|
< Earlier Kibitzing · PAGE 4 OF 4 ·
Later Kibitzing> |
| Nov-18-10 | | Patriot: <kingfu> <Why is it a principle to go for two bishops?> Having the "bishop pair" (one side has both bishops, while the other does not) is worth a ~1/2 pawn advantage on average, according to research by GM Larry Kaufman. This depends on other factors such as pawn structure. Both bishops control both color complexes, which is the basic reason behind it. Whenever an enemy knight can be moved that results in losing the bishop pair...this is generally regarded as a threat and often avoided by strong players. |
|
| Nov-18-10 | | Stormbringer: Did anyone throw some silicon at this? I'd be interested to see what the actual best defense is, since it looks like Black fumbled. |
|
Nov-18-10
 | | kevin86: White's 36th move reminds me of a manoever in checkers when a player "forks" the opposing player by interposing a piece between two adverse pieces. (make believe that the bishops are checkers):
click for larger view/. Now interpose a checker to e4. |
|
| Nov-18-10 | | trguitar: I'm not so sure this is a win for White. After 33. Bxb7 Nxb7 34. a6 Bb8 35. Bb6 b4 how does White proceed? This stops the threat of Bc7. It seems to me that the pawn will not be so easy to promote. |
|
| Nov-18-10 | | bengalcat47: kevin86 The checkers maneuver you are referring to is known as "the breeches." It has occured rather often even in master checkers play.
As for the game continuation this gamr reminds me of Tartakower vs. Domenech, 1934. Tartakower uses a similar Bishop sacrifice to force Pawn promotion on the Queen side. |
|
Nov-18-10
 | | Jimfromprovidence: Just a follow up to the earlier discussion of what is the optimal line for black after 35 Bb6. It has to be 35...b4 (by not allowing 36 b4 as discussed by several kibitzers). Specifically, this is what happens if 35…f6?! is played. 36 b4 follows, (preventing …Nc5 and allowing Bc7). Now, if 36…Kf7, then 37 Bc7 Nc5 38 bxc5 Ba7 39 c6 looks very good for white. 
click for larger view |
|
Nov-18-10
 | | agb2002: White has the bishop pair for a bishop and a knight. Black threatens ... Nxe4 and ... Bd2, tying the bishop to the defense of the a-pawn. The pawn on b7 stops the a-pawn. This suggests 33.Bxb7 Nxb7 (otherwise drops a pawn) 34.a6 Bb8 35.Bb6 (35.Bc7 Nc5 (35... Ba7 36.axb7 + -) 36.Bxb8 Nxa6 37.Bd6 f5 38.Kf1 Kf7 39.Ke2 Ke6 40.Bf8 Kf7 41.Bd6 Ke6 repeats moves; 35.axb7 Kf8 36.Kf1 (36.Bh4 f5 37.Bg3 f4) Ke8 37.Ba5 Kd7 38.Ke2 Kc6 39.Bb4 Be5 40.b3 Kxb7): A) 35... Nd6 36.Bc7 Ba7 37.Bxd6 f5 38.b4 Kf7 39.Bc5 Bb8 (39... Bxc5 40.bxc5 b4 41.a7 b3 42.a8=Q b2 43.Qa2+ + -) 40.a7 Bxa7 41.Bxa7 + - [B]. B) 35... Kf8 36.f4 Ke7 (36... Bxf4 37.Ba7 + -) 37.Kg2 Kd7 38.Kf3 Kc6 39.Bd4 B.1) 39... f6 40.Kg4 Kxb7 (40... g6 41.Bxf6 and 42.Bg7 + -) 41.Kf5 and 42.Kg6 + -. B.2) 39... g6 40.Be5
B.2.a) 40... Kxb7 41.Bxb8 Kxb8 42.Ke4 Kc7 (42... f6 43.Kd5 + -) 43.Ke5 Kd7 44.Kd5 winning the b-pawn and the game. B.2.b) 40... Bxe5 41.fxe5 Kxb7 42.Ke4 Kc6 43.Kd4 and White seems to have the better ending. |
|
Nov-18-10
| | agb2002: My line B is completely wrong due to a peculiar hallucination: White played axb7 (omitted) after Kf8 and Black didn't reply! |
|
| Nov-18-10 | | David2009: 
click for larger view
<Jimfromprovidence: Just a follow up to the earlier discussion of what is the optimal line for black after 35 Bb6. It has to be 35...b4 (by not allowing 36 b4 as discussed by several kibitzers). > Crafty EGT disagrees and prefers activating its King safely with 35...Kh7. If 36.b4 Kg6 37.Bc7 Bxc7 38.a7 Be5 39.a8=Q Nd6 40.Qd5 Kf5 and we reach the diagram above. How does White propose to break down the EGT's fortress? Not easy! The Crafty EGT link is in my previous post: Kasparov vs A Georgiadis, 1996 |
|
| Nov-18-10 | | YCP: Someone please explain why is 35. Bb6 such a good move? |
|
| Nov-18-10 | | muralman: This one was an easy one for me. Or, I should say Black made it easy. The drama obviously revolved around the A pawn. I could see that a successful promotion of the A pawn would supersede the survival of either of my minor pieces. So, I proceeded forth. Sacrificing my knight was a given. That allowed my A pawn to move forward menacing the knight. The black Night is lost wherever he moves. Move he must. I just didn't see the black Bishop allow my pawn promotion. I think it would have been an awful lot harder had he not, and moved to A7. Frankly, I think |
|
| Nov-18-10 | | gropek: <YCP> read the previous posts, they explain all the good variations |
|
| Nov-18-10 | | David2009: <trguitar: I'm not so sure this is a win for White. After 33. Bxb7 Nxb7 34. a6 Bb8 35. Bb6 b4 how does White proceed?>
click for larger view Kasparov vs A Georgiadis POTD with 35...b4, White to play White can safely win a Pawn starting 36.b3. Setting the diagram position up in Crafty End Game Trainer, http://www.chessvideos.tv/endgame-t...
the EGT defends with 36...Kf8 37.axb7 Ke8 38.Bc5 h5 39.Bxb4 Kd7 40.Kf1 and now after 40...Kc6
click for larger view White (to play) still has work to do to win. The ending is similar to that discussed in my earlier posts (see Kasparov vs A Georgiadis, 1996). Instead the EGT inexplicably plays 40...Kc7 41.Bc3 g6 allowing White to liquidate into a winning K+P ending with 42.Be5+ Kxb7 43.Bxb8 Kxb8 44.f4! I have posted instructions on setting up positions using Crafty EGT on the crafty chessforum. |
|
| Nov-18-10 | | rapidcitychess: Does 33.Bc7 not win? |
|
Nov-18-10
 | | Sastre: <33.Bc7 Nxe4 34.Bxf4 Nc5 35.b4 Na6> is good for Black. |
|
| Nov-18-10 | | wals: Got it! I followed the thinking of the mighty Kasparov. Wow. Rybka 4 x 64
depth: 20 : 7 min :
Black error
(+1.43):10...Nf5. Best, Bc5, +0.79.
depth: 18 : 4 min :
White error
(+0.78):11.Qh3. Best, Qe1, +1.42.
depth: 25 : 5 min :
Black blunder
(+1.34):23...Bb4. Best, Qxf3, =0.00.
depth: 23 : 4 min :
White blunder
(=0.00):24.Red1. Best, Qxd5, +1.34.
depth: 23 : 14 Min :
Black blunder
(+1.85):27...Rd2. Best, Rxb2, =-0.12.
depth: 27 : 4 min :
Black blunder
(+3.73):35...Nd6. Best, b4, +2.29.
depth: 22 : 3 min :
Black blunder
(+9.13):37...Nc4. Best, Kf8, +3.92.
depth: 27 : 5 min.
White error
(+8.57):39.Qc6. Best, Qe4+. +9.14.
Black resigned. |
|
Nov-18-10
| | Jimfromprovidence: <David 2009> <Crafty EGT disagrees and prefers activating its King safely with 35...Kh7. If 36.b4 Kg6 37.Bc7 Bxc7 38.a7 Be5 39.a8=Q Nd6 40.Qd5 Kf5 and we reach the diagram above. How does White propose to break down the EGT's fortress? Not easy!> I can't beat it. How did Crafty know to play 37...Bxc7, which loses more material? |
|
| Nov-18-10 | | Old Wolf: Nobody has looked at the ending after black's best line yet (as far as I can tell) and it is interesting :) 35. Bb6
35. ... b4! (Otherwise 36.b4 stopping the ..Nc5 tactic) 36. b3! (fixing the pawn on b4)
36. ... f5 (or ..g5)
37. axb7 .
Now, Black has to meander his king to c6. White first of all picks up the b4-pawn. Now, white threatens this tactic to get his bishop to a7: f4 Bxf4; Bd4 Bb8; f4 Bxf4; Ba7.
To stop this Black has to play ..f4 or ..g5, both of which leave an easy kingside invasion for either White's bishop or White's king. We end up with White having passed pawns on both sides (the b3 pawn), at which point I think we can safely say that it is a clear win with same-coloured bishops :) |
|
Nov-18-10
 | | Gypsy: The opening sequence <33.Bxb7 Nxb7 34.a6...> is clear thus far. For that matter, also <34...Bb8> seems pretty forcing. Now the obvious choicer are (1) <35.axb7>, (2) <35.Bc7>, and (3) <35.Bb6>. -- (1) The <35.axb7> snags immediately on <35...f6!>. Were we able to play Be8-h3-g2, things would have been rather easy. But, after the f6, the endgame does not really seem winnable. -- (2) And the <35.Bc7> snags on <35...Nc5 36.Bxb8 Nxa6 ...>. -- (3) Thus, <35.Bb6...>. Here, the threat is to play either axb7 or Bc7 after enough preparation. Key branches from this point go: (3.1) <35...Kf8 36.b4(!) Ke7 37.Bc7...>. Curtains, because <37...Nc5> is no longer a viable option; <38.bxc5 Bxc7 39.a7...>. (3.2) <35...b4!? 36.axb7 ...> Now, this endgame already looks winnable alright. Here the black b-pawn is too far forward and will fall. (3.3) So <35...Nd6> seems to be the only thing left for Black. However, after <36.Bc7...>, either the a-pawn promotes or White ends up a piece <36...Nc8 37.Bxb8 ~ 38.a7 Nxa7 39.Bxa7...> White wins in all three lines. This enumerates the tactical skeleton of the combo, I think. |
|
| Nov-18-10 | | WhiteRook48: i got the first two moves |
|
| Nov-18-10 | | James Bowman: Well it takes only a few seconds to realize the a pawn was going to score a touchdown, so the first two moves were automatic. I did not calculate the entire line but even removing the bishop was a natural move and I gave myself a passing grade. Nice puzzle. |
|
| Nov-18-10 | | BOSTER: <trgultar> <I'm not so sure this is a win for white>.
33.Bxb7 Nxb7- only because we are all greedy.
34.a6 Bb8
White has no any desire to take knight on b7, because to move pawn b7-b8 you need to put white bishop on b8-h2 or a7-b8 and have support for it. So "a" pawn has to stay on "a" file.
35.Bb6- the only move to create the net for black knight and at the same time very tricky.
35...b4
And now we have the position on diagram with white to play.
click for larger view.
If now 36.Kf1 f6 37.Ke2 Kf7
38.Kd3 Ke6 39.Kc4 Nd6+ 40.Kxb4 Nc8
or 39.a7 Bxa7 40.Bxa7 Kd5
I don't think that white can easy to win. |
|
| Nov-18-10 | | randomsac: I was confused at first, but then I tried to promote a pawn. And I found Bxb7 to let the a pawn through. It played out similarly to what I thought. |
|
| Nov-19-10 | | timothee3331: 35...b4! (otherwise 35...Kf8 36.b4 and 37.Bc7!) 36.b3! Kf8 37.axb7 and then 38.Bc5 and this endgame might be won though i'm no sure |
|
Nov-19-10
| | Gypsy: There seems to be some uncertainty whether the endgame after <33.Bxb7 Nxb7 34.a6 Bb8 35.Bb6 b4!?> and <36.axb7...> 
click for larger viewis easy enough for White to win.
Several people have thought that a finesse, 36.b3, is needed here; but even then a subset is still not convinced of the outcome. (While the 33.b3 is alright and it does not spoil anything, it does not really help White with the win either.) But winning from the above diagram is easy enough.
<36...f6>
The alternatives are worse. For instance, <36...f5 37.f4 g5> (lest Bb6-d4-e5) <38.fxg5 hxg5 39.Bd8 g4 40.hxg4 fxg4 41.Bh4 ~ 42.Bg3...>. <37.Kf1 Kf7 38.Ke2 Ke6 39.Bc5...> and, say, <39...b3 40.Kd3 Kd5 41.Bf8...>. 
click for larger viewWhite now wins a multitude of pawns for the one on c7. Black position quickly collapses. |
|
 |
|
|
< Earlier Kibitzing · PAGE 4 OF 4 ·
Later Kibitzing> |
Kibitzing
sac: 33.Bxb7