|
< Earlier Kibitzing · PAGE 3 OF 3 ·
Later Kibitzing> |
| Dec-17-14 | | whiteshark: <33.Nf4+?> Add me to the <lack of accuracy> group. |
|
| Dec-17-14 | | Sally Simpson: Good post Once, just read it.
We chase the same theme. Except I appear to have dug in and back tracked to look at 33.Nh4+ (33.Nf4+ was also my first choice though I admit at first I thought it probably does not matter which square the Knight gives a check on.) "Knights on the rim often win."
Spread the word.
Hi Obit,
"What's better, a mate in six that's a little tricky to calculate, or a three-move combo that reaches a dead won position?" Go for the mate, always go for the mate. The game is over. There are 100's...nay, 1000's of games that have been lost in 'dead won positions'. Never underestimate the power of the human brain to cock things up. |
|
| Dec-17-14 | | TheaN: Wednesday 17 December 2014 <32.?> This puzzle position proves one simple rule. Unless heavily defended, a knight and queen always mate. Coordination between the queen and other pieces is way more precise than with the knight, because they usually interfere. A knight flawlessly augments the queen's line dominance with the only other move in the game, the knight jump. The amazon piece (RNB) in chess variants with different armies would prove this in a similar situation as this puzzle. How does this translate to this puzzle then: the knight is currently not in play. White is aiming at the black king, but the g-pawn is in the way. White can kill two birds with one stone by sacrificing the rook: <32.Rxg6+ Kxg6 33.Nf4+> Nh4+ works too but the mate is a bit more sophisticated: it does keep the king out of f5, but this is no problem <33....Kf5 34.Qe6+ Kg5 35.h4# 1-0> and white proves that indeed, a queen and knight combination is powerful. To get back to what I initially said: an amazon in the similar situation would mate. If <either> white piece at any point in the combination would be an amazon with the other pieces <removed> it would still be mate: the queen or rook: <32.RNBf7#> or the knight: <32.RNBf4+ Kg7 (g5/Kh7 33.RNBf7#) 33.RNBe6+ Kh7/h6 34.RNBf7#>. |
|
| Dec-17-14 | | TheaN: Well that is embarassing. Realizing Nh4+ was mate I still went with 33.Nf4+? and completely missed black's option of <not> going to f5. In a way, I did solve the puzzle saying that Nh4+ mates and I presented the variation in which white does mate after 33.Nf4+. On the board, I would have reconsidered 33.Nf4+ or Nh4+ after the rook sacrifice. Still it is pretty lazy to write a whole comment describing the phenomena and then failing the puzzle. I will keep my initial post up for... illustrative?... purposes. |
|
Dec-17-14
 | | OBIT: <Sally> Yes, and when is the human brain more likely to cock things up, when it is trying to calculate a six-move line with some reasonable-looking moves that don't work (e.g. Nf4+), or when it is up material and maintaining an attack? Also, to those who think the passed pawn is a problem after 32. Nh4 Qa1+ 33. Kg2 Qb1, I see no such issues if White continues with 34. Qf7. Mate is threatened by Rxg6+. If Black plays 34...Nf8, then 35. Qf4+ leads to a quick mate. Anything else Black tries to stop the mate just loses a ton of material. |
|
| Dec-17-14 | | JG27Pyth: Nf4 *sigh* |
|
| Dec-17-14 | | TheaN: For what it is worth, I did see the entire combination up till mate at the 37th move in the Nh4+ line. This is <exactly> the reason I went with the Nf4+ line because in the wrong variation it mates on move 35. This comes back to the recent question "once you see a forced mate, do you need to look further?". Well, not seeing a defense against the forced mate would be another one. |
|
| Dec-17-14 | | dfcx: The first move is easy to see,
32.Rxg6+ Kxg6
Here white has several choices
Nf4+, Nh4+, f4, h4. etc. f4 and h4 can be ruled out since white needs to keep checks going.A. 33.Nf4+ Kh6 34.Qe6+ Kg7 and white has no better follow up than drawing with perpetual checks. B. 33.Nh4+ Kh6 34.Qe6+ Kg7 (Kg5 35. f4#) but now white has 35. Nf5+ and will mate in two. B1. 35...Kf8 36. Qe7+ Kg8 37. Qg7#
B2. 35...Kh2 36. Qh6+ Kg8 37. Qg7#
Good Wednesday puzzle! |
|
| Dec-17-14 | | Sally Simpson: Hi Obit.
Go for the mate. That is why we do these puzzels, to encourage and stimulate out the box thinking. Mating combinations are easiest to play. There are no afters to judge, no stings in the tail. It is over. Get into that habit and you will never feel the burn or experience the misery of losing a game where you had a chance to checkmate an opponent.
I've seen better 'winning positions' than the one that could appear in this game lost. The aim of the game is checkmate.
That is why Baadur Jobava the player of the White pieces here.... 
click for larger view....went for the mate with 31.Qe7+ instead of simply taking the Queen with 31.Qxa7. |
|
| Dec-17-14 | | Castleinthesky: I'll give myself 1/2 credit, I chose Rg6, but Nf4, which leads to a draw. |
|
Dec-17-14
 | | varishnakov: 32.RxP+ KxR 33.N-R4+ K-R3 34.Q-K6+ K-N2 (34...N-B3 35.QxN+ K-R2 36.Q-B6 mate; 34...K-N4 35.Q-N6 mate) 35.N-B5+ K-B1 (35...K-R2 36.Q-N6 mate) 36.Q-K7+ K-N1 37.Q-N7 mate |
|
| Dec-17-14 | | patzer2: There's more than one way to skin the cat for a winning solution to this Wednesday (32.?) puzzle. The fastest route to mate, which I missed, is 32. Rxg6+! Kxg6 33.
Nh4+! Kh6 34. Qe6+ Kg7 35. Nf5+! Kf8 (35... Kh7 36. Qf7#) 36. Qe7+ Kg8 37. Qg7#. However, the computer (i.e. Fritz 12 @ 22/51 depth) indicates White also wins with 32. Ne3 (+16.57), 32. Nh4 (+15.13), 32. Nf4 (+10.21), 32. h4 (+3.82) and 32. Qf7 (+1.64). In my look at the 32. Rxg6+! line, I got all the way up to 34. Qe6+ where Black resigned. However, in looking two moves ahead, I couldn't see the win for White after 34...Kg7 (diagram below) 
click for larger viewwhen White wins easily with 35. Nf5+! Kf8 (35... Kh7 36. Qf7#) 36. Qe7+ Kg8 37. Qg7#. So I switched to 32. Nh4! which might be an even more difficult win. One possibility here (i.e. with 32. Nh4!), as verified with Fritz 12, is 32. Nh4! Qa1+ (32... Rg8 33. Nf5#) 33. Kg2 Qb1 34. Qf7 Nf6 35. Qxf6 Rg8 36. Re7 Be8 37. Qf4+ g5 38. Nf5+ Qxf5 (38... Kg6 39. Qd6+ Kxf5 40. Qe6#) 39. Qxf5 Bg6 40. Qf6 h4 41. Kf3 b2 42. Kg4 b1=Q 43. Qxg5#. P.S.: Really well played game by White. Black seems to go wrong early in the opening with 7...b5?! After 8. 0-0 a6? 9. axb5  White is in control. |
|
Dec-17-14
 | | kevin86: White gives up two rooks to mate, a bargain at twice the price. 32♖xg6 is the quick gateway to mate! |
|
| Dec-17-14 | | patzer2: Maybe 7...b5 isn't so bad. After 8. 0-0 Qb6 (instead of 8...a6?) the Black position is playable. In the only game with this line in the chessgames.com OE, Black managed a draw in the game J Hvenekilde vs H Lahlum, 2006. . |
|
| Dec-17-14 | | BOSTER: <Once: do we play Nh4+ or an f4+?
Understanding that tell us something quite unexpected about the nature of knights>.
My opinion the Knight on h4 or f4,or even on b1 has the same <nature>.
The key word for the solution is <coordination>.
In <POTD> pos. Knight on h4 has better coordination with the queen on e7 than Knight on f4. |
|
Dec-17-14
| | OBIT: <Sally S> I'm sure Jobava saw the whole mate when he played 31. Qe7+, otherwise, as you say, he'd have played 31. Qxa7. I'd have probably played 31. Qxa7 myself, figuring I've just won queen for rook. Most likely, my continuation would be 31...Ra8 32. Qc7 b2 33. Re1 Ra1 34. Qxc6 Rxe1+ 35. Nxe1 b1=Q 36. Qxd7+, and here I'd be thinking, "I'm up a piece and a pawn. No problem winning this." Yes, of course, a forced checkmate is always the best solution. It guarantees the win, and it is the most esthetic finish. In the real world, though, things can get more complicated, especially when the clock is a factor. For example, suppose you have the position at move 32 with 15 minutes on your clock. You see a way to maintain a big advantage. That's your Plan B. You also have a Plan A: that promising rook sacrifice. However, you certainly don't want to make the sac if it doesn't force checkmate. How much time do you want to spend trying to calculate the attack to checkmate? Remember, if you don't find the mate, you need to go back to Plan B, and you'll want to leave yourself with enough time to play out the game. |
|
| Dec-17-14 | | Longview: I saw the rook sacrifice fairly quickly, looking for forced lines. I instinctively, like others wanted to keep coming with Nf4 and discovered the strength of Nh4 second. It is always interesting what comes up when you "look for the better move". I stopped with 34.Qe6 but seeing the payouts, I should not have been so arrogantly lazy. I agree with <Patzer2> that white played well but in going through the initial 7-9 moves I thought Black was the aggressor and White reacted in reasonable fashion. Black seemed to me to have erred when he moved 15....b4 instead of putting his King in a safer spot and getting on with mobilization of the Queenside. I don't know that he could have won but it just seems a better principle to follow. As always, instructive comments guys and gals. |
|
| Dec-17-14 | | Sally Simpson: Hi Obit
I can see where you are coming from and we agree taking a checkmate and posting a no nonsense win is the way to go. Instead of having 15 minutes left, suppose you were down to 2. Then Black may try to time you off the board. The other good thing about checkmate is the clock is stopped. (Infact it does not matter even if your flag drops after you have given checkmate. ) There will be times when the only option is to go for the mate. Getting into the discipline of looking for the best shots will benefit your game. Try not to settle for a technical win when you can wrap it up sooner. That is a bad habit.
The players here who have missed a mate and blown it will be nodding. |
|
| Dec-17-14 | | Conrad93: Okay, the sacrifice itself is vey easy to see, but the eventually mate takes some calculation. The two star rating is very misleading.
I've seen Friday puzzles easier than this |
|
| Dec-17-14 | | Edeltalent: <Chess Dad> <OBIT> <Sally Simpson>:
Yasser Seirawan's take on the matter: "If you have the choice between taking the queen or delivering checkmate - always take the queen, it might not be checkmate!" ;-) |
|
Dec-17-14
| | OBIT: <Edeltalent> LOL on the Sierawan quote. Yasser and I must think a lot alike, as his comments often hit home with me. I don't recall ever losing a game where I was a queen up, so I'm thinking that winning a queen is more reliable than trusting my abillity to calculate a mating combination with 100% accuracy. I can recall a tournament game I played earlier this year where I sacrificed a piece for a wild attack that I was sure would work. A few moves later, my opponent played O-O-O. I had to double-check my score sheet. Really, my opponent can castle here? That move is legal? |
|
Dec-17-14
 | | gawain: A pleasing puzzle. There are some ways for White to go astray, but I found the true path rather quickly. Hooray! |
|
| Dec-18-14 | | Sally Simpson: Hi Obit.
Not convinced regarding the Sierawan quote.
Here:
Gurevich - Seirawan, Estes Park 1987
click for larger viewSierawan had the chance to take the White Queen but went for the mate instead. Final position.
click for larger viewA case of do as I say, not do as I do.
D Gurevich vs Seirawan, 1987
No Kibitizing yet on that page - you can be the first! |
|
| Dec-18-14 | | Kinghunt: <Sally Simpson> Almost certain there is a score error - 34. Qh5 is just ludicrous, as is not exchanging on e5 on the following move. I bet 34. Qh6 was actually played, and the white queen was never hanging. |
|
| Dec-18-14 | | Sally Simpson: Hi Kinghunt.
So the points in favour of not carrying out your job correctly (i.e checkmating the King as per the aim of the game.) is an unsourced quote and a game score error. :) I suspected straight away the score was wrong. It must be. But it is an amazing coincidence. |
|
 |
|
|
< Earlier Kibitzing · PAGE 3 OF 3 ·
Later Kibitzing> |
Kibitzing
