Recent B&N mate at GM level.

Originally posted by philidor position
I completely agree with Silman. .

I completely disagree with silman. not only because messing up KNB the rare occasion you get it will make you rip your hair off in frustration (not to mention the smirk on your opponent's face when he sacks a rook to stop your last pawn thinking you can't pull KNB off), but because learning it will improve your understanding of N+B coordination like nothing else.

you don't learn KNB to be able to mate, you learn it because it'll improve your skill & vision in all positions with a knight and a bishop (also when they're not on the same side).

learning both KNB and KQkr have helped my game immensely.

Originally posted by greenpawn34
You should try it v a computer, that is hard.

about that, excluding the typical human piece-dropping blunders, I think both KNB and KQkr are actually easier against engines. why? -because engines are very predictable in both, they'll play almost identically and skip most 'bad lines' so you never ever have to find the often obscure refutation. I can mate fritz or rybka in KQkr in less than a minute (well okay, I screw it up sometimes), but even an average human player can give huge trouble because of the random 'inferior' lines. against a tablebase the mating is even faster because it's even more mechanical.

Originally posted by wormwood
I completely disagree with silman. not only because messing up KNB the rare occasion you get it will make you rip your hair off in frustration (not to mention the smirk on your opponent's face when he sacks a rook to stop your last pawn thinking you can't pull KNB off), but because learning it will improve your understanding of N+B coordination like nothing ...[text shortened]... they're not on the same side).

learning both KNB and KQkr have helped my game immensely.

silman bases his ideia on the fact that there are things more worthwhile to learn, he says it ends up being a waste of time, let me quote him:"When studying this book please remember that I've deliberately left out many endgames. Why? Because I don't feel they are important to players under the 2400 level. For example, I heretically decided not to inclode B+N vs K because it's far from easy to master, and it occurs very rarely in over-the-board play. In fact I've only got it once in my entire career, while IM John Watson and IM John Donaldson never got it at all! Is such a rarity really worth the two or three ours it would take to learn it? I say no. Ultimately, this is what Silman's complete endgame course is all about: Learning what's useful, and devoting the rest of your precious study time to other areas of the game" when he puts it this way I must say I quite agree with him, I mean I have so many more and more important things to learn and I don't want to be stuck studying all endgames for months and months to come! Tactics and strategy that's what I really must study (even though I must say I like studying endgames). Let me add, the book is very well suited for players who are terrible at endgames and really want to improve, and less well suited for those who know all about endings already, its a manual for the starting endgame student, not for masters.

Originally posted by orion25
silman bases his ideia on the fact that there are things more worthwhile to learn, he says it ends up being a waste of time, let me quote him:"When studying this book please remember that I've deliberately left out many endgames. Why? Because I don't feel they are important to players under the 2400 level. For example, I heretically decided not to inclode B+ about endings already, its a manual for the starting endgame student, not for masters.

KNB is not that hard, and silman's reasoning seems more like making an excuse for a stupid omission.


I wonder how he performed the one time he got a KNB? I hope he didn't draw it, because that would force up the only rational advice: "get a proper endgame book and learn it." 🙂

Hi Guys.

Heinzcat and I went to a lot of trouble to get Russ to put on
the feature where you can play out a game from a set position.

I supplied all the pleading and the reason why we needed it.
Heinzcat wrote the code.

It's not difficult to do.
Rememner that you must make it so it is White has to go first.
(Heinz screwed that bit up).

It is an intereting thread but I fear some guys will not want to watch
100+ moves till the position under discussion is reached.
I know you can go to then end click backwards but in this case
that will be over 50 moves.

If you have read the thread you will now know the proceedure.
Let us watch GM Epishin (2567) mess it up from the moment
there were only KNB v K is on the board.

Re: K & Q v K & R.
Agree totally humans will play humans moves but knock off the
tablebase from the box. Let it 'think' for itself.

Walter Browne once lost a $50 bet he could do it in five muntes and lost.
It's not that easy.

OK let's guffaw at Epishin.
He chases him away from the right corner and bottles him up
in the wrong corner. You can almost feel the desparation.

Originally posted by greenpawn34
OK let's guffaw at Epishin.
He chases him away from the right corner and bottles him up
in the wrong corner. You can almost feel the desparation.

oh man, that was painful. 🙂 you'd think he'd realize it can't be done in the wrong corner when the king takes one of the knight squares for checking the corner, and the knight can't hop into the other without obstructing the bishop, but no. he just keeps trying to magically stumble into the mate, like fly buzzing against a window... embarrassing as hell.

now if watching that doesn't get a person to learn KNB, I don't know what will.

Heinzcat and I went to a lot of trouble to get Russ to put on
the feature where you can play out a game from a set position.

It's not difficult to do.

How do you do this for future reference? Is there a better way than cutting off half the moves from the pgn? (the one I copied from chessgames had a lot of bracketed annotations which i didnt want to touch).

I dont think its a major problem (or that time consuming) to learn the W pattern of knight moves to drive a king from the wrong corner to the correct one, I think it would be more difficult to force the king away from the centre in the first place.

Originally posted by mathmo
How do you do this for future reference? Is there a better way than cutting off half the moves from the pgn? (the one I copied from chessgames had a lot of bracketed annotations which i didnt want to touch).

I dont think its a major problem (or that time consuming) to learn the W pattern of knight moves to drive a king from the wrong corner to the correct ...[text shortened]... e, I think it would be more difficult to force the king away from the centre in the first place.

a fen has two relevant tags for setting up a position, the FEN and SetUp -tags.

[FEN "8/8/8/8/4K3/2kn4/5b2/8 w - - 0 1"]
[SetUp "1"]

(clipped from greenpawn's post as an example. also, anytime you see something 'special' in a message, you click reply and view from the quoted message how it was done.)

SetUp "1" is redundant

Originally posted by heinzkat
SetUp "1" is redundant

as I remember winboard choked on its omission (there's no 64bit version so I can't check), so if the pgn is supposed to work 'everywhere' its probably best to add SetUp tag as well though...

So its
[p gn]
["fen here"]

pgn here
[/p gn]

EDIT: i tried it and it works without the setup1.

Thanks

Originally posted by wormwood
as I remember winboard choked on its omission (there's no 64bit version so I can't check), so if the pgn is supposed to work 'everywhere' its probably best to add SetUp tag as well though...

I was talking about getting it to work on RHP, any other place needs the SetUp thingy.

That is very painful for a GM ... He forgot about (... or never heard of) the "W" manoeuvre by the knight.

But then again ... if you haven't studied this endgame and still manage to find the correct moves sitting behind the board, you'll know for yourself that you are a genius.😲

My advice: study this endgame. If you manage to master it ( a good teacher or a good book will do the trick) then it is really interesting to play all kind of variants.

Here is a really cristal clear explanation of this B&N vs K endgame. If you want to master this endgame, then I can advise you to study this vid over and over again.

http://www.chessvideos.tv/forum/viewtopic.php?t=802

You can practice this endgame at;

http://www.chessvideos.tv/endgame-training/bishop-knight-checkmate.php

Originally posted by wormwood
I've had KNB three times in blitz so far (with up to 25s time left), screwed up all of them. the closest one got flagged one move to mate.

also had one Q vs. R mate with 18s left, panicked and totally screwed it up.

it's been at least half a year (probably even a full year) since I've even looked at KNB or KQkr, so I thought I'd go and try a few against the ICC training bots. scored 7/7 wins in KNB, but it was quite slow especially on the first couple of them. the first one took at least 3 minutes, the last ones maybe a little less than a minute. 5/6 wins in KQkr, also quite slow. the one draw was because I couldn't figure out how to crack the 4th rank 'defense' in time. got that one on latter occasions though.

anyway, it was nice to see that the basic techniques have been retained after the initial heavy drilling. maybe not at blitz speed (which I consider to be 10s-30s), but fast enough to ever having a problems in slow chess.

just thought I'd mention it here, for anyone who's thinking about whether the time they've invested in these endings will have a lasting effect or not.

Originally posted by wormwood
it's been at least half a year (probably even a full year) since I've even looked at KNB or KQkr, so I thought I'd go and try a few against the ICC training bots. scored 7/7 wins in KNB, but it was quite slow especially on the first couple of them. the first one took at least 3 minutes, the last ones maybe a little less than a minute. 5/6 wins in KQkr, also ...[text shortened]... t whether the time they've invested in these endings will have a lasting effect or not.

I was wondering about that. thanks.