< Earlier Kibitzing · PAGE 2 OF 2 ·
|Apr-11-09|| ||Manic: <al wazir> Although I can't give you complete analysis in your variation with 32...Qa1 not 35...Bxc2+ but 35...Bf3#. In fact, perhaps 31...Qh5 immediately saves black (thanks for <cyclon> for putting that idea in my head and pointing it out and perhaps what <AdrianP> was talking about. Sorry for ruining it but it was already stated before me).|
|Apr-11-09|| ||RobertLangdon: I missed 32.hxg7.I thought there is nothing better then a draw for white.So I was satisfied with 32.Bxf7 Rxf7(if ...Kh8 33.hxg7+ Kxg7 34.Qg6+ Kh8 35.Qh6#) 33.Qe8+ Kh7 34.Qxf7 Qa1 35.Qf5+ Kg8 36.Qe6+ and perpetual.|
|Apr-11-09|| ||butilikefur: 35. Qf6 looks much stronger - the best reply is probably 35...Qxe5 36. Qxe5 Bd3+ 37. Kg2 Rxf7 and 38. Qe6 is winning (for example, 38...Kg7 39. d7 Bf5 40. Qxf7+). If 35...Rxf7 36. Qxf7+ Kh6 then 37. Bf4+ drops the queen. 35...Bd3+ 36. Kg2 Be4+ 37. Kg3 Qa3+ 38. Kh4 39. Qh3+ Kg5 and Black cannot stop the mate.|
|Apr-11-09|| ||Once: <al wazir> I went the same way as you, with a fond dream of something like 31. Bxc4 Bxd1 32. hg|
But Fritzie shows that it doesn't work. After 31. Bxc4 Qh5! spoils the fun.
click for larger view
White is busted. Black threatens Qh1# and Qxh6, as well as adding more preotection to the f7 square. Herr Fritz says -4.1 and who am I to argue? (I always imagine Fritz talking like Arnie in his terminator days, but maybe that's just me).
But I think with 31. Bxc4 we had the right idea, even if we got the wrong continuation. The Rd1 is relatively unimportant in this position - what matters is hammering away at the pinned Rf7. 31. Bg6 is the way to do this, since it has the side benefit of keeping the black queen out of h5.
I just wish I had spotted it! Kudos to anyone who works this one out OTB and without T1000 assistance today.
|Apr-11-09|| ||apoka: When I first looked at the puzzle, White did not look very strong with the bishop and rook under attack and only very limited king safety. I found hxg7, but missed the refutation Qh5 (which is probably the only move saving Black). I did not really have time to look in detail at other candidate moves like Bxc4, Be5, Bg6, h7+, d7. What is interesting, is that in a real game we would be more likely to spot Qh5 and the dangerous attack coming with it. In a puzzle, you tend to pay less attention to threads by your opponent (apart from obvious one move threads).|
|Apr-11-09|| ||chesskingqph: It seems that Black blundered with 20...fxg4. Shredder suggests 20...Rdf7 instead. The pawn e6 is now protected, White still needs to move its g4 rook somewhere in the g pile, so 21.Rg6 seems to be the better move compared to 21.Rg5. Black then play 21...f4 to prepare to take the pawn d6, the pawn is lost now; 22.Bh4 Qxd6 then it seems balanced. It is hard for White to continue its attack on the kingsize while Black has chances to counter on both sizes. What do you think?|
|Apr-11-09|| ||PolishPentium: After the text moves, why not 35...Bd3+ for Black? Seems to lead to further checks that may in fact win...|
|Apr-11-09|| ||butilikefur: Hey, 35. Bg6+ might win on the spot.
35...Bxg6 36. Qe7+ Kh6 (36...Rf7 37. Qh4+ Kg8 38. Qh8+ mate; 36...Bf7 37. Qh4+ Kg6 38. Qxg4+ Kh6 39. Qf4+ Kh7 40. Qf5+ Bg6 [40...Kh6 41. Bf4+ wins the queen, as with 39...Kh5 40. Qf5+] 41. Qh3+ Kg8 42. Qh8+ Kf7 43. Qf6+ followed by mate)
37. Qxf8+ Kh5 (37...Kg5 38. Bf4+ Kh4 39. Qf6+ Kh3 40. Qxg5 Qa8+ 41. Ke2 and White should be won)
38. Qh8+ Kg5 39. Bf6+ (here 39. Qf6+ looks like a bad choice after 39...Kh5 40. Bf4 Qf5 [40...Bd3+ 41. Kg2 and Be4+ and Qd5+ and both useless, followed by Kh2 and Kg1 respectively] 41. Qh8+ Bh7 42. Qg7 Qb1+ [not 42...Qxf4 43. Qxh7+ Kg5 44. Qg7+ Kh4 45. Qe7+ Kh5] 43. Kg2 [44.Ke2 Bd3+ 45. Kd2 Qc2+ 46. Ke3 Qe2+ 47. Kd4 Qe4+ 48. Kc5 Qxf4] 43...Be4+ 44. f3 Bxf3+ 45. Kf2 Qc2+ 46. Bd2 [46. Ke3 Qd3+ 47. Kf2 Qe2+ 48. Kg1 Qg2+ mate] 46...Qxd2+ 47. Kg3 Qxd6+)
39...Kf5 (39...Kf4 40. Qh6+ Kf5 41. Qg5+ wins the queen, 40...Kf3 41. Qe3+ is mate) 40. d7 and Black has no checks with White clearly winning.
|Apr-11-09|| ||OhioChessFan: Whew. I had an easy time of it this week until today. Seting up the shot Bxc2...very impressive.|
|Apr-11-09|| ||piever: Ouch, after 31. hxg7 I considered 31..Bxd1 to be the best black move. It isn't, but I wonder how many player with black would play the right move here... It's a very nice puzzle after 31. hxg7: black to move and draw.|
Really impressive game by Shirov!
|Apr-11-09|| ||johnlspouge: Saturday (Very Difficult)
Shirov vs Movsesian, 2005 (31.?)
White to play and win.
Material: B for R. The Black Kg8 has 1 legal move. The White Qe6 pins Rf7 to Kg8. Black has 2 principle threats: …cxd3 and …Bxd1. The Black battery Rf8 and Rf7 are aimed at Kf1, obscured by the Black Bf3 and the White Pf2. The White Bd3 can discover support by Rd1 for Pd6 or support Ph6-h7+. The Black Qa5 is loose, but eyes e1 and d8, the queening square of Pd6. The White Kf8 is vulnerable to Qa5-e1+, Bf3-e2+, and Bf3-g2+, but is in no immediate danger.
Candidates (31.): h7+, hxg7, Bxc4, Bg6, d7
I went for 31.hxg7, missing that Black had an adequate defense in
31…Qh5 32.gxf8=Q+ Kxf8
|Apr-11-09|| ||karnak64: What a wild game. Beyond my depth, but fun to play over starting around move 19.|
|Apr-11-09|| ||CHESSTTCAMPS: This must be a desperate situation - white is down an exchange and has a rook and a bishop en prise. But this is Shirov and this is expected! In fact, a sober assessment of the situation reveals a big advantage to white. White's magnificently placed queen is the boss, essentially tying down two rooks. White's goal must be to break down the illusory rook shelter and get to the black king. The rook on d1 can be a greek gift - clearly it's the speed of the attack that matters. I looked at three candidates: Bxc4, hxg7, and Bg6. When I looked at 31. hxg7, I saw that 31...Qh5 might be troublesome. In fact, there are compelling tactical reasons for Bg6 other than preventing Qh5, but it is interesting to note that with no entry points for the black queen, white does not need to worry about a counterattack. |
When I focused on 31.Bg6, the calculation clarified:
A. 31...Bxd1 32.hxg7! Kxg7 (forced) 33.Bxf7 Rxf7 34.Be5+ Kg8 (Kf8 35.Qc8+ forces mate next) 35.Qe8+ Kh7 (Rf8 36.Qg6#) 36.Qxf7+ Kh6 37.Bg7+ Kg5 (Kh7 38.Bf8+) 38.Qf6+ Kh5 39.Qh6#
A.1 33....Qg5 (to prevent Qg6+) 34.Be5+ Kh7 35.Bf6 Qf4 (Qh6 36.Qf5+ mates next) 36 Bg8+! Rxg8 37.Qf7+ Kh6 38.Bg7+ Kg5 39.Bh6+ wins the queen with check.
A.2 34...Qxe5 (to prevent the mate shown in the main line of A) 35.Qxe5+ Kg8 (Kf8 36.Qh8#) 36.Qd5 Bxc2 37.d7 wins the rook and the game.
B. 34...gxh6 35.Bxf7+ Rxf7 36.d7 Bxd1 37.Bh4 forces promotion.
C. 34...Qg5 35.hxg7 Kxg7 36.Bxf7 Rxf7 (Bd1 37.Be5+ transposes to A.1) 37.Be5+ Kg8 (Kf8 38.Qc8+ mates next) 38.d7 Bxd1 39.Bf6 Qxf6 39.Qxf6 Rxd7 40.Qe6+ Rf7 41.Qxc4 The ending is tricky, but the c-pawn should decide.
There's probably something better in line C, but I can't seem to find it.
Time to check.
|Apr-11-09|| ||WhiteRook48: I chose 31. hxg7, not thinking to sac a rook. Who would without a computer engine?|
|Apr-11-09|| ||CHESSTTCAMPS: Oops - I messed up the move numbering. Lines B and C should begin at black's move 31. It also seems that I missed 33... Bxc2 played in the game.|
|Apr-11-09|| ||kevin86: What a b rilliant finish! Who ever would have dreamed that the bishop at c2 would have been captured after a skewer at h7!!|
|Apr-11-09|| ||agb2002: Black is an exchange up and threatens 31... cxd3, 31... Bxd1 and 31... Qh5-Qh1#. However, White's queen, bishops and pawns on d6 and h6 may create counter-chances while meeting some threats, for example, the mate on h1. This suggests, 31.Be5 and 31.Bg6. In the case of 31.Be5:|
A) 31... Bxd1 32.hxg7
A.1) 32... cxd3 33.gxf8=Q+ Kxf8 34.Qc8+ and mate next.
A.2) 32... Rd8 33.Bg6 Qxe5 (33... Rdd7 34.Qe8+) 34.Qxf7#.
A.3) 32... Qxe5 33.gxf8=Q+ Kxf8 34.Qxe5 + - (34... cxd3 35.Qh8#).
B) 31... cxd3 32.hxg7
B.1) 32... Bxd1 transposes to A.1.
B.2) 32... Rd8 33.Qh6 (33.d7 Qb5) Rfd7 34.Qh8+ (34.Qe6+ seems to draw) Kf7 35.Qh7 Ke6 (35... Rg8 36.Qf5+ Ke8 37.Qe6+ and mate next) 36.g8=Q+ Rxg8 37.Qxg8+ Kf5 (37... Kxe5 38.Qg5+) and the king seems to escape.
C) 31... Bd5 32.Bh7+ Kxh7 33.Rxd5 Rxf2+ 34.Kg1 Qb6 doesn't look good.
In the case of 31.Bg6, avoiding 31... cxd3 and 31... Qh5 and threatening 32.hxg7 Kxg7 33.Bxf7 Rxf7 34.Be5+:
A) 31... gxh7 32.Bxf7+ Rxf7 33.Be5 (33.Rb1 Qh5 34.Rb8+ Kh7)
A.1) 33... Bxd1 34.Qe8+ and mate next.
A.2) 33... Qd8 34.Qg6+ Kf8 35.Re1 and the black king won't survive.
A.3) 33... Qb5 34.Qg6+ Kf8 35.Re1 c3+ 36.Kg1 is similar to A.2.
B) 31... Bxd1 32.hxg7 Kxg7 33.Bxf7 Rxf7 34.Be5+ Kg8 (34... Kf8 35.Qc8+) 35.Qe8+ Kh7 36.Qxf7+ Kh6 37.Bf4+ Qg5 38.Qf6+ wins.
Time to post and check.
|Apr-11-09|| ||agb2002: It seems I spent too much energy in 31.Be5 because I didn't even see 33... Bxc2. Better luck tomorrow.|
|Apr-11-09|| ||Jimfromprovidence: Here is the position after 29 h6. Notice that if black can safely get a rook over to e8, he can pin white’s queen.
click for larger view
So, 29….Bf3+ was the losing move. It wasted a tempo and gave the king a safe square at f1. Needed was 29…Qd8, threatening Re8, pinning white’s queen.
click for larger view
Now 30 hxg7 is a losing move for white after 30...Re8.
If white now follows with 30 Bg6, it looks like white can win no more than a pawn out of the exchange.
|Apr-11-09|| ||al wazir: <Manic, Once>: Yes, 31...Qh5 not only refutes my analysis, it makes it pointless. Thanks.|
|Apr-11-09|| ||SamAtoms1980: Well, this wasn't a good day for me at all. I looked at this position for a quarter of an hour and the best I could come up with was <CENSORED FOR FAILING RYBKA BLUNDERCHECK>.|
|Apr-11-09|| ||Lord Osiris: PolishPentium
36.Kg2...Be4+ (blacks only check)
37.Kg3...Qa3+ (again blacks only check)
38.Kh4...Qh3+ (and again the only check)
39.Kg5...Qh6+ (need I say it again?)
40.Kxg4 and black has no more checks left and mate is coming or loss of material equal to game over.
|Apr-12-09|| ||TheBish: Shirov vs Movsesian, 2005|
White to play (31.?) "Very Difficult"
Leave it to Shirov to sort out the complexities here!
After much deliberation, I came to the conclusion that the solution must be
31. Bxc4! and after 31...Bxd1 not 32. Qg6? (threatening mate) Qa1! and Black simultaneously prevents the mate and threatens one of his own!
Instead, the key is 32. hxg7! forcing 32...Kxg7 and then 33. Be5+ wins after 33...Rf6 (or 33...Kh7 34. Bd3+ Rf5 (forced) 35. Bxf5+ and a quick win) 34. Bxf6+ Rxf6 35. Qe7+ Kg6 36. Bd3+ Rf5 (or 36...Kg5 37. Kg7+ wins the rook) 37. Qe6+ Kg5 38. Qxf5+ Qxf5 39. Bxf5 Kf6 (or 39...Kxf5?? 40. d7 and the pawn will queen) 40. c4 and White is winning the endgame, although there is still a little work left. Black will have to abondon the g4 pawn to stop White's two connected pawns, after which White will have three pawns against one and an easy win.
I'm sure I'm missing something here, maybe a quicker win, but it's late in the game already (already Sunday EDT, but still Saturday here in the west), and time to see the game and move on to Sunday's puzzle.
|Apr-12-09|| ||SetNoEscapeOn: They should have a separate category for these tests: 7 stars for those originating on Planet Shirov.|
|Apr-12-09|| ||patzer2: For the Saturday April 11, 2009 puzzle solution, White undermines the helpless King position with 31. Bg6!!|
< Earlier Kibitzing · PAGE 2 OF 2 ·