< Earlier Kibitzing · PAGE 37 OF 37 ·
|Feb-17-11|| ||Penguincw: < NGambit: <Penguincw:It's the only game for the match with a sacrifice (a winning one).>
Couldn't be more wrong. Anand won Game 4 with a beautiful knight sac. > What I meant was a "chessgames.com sacrifice".|
|May-18-11|| ||qqdos: In Chess Informator 109 this game was voted the Best Game of Ch. Inf. 108 and 2nd Best Novelty (D87) behind Anand's win in the 4th Game of the match (E04). Adrian Mikhalchishin gave Topalov 10 points in both categories!|
|Jun-07-11|| ||DrMAL: When watching this game I was wondering why black had not played Bb7 earlier, such as 16.Bb7 instead of 16.Qd6 and when move 23 came around I was still thinking 23...Bb7 would be best with 23...Bd7 second. I thought 23...Kf7 was dubious and 24...Kxf6 instead of 24...Qxf6 seemed odd as well.|
Deep analysis by Rybka 4.1 below shows 23...Bd7 as best with 23...Bb7 as second. Perhaps this is why Vishy had not played Bb7 earlier, since he stated he prepared 23...Bd7 before the game but got move order backwards. I was stuck on 23.Bb7 as best because I thought it should have been played earlier. Interestingly, Rybka evalulates 23...Kf7 as black's third best:
[-0.21] d=21 23...Bd7 24.Bc4+ Nxc4 25.Rxc4 Rac8 26.Qxd4 Qxd4+ 27.Rxd4 Be8 28.Ng3 Rc1+ 29.Rf1 Rc2 30.Rf2 Rc5 31.Rfd2 Kf7 32.Nh5 Rc6 33.g4 Rh8 34.Kg2 Rh6 35.Kg3 Rc3+ (3:23:50) 1422817kN
[+0.70] d=21 23...Bb7 24.Nxf6+ Qxf6 25.Rg3 Rac8 26.Rxg5+ Kf7 27.Be2 d3 28.Bh5+ Ke7 29.e5 Rxc1+ 30.Qxc1 Qc6 31.Qa3+ Qc5+ 32.Qxc5+ bxc5 33.Rg7+ Kd8 34.e6 Nc6 35.Rxb7 Rxf5 36.Rd7+ Kc8 37.Bg4 Rg5 38.Bh3 Re5 (4:55:50) 2250310kN
[+2.02] d=21 23...Kf7 24.Rh3 Rh8 25.Bc4+ Nxc4 26.Rxc4 Bb7 27.Rxd4 Rae8 28.a4 Bc8 29.g4 a6 30.Rb3 b5 31.axb5 axb5 32.Qd3 Rxh5 33.gxh5 Qf4 34.Qxb5 Qg4+ 35.Kh2 Qh4+ 36.Kg2 Qxh5 37.Qc4+ Kg7 38.Rg3 Qh4 (7:02:49) 3132190kN
At the end of the above line for 23...Kf7 white is clearly better but the position does not look hopeless. After 23...Kf7 Rybka evaluates 24.Rh3 and 24.Nxf6 as nearly identical where white played the latter. Black choice of 24...Kxf6 instead of 24...Qxf6 was indeed a second mistake as Rybka shows:
[+2.48] d=20 24...Qxf6 25.Bb5 Qe7 26.Rh3 Kg8 27.Rh5 Bxf5 28.Rxg5+ Kh8 29.Qxd4+ Qf6 30.Qxf6+ Rxf6 31.exf5 a6 32.Rh5+ Kg7 33.Rc3 Rff8 34.Rg3+ Kf6 35.Rg6+ Ke5 36.Re6+ Kd4 37.Bf1 b5 38.Rh4+ Kc5 39.g4 Rfe8 (0:50:24) 445464kN
[+3.75] d=20 24...Kxf6 25.Rh3 Bd7 26.Rh6+ Kf7 27.Qxg5 Rg8 28.Rh7+ Rg7 29.Rc7 Kg8 30.Rxg7+ Qxg7 31.Qxg7+ Kxg7 32.Rxd7+ Kf8 33.Rxd4 Nc6 34.Rd6 Ne5 35.Kf2 Nf7 36.Rf6 Kg7 37.Re6 Rc8 38.Ke3 Kf8 39.e5 Rc5 (1:06:19) 576799kN
While computers are not by any means an end-all in chess at least they help gain better post-game understanding. It seems that 23...Kf7 was not yet decisive until 24...Kxf6 was played, forming a compound error often seen in chess. Once one dubious move is made a player often realizes it soon afterwards, gets confused or upset, and makes more mistakes.
Looking earlier in the game, Rybka does slightly favor 16...Bd7 over 16...Qd6 but I suppose the difference is negligible and a matter of taste:
[-0.10] d=20 16...Bb7 17.f4 Rc8 18.Rxc8 Bxc8 19.Qb2 f6 20.Nxd4 Qd6 21.Be2 Re8 22.e5 fxe5 23.Nf3 Qc5+ 24.Kh1 Nc4 25.Qa1 Kg8 26.Rc1 b5 27.Nxe5 Qf2 28.Bxc4+ bxc4 29.Rf1 Qc5 30.Qc3 Rd8 31.Qxc4+ (0:21:41) 142460kN
[+0.00] d=20 16...Qd6 17.f4 f6 18.f5 Bd7 19.Ba6 Rad8 20.Nxd4 b5 21.Qxa5 Qxd4+ 22.Rf2 Qxe4 23.fxg6 hxg6 24.Rd1 Qc6 25.Rfd2 Bg4 26.Rxd8 Rxd8 27.Qxd8 Bxd1 28.Qxd1 Qxa6 29.Qd7+ Kg8 30.Qe8+ Kh7 31.Qd7+ Kh6 (0:23:49) 156035kN
However, note in this latter line 18...Bd7 is recommended over 18...Qe5 that black played, in case someone wants to analyze further. Either way, it seems clear that black delayed developing his light square bishop to d7 or b7 entirely too long.
|Jul-13-11|| ||qqdos: <DrMAL> see Podolchenko vs Howells, 2010 Olympiad where Black played 16...Bb7 in an entertaining game and lost. Someone described the result as "Podolchenko's Immortal"!|
|Jul-14-11|| ||DrMAL: Thanx I will examine that game and, if I have anything useful, will comment there. Doubt 16...Bb7 had much to do with the loss, it still seems intuitively (and, according to Rybka analytically) best. But the resulting mistake that must have occurred in that "Immortal" should be very interesting, cheers.|
|Jul-14-11|| ||DrMAL: I wanted to return to this game because the quandary about move 16 (and Vishy's explanation that he purposely planned later Bd7) still interested me.|
Checking with another engine, 16...Bb7 is still preferred:
click for larger view
Houdini_15a_x64: 27/66 32:39 8,139,953,578
+0.11 16. ... Bb7 17.f4 Rc8 18.Rxc8 Bxc8 19.Qb2 f6 20.Nxd4 Qd6
+0.00 16. ... Qd6 17.f4 f6 18.f5 Qe5 19.Rf3 g5 20.Rg3 Bd7
Moreover, in the sharper line with 16...Qd6 Houdini (like Rybka above) recommends Bd7 much earlier than on move 23.
It was clear WC Vishy thought his counterattacking plan starting with 16...Qd6 was best and I am not about to argue with him LOL. I actually agree with him, as 16...Bb7 albeit safer leads to a somewhat drawish position. My effort in questioning it was directed to better learn the benefit versus danger of such attacking plans, I tend to play safer and am trying to improve.
The position after 23.Rf3 demands extreme accuracy. My initial hangup on 23...Bb7 would have probably lost at this level, it was more of a positional move not accurate enough for the tactical sharpness here.
On move 23 something obviously had to be done about white's development (the bishop!). So it is difficult for me to accept Vishy's explanation that, after all the preparation into entering such a position, he forgot to make at least one of two totally obvious choices (23...Bd7 or 23...Bb7), but we all make mistakes
But also clear was Vishy's idea behind 23...Kf6 of getting his king away from black's knight and at the same time opening the way for his rook(s). If Vishy had played 23...Bd7 Topalov may not have even won let alone get some big prize for this game. As shown in my first post after 23.Rh3 white does not have any sort of advantage, possibly the other way around.
I will look at an alternative game Evgeniy Podolchenko vs D Howell, 2010 where black played 16...Bb7 to see what made it "Immortal" LOL, cheers.
|Mar-22-13|| ||Conrad93: Everything seemed to go wrong right after 19...g5?|
|Apr-13-13|| ||Oxnard: <garrykasparov> 30. ... Kxc8 31. Qc1+ Nc6 32. Bxc6 Qe3+ 33. Qxe3 dxe3 34. Bxa8 and won endgame. There is no mate.|
|May-26-13|| ||SamAtoms1980: I remember watching this as it unfolded live, at a time of day when I should not have been awake (heheh).|
24.Nxf6 Kxf6 (24....Qxf6 25.Rc7+) 25.Rh3 Rg8 26.Rh6+ Kf7 27.Rh7+ Ke8 28.Rcc7 was as far as I remembered, after 28....Kd8 29.Bb5 with the decisive threat of 30.Rxc8+ and 30.Rcd7+.
|May-26-13|| ||M.Hassan: "Insane"
White to play 24.?
White is a pawn down
White gains a pawn
Again White gains a pawn.
I can only say that White can win because his pawns has more space
Time to check
|May-26-13|| ||bachbeet: didn't come close to solving this.|
|May-26-13|| ||offramp: Superb photo. High drama!|
|May-26-13|| ||lost in space: 24. Nxf6! My memory is still working; 2 years are not enough to forget such a hammer move.|
|May-26-13|| ||Patriot: The only candidate I see that "might" win is 24.Nxf6.|
24...Qxf6 25.Rc7+ Ke8 26.Bb5+ Kd8 27.Qc1 is unclear.
24...Kxf6 25.Rh3 looks strong, weakening g5.
This looks tough!
|May-26-13|| ||howlwolf: I came up with Nxf6 and Rh3 very quickly so the game may have been lingering in the recesses of my memory. Glad something worthwhile is stuck in there.|
|May-26-13|| ||Xeroxx: I remember this game.|
|May-26-13|| ||Jimfromprovidence: This is not a game for mere mortals.
There are so many great concepts at work here, after 24 Nxf6, below, but I could not get there; to me these threats do not reveal themselves immediately.
click for larger view
Now, after 24 Nxf6, above, white threatens 25 Ng4 and then 26 Nh6+.
click for larger view
After the text 24...Kxf6 25 Rh3, above, white's rooks control both the c and h files plus he has connected passed pawns.
He also as threats of putting both rooks on the 7th rank as well as Qxg5.
I kind of liked 25...Qf4, below as a response.
click for larger view
It does not work but it's a nice side exercise to figure out why not, for those inclined to take a look at it.
|May-26-13|| ||Patriot: <Jimfromprovidence> I figured out some of it and double-checked with Houdini. 26.e5+ Kxe5 27.Re1+ Kd5 28.Be4+ etc. I thought 26...Kxe5 may be best so I pressed on with that. And 27...Kd5 seemed best.|
I looked at 26.Rh6+ first and it just didn't feel right. 26.e5+ threatens to tighten the wrench via a strong pawn roller. I'm not sure how you try to calculate with so many options, but what I do is "best guess" my way through a line and then try to punch holes in it by working my way backwards.
I don't see how any player could calculate all the twists and turns of positions like this--especially during a game. 25...Qf4 is a reasonable response--it's quiet and is a nice attempt to neutralize white.
|May-26-13|| ||agb2002: White is a pawn down.
The black queen protects the square c7, the pawn on f6 protects g5 and the black rooks are still disconnected. These details suggest 24.Nxf6, weakening the black king's shield or trying to divert the black queen:
A) 24... Kxf6 25.Rh3
A.1) 25... Qf4 26.Rh6+
A.1.a) 26... Ke5 27.Qxa5+ bxa5 28.Rc5#.
A.1.b) 26... Ke7 27.Qb4+
A.1.b.i) 27... Kd8 28.Qxf8+ Kd7 29.Rh7#.
A.1.b.ii) 27... Kd7 28.Rh7+ and mate soon.
A.1.b.iii) 27... Ke8 28.Bb5+ Bd7 (28... Kf7 29.Rf6+ and mate in three) 29.Re6+ Kf7 30.Qe7+ and mate in two.
A.1.b.iv) 27... Kf7 28.Rf6+ and mate in three.
A.1.c) 26... Kf7 27.Rc7+
A.1.c.i) 27... Qxc7 28.Rh7+ and 29.Rxc7 + - [Q vs R+N], still with a winning attack.
A.1.c.ii) 27... Bd7 28.Rxd7+
A.1.c.ii.1) 28... Ke8 29.Bb5 Rg8 (29... Qxd2 30.Re6#) 30.Re6+ Kf8 31.Qb4+ and mate next.
A.1.c.ii.2) 28... Kg8 29.Rhh7 Rfd8 (29... Qxd2 30.Rcg7#) 30.Qxf4 gxf4 31.Rdg7+ Kf8 32.f6 + -.
A.1.c.iii) 27... Ke8 28.Qc2 Bd7 29.Rhh7 Rd8 30.Bb5 seems to win.
A.1.c.iv) 27... Kg8 28.Qxf4 gxf4 29.Rhh7 Rd8 30.Rcg7+ Kf8 31.f6 + -.
A.1.d) 26... Kg7 27.Rg6+ Kh8 (27... Kh7 28.Qxf4 gxf4 29.Kf2 + -; 27... Kf7 28.Rc7+ is similar to previous lines) 28.Qd1 Qh4 29.Bb5 Rd8 30.Rc3 g4 31.Rxg4 Qxg4 32.Qxg4 dxc3 33.Qh4+ Kg7 34.Qxd8 + -.
A.2) 25... Kg7 26.Qxg5+ wins.
A.3) 25... Rg8 26.Rh6+ Kf7 (26... Ke7 27.Qb4+ seems to make things easier for White) 27.Qe2 with the threats Qh5+ and Bc4+.
B) 24... Qxf6 25.Rc7+
B.1) 25... Kg8 26.e5 Qd8 (26... Qxe5 27.Qxg5+ and mate in two) 27.f6 Qxc7 28.Qxg5+ Kf7 (28... Kh8 29.Rh3+ Qh7 30.Qg7#) 29.Qg7+ and 30.Qxc7.
B.2) 25... Ke8 26.Bb5+ Kd8 27.e5 Qxe5 28.Rh7 with the threats Qxg5+ and Rd3.
Not very clear in many lines but this is all I can do today.
|May-26-13|| ||PinnedPiece: I don't understand why after 30.Kxc8 the game is resignable.|
<garrykasparov>: ...30....Kxc8 31.Qc2+ followed by Qc7# or ...30.Rxc8 31.Rd7+ ...31.Ke8 32.Rxd4+ followed by Rxe4.
Umm, make that 31.Qc1+ I do believe. If that is the scenario, Black will be able to trade queens, and have two Rooks against a Rook and a Bishop.
|May-26-13|| ||PinnedPiece: Ahhh....Bxa8 is also in the cards....
Topalov was at the top of his love of the game during this stint, to play with this nerve.
|May-27-13|| ||pittpanther: After 29 Bb5 why can't black play 29 a6 in reply? If white plays 30 Rxc8 black can play Kxc8 and the queen covers the c7 square. I assume I am missing a simple response.|
|May-27-13|| ||Nerwal: <After 29 Bb5 why can't black play 29 a6 in reply? If white plays 30 Rxc8 black can play Kxc8 and the queen covers the c7 square. I assume I am missing a simple response.>|
It's not that simple. After 29. ♗b5 one of white's idea is to remove the guard of the d4 pawn, so after 29. ♗b5 a6, 30. ♖ce7 hitting the ♕e5 is the correct move, but the tactical details behind it require a good amount of precision...
|Nov-03-13|| ||Ulhumbrus: An alternative to 16...Qd6 is 16...Nb7 heading for c5|
|Nov-12-13|| ||thedoctor46: Ninja moves...24.Nxf6|
< Earlier Kibitzing · PAGE 37 OF 37 ·