< Earlier Kibitzing · PAGE 3 OF 3 ·
|Jan-15-08|| ||playground player: It's not a difficult puzzle if you play the Albin Countergambit and are familiar with Lasker's Trap (1.d4, d5 2. c4, e5 3. dxe5,d4 4. e3,Bb4+ 5. Bd2,dxe3 6. BxB4,exf2 -if White plays Kxf2, he loses his Queen--7.Ke2, fxg1=N+ 8.Rxg1,Bg4+, bagging the White Queen). This pattern is very similar.|
|Jan-15-08|| ||crwynn: <Yes, IMO if you didn't notice the 12.cxb7 idea and its refutation then you can't claim to have solved this one.>|
Not exactly - I think 12...Bb7 13.Bb5+ c6 would also win eventually, although it's pretty stupid. Unlike the famous line in the Danish Gambit (it has to be famous if even I know it) 1.e4 e5 2.d4 ed 3.c3 dc 4.Bc4 cb 5.Bb2 d5 6.Bd5 Nf6 7.Bf7 Kf7 8.Qd8 Bb4, regaining the queen in the same way. In that line White sacks a piece to win the Q so Black can do likewise to regain it; not so here.
|Jan-15-08|| ||crwynn: Well, 13...Qd7 actually, why give up the c-pawn|
|Jan-15-08|| ||DarthStapler: Another easy one.|
|Jan-15-08|| ||dzechiel: <MAJ: However, here's a completely different follow-up by black that is kinda cute...>|
Very interesting! Thanks for taking the effort to publish this line with diagrams, I really enjoyed it.
I'm in agreement with you that the human player would not bother to search for this line once the win of the piece was in hand.
|Jan-15-08|| ||benveniste: White can avoid the loss of the queen with 10. Bd2, but it's still a massacre. Instead, black wins at least two full pieces.|
|Jan-15-08|| ||wals: So what to do. 9...f2+ 10.Kxf2 Bc5+ 11.Ke1 Qe7+
all seems a bit long winded
how about 9...Qe7+ 10.Bf-e2 f3xe2 11.Qxe2 Bc8-g4
Thats my best shot................lets have a look
What a pathetic effort....missed the boat completely
|Jan-15-08|| ||penarol: Nice puzzle! I think yesterday´s was tougher (and very nice too!)|
|Jan-15-08|| ||MiCrooks: It is often the case on here that the best defense is not played. However, the idea that the difficulty of the problem is somehow related to how quickly a computer zeros in on the correct solution seems odd to me.|
A computer quickly sees that after Bxc3 it cannot take back and starts looking for the best way to squirm. The human looks at it and says if I can't take back, and I have no threats of my own then I am just dead in the water. So what is easy for a human takes longer for a computer as it tries to decide which of the many relatively equal losing lines is best!
In this case, as in others, if you want to attempt this sort of thing you need to only see how long it takes a computer to find the first move. For a human, this is insufficient as often the "try" is obvious but the proof is in finding the correct follow up.
|Jan-15-08|| ||MostlyAverageJoe: <MiCrooks: ... the idea that the difficulty of the problem is somehow related to how quickly a computer zeros in on the correct solution seems odd to me.>|
It may be odd, but there appears to be a reasonable correlation between the difficulty computed by the engine and the difficulty designated by CG. Furthermore, on many occasions when the puzzle solving crowd's opinion was "too easy" or "too difficult", the engine agreed with such assessment (the most recent case of such agreement happened yesterday).
In very few cases, the perception and engine analysis disagreed (people said it was too easy, while engine had a hard time, or vice versa).
Only Sat/Sun puzzles (and some endgame puzzles) frequently defy the classification with any regularity - usually beacuse they require more computing resources than my method allows.
One of these days, I will finally write the code for automated difficulty analysis and run the results through a more rigorous statistical analysis. For now, I am still evolving the approach, although I've been playing with this idea for about 10 months now.
|Jan-15-08|| ||patzer2: For today's Tuesday puzzle solution, White plays the deflection combination 9...Bxc3+ to remove the guard after 10. bxc3 f2+! and win the Queen after either 11. Kxf2 Qxd1 or 11. Ke2 Bg4+ (skewer tactic).|
|Jan-15-08|| ||pawnofdoom: I think this puzzle was easier than yesterdays. It's always easier to find checks on the puzzles, and the first two checks I considered were f2+ and Bxc3+|
|Jan-15-08|| ||TheaN: 2/2
Why don't I have these nice miniatures XD?
9....Bxc3+ 10.Bd2 with a lost game as: 10.bxc3 f2+! 11.Kxf2 (Ke2 Bg4 12.Kxf2 Bxd1 is actually worse) Qxd1 12.Bb5! (cxb7 Qc2+ with Bxb7) Qc2+! (Qxh1 13.cxb7+ with difficult variations) 13.Kg1 bxc6 wins as 14.Bxc6+ Bd7 15.Bxa8 is no compensation for the lost development and Q<>R.
|Jan-15-08|| ||FSR: White's queen looks like it's about to fall off, but the immediate 9...f2+? (similar to the classic Albin Countergambit trap) doesn't work since White's knight also guards the queen. So let's kill the knight first: 9...Bxc3+! 10.bxc3 (perhaps 10.Bd2 is better, but that would still leave White a piece down and dead lost) f2+! and now 11.Ke2 Bg4+ or 11.Kxf2 Qxd1 12.cxb7 (it's cheapo time) Qc2+! (12...Bxb7?? 13.Bb5+) and 13...Bxb7 wins.|
|Jan-15-08|| ||dzechiel: <TheaN: 9....Bxc3+ 10.Bd2 with a lost game as: 10.bxc3 f2+! 11.Kxf2 (Ke2 Bg4 12.Kxf2 Bxd1 is actually worse) Qxd1 12.Bb5! (cxb7 Qc2+ with Bxb7) Qc2+! (Qxh1 13.cxb7+ with difficult variations) 13.Kg1 bxc6 wins as 14.Bxc6+ Bd7 15.Bxa8 is no compensation for the lost development and Q<>R.>|
After 12 Bb5 Qc2+ 13 Kg1 shouldn't black play 13...b6? This seems to save a rook.
|Jan-15-08|| ||zb2cr: Quick and simple--and as usual, <dzechiel> explained it in the most concise way.|
<keypusher>--It's odd that you had difficulty finding the preliminary capture of the Knight. I guess the explanation is that you were explicitly looking for mating continuations; once more, an illustration of the psychological principle that you can't see a combination that you're not looking for.
|Jan-15-08|| ||zb2cr: Okay, for all those with chess game collections, here's an opportunity to help <chessgames.com> clean up their collection. This game is a near duplicate of the following:|
R Trattner vs Gibbs, 1955
(In the latter case, it's given as a White loss following the Qhich King takes Pawn and ... Qxd1. Following <cg.com>'s admonition re: duplicate games, how can we decide which game score is the correct one?
|Jan-15-08|| ||Eggman: <CRWynn> Bottom line: you have to see the 12.cxb7 idea, and (a) refutation, whether you choose the far superior 12...Qc2+ refutation or not.|
|Jan-15-08|| ||zenpharaohs: OK the correct answer is indeed 9 ... Bxc3
But the rest is not as in the game line.
9 ... Bxc3
At this point black has at least four lines which win (maybe more, at some point you stop looking for more). Simplest and perhaps weakest is
10 ... Qxd2+
11 Qxd2 f2+
12 Kd2 or Ke2
12 ... Bdx2
13 Kxd2 Ne7
14 cxb7 Bxb7
Black is up a knight and white has nothing up his sleeve.
10 ... Bxd2+
11 Qxd2+ Kxd2
which leads to about the same advantage for black.
Better still is
10 ... Qh4+
11 g3 Qe7+
this motif can occur in other lines, but here it is clearest;
12 Kf2 Nf6! (threatening Ne4+, etc.)
13 Qxf3 Bxd2
14 h3 O-O
15 Kg1 bxc6
Black is up knight, bishop, and pawn.
But best of all is
10 ... f2+
11 Kxf2 Qxd2+
12 Qxd2 Bxd2
13 bxc3 bxc6
14 Rd1 Ba5
15 Bc4 Nf6
16 h3 O-O
This is within an eyelash of the previous line's position but the queens are off the board. Since white has been able to get the king out of the center in both of these lines, it's better for black to have exchanged the queens. This line continues
17 Kg1 Bf5
18 Kh2 Rfd8
19 Bb3 Nd5
20 Bxd5 Rxd5
21 Rxd5 cxd5
black has two bishops and a pawn and an easily won game.
So there is no way for black to "win the queen".
The continuation of the game line is probably
10 bxc3 f2+
11 Ke2 Bg4+
12 Kxf2 Qxd1
13 cxb7 Qc2+
14 Kg3 Rb8
15 Be3 Nh6
Black has queen and knight to bishop and pawn.
|Jan-15-08|| ||GannonKnight: Surprised to read that many think that this puzzle is too hard for a Tuesday. It took me less than a minute ... and I missed yesterday's puzzle.|
|Jan-16-08|| ||zooter: <Eggman: <CRWynn> Bottom line: you have to see the 12.cxb7 idea, and (a) refutation, whether you choose the far superior 12...Qc2+ refutation or not.>|
I know its a little late and may not be noticed, but seeing 13.bxc7 Qc2!+ is not really required for a Tuesday puzzle. And anyways, if the position comes to that, it must be pretty obvious that the queen needs to escape with check before capturing the white pawn
And also, as somebody else pointed out, even without playing Qc2+ black still wins by capturing the bishop
|Jan-16-08|| ||TheaN: <dzechiel: After 12 Bb5 Qc2+ 13 Kg1 shouldn't black play 13...b6? This seems to save a rook.>|
Of course. Slight inconsistency in analysing, but I would've played b6 on the board. Not that bxc6 does not win, though XD.
|Jan-17-08|| ||Eggman: <Zooter> I think you're missing the point. You have to solve the puzzle. You don't get full credit for coming up with a sloppy half-solution just because it's Tuesday. If you thought that 11...Qxd1 was the end, then you didn't solve the puzzle. You ought to have noticed the potential discovered attack on the Black Queen and ought therefore to have considered 12.cxb7 (or 12.Bb5, by the way). In a given instance (say, during a real game), a move like 12.cxb7 might be winning, so it has to be seen. After all, in a given instance there might not be a refutation like 12...Qc2+ (or even 12...cxb7 13.Bb5+ Qd7) available.|
|Jan-17-08|| ||Eggman: Further to the above:
For example, after 1.e4 e5 2.Nf3 Nc6 3.Bc4 h6? 4.d4 exd4 5.Nxd4 Nf6 6.Nxc6 dxc6?! (see below)
click for larger view
It makes no sense to play 7.Bxf7+! unless you see that after 7...Kxf7 8.Qxd8 Bb4+ 9.Qd2 Bxd2+ 10.Nxd2! White emerges with an extra pawn. If you played these moves but without seeing 10.Nxd2 coming ahead of time, then you were just lucky - this time.
|Jan-18-09|| ||WhiteRook48: lesson: never open the d-file!!|
< Earlier Kibitzing · PAGE 3 OF 3 ·