< Earlier Kibitzing · PAGE 2 OF 2 ·
|Dec-26-08|| ||Samagonka: The pinner is the winner once again.|
|Dec-26-08|| ||redsincebirth: After 28...Bg5
29.Rxh8, Qxh8 (...Kxh8, 30.Nxf7, black queen is forked)
31.Nf5, and black position is lost.
|Dec-26-08|| ||Alphastar: The two central pawns after Nxe5 dxe5 Bxe5+ are enough compensation for the piece already for me. So that's where I ended my calculation.|
|Dec-26-08|| ||goodevans: <dzechiel: … this seems more than "Difficult" to me.> My thoughts too, initially, but if you look at the solution in a different way then perhaps it isn’t so hard after all. With the benefit of hindsight I asked myself if it made sense to look at 28 Nxe5 as an IPS (Intuitive Positional Sacrifice). After 28 … dxe5 29 Bxe5+ f6 30 Bc3 black’s in a real bind:|
- The pin is doubly strong because the K and R are both on the same diagonal.
- Black can’t exchange rooks or play 30 … Rf8 because conceding the h-file lets the Q in at h6.
- Black’s going to find it difficult to unpin his king for similar reasons.
- The light squared bishop is stuck to the c8-h3 diagonal to prevent Ng4.
White has several threats, including …
- The immediate 31 d6.
- Exchanging rooks then playing Rh1 followed by gxf6 and Qh6.
- Bg4 followed by an exchange of bishops and then Ng4.
Looked at this way it’s a strong case as an IPS, but does black have any defensive resource? As far as I can see any passive defensive moves simply allow white to build up more and more pressure, but black can also force the issue with … Na4 (as in the game) so perhaps that’s the only line where white needs to calculate really deeply.
|Dec-26-08|| ||fporretto: The White pawn at g5 appears to be critical to all the forcing continuations. What if Black continues 28...Rxh1; 29 Kxh1, f6; allowing the Knight on e5 to hang?|
Oops, wait: 30 gf, Bxf6; 31 Nf5+, gf; 32 Qh6+ wraps it up. 'Scuse me please.
|Dec-26-08|| ||MaxxLange: <fporretto> right, that's why I now think 28...Rxh1 is bad: the Nf5+ and Qh6+ idea|
|Dec-26-08|| ||ounos: If <27> ... Bxg5 then it seems that 28. Nxg5 Qxg5 29. Nf5+ Kf6 30. f4! exf4 31. Qc3# is the basis of the refutation. Nice trick, which allowed the important 26. g5, which allowed the beautiful game combination...|
|Dec-26-08|| ||JG27Pyth: Solving this to the level of certainty accuracy and clarity that most of us attempt -- would make this an _insane_ puzzle IMO. |
How deep do you need to go to get to difficult? I don't know. Maybe to Bc3...
It was not hard to see that white had some kind of play attacking e5 and using the long dark diagonal and the open h-file -- I examined both Nxe5 and Bxe5 as my only serious candidates -- I ended up rejecting both. I rejected Nxe5 on the basis of:
28.Nxe5 Bxg5 at which point I couldn't see any clear process for White -- I suppose I still like white better from there... d6 looks very weak... but where's the win?
Am I missing something obvious in this line?
|Dec-26-08|| ||newzild: I saw this:
But I found it very difficult to visualise a continuation after this, as black's reply isn't forced. It seems clear that white has enough for a minor piece, but after 30...Na4, white is virtually forced to sac the exchange as well, after which he is down a whole rook.
Also, black doesn't have to capture the knight. He can play 28...Bxg5. I pondered this for a while. However, it seems white can play 29.Nxd7, winning a pawn, as 29...Qxd7 allows 30.Nf5+! and 29...Nxd7 allows 30.Bxd6
|Dec-26-08|| ||Utopian2020: Black should have played 30...Rxh1 then 30.Rxh1 Na4|
|Dec-26-08|| ||johnlspouge: Friday (Difficult):
Epishin vs Gheorghiu, 1993 (28.?)
White to play and win
Material: Even. The Black Kg7 has 2 legal moves, both on the back rank. The White Ra1 has a semi-open file, while Rh1 faces the Black Rh8. The closed center yields White a permanent spatial advantage and cutting Nb4 and Nb6 off from the K-side. White has a local superiority on the K-side, making a sacrificial attack possible. The White Nf3 and Bg3 both attack Pe5, so White can make the closed center mobile at the bargain price of a piece for 2Ps. As the player with greater mobility, the opening of the position will favor White, activating Nd3, Qd2, and Be2.
Candidates (28.): Nxe5
28.Nxe5 dxe5 [else, drop a center P]
29.Bxe5+ (threatening 29…Kg8 [or Kf8] 30.Rxh8#)
This anti-positional move opens up the Black K-position and confirms the initial candidate.
<[At 15 plies, Toga II 1.3.1 evaluates
30.Rxh8 at +2.51 Ps (time 7:23),
my move 30.gxf6+ at +2.70 Ps (time 12:15),
and the showy game move 30.Bc3 at +1.95 Ps. (time 7:23)]>
30.gxf6+ Bxf6 31.Rxh8 (threatening 32.Rxd8 winning Q+2Ps for N)
The zwischenzug 31.Rxh8 gives White the h-file. In principle, the Black Qd8 could flee, but there is a mate:
(1) 31…Qe2 [or Qc2] 32.Nf5+ Bxf5 [or gxf5]
<[Toga evaluates best play, refusing Nf5, in excess of +20 Ps.]>
32.Qh6+ Kf7 33.Qh7+ Bg7 34.Qxg7#
Black has 3 possible captures, the first for completeness but infeasible:
(2) 31…Bxe5 32.Rxd8 Rxd8
White has Q+2Ps for B+N.
(3) 31…Kxh8 32.Qc3 (threatening 33.Bxf6+)
[Bxe5 33.Qxe5+ Kg8 34.Rh1 (threatening 35.Qh8+ 36.Rh7#)]
33.Rh1 (threatening 34.Bxf6+ Qxf6 35.Rh7+ Kxh7 36.Qxf6)
The Black position has collapsed.
(4) 31…Qxh8 32.Rh1
(threatening 33.Qc3 and the carnage in Variation 2)
<[Toga gives +1.35Ps to my 32.>Rh1< Qxh1+; and over 2.5 Ps to the best play variation
32.>Bxf6+< Kxf6 33.Ng4+ Bxg4 34.Bxg4 Kf7]>
|Dec-26-08|| ||johnlspouge: < <JG27Pyth> wrote: [snip] How deep do you need to go to get to difficult? I don't know. Maybe to Bc3... >|
To my surprise, Toga evaluated 30...Bc3 as less than the best. Personally, I do not regard game variations as gospel, although I also now recognize that the evaluation functions in chess programs do not give sufficient weight to the boredom and frustration humans feel in a drawn-out loss.
<[snip] 28.Nxe5 Bxg5 at which point I couldn't see any clear process for White [snip] where's the win?
Am I missing something obvious in this line? >
Not particularly. I did not see anything immediately, so if you had been playing Black against me, <Pyth>, 28...Bxg5 might have made my heart skip a beat.
After 28...Bxg5, Toga evaluated the best play
29.Rxh8 Qxh8 30.Nxd7 Nxd7 31.Nf5+ Kg8 32.Qxg5
at +5.0 Ps to White, but the psychological edge would have been all yours, <Pyth> ;>)
|Dec-26-08|| ||johnlspouge: < <dzechiel> wrote : OK, this seems more than "Difficult" to me. [snip] It looks almost "insane" to me. >|
My impression, subjectively and based on and Toga's times, puts the puzzle at a very solid Saturday, so I agree.
I know I am tempting <MAJ> to rap my knuckles again for evaluating difficulty.
"Thank you, sir. May I have another?"
- in the sense of both movies: "Animal House" and "House of Games"
|Dec-26-08|| ||akapovsky: I thought Nxe5,exd5,Bxe5+,f6,Rxh8,Qxh8,Rh1,Qf8,gxf6+,Qc3 and now he can't take the bishop because he loses his queen after Bxe5, Qxe5+,Qf6,and Rh7+!|
|Dec-26-08|| ||al wazir: <boringplayer>: You're right about 28...Bxg5 29. Nxd7 Qxd7 30. Nf5+ Qxf5 being safe for black.|
I still think the win is unclear after 28...Bxg5, despite a number of comments since my post, including Toga II's (via <johnlspouge>) 29. Rxh8 Qxh8 30. Nxd7 Nxd7 (what about 30...Bxe3 ?) 31. Nf5+ Kg8 32. Qxg5.
|Dec-26-08|| ||stacase: Move 28 & 29 OK, move 30? I considered backing the Bishop up, but went with 30 Pxf6+ Oh well!|
|Dec-26-08|| ||Lucid Faia: <Once again I'm a little quick to question, I see the queen could be taken.>|
And not only that, but checkmate is inevitable.
|Dec-26-08|| ||Patriot: In my opinion, this qualifies as "insane".
28.Nxe5 dxe5 29.Bxe5+ f6 30.Bc3
So far white is down a knight for 2 pawns and white makes this quiet move. Chess101 teaches that a knight is worth about 3 pawns, so there had better be something good to follow up with. But what does Bc3 threaten that could make up for the material deficit? Nothing at all directly. 31.d6 is one of the threats I missed. This maintains the reason this line should even be analyzed at all--not that it's the only threat. For if allowed: 31.d6 Bxd6 32.Bxf6+ scoops up serious material for example.
Notice that if this initiative can not be maintained, it is white who is technically lost.
|Dec-26-08|| ||JG27Pyth: <at +5.0 Ps to White, but the psychological edge would have been all yours, <Pyth> ;>)|
I've played just that scenario far too many times! It's rather my style... I've mastered the art of seizing the initiative with bold forcing moves that give my opponent no choice but to accept long term positional advantages!
It's quite horrible too, when the truth makes itself clear and the psychological advantage melts like the wicked witch of the west into a shreiking, steaming, widening pool of hopelessly lost variations.
|Dec-26-08|| ||johnlspouge: < <JG27Pyth> wrote: [snip] I've mastered the art of seizing the initiative with bold forcing moves that give my opponent no choice but to accept long term positional advantages! >|
Ahh, I see. You have crossed chessplaying with Woody Allen's skill in the martial arts: "Then, I hit him in the knee with my groin..."
|Dec-26-08|| ||agb2002: White should take advantage of the position of the black knights, far away from the defense of the black king. There are several options, namely 28.Nxe5, 28.Nf5+, 28.Rxh8, etc. After checking them rather superficially, I think I’d try|
A) 28... dxe5 29.Bxe5+ f6 30.gxf6+ Bxf6 31.Rxh8
A.1) 32... Kxh8 33.Ng4
A.1.a) 33... Bxg4 34.Qh6+ Kg8 35.Qxg6+ with a winning attack.
A.1.b) 33... Bxe5 34.Qh6+ Kg8 35.Qxg6+ Bg7 36.Nh6+ Kh8 37.Nf7+ winning the queen.
A.2) 32... Qxh8 33.Bxf6+ Kxf6 34.Ng4+ Bxg4 35.Bxg4 threatening Qf4+ and Rh1 with two pawns and attack for the piece.
B) 28... Bxg5 29.Nxd7
B.1) 29... Nxd7 30.f4 followed by Ng4, Qc3+, e5, etc.
B.2) 29... Bxe3 30.Qxe3 Nxd7 31.Ng4 threatening Rxh8.
Time to post, check and go to bed.
|Dec-26-08|| ||c o r e: I can't say that I <solved> this puzzle. But I decided on 28.Nxe5 in about a minute. I didn't bother to analyze every continuation, but it seemed that white's attack, even if it were sometimes quiet, would be strong enough given the circumstances. Specifically I could not find a way for black to escape the eventual loss of f6, and thus the total collapse of his position.|
|Dec-26-08|| ||njchess: I got this one one if only because there is no other meaningful move for White other than Nxe5. And, as for Black, his responses are more or less forced.|
30. Bc3 is probably the most difficult move to see if only because it doesn't seem particularly natural to move a piece backwards. But the pin must be maintained in order to get the passed pawns, which was the point of Nxe5.
Black doesn't have any real coordinated counter play here, so White can simply press the attack. To be blunt, that pretty much sums up this game.
|Dec-26-08|| ||MostlyAverageJoe: <dzechiel> & <johnlspouge> Yeah, this was pretty much insane. |
It takes many forward/backward slides to get a decent valuation to the < 30. Bc3> move, which ends up eventually around +4.25. The required analysis is proabably about 25 plies deep, and finding that move with my usual engine is pretty much intractable within the parameters I use for evaluations. But I never really worked out the methodology for the higher end of the difficulty scale.
One thing is certain - definitely more difficult than your average Friday puzzle.
|Dec-27-08|| ||zenpharaohs: It won't surprise anyone that Rybka 3 took a long time and a lot of playing into lines to really nail down what happens.|
It really comes down to
28 Nxe5 dxe5
29 Bxe5 f6
30 Rxh8 Qxh8
31 gxf6 Bxf6
32 Bxf6 Kxf6
33 Rh1 Qe8
34 Ng4 Bxg4
35 Qf4+ Kg7
36 Bxg4 Nd3
37 Qh6+ Kf6
click for larger view
Black is busted at this point - White got an attack for his Knight; Black has not much up his sleeve. At this point, Rybka identifies quite several alternatives. None of them are good though. At a depth of 14 plies, the values is given as +4.45.
It took most of the rest of the day to compute through these alternatives.
< Earlier Kibitzing · PAGE 2 OF 2 ·