|
< Earlier Kibitzing · PAGE 2 OF 2 ·
Later Kibitzing> |
Feb-12-05
 | | patzer2: Today's puzzle solution 85...Nf5!
illustrates why two Knights require the assistance of an opposing Pawn to mate.
An important side line, with a trap to avoid, is 85...Nf5! 86.h6 Nc3! (not 86...Nxh6?? = and there is no longer a mate) 87.h7 Nd4 88.h8Q Nb3#. |
 |
| Feb-12-05 | | greystar69: An interesting question is if black got it wrong and allowed the pawn to queen, would he still be able to draw with the two knights. Presumably he would need to position them to defend each other then run his king around them. |
|
| Feb-12-05 | | cuendillar: He most likely would manage a draw. There do be positions in which the queen wins, but not too many. The knight's don't need to defend each other, just keep the enemy king away, while guarded by the king. |
|
Feb-12-05
 | | kevin86: Two knights can win against a pawn,but it is not easy-one knight must block the pawn while the other must work with his king to literally corner his opponent-then,the other knight must race the pawn to finish off the mate-before the pawn can queen and disrupt things. |
|
| Feb-12-05 | | InfinityCircuit: <somethingstrong> Thanks for the information, even if you posted it a year and a half ago. I wasn't aware of the troitsky line. <cuendillar> I agree, but in the end you never know. This is why basic calculation skills are critical in endgames. The same goes for pawn endgames. |
|
| Feb-12-05 | | Whitehat1963: Amazing. I saw a Saturday puzzle in a flash. But I must confess to cheating. Capablanca gives an almost identical puzzle in "Chess Fundamentals." Ironic, isn't it? The pawn dooms the king against two knights, because alone, the king cannot be mated by king and two knights. |
|
| Feb-12-05 | | maxundmoritz: After 69.Ka4 (when the 50 move clock starts ticking) it's # in 80. 69.Ka2 (as played in the game) is #31, 69.Ka3 #71, 69.Kc2 and 69.Kb2 #75. All according to Fritz with Tablebases. |
|
| Feb-12-05 | | aw1988: Actually, I normally don't try solving puzzles here unless I got it from the day before, but here I knew the answer instantly. I think chessgames put this as Saturday because most people would see NN vs P and instantly give up. I think that's the main difficulty here- trying! |
|
| Feb-12-05 | | OktHorse: Just a note: 86... Ne3 also works for the original problem.
87. h6 Nc4 88. h7 Nc3+ 89. Ka1 Nd2 0-1. |
|
| Feb-12-05 | | Backward Development: whitehat:
I LOVE your avatar. almost makes me want to join... |
|
| Feb-12-05 | | who: i don't get it 62.Be2??
why not 62. Nf7 |
|
| Feb-13-05 | | beatgiant: <who>
On 62. Nf7, perhaps Black can try for a mating net with 62...Ne4 63. Nxh6 Kb4, for example 64. Bg4 Nbc3+ 65. Ka1 Nc5, threatening 66...Nb3#. |
|
| Feb-13-05 | | Whitehat1963: <Backward Development> Thanks. And of course it goes without saying, "I don't get no respect" over the chessboard! |
|
| Jul-30-05 | | blackjacki2: Isn't this game theorteically drawn? I'm guessing white made some error that got him mated? |
|
| Aug-11-06 | | jdoliner: <blackjacki2> no i think this endgame actually is a win for the K-N-N. Provided that the pawn doesn't start too close. I've let a computer play it out before and it won with the two knights |
|
| May-14-08 | | who: <beatgiant> good point. What about 62.Bf3 |
|
Jul-09-08
 | | 4tmac: It seems to me that this is a win even though it is a win in 80 after move 68 because white will be forced to advance his pawn to avoid mate at about move 100. (30+ moves after last capture) Also, both players move 69 were poor. WHITE was correct to "stay away from the KNIGHTS" by moving to the A file but he should stay away from the corner with 69. K-a4! After 69. K-a2?! I don't know why BLACK would not play K-c3 (K-e3?) |
|
| Feb-08-09 | | WhiteRook48: obvious finish |
|
Apr-27-09
 | | David2009: 
click for larger view
This is the position after Matanovic (Black) has played 69... NxNa6. Bisguier lost this ending because (a) he was unaware of danger early on (e.g. he voluntarily played Nf7 x Ph6 to reach the diagram position) and (b) he sought shelter in the wrong corner. With the BN at h6 and WP at h5, he
can be mated when confined to the a1 corner but (with reasonable care) not when confined
to the a8 corner. Reason: it takes 3 moves for the Na6 to check a K at a1, but 4 moves
to check a K at a8. If he had headed for the a8 corner he can still be mated with best play,
but the win is really hard to find over the board. The losing moment came at move 76: Ka3?
lost quickly by force, Ka5! is a much better try.
With best play on both sides (Nalimov data base http://www.lokasoft.nl/tbweb.htm) one reaches 
click for larger view with Black to play (White has just played Kc8-d8).
The winning move is 90 ... Nf5! 91 h5-h6(best) Nh7 and with best play on both sides White returns briefly to the a8 corner;
is winkled out by Nd7+ when at b7 with BK at b6; is driven towards the h8 corner;
is offered but refuses the Nh7 as a sacrifice; and is finally mated in either the h8 or h1
corner. The final stage is similar to that given by Resignation Trap in Game Collection: Two Knights versus Pawn. The whole ending narrowly avoids the 50-move-rule draw because the Pawn on h5 has to move Matanovic played the ending logically except for one lapse: after 69 Ka2? in the first diagram, Kc3! confines the
K near a1 whilst Ke3? as played should have let White escape to a8. |
|
Jul-26-09
| | 4tmac: Early on in this interesting ending, it appeared a1 was the better corner. This corner also has the advantage of some lines "Q with check" (g Pawn). Here, the Nh6 is "closer" to a1 than a8 & no check. Very tricky. |
|
Jul-26-09
| | 4tmac: With best play after move 75 (76.Ka5) seventysome ply later(!) 
click for larger view ...Ke6! is the only move to win. White's best is Kh5 heading for h1. In fact if he ventures too far back up the h file 
click for larger view ...Kf3! wins quickly by forcing white to h8 using only the King. |
|
Oct-08-09
 | | GrahamClayton: <somethingstrong>the longest win for king & 2 knights v. king & pawn is 115 moves with perfect play. occasionally, exceptions to the 50 move rule apply in these type of situations. additionally, the troitsky line (a4,b6,c5,d4,e4,f5,g6,h4) describes the farthest a black pawn can be advanced while being blocaded by a knight and white still has a forced win no matter where the kings are. Somethingstrong,
I think that the exceptions to the 50 move rule have been abandoned - the 50 move rule applies to all endings. Here is an excellent summary of the "Troitzky Line":
http://www.chesscafe.com/text/muell... |
|
Oct-08-09
| | GrahamClayton: <cuendillar>He most likely would manage a draw. There do be positions in which the queen wins, but not too many. The knight's don't need to defend each other, just keep the enemy king away, while guarded by the king. cuendillar,
The worst situation for the player with the knights is when they are self-defending themselves, and can't move, eg 
click for larger viewWhite wins by 1. d6  c8 2.e7 b7 3.d8 a6 4.c7 b5 5.d5 ka5 6.b2 a6 7.b3 a5 8.b1 a4 9.c6 a3 10.b5 |
|
| Oct-08-09 | | nescio: One of my friends finally found out what happened during this strange game which intrigued us for years. Donner describes why Bisguier gave away the game this way in the article "A hole in the board" in Schaakbulletin no. 159, February 1981. The article can also be found in Donner's book The King, page 340. |
|
| May-23-11 | | scutigera: <Graham Clayton et al.>: The current Laws of Chess at the FIDE website include no exceptions to the fifty-move rule. Perhaps they'll change it again one day, but the long-term trend is to hurry things along. |
|
|
< Earlier Kibitzing · PAGE 2 OF 2 ·
Later Kibitzing> |
and 
buttons, but it's much easier to simply use the left and right arrow keys on your keyboard.