< Earlier Kibitzing · PAGE 3 OF 3 ·
|May-31-08|| ||ezmerin: Hmm, chezzy, isn't that simplier?: 30...Qg4+ 31. Ke3 Bf4+ 32. Kd4 Bxc1+ and White is a full rook down.|
|May-31-08|| ||kevin86: A fine finish! The rook sac looked automatic. The real secret is that the threat of a discovered check was better than the check itself.|
|May-31-08|| ||johnlspouge: Saturday (Very Difficult): Black to play and win.
Material: Down B for N+P. Black has recompense for the lost P in the corresponding open g-file, on which Rg7 faces Pg2 and Rg1. The two Black Bs, Bf6 and Bf5, are active on adjacent diagonals, attacking 2 singly defended pieces, Nc3 (defended by Rc1) and Bd3 (defended by Qd1). Black has 2 inactive pieces, Ra8 and Nd7, but both can be activated rapidly, through the moves Rag8 (or Re8) and Ne5. The Black Qh6 pins Ph3 to Kg2, so the square g4 is unprotected. Black must respond to the threat 23.Bxf5. The weak point in the White K-side is Ph3, because the sacrifice 22…Rxg2+ can undermine the support from Pg2.
Candidates (22…): Rxg2+
The sacrifice brooks no refusal, because of
23.Kh1 Qxh3+ 24.Nh2 Qxh2#
White has only one feasible way to accept the sacrifice.
23.Rxg2 [Kxg2 Qxh3#]
Candidates (23…): Qxh3+, Bxh3, Bxc3
24…Bxg2+ 25.Kxg2 Bxc3 26.Rxc3
[else, down a P with an exposed Kg2]
26…Qg7+ 27.K moves Qxc3
leaving Black up R+P for B, with the White K exposed. White cannot move Rg2 without loss of material (e.g., 24.Rg1 Bf1 leads to mate), so White can either move Nc3 or reinforce its protection with Qd1. If Qd1 reinforces protection of Nc3, however, Nf3 hangs, leading to
24…Bg5+ 25.Kg1 Bxf3
and the mate threat 26…Qh8# ensures Rg2 cannot move, leading to a Black win. Thus, White must move Nc3. Because 24.Ne2 Bg5+ loses as above,
At this point, I realized I had lost my way (with 24...Bxg2+ instead of 24.Ne5, which I planned next move).
Like anyone who actually calculated some lines, it was obvious to me that the position is very delicate, with transpositions fatal and many trails of breadcrumbs leading to a loss. Polgar's ability to thread past all the dangers over the board is amazing.
|May-31-08|| ||johnlspouge: <<kevin86> wrote: [snip] The real secret is that the threat of a discovered check was better than the check itself.>|
One of my favorite chess sayings (probably from Irving Chernev) is that a threat is often more powerful than its execution.
And Life imitates Chess...
|May-31-08|| ||TrueBlue: OK, I found it very difficult. Maybe because I didn't find the right line :) I found a slew of lines that give advantage, but white has way too many defensive options, so nothing forcing. Here is how I see it.|
Rxg2 is a must. Kxg2 leads to losing the queen or checkmate, so Rxg2 is forced response. Now black plays the great move Bxh3. And here it gets tricky, white has too many responses.
Idea is that black should play Ne5 (replied by Ne4) and then Rg8 to crush white. Black can easily get the queen on the queen side, so white king has nowhere to run.
|May-31-08|| ||TrueBlue: OK, I think I saw some of the ideas played in the game, but white response was pathetic :) Ne4 prevents Be4 later, so it seems like a horrible move to me.|
|May-31-08|| ||Marmot PFL: <johnlspouge> I also missed the power of Ne5, which is the key to the position opening up discovered checks to win the white queen and threatening mates after f3.|
I am sure the quote about the threat being stronger than the execution goes back at least to Nimzovich if not earlier. Winter says it was used before 1900 in a book by James Mason.
|May-31-08|| ||Geronimo: <piever> I don't know if I'm "more experienced" than you, but I too was able to calculate about three moves in. I liked what I saw and OTB would have taken the plunge into instinct over calculation. What does that mean? It probably just means that I've integrated a few of Nimzowitsch's principles of positional play into my limited chess universe. It also reflects that I play for pleasure instead of for high-stakes at world-class level, and thus I have less incentive to calculate to the limit on every critical position. Sometimes I even love losing to a stronger player, something someone in her position would never do....|
The fact that Judit was 14 when she played this is enormously impressive. Perhaps Chernin (no dope, he) underestimated the young prodigy, or maybe he simply got outplayed. In any case, I'm sure that someone with x00 points on me has a deeper "spot check" on calculation, namely that even at her tender age she had the ability to go deeper than three moves before just throwing up the hands and taking the plunge..
|May-31-08|| ||lost in space: had a second look to the puzzle after seeing the game and a few posts.
The critical position (after move 23) looks like this:
click for larger view
A: 24. Kg1? Bxg2 25. Kxg2 Bxc3 26. Rxc3 Qg7+ 27. Kh2 Rg8 28. Nh4 Qxc3
B. 24. Ne4 (seems to be the best) Ne5 (same for this move) 25. Nxe5 (25. Nxf6?? Nxf3 26. Qxf3 Bg4+ and the queen is gone) 25...Bxe5 26. Kg1 (what else?) Bxg2 27. Kxg2 Rg8+ 28. Kf1 f3! 29. Qxf3 (no alternative) Qxc1 and Black is much better.
click for larger view
|May-31-08|| ||Terry McCracken: How many of you have seen games switched and all the messages lost?|
E Mortensen vs L Karlsson, 1988
|May-31-08|| ||MostlyAverageJoe: <Terry McCracken: How many of you have seen games switched and all the messages lost?>|
I did. The puzzle got switched at least twice; it was reported in one of the lost messages as having started out as this one, then the one you mentioned (which would be way too easy for Saturday), then back to this one. On chessgames.com chessforum CG says that they were fixing some bugs yesterday that messed up the opening of the day link, so possibly the puzzle was affected as well.
|May-31-08|| ||MostlyAverageJoe: After overnight run at 20 plies per move, Hiarcs came up with this line as best for both sides after 24. Kg1|
24 ... Bxg2 25. Kxg2 Bxc3 26. Bf5 <now it diverges from the lines proposed by <Alphastar> & <patzer2> > Rd8 27. Qc2 Bxb4 28. Qb2+ Nf6 29. Rg1 Rg8+ 30. Kf1 Rxg1+ 31. Nxg1 a5 32. Qe2 Bc5 33. Qe7 Ng8 34. Qf7 Qg7 35. Qh5 c6
click for larger view
The above evaluates at only (-0.94), with this likely continuation:
36. dxc6 bxc6 37. Nh3 <the rest is less reliable ... Nh6 38. Qe8 Qg8 39. Bd7 Qxe8 40. Bxe8 d5 41. cxd5 cxd5 42. Nxf4 Ng4 43. Nxd5 Nxf2 44. Kg2 Ne4 45. Bc6 Kg7 46. Nf4 Ng5>
A quick analysis forward indicates a possible divergence from that continuation as early as at the 4th ply, with white likely to recover one pawn and threaten another, so the apparent pawn majority that black is enjoying is not likely to last long. The evaluations seem to be drifting towards drawish with further analysis.
<<< BACKSLIDE >>>
I did a backslide to the divergence point, to see whether <patzer2: 26. ... Rg8+ ...> or <Alphastar: 26. ... Bxb4 ...> lines are better for black. Here's the position after 26. Bf5 (black to move):
click for larger view
At this point, what is needed to win the game for the black is a PLAN, something that engines cannot provide. The above position is another of the engine-killers, where the progress of the analysis becomes very slow.
After sliding back from the mainline shown at the beginning, 26...Qg7+ is considered the best for black (-1.00), with 26...Nb6 second best (-0.95), but this is only after 18 plies of analysis (which took 150 minutes of computer time to calculate).
The line <patzer2: 26. ... Rg8+ 27. Kf1 Bb2 28. Rb1 Ne5 29. Nxe5 Bxe5 30. Qf3 ... > evaluates at about (-0.75)
The line <Alphastar: 26. ... Bxb4 27. Bxd7 Qg7+ 28. Kh2 Qxd7 29. Qd4+ Qg7 30. Qxg7+ Kxg7> contains inaccuracies played by the white and indeed should be winnable by black. Possible improvements for white shown during backslide: 30.Rb1, 29.Ng5 (MUCH better than Qd4+), 28.Kf1 (small improvement).
I'll do a forward slide after 26. ... Bxb4 to see what it leads to.
Based on computer analysis only, I am still not sure whether this game could not have been drawn. Engines are not likely to provide a conclusive analysis in reasonable time. Human insight at high playing level (i.e., not mine :-) would be needed.
If anyone would like to verify the above analysis with Rybka, it would be appreciated.
|May-31-08|| ||TrueBlue: MAJ, those lines get pretty complicated. That's why I wouldn't believe anyone who said they solved today's puzzle. I like Alphastar's line up to Qxd7. and then 29 Ng5 as you wrote. Possible continuation is: Bc5, Qf3 Bd4, Qh5, Bxf2 . But calculating so many moves ahead is just silly :)|
|May-31-08|| ||DarthStapler: I got the first move|
|May-31-08|| ||johnlspouge: <<Marmot PFL> wrote: <johnlspouge> I also missed the power of Ne5, which is the key to the position opening up discovered checks to win the white queen and threatening mates after f3.>|
I feel I am in good company, then.
<I am sure the quote about the threat being stronger than the execution goes back at least to Nimzovich if not earlier. Winter says it was used before 1900 in a book by James Mason.>
Then I might have read it in one of Nimzovich's books. I always appreciate correction, gently delivered, so thanks, <Marmot PFL>.
|May-31-08|| ||Marmot PFL: Nimzo quote from 1933 article-
Black played 20…Qe7, and Nimzowitsch wrote:
‘This retreat by the strongly-posted queen is an excellent move. The pawn at e5 is now heading for its sad fate. It should be noted that Black has continually operated with the threat of …Qxa2 without executing it (the modern doctrine, which says that the threat is stronger than the execution).’
|May-31-08|| ||johnlspouge: <<Marmot PFL> wrote: Nimzo quote from 1933 article- [snip] (the modern doctrine, which says that the threat is stronger than the execution).>|
It's a dull old day when you don't learn something. So, you have established Nimzovich used the phrase as early as 1933, although it might go back earlier than 1900. I checked with Bartlett's and the Oxford Dictionary of Quotations, but the 'net was more helpful than they were this time.
"Nothing new under the sun..." as my neighbor Charlie used to say. (Relax - the origin of that one I do know ;>)
Incidentally, as long as we are "chatting", in a previous forum I asked whether your avatar indicates you are a climber, but you might have missed the question...
In any case, thanks, <Marmot PFL>.
|May-31-08|| ||234: Friday puzzle <29. ?> May-30-08 Hebden vs E Prie, 1984|
|May-31-08|| ||Marmot PFL: < johnlspouge> Love mountains, but haven't done any climbing in years, and that at beginner level. Drawback of living in the Midwest.|
I had a flag avatar for a while, but the Armenians demanded that I change it.
|May-31-08|| ||Terry McCracken: <MostlyAverageJoe> Thanks for letting me know what happened.|
|May-31-08|| ||Octal: I saw the rook sac as I noticed the g-pawn was holding the h-pawn, which is threatened with heavy pieces. I was thinking about 1. ... Rxg2+ 2. Rxg2 (everything else quickly mates) 2. ... Qxh3+, but didn't know here to go from there.|
|May-31-08|| ||patzer2: <MostlyAverageJoe> Thanks for taking the time to do the analysis. Positions like today's puzzle (22. ?) stretch the limits of human and computer Chess analysis, and I appreciate your diligent efforts today in trying to discern whether it's really winnable. I also appreciate your good comments and insight on what it would take to win such a position.|
|May-31-08|| ||zenpharaohs: Wow. I spent well over an hour on this. And, depite this having been a very bad week for me, I am feeling quite vindicated by that. I couldn't find anything better than Rxg2+, but this move is very far from decisive. In the game line, the opponent makes mistake after mistake to make Rxg2+ look strong.|
But Rxg2+ is just the best move, and the game is very far from over.
22 ... Rxg2+
23 Rxg2 Bxh3
24 Kh1 or Kg1
24 ... Bxg2+
click for larger view
Rybka values this as -0.77 at 20 ply depth.
Of interest is that it took Rybka over a minute and a half to find
22 ... Rxg2+
when I finally handed the problem to her. This is much longer than usual, certainly the longest I remember for these problems. Typically Rybka has the answer in less than a second.
|Aug-05-08|| ||newzild: In an earlier post I said I couldn't believe a player as good as Chernin would exchange his dark-squared bishop for a mere knight and pawn in a Kings Indian.|
After consulting the "book", it apears that black starts getting the upper hand as early as 8.Be3 Nh5, which is good for black. (47 per cent wins, 31 per cent draws and 22 per cent losses for black, according to the free openings database at www.365chess.com)
White coulda played 8.Bg5, which results in 54 per cent wins for white.
|Jan-09-09|| ||plang: Gallagher after 14..Nc5: " For a pawn White has forfeited his right to castle, parted company with his important dark-squared bishop and opened files on the kingside for the Black rooks; a heavy price by anyone's standards. In addition, with the extra pawn defending his king it will only be in the ending that he can hope to derive any benefit from it. The odds are heavily stacked against him arriving there with the rest of his army intact." 9 Nh2 was a standard move at the time but it seems vulnerable to the type of sacrificial attack that Polgar used here.|
< Earlier Kibitzing · PAGE 3 OF 3 ·