the FEN?. Does anybody know?> There's an extra space or a hard return in the line "1Q4P1". When you take that out, the FEN works. 
click for larger view |
|
| Dec-09-10 | | JohnBoy: <Englishman: ... Dziuba's inspired swindle> I don't see any inspired swindle here. All I see is a blown game by black. What are you thinking of? |
|
| Dec-09-10 | | Marmot PFL: Fairly obvious sac but i didn't understand Qg3. if white declines why not just 38...Bd7 with an extra pawn? |
|
| Dec-09-10 | | Toliman: Yes, 39. ..Bd7! |
|
| Dec-09-10 | | Patriot: I couldn't see a forced mate so I noted material is even and didn't see any threats that stand out for white. The move I wanted to play is the obvious 38...Bxh3. 39.gxh3 Qxh3+
A) 40.Qh2 Qxh2+ 41.Kxh2 Rxc3
B) 40.Kg1 Rxc3 41.Qxc3 Qg4+ followed by 42.Qxd1
I had to refine line B since I first looked at 41...Qg3+ 42.Kf1/Kh1 Rxc3 43.Qxc3 Qf3+ and 44...Qxd1. BUT all the checking helps white. 42.Kf1 Rxc3 43.Qxc3 Qf3+ 44.Ke1 saves the rook so 41...Qg4+ is simpler. Having said that, I thought there might be more in the position but it looks winning for black. For example, 38...Bxh3 39.gxh3 Qxh3+ 40.Kg1 Qg4+ 41.Kh2 Rxc3 42.Qxc3 Qxd1 43.Qxc5 Qxd3 44.Qxd6 Qxe4 and white still looks dangerous because of the d5-pawn. I never considered 39.Kg1 because it doesn't look like the move could possibly refute anything. The reason is that in an even material position with no threats for white, it seems to just drop a pawn for nothing. As it turns out, 39...Bxg2! puts black in a similar position as if white accepted the sacrifice. I don't know if 39...Bg4 is truly a blunder or black blundered at another point that caused his demise. I played over the rest of the game quickly but didn't analyze it further. |
|
Dec-09-10
 | | patzer2: In assessing today's Thursday puzzle position (38...?, Black to move), two big weaknesses jump out in White's position: 1. A weakly defended King.
2. Weakly defended pieces, with only one piece protecting most of them. This suggests a demolition of pawn structure (weak king position), deflection (removing the guard or defender) or a double attack (once the guard is removed) might be possible. Well, as it turns out, none of those tactics work by themselves. So, what if we put them together? Well, what do you Know? It works! Beginning with the demolition 38...Bxh3!, White's house of cards falls flat after 39. gxh3 Qxh3+ 40. Kg1 (40. Qh2 Qxh2+ 41. Kxh2 Rxc3) 40...Rxc3! 41. Qxc3 Qg4+ 42. Kf2 Qxd1 with a decisive two pawn advantage. P.S. Haven't run this through the computer yet, but it seems obvious enough. I'll put it in my demolition of pawn structure combination collection(s), since these usually involve multiple tactics. |
|
| Dec-09-10 | | M.Hassan: <Phony Benony>:Thank you for mentioning the reason and opening my eyes! |
|
| Dec-09-10 | | gofer: Before the school run, I looked at Bxh3, f6, f5, Rxc3, Nc4 and discarded them all. After the school run I looked at
Nxe4 and so realised that Rxc3 is actually just what the doctor ordered! But in order to make Nxe4 work we need to
"remove" two defenders Nc3 and Bd3. Nc3 is easy, but what about Bd3? How can we do this?! I then looked at the pin along the 3 rank and realised that after Bxh3 we can actually attack Rd1! So lets start with a bishop sac.
38 ... Bxh6
39 gxh3 Qxh3+ (white has no alternative giving up the pawn is very but probably the best alternative given what follows!!) 40 Kg1 (Qh2 Qxh2 Kxh2 Rxc3 winning) ...
Now we go back to theme number 1 the rook sac to remove the defenders on e4 40 ... Rxc3 (again white has little choice giving up the knight cheaply means that white has lost two pawns for nothing) 41 Qxc3 Qg4+
and now white can see the light! Unfortunately its a freight-train not daylight; a) white has no immediate threats on black's king which can stay safe and sound on g7 for ages. Also Qc8+ is not available yet! and b) The white king needs to move only f2 and h2 are available as f1 and h1 allow Qxd1 without loss of tempo. 42 Kh2 Qxd1
Black gains two key pawns for no loss and though white can launch a counter attack it will have to be something pretty special to
gain back the two pawns and start an attack when white's king is so exposed... ...time to check! |
|
Dec-09-10
 | | kevin86: Two contradicting saying are these:"In for a penny,in for a pound" "Don't throw good money adter bad"---too bad black followed the wrong one. I saw the two sacs at h3 and c3-black took the former,but backed away on the second attempt to sac the bishop at g2. |
|
| Dec-09-10 | | gofer: Dang! I thought about Kg1 but disregarded it thinking that it wasn't winning, but obviously black can hash things up - if they are not careful. Having seen Qg4+ I should have seen Bxg2, but I have to confess I didn't. In response to ideas like 39 Rd2. I think the simple Bf5+ gives black the advantage as black will come out of the sequence two pawns up... 38 ... Bxh3
39 Rd2 Bf5+
40 Kg1 Bxe4 |
|
Dec-09-10
| | patzer2: Well, I missed the followup possibility 39. Kg1 Bxg2! 40. Kxg2 Rxc3 41. Qxc3 Qg4+ 42. Kf2 Qxd1 . Since it's so similar to the original combination, I'm surprised a strong Master missed it. I wonder if time trouble, trying to reach the 40th move, was a problem. I didn't see it in my original calculation, but would hope I would have caught it when I reached that point (39...?) over the board. Having jeopardized the win with 39...Bg4? (instead of 39...Bxg2!), Black gave up his last chance for a strong advantage with 40...Qg3? Instead, 40...Bd7 , protecting the pawn on b5, would have preserved some winning chances. |
|
Dec-09-10
 | | chessgames.com: Please note that there is a choice of move order here: you can play 39...Bxg2! and follow it up with ...Rxc3!, or you play 39...Rxc3! and follow it up with ...Bxg2! Either move order works the same; the computers say that one variation transposes into the other. |
|
| Dec-09-10 | | DarthStapler: I got the first move and the gxh3 variation but I didn't consider Kg1 |
|
| Dec-09-10 | | howlwolf: I didn't consider 39 Kg1 as white's defense but saw quickly that Bxg2 wins. Onischuk must have been in time pressure. |
|
| Dec-09-10 | | WhiteRook48: missed it |
|
Dec-09-10
| | ajk68: I knew there must be a rook-king fork with the queen, but I kept missing Qg4+. Any other series of checks allows the king can make it over to the rook. |
|
| Dec-09-10 | | wals: Found Bxh3 O.K but not the continuation nor the advised move 39...Bxg2. Rybka 4 x 64
Blunders only
depth: 21 : 6 min :
White
(-2.80):38.Rd1. Best, Qd2, -0.06.
depth: 19 : 5 min :
Black
(-1.61):39...Bg4. Best, Bxg2 -2.80.
depth: 22 : 5 min :
Black
(=0.00):42...Rb8. Best, Qe3+, -0.94.
depth: 20 : 5 min :
Black
(+1.41):43...h4. Best, Qe3+, =0.00.
depth: 23 : 6 min :
Black
(+2.81):45...Qxf2. Best, Kh6, +1.73.
depth: 22 : 5 min :
Black
(+4.31):46...f5. Best, Kf6, +2.81.
depth: 25 : 6 min :
White
(+1.86):48.Rh1. Best, Ra1, +4.27.
depth: 25 : 5 min :
Black
(+6.05):48...h3. Best, f4, +1.86.
depth: 20 : 5 min :
White
(+3.75):50.d6. Best, Ra1, +6.60.
depth: 23 : 6 min :
Black
(+10.73):54...Kc7. Best, f4, +3.87.
depth: 19 : 5 min :
Black
(+30.85):56...Rh8. Best, Bxb5, +10.52.
depth: 13 : 3 min :
Black
(+#12):57...Rxh5. Best, Rd8, +16.74.
depth: 16 : 25 min :
58.b6+
1.  (#10): 58...Kxc6 59.d8Q[] Rh2+ 60.Kg3[] Re2 61.b7 Re3+ 62.Kf2[] Re2+ 63.Kf3[] e4+ 64.Kf4[] Rf2+ 65.Ke5[] Rb2 66.Rc7+ Kb5 67.b8Q+ Ka6 68.Ra7# 2. (#9): 58...Kd6 59.d8Q+ Kxc6 60.b7 Rh2+ 61.Kg3 Rg2+ 62.Kxg2 Bd5+ 63.Kg1 Kc5 64.b8Q Kd4 65.Qxe5+ Kd3 66.Rc7 f4 67.Qdxd5# |
|
Dec-09-10
| | Phony Benoni: <chessgames.com: Please note that there is a choice of move order here: you can play 39...Bxg2! and follow it up with ...Rxc3!, or you play 39...Rxc3! and follow it up with ...Bxg2! Either move order works the same; the computers say that one variation transposes into the other.> 
click for larger viewThat's very interesting. In my opinion, from the human point of view 39...Bxg2 is the clearer move, even though they are equivalent in Silicon Valley. The reason is that 39...Bxg2 followed by ...Rxc3 leaves White fewer choices, hence fewer lines for Black to calculate. After each move in this sequence, White has only the choice between capturing or being material down. After 39...Rxc3 40.Qxc3 Bxg2 material is roughly even, so White is not forced to take the bishop. Computer analysis shows that none of his extra options are any good, but for the human this would represent an extra burden of calculation and additional chances for error. |
|
| Dec-09-10 | | Pawnage: <Phony Benoni> No, Bxg2 is clearer here, too. ;) |
|
| Dec-09-10 | | rusticbull: this puzzle looks not so appealing |
|
 |
|
|
< Earlier Kibitzing · PAGE 3 OF 3 ·
Later Kibitzing> |
Kibitzing
