Apr-12-07
 | | Honza Cervenka: Why not 46...Bxe5? |
|
| Nov-01-08 | | gambitfan: <Honza Cervenka: Why not 46...♗xe5?> 46... ♗xe5 47 ♖a5
click for larger view
after 47 ♖a5
Black ♗e5 is pinned by the White ♖a5 and cannot be protected by any Black piece ; therefore it is lost ! |
|
| Nov-01-08 | | hcgflynn: <gambitfan>: 47. bd5? or 47. rg3+ and after re3 and kxh4? |
|
| Nov-01-08 | | gambitfan: Addendum :
Black might be wanting to interpose ♗d5 apparently protected by ♙e6 47... ♗d5 48 ♖x♗d5! e6x♖d5 49 ♖f5+ ... 50 ♖x♗e5
And White just won the quality : ♗♗ // ♖
White have now ♖ + ♙♙ against ♖ + ♙, which gives them good chances to win or at least to draw... |
|
Nov-01-08
 | | SwitchingQuylthulg: <gambitfan> It would appear that, in spite of your long stay at CG.com, you haven't yet learned the Basic CG.com Rule #1: <Honza Is Never Wrong>. (If Honza suggests a move that computer analysis doesn't find optimal, you know it's time to buy a better computer.) There is no refutation of 46...Bxe5 - if 46...Bxe5 47.Ra5 Rg3+ 48.Kh2 Bd5, not only is White threatened by a killing discovered check (if 49.Rxd5 exd5 50.Rf5+ Rg5+!), but the h4-pawn is also going to drop. |
|
| Nov-01-08 | | gambitfan: < hcgflynn: 47... ♗d5 or 47... ♖g3+ 48 ♔h2 ♖e3+dec 49 ... ♔xh4 > I answered by anticipation your first proposal 47... ♗d5 : no good ! White win the quality ♗♗ // ♖ Let's have a look at your second proposal
47... ♖g3+ 48 ♔h2 ♖e3+dec 49 ... ♔xh4
After 47... ♗d5 or 47... ♖g3+ 48 ♔h2 ♖e3+dec White are to move : 
click for larger viewWhat do you think of 49 ♖f4! ?
Now Black are to play...
click for larger viewWith 49 ♖f4! White interpose ♖f4 exploiting the pinning of ♗e5 and protecting ♙h4 no loger protected bi White ♔ |
|
| Nov-01-08 | | gambitfan: In short, looking at my last diagram, we can see that 49 ♖f4! 1. interposes ♖f4 between the pinned ♗e5 which cannot take ♖ and ♖h2 2. protects ♙h4
3. threatens ♗c4 (but ♗f4 is also pinned
As long as ♗e5 is pinned White ♔ can walk on dark squares... What can Black do ?
49... ♖e2 50 ♔g1 (now ♖f4 is un-pinned) ♖e1+ 51 ♔f2 
click for larger viewBlack are to play... |
|
| Nov-01-08 | | gambitfan: Conclusion : thanks to the brilliant move 49 ♖f4! Black lost at least the quality ♗♗ against ♖ and at worst a ♗. Do not you agree ? |
|
| Nov-01-08 | | gambitfan: And I have no computer ! All these modest reflexions are the pure product of my brain... And no human being is infaillible, even the Pope !
In conclmusion, Karpov was right to play 46... ♗h6 and not 46... ♗xe5 |
|
| Nov-01-08 | | hcgflynn: <gambitfan>:46. - bxe5 47. ra5 bd5 48. rxd5 rg3+ 49. kh2 exd5 and now if 50. rf5+ then rg5+! or:
46. - bxe5 47. ra5 rg3+ 48. kh2 re3+ 49. rf4 bd5 and now what? |
|
| Nov-01-08 | | gambitfan: <hcgflynn: 46... ♗xe5 47. ♖a5 ♗d5 48. ♖xd5 ♖g3+ 49. ♔h2 exd5 and now if 50. ♖f5+ then ♖g5+! or:
46... ♗xe5 47. ♖a5 ♖g3+ 48. ♔h2 ♖e3+ 49. ♖f4 ♗d5 > OK. Brilliant.
Karpov should have played 46... ♗xe5 |
|
| Nov-02-08 | | gambitfan: ENDGAME :
♗♗ + ♙ (7.5 points) // ♖ + ♙♙ (7 points)
♗♗ + ♙ prevailed over ♖ + ♙♙
click for larger view61 ♔xg2...
After 7 moves we have :
click for larger view68... ♗e7
0-1
The truth is that I do not see very clearly the gain for Black... It seems pretty obvious that White are losing ♙b7 but I do not see any clear reason for White to resign at this stage of the game... Who can help me understand ? |
|
| Nov-02-08 | | gambitfan: After 68... ♗e7 we have ♗♗ + ♙ (7.5) // ♖ + ♙ (6)
Endgame Explorer gives : 30 games / 53% wins / 47% draws After almost sure pawn's loss by White we have ♗♗ + ♙ // ♖ Endgame explorer gives 15 games / 80% wins / 20% draws All these elements don't tell me why White resigned as early as the 59th move... |
|
| Nov-02-08 | | Eyal: <gambitfan> Perhaps Garcia-Martinez could have tested Karpov's technique for a bit, but after 68...Be7 it's clear White is going to lose his last pawn (Bd6 is coming, and 69.Re8 doesn't help because of a bishop check on c5 or h4), and once that happens it's definitely a theoretical win for Black - if he coordinates his king and bishop pair correctly, White cannot stop the e-pawn from queening. Since it's no more than 6 pieces, you can check this out with a tablebase (http://www.k4it.de/index.php?topic=...) |
|
| Nov-02-08 | | gambitfan: <Eyal>
Thanks for your comment...
I checked with the link you gave me...
They say "White lose in 35 moves!"
Not such a quick loss ! |
|
| Nov-02-08 | | gambitfan: @ Eyal
Have a look at this game :
D Gurevich vs K Gudmundsson, 1982
78th move : White obtain ♗♗ + ♙ // ♖
It took White as many as 47 moves to win the game - resignation by Black at the 125th move ! ♗♗ + ♙ // ♖ may well be a "theoretical win" ; if it took 47 moves to win, it is very close to a Draw... |
|
| Nov-02-08 | | Eyal: <They say "White lose in 35 moves!" Not such a quick loss!>
Yeah, but you should remember that the 50 moves rule doesn't apply here, as long as there are pawn moves thrown in once in a while... so in fact there's a very wide margin of error for the side with the pawn. And the basic technique of coordinating the BB+K and pushing forward the pawn isn't so difficult, I think, even it takes a bit of time. The end of this Gurevich vs Gudmundsson game, btw, looks a bit strange - it seems that White could have done it much quicker... |
|
Dec-12-19
 | | woldsmandriffield: 
click for larger viewTo win, Black must drive the White Rook away from b8 and then capture ..Bxb7. This plan only works if a one-two can be put into effect: 1) attack the Rook by playing Ba7 or Bd6 2) when the Rook in turn attacks the dark squared Bishop, give a check with the Bishop. Evidently the plan requires the White King to be on a dark square. It follows, therefore, that from the DIAGRAM a move to a light square (67 Kh5, 67 Kh3) draws whereas a move to a dark square (67 Kh4 or 67 Kg3) loses. Garcia Martinez played 67 Kg3 and lost. |
|
Dec-12-19
| | woldsmandriffield: After 67 Kg3 Bg5 if White had returned to a light square with 68 Kg4, then simply 68..Bf6 69:Kh3 Kf5 Z. In the starting position, after 67 Kh3 Bg5 68 Rg8 or 67 Kh5 Bc5 68 Rc8 |
|
Kibitzing
sac: 26...dxe4