ErictheRed wrote on 05/09/13 at 00:00:26:


But she actually had two chances to find the right move (82.Ne2! and 107.Ng4!) and missed it both times. Equally hilarious is the moment on move 97 where she gets back to the starting position for the W manoeuvre again and hesitates for seven moves before executing it, because she realizes she's about to screw it up again.
Actually, Silman did include a tutorial on the B+N mate in the first version of his endgame book (Essential Chess Endings Explained Move By Move); that's where I learned it from and I still haven't found a better explanation on how to do it. Funny that he changed his mind about it when writing his "complete" endgame book.
Some people have made comparisons between this mate and Q vs R or 2N vs P, but winning the latter two against perfect (tablebase) defence is significantly more difficult than B+N.

Thanks, TalJechin. Lilienthal was it. My memory fails me, unfortunately more often than I'd like too, nowadays.

fling wrote on 05/07/13 at 13:31:52:


Apparently so:
"Even when the position is a theoretical win, it is very complicated and difficult to play correctly. Even grandmasters fail to win it. Andor Lilienthal failed to win it twice in a six-year period, see Norman vs. Lilienthal and Smyslov vs. Lilienthal. But a fine win is in a game by Seitz, see Znosko-Borovsky vs. Seitz (Giddins 2012:26).
Two knights versus pawn is sometimes called the "Halley's Comet" endgame.[3]"
http://en.wikipedia.org/wiki/Two_knights_endgame

From a practical view, all you need is de la Villa's 100 endings. Though confidence-wise it's good to not only know it's winning but also how to win it. She's probably not alone, though, iirc some male GMs have also failed to win it over the years...
 
Not to mention Browne vs the TB in Q vs R, even in the rematch after preparing, he only won the rook on move 50! 
Maybe human thinking rely too much on pawn structure, and without pawns it's harder to orientate oneself?

Btw, these kind of rare endings can occasionally become quite important as they're the stubborn defender's last salvation. Iirc, there was a 2N vs p ending in one of the Karpov-Kortchnoi matches that ended in a draw - not sure if it was winning at some point.

Vass wrote on 05/06/13 at 20:22:23:

Yesterday the Women Worldchampion wasn't able to mate with bishop and knight although more than sufficient time on the clock.

So knowing a lot of endgames isn't a necessity to become (women) worldchampion.  I believe that the endgame has only a minorimpact on the total performance of an OTB player despite what teachers are telling us.

MartinC wrote on 05/08/13 at 13:13:18:


I've often heard stuff like this about studying certain technical endgames.  I must say I've never really understood it.  Do people look at an opening variation once and expect to remember it years later?

As said earlier the traditional way to do the BN v K mate is pretty simple once you know it.  Essentially it comes down to remember three key positions.  Once you know that a few minutes top up every now and then is sufficient to keep the skills sufficiently polished.  I've found that so anyway.

I last did my revision the day after I heard of the World Champion's fail.  It took me half of my journey to work - did it on my phone.  Hardly onerous.

(The revision, btw, was so I could do it in 1-2 minutes max.  I reckon I would probably be successful with a 30 second increment and a small timebank although if and when I have to do it at the end of a long session/tired/hungry/anxious/busting for a wee/being attacked by a swarm of killer bees/whatever that might turn out not to be the case.  I suppose you never really know until you're in the situation)

I wouldn't be at all surprised if I never got to play this out in a real game in the rest of my chess life, but that's not really the point.

ErictheRed wrote on 05/08/13 at 18:49:25:


He did. Having looked at it a bit last autumn I disagree with that. The method to get the defending king from the "wrong" to the "right" corner is so easy almost everyone can learn it, while chasing the king to (any) corner first can be improvized by reasonably strong players. This mate has also gotten this special cultural status as an indicator of endgame skill, which is in itself a reason for learning it.

It seems many people reason that because it's rare and many strong players have failed to win it, it must therefore be very hard to learn. But in this case that's not true.