< Earlier Kibitzing · PAGE 3 OF 3 ·
|Feb-06-08|| ||JG27Pyth: <Funicular: Perhaps it's because it's 3:48 am, but I'm unable to see any effective black counterplay, so i'd open the H column to control it myself>|
That's exactly how I read the position too... I didn't find g3, (although I think I would have seen it OTB.) I saw Ng4 immediately, then, discarded it thinking it would open the h file, and came back to Ng4 when I saw that not only would white win the N, but was better positioned to take advantage of the open h file anyway... although Keres playing g3 contradicts that assessment I suppose... I guess it's a case of, "Look, Viktor, I'm going to win the N and you get _nothing_, you can't even play for complications...
|Feb-06-08|| ||zb2cr: Okay, the Black Knight is badly pinned. 37. Ng4 adds pressure--but Black can swap, BxN. Bummer. Oh! wait a minute!
After the Pawn capture, White threatens to advance g5. I see no constructive way for Black to oppose this.|
|Feb-06-08|| ||MostlyAverageJoe: This is one of those puzzles that are much easier for humans than for computers and defy my evaluation method (according to which this was an average Thursday (!!!) puzzle and more difficult than yesterday). Personally, I agree that it was easy.|
Part of this is because the actual material gain, although unavoidable, happens 9 plies deep at the earliest; another reason is perhaps in that noticing the pinned N focuses human thinking on capturing it, while engines perhaps still look at all other possibilities.
Incientally, the best continuation for white in the line played in the game is NOT to take the knight immediately (after all, it is not going anywhere, is it?), but to pile up more threats.
Here's one possible line:
37. Ng4 Bxg4 38. hxg4 h3 39. g3 <Qc8> 40. g5 Qf5 41. Rde1 Re8 42. Qd4 Ra5 43. Rxe4 Rxe4 44. Rxe4 Ra2+ 45. Re2 Rxe2+ 46. Bxe2 Kh7 47. gxf6 <finally!>
click for larger view
37. Ng4 Bxg4 38. hxg4 h3 39. g3 <Qd8> 40. g5 Re8 41. Rde1 a5 42. Rxe4 Rxe4 43. Rxe4 Rc8 44. f5 a4 45. Rh4 gxf5 46. gxf6+ <at last!>
click for larger view
Both of the above are crushing with evals > 5.50; they were derived in a *quick* forward scan with Hiarcs so improvements are possible for both sides.
On the other hand:
37. Ng4 Bxg4 38. hxg4 h3 39. g3 Qd8 40. g5 Re8 41. gxf6+ <greedy white!> Qxf6 42. Qxf6+
click for larger view
is less forcing (<4.00), although just as winning in the long run.
Oh, well, another difference between engines and humans, I suppose. I guess many players would grab the knight early.
|Feb-06-08|| ||YouRang: I found it easier than I normally expect for a Wednesday.|
Certainly the most glaring problem for black is his pinned knight. Black is curiously helpless to either unpin the knight or bring other defenders to its aid.
Naturally, white wants to attack this knight, and there's one way: 37. Ng4. Sure, the knight can be captured by a bishop, but then it's recaptured by a pawn which itself threatens to attack & win the knight. Nothing black can really do about it.
I pretty much saw the same line that was played, except I didn't see any harm in 39. gxh3.
|Feb-06-08|| ||playground player: Just one question: Why would you ever use the Sicilian Defense against someone known as "The Sicilian Killer"?|
|Feb-06-08|| ||kevin86: I thought I didn't have a clue on this one-until I found the solution. Black's real weakness in the puzzle position, is the immobility of the knight. It cannot move now-nor is it possible to see a time that it can ever be moved. White needs to attack it. The knight can attack it at g4,but will be exchanged. The pawn,now on the g-file is the key. Black cannot stop it from getting to g5 and winning the staid knight. KISMET!|
|Feb-06-08|| ||Halldor: Attack on a pinned piece.|
|Feb-06-08|| ||jovack: Korchnoi forgot about the pawn that would be in position to attack the black knight after his bishop took the white knight.|
Medium/Easy is appropriate.
|Feb-06-08|| ||chopin4525: Quite easy today :)|
|Feb-06-08|| ||dzechiel: <MAJ> The reason that I would consider grabbing the knight early is that it forces the exchange of queens and will help to eliminate the complications that are inherent in queen endings. Yeah, you get to keep another pawn by delaying the capture, but as you said, both are winning. So which is "easier" to win for the organic player?|
|Feb-06-08|| ||patzer2: For today's Wednesday puzzle solution, Kere's 37. Ng4! begins a slow but decisive pile up on Korchnoi's pinned Knight.|
|Feb-06-08|| ||MostlyAverageJoe: <dzechiel> Oh, I agree that early exchange is easier for a human player, no question about it, especially OTB.|
I'd play it just about automatically (QxN to force Q exchange, but note that the engines prefer PxN and keep the Qs on board), but I keep wondering whether GM-level player would do the same. I recall reading advice that in a winning position one should not hurry, but consolidate and constrain the opponent to make sure that the final blow can be delivered without surprises.
Also: would you also opt for early exchange in a correspondence game, with more time to analyze?
|Feb-06-08|| ||baseballplayer: I saw all the way up until Kortchnoi resigned, and decided that white was winning. It took me about five minutes though. A little bit long for a Wednesday.|
|Feb-06-08|| ||DarthStapler: I got it, g5 wins the knight|
|Feb-06-08|| ||wals: Medium/Easy - white to play
Cursory glance - move the knight the king is in check
thats a start.Ng4 would be taken by Bc8
b4 necessitate a rook move b5? B x b5 p x b Rd1-a1
b4 rook c7 no advantage
Qd4 Rh5 no advantage
37.Rd4 Bf5 38.N x f5 g3xf5 no continuation
37 Ne3-g4 ...B x g4 38h3xg4
g4 is a winner
A winner - there is still some life in the old brain cell
|Feb-06-08|| ||uuft: Yesterday was Super Tuesday. Today is just Plain Wednesday :)|
|Feb-06-08|| ||dzechiel: <MAJ: Also: would you also opt for early exchange in a correspondence game, with more time to analyze?>|
Probably so. The only way I would leave the queens on the board would be if:
1) removing the queens increased my opponents drawing chances for some reason
2) I could see a concrete way to end the game sooner by leaving the queens on the board
I am a firm believer in consolidating your position once you achieve a material advantage, and most often that means trading material off because it increases your percentage of advantage.
|Feb-06-08|| ||OhioChessFan: I think Keres played 29. Qe1 anticipating the weakness of the Knight. When Korchnoi obliges with Kg7, I bet Keres was licking his chops. Her Majesty moves several times but always has a spot on the long diagonal in view, though usually with a piece between her and her destination. She eventually gets to c3, and the blunder Bc8 seals the fate of the pinned Knight. Many of the earlier moves remnind me of the computers in the 80's and 90's with their seemingly random maneuvers.|
|Feb-06-08|| ||eternaloptimist: Yet another example of a blunder by Korchnoi (Bc8??). One of the key moves in this game was 24.Ng3, provoking 24....g6 to prevent the N from getting to f5 & thus attacking the weak d-pawn. 33....h4? only created a weakness on g4, which Keres took advantage of 37.Ng4. Another example of his blundering is Petrosian-Korchnoi Leningrad 1946 (Dutch Defense). His d-pawn is pinned & Petrosian's Q wins Korchnoi's N because Korchnoi's Q is forced to stop guarding the N on the last move of the game (a dismal final position). Also, Petrosian-Korchnoi Curacao ct 1962 (English Opening) is an !? game. 7....Qa5 only lost time because Petrosian kept attacking her. Eventually, Petrosian won the exchange by a N fork (a R for a N). I think some of Korchnoi's biggest faults were (sometimes): 1)He didn't look far enough ahead. 2)He didn't try to figure out his opponent's plan (or couldn't). 3)His positional understanding was suspect in certain positions which led to tactical shots by his opponent. I think these were the main reasons he never became world champion.|
|Feb-06-08|| ||crwynn: I suspect 36...Qd8 37.f5 leaves White with a big advantage, which was why 36...Bc8 was played; I think taking a5 was the real mistake.|
|Feb-07-08|| ||johnlspouge: <CRWynn>: "I suspect 36...Qd8 37.f5 leaves White with a big advantage, which was why 36...Bc8 was played"|
Toga II 1.3.1 gives the best response to 36...Qd8 as
[ply 16/46, time 00:53, value to White +0.50]
36…Qd8 37.Ng4 Qd8 38.Nf2 Kh7 39.Nxe4 Nxe4 40.Rxe4 Rc7 41.Qa5 Kg7 42.Kg1 Rc5 43.Qxd8 Rxd8
Interestingly, it gives better value to your line
[ply 15/53, time 02:38, value to White +0.96]
36…Qd8 37.f5 gxf5 38.Qd4 Rh5 39.Rf1 Rc7 40.Ref2 Kg6 41.Kg1 Rg5 42.Nxf5 Rxf5 43.Rxf5 Qxf5
|Feb-07-08|| ||znprdx: By the time the game got to the problem position it had been over for nearly a dozen moves. <eternaloptimist:>'s observations are quite interesting - |
Korchnoi desperately needed to form a plan, (any plan!) to survive to time control with hopes for a draw. 28...Black to move: the hopelessly misplaced King Bishop can only redeem itself by exchanging with White's Queen Bishop.
The d6 pawn cripples Black's maneuverability in fighting for control of the dark squares. 28...Bx[B]e3 29. Nx[B]e3 surely offers more hope than h5,(perhaps the losing move)which might be best for the knight. 29...Rc5 followed by a timely Qd8 seems to be the only hope.
|Feb-07-08|| ||eternaloptimist: <znprdx:> <eternaloptimist:>'s observations are quite interesting>. Thank you!! I try to make them interesting. I agree with you when you said the <d6 pawn cripples Black's maneuverability>. It took Korchnoi 5 moves to get his B on a open diagonal (f8 to a7). This wasted time. I also noticed that Korchnoi never played ...d5 in the game, and this is the reason that his position was so cramped. That is one of the most important goals in the opening in the Sicilian. It is very tough to equalize against a great player (which Keres undoubtedly was) without playing it.|
|Feb-07-08|| ||eternaloptimist: Now that I have looked @ the position more, I really think that Korchnoi could have gotten a draw here w/: 36....♖e8! This neutralizes the pressure on the e-file. If white was allowed to double his rooks on this file (unopposed), it could have been trouble for black. It enables black to get rid of one of white's rooks. 37.♘g4 (Actually 37.♖f2 is better, but to show that ♘g4 doesn't work anymore I'll show this line). 37....♕d8 (korchnoi's move here ....♗c8?? lost the game for him because it cut off his ♕ from getting to d8 to guard the ♘). 38.♙b4 ♖c7 39.♖de1 ♗c8 40.♘f2 ♙e3 41.♖xe3 ♖xe3 42.♖xe3 ♗f5 43.♘e4 ♗xe4 44.♖xe4 and korchnoi would be fine here. As for the 37.♖f2 line,....♕d8 38.♙f5 ♙g5 & the f-file stays closed so white can't load up on it. <CRWynn: I suspect 36...Qd8 37.f5 leaves White with a big advantage, which was why 36...Bc8 was played; I think taking a5 was the real mistake.> Here the ♕ moves to d8 one move earlier & the pawn moves to f5 one move earlier than in my line above. His statement is not correct though because you can still play ....♙g5 & stop the attack as in the line above. Also, 34....♖xa5 was the correct move & not the losing move.|
|Feb-08-08|| ||eternaloptimist: <eternaloptimist:33....h4?> I will correct myself here. h4 was not a bad move. It just created a weakness on g4, which Keres took advantage of because of the howler Bc8??. Keres won this game because of that move & because he provoked both <24....g6> (weakening the a1-h8 diagonal) & 33....h4.|
< Earlier Kibitzing · PAGE 3 OF 3 ·