|
< Earlier Kibitzing · PAGE 3 OF 3 ·
Later Kibitzing> |
Oct-15-09
 | | Check It Out: I see 21.Qxd5 and the knight can't take back due to 21...Nxd5 22.Re8+ Kg7 23.Nh5#, so white has won his pawn back with all sorts of threats. This doesn't seem like enough of a solution, but let's see... |
|
Oct-15-09
 | | al wazir: I saw 20. Qh5, saw that it wasn't forcing enough, then saw 20. Qxd5, but didn't see 20...Be6 as a viable reply. What I did see was 20...Qd4. But after 21. Qa8, black has no good way to defend the ♗: if 20...Qd7/Qd8, then 21. Ba6. I score puzzles by degree of difficulty: 1 point for Monday, 2 for Tuesday, 3 for Wednesday, etc., so the maximum possible total score for a week is 28. The only way to get full credit for a solution is to find winning lines against ALL of the opponent's tries. Today I give myself 2 points out of a possible 4. |
|
| Oct-15-09 | | e4boy: 21 ... Qc7 and 21 ... Bd7 seem to be Black's best. They are the reason I had concluded that I had not solved the puzzle although I felt that 21 Qxd5 was the move I would have made in a game. |
|
| Oct-15-09 | | remolino: Thursday, Medium. White to play, 21.?White has a pawn less. Motifts:
Black's rook is inactive; white's pieces are more active and well placed; bishop covers important squares around the black king; Black's back rank is week; bishop on back rank is vulnerable; Knight is overloaded, covering both back rank and d5 pawn; d5 is weak; Black king is open, can only escape to g7 besides squares on back rank. Tried both Nh5 and Nf5 but those lead nowhere. Allowing exchange of queens does not seem intuitive as we want to attack the open king position and take advantage of our much higher piece activity. Therefore:
21. Qxd5! and white wins a pawn, centralizes the queen and maintains many threats on the back rank. Not sure Black can prevent the loss of a piece. Of course Black cannot take the queen:
21...Nxd5, 22.Re1+ Kg7, 23. Nh5++
=====
Time to check. |
|
| Oct-15-09 | | VincentL: I saw Qxd5 and the main follow up in a few minutes, not instantly. So Thursday level is about right for this part. But to find the winning line in all the variations in which black declines the queen is tough - Saturday if not Sunday level. In some lines it appears that white only emerges a pawn to the good after several moves. |
|
| Oct-15-09 | | David2009: Thursday's puzzle E Berg vs Leko, 1995 White to play 21? Medium 21 Qxd5 hoping for 21...Nxd5 22 Re8+ Kg2 23 Nh5#.
If 21...Bb7 22 Qxc5 (not 22 Qd8+ Kg7 23 Re8 Rb1+ 24 Bb1 Ba6) Rxa2 24 Bc4 with dangerous threats: e.g.
24...Qd2 25 Qe7 and the B now forks the Ra2 and Pf7.
Time to check and verify this analysis. Over the board I would consider the position further after
21...Bb7
====
Black played Be6. Time to check ove the board/ read others' comments. |
|
| Oct-15-09 | | jfshade: Well, it jumped out at me pretty quickly that the only thing preventing the Re8 Nh5 mate was that Knight, and that Qxd5 would either decoy it off or else win the P and leave the white Q in a good attacking spot. But after spending a lot of time running through variations following black's 21st Qc7, Bd7 or Be6, I couldn't see (not that I can see that far) white doing more than maybe taking another pawn. Still, Qxd5 is a pretty cool move and I hope I might have found it as quickly in a game situation. |
|
| Oct-15-09 | | johnlspouge: < <goodevans> wrote: [snip] the trick is knowing when to stop. [snip] As <TheaN> points out,it's not always necessary to analyse to the final conclusion simply to establish what the best move is. > < <Patriot> wrote : Great points! > Indeed. Why bother analyzing past 21.Qxd5, when you have won a P, and a check is available to get the Qd5 out of any jam? (Unless, like <agb2002>, you regard the puzzle as training for board vision.) Certainly, over the board at least, you can only make the best move, so efficiency dictates terminating calculation, after you know what the best move is. |
|
Oct-15-09
 | | Pawn and Two: I was able to find 21.Qxd5 fairly quickly, especially since White has no good alternative. However, as Black will decline to play 21...Nxd5?, and has the defensive options 21...Qc7!, 21...Bd7 or 21...Be6 available, the position became too difficult for me to verify a win. It turned out the position after 21.Qxd5 was even too difficult for my Fritz to verify a win for White! Fritz indicated that after 21.Qxd5, 21...Qc7! was Black's best defensive try, giving Black good drawing chances. Here is Fritz's analysis after 21.Qxd5: (.80) (23 ply) 21...Qc7! 22.Qf3 Qf4 23.Qc6 Bd7 24.Qxc5 Qb4 25.Qxb4 Rxb4 26.c4, (.79) (24 ply) 26...Ra4 27.Re2 Be6 28.Rc2 Ra5 29.Ne2 Nd7 30.Nd4 Ne5 31.Nxe6 Nxd3, (.70) (26 ply) 32.Nd8 Kg7 33.f3 Kf6 34.Nc6, (.55) (26 ply) 34...Ra6 35.Nb8 Nb4 36.Rb2 Ra4 37.Nd7+ Ke7 38.Nc5 Ra5 39.Rxb4 Rxc5, with good drawing chances for Black. Improvements can probably be found for both sides in this long variation. However, Black appears to have good drawing chances after 21...Qc7!. Can anyone verify a win for White after this move? |
|
| Oct-15-09 | | David2009: As the game E Berg vs Leko, 1995 went, 23...Rb6 seems tougher than 23...Kg8. It prevents the mating combination. After 24 Qxc5 Qd2
25 Qe5 Kf8 26 a4, White is a solid Pawn ahead but Black is still in there. |
|
| Oct-15-09 | | Justawoodpusher: I was pretty sure that 21.♕xd5 is a good move and I analysed the defence of 21...♕c7. I did not see a clear win and gave up seeing no better 21st move for white.
Fritz however shows that my analysis was not so bad at all, even when I haven't seen all the following lines. So I count this for me as "half solved". |
|
| Oct-15-09 | | Keith Dow: 23 ... Kg8 was the losing move.
23. ... c4 gives black a slight advantage. |
|
| Oct-15-09 | | WhiteRook48: 21. Qxd5! |
|
| Oct-15-09 | | ruzon: I did not see the reply 21...Be6, but I knew 21. Qxd5 was the right move. I won't pretend I know how to win when I am winning, which is when I often lose. |
|
| Oct-15-09 | | TheBish: E Berg vs Leko, 1995 White to play (21.?) "Medium"
White is down a pawn. On top of that, Black is threatening to trade queens. I noticed that if White could divert the knight on f6 somehow, then Re8+ could cause problems. Indeed, 21. Qxd5! Nxd5? 22. Re8+ Kg7 23. Nh5# ends matters. So it looks like at least wins a pawn to restore the balance, but let's see where it goes. 21. Qxd5! White now threatens 22. Qd8+.
A) 21...Nxd5? 22. Re8+ Kg7 23. Nh5#.
B) 21...Be6 22. Qd8+ Kg7 23. Qe7! (threatening Rxe6) leads to: B1) 23...Bxa2? 24. Qxf6+! Qxf6 (or 24...Kxf6 25. Nh5#) 25. Nh5+ Kf8 26. Nxf6, winning a knight. B2) 23...Rb6 24. Qxa7 wins a key pawn, giving White a passed pawn. B3) 23...Kg8 24. Bg6! Rxa2 (or 24...fxg6 25. Rxe6 Rb1+ 26. Kh2 Rb6 27. Qd8+ Ne8 28. Rxe8+ Kg7 29. Re7+ Kf6 30. Qf8#) 25. Nf5 Bxf5 26. Qxf7+ Kh8 27. Qxf6+ Kg8 28. Re8#. C) 21...Qd4 22. Qa8 Qd8 23. Bf5 Nd7 24. Rd1 Rb7 (24...Rb8 25. Qxb8) 25. Ne4! Kf8 26. Nxc5 Rc7 27. Rxd7 Rxd7 28. Bxd7 followed by 29. Qxc8 leaves White two pieces up. D) 21...Kg7 22. Qxc5 Rb8 (22...Be6 23. Qc3 Rxa2 24. Nh5+ wins the queen) 23. Qc3 Kg8 24. Nh5 wins a piece, since 24...Nxh5 25. Re8 is mate. Probably missing some critical lines, but time to see how this went down. |
|
| Oct-15-09 | | TheBish: I had 24. Bg6!? instead of 24. Bf5 (although I looked at that), missing the point that 24...Bxf5 25. Qxf6 forks the rook and bishop, and wins a piece. But I think that White's position is so good that my move wins as well, although not as directly. |
|
| Oct-15-09 | | Athamas: Well as far as the game went, I don't agree with <Keith Dow> 23...c4
This move seems to lose pretty badly. It might even be worse than the game move of Kg8. I think the best chance black has is 23...Rb6, but it still leaves white with an advantage, not black. This is the continuation I see if c4 was played. 24. Rxe6 cxd3 25. Rxf6
Now what? Black's lost a whole knight and is threatening mate. 25...Rb1+ 26. Kh2 Rb7 27. Rxf4 Rxe7 28. Nf5+ Kf6 29. Rf3 Now black loses a pawn as well. Down a knight and a pawn, I think you can safely say white has the advantage. |
|
| Oct-15-09 | | butilikefur: <21. Qxd5 Be6> with other possible defenses <22. Qd8+ Kg7 23. Rxe6 Rb1+ 24. Kh2 fxe6 25. Qe7+ Kh8 26. Bf5 exf5 27. Qxf6+ Kg8 28. Qg6+ Kf8> and I don't see the win. |
|
| Oct-15-09 | | butilikefur: It's pretty amazing that after 23...Bd7, 23...Bd5, etc. 24. Qxf6+ wins. |
|
| Oct-15-09 | | patzer2: Peter Leko was a 15 or 16-year-old GM at the time of this game. I suspect, he overestimated his attacking chances with 21...Be6? Perhaps a now older and wiser Leko would not hesitate to play the strong defensive move 21...Qc7! with good drawing chances, as indicated in the Fritz analysis provided by <Pawn and Two>. |
|
| Oct-15-09 | | patzer2: Berg's 24. Bf5! is an impressive move in its own right, attacking Leko's once seemingly strongly posted but now overworked Bishop on e6 with multiple threats. If 24...Nd5??, then our familiar mate threat 25. Qd8+ Kg7 26. Nh5# is renewed. If 24... Bxf5, then White wins after 25. Qxf6 when 25...Rxc2?? (25... Rb8 26. Nxf5  ; 25... Be6 26. Qxb2 ) 26. Re8+ Kh7 27. Rh8# is one amusing possibility. If 24...Rb8, then 25. Qxf6 also wins when play could go 25...Qd4 26. Qxh6 Qg7 27. Rxe6! Rb1+ 28. Kh2 fxe6 29. Qxe6+ Kf8 30. Qc8+ Ke7 31. Qxc5+ Ke8 32. Qc8+ Ke7 33. Be6 Qg6 34. Qd7+ Kf6 35. Nf5 . If 24...Rb6, then 25. Qxf6 wins a piece and more as 25...Bxf6?? leads to 26. Re8+ Kh7 27. Rh8# (or 27. Qxf7#). |
|
| Oct-15-09 | | patzer2: Berg's followup to 24. Bf4! Qd2 with 25. Rxe6! leads to an instructive mating finish. The initial threat is 25...fxe6 26. Bxe6+ Kh8 27. Qf8+ Kh2 28. Bf5#. In the final position, it's mate in two after 28...Ne8 29. Rxe8+ Kg7 30. Rg8# or 28... Kg7 29. Qxf6+ Kf8 (29...Kg8 30. Re8#) 30. Qh8#. |
|
| Oct-15-09 | | Athamas: I don't think 21...Be6 is deserving of a ?
It seems only slightly weaker than Qc7 to me... although interestingly I think 24. Nf5 is also worth looking at |
|
| Oct-15-09 | | patzer2: <Athamas> You may have a point there. I suppose if Leko had found 23...Rb6! 24. Qxc5  (instead of 23...Kg8?) he might have survived the complications and justified 21...Be6. However, IMO his defensive task is easier and survival chances are much better after 21...Qc7! |
|
| Oct-22-09 | | Formula7: I considered 21.Qxd5 but couldn't find a win after 21...Qc7. It still looked like the best move, which it is. |
|
 |
|
|
< Earlier Kibitzing · PAGE 3 OF 3 ·
Later Kibitzing> |
Kibitzing
