| |
| From | Message |
|---|
|
 Creativity in Endgames...
It has long been my belief that endgames have a lot more to offer in terms of interest, excitement, and opportunities for creative ... erm ... creativeness, than many, especially beginners, appreciate. True, the gods in their wisdom placed before the endgame the opening and middle games, but they also, be it noted, made the endgame very often much longer drawn out than the other phases. If the game ends before the endgame, that is a signal that someone was decisively outplayed early on. An endgame struggle bespeaks a much more evenly balanced fight, and hence the more interesting in terms of a battle than, say, a middle game win, which can be revelatory in terms of discovering tactics and how pieces work together.
Endgames are not without their combinations, but I thought I'd begin this thread with a pick-up game played several years ago. Black (Me) had played fairly indifferently in the opening and had reached this position after 22 moves:
w
Black is two pawns down, but all his pieces were aimed towards the White King. After White's move, the game flared up into a wild tactical melee:
23.Re4?! Qa3+ 24.Kb1 Bxb3!!
Black's assault columns come storming in!
25.cxb3 Rxb3+!!
Has Black gone completely crazy??
26.axb3 Qxb3+ 27.Kc1 Bxc3!
White's King defences have been completely stripped. How can he save the homeless monarch now?
28.Rd8+!! ...
Only by decoying the enemy rook away from the c-file...
28...Kg7!
But Black isn't complying! A savage battle now rages on the c- and d-files.
29.Red4 Rc5 30.R4d5! Rc4!
31.Qxc3+!! Rxc3+ 32.Nxc3 Qxc3+
As so often happens, the tactical battle has burnt itself out, and the game enters an endgame that is, materially at least, dead level: 2R plus 3 pawns vs Q plus 4. Black has something of the initiative, and the extra pawn is passed, but it seems hard to believe this will end in a decisive result. Nor does it look all that likely to be full of incident, but ... well, let's see.
w
33.Kd1 a5
The pawn advances whilst the opportunity to do so exists...
34.Rd2 a4 35.R8d3 Qb4
36.h3 a3 37.Ke2 ...
It looks as though White has successfully halted the a-pawn's onrush. What can Black do now?
37... Qe4+ 38.Re3 ...
A King move would have led to 38...a2!
38... Qc4 39.Kf3 a2
40.Rd1 ...
With his back to the wall, White finally halts the a-pawn, and has merely to arrange for its capture. How can Black create more chances? Black would kinda like to play ...Qb1 right now, but of course that isn't possible. But the idea does lead to the discovery of a remarkable manoeuvre not altogether unknown in endgames: the "ladder". But instead of ascending the ladder, this Queen descends:
40... Qc6+! 41.Kg3 Qc7+!
42.Kf3 Qb7+! 43.Kg3 Qb8+!!
A finesse in the hope that the King would return to f3, whereupon Black plays 44...Qb1!
44.f4 ...
"Nothing doing!" says White.
44... g5 45.Rf3 gxf4+!
The pawn on f4 may not be retaken, on account of 46.Rxf4?? Qb3+ winning the d1-rook.
46.Kh2! Qb2 47.Rff1 f3
Now Black is attacking on both wings. Already it is clear that White's chances of survival have become very slender over the last half-dozen moves...
48.Rg1 Kh6 49.Kg3 fxg2!
Even here, the pawn is inviolate. White carries on the fight as best he may, but he is already quite lost.
50.Rge1 Kh5!
Hoping to induce 51.Rd5+ f5 52.Rxf5+ Kh6, but White is not having with that!
51.Rc1 f5 52.Red1 Kg5
53.h4+ Kh5 54.Ra1 ...
Here it is not easy to see how Black can progress. Black's "non plus ultra" line of rooks appears to have stopped Black's pawns cold.
54... Qb7!
Naturally, White has to leave alone the a-pawn (55.Rxa2?? Qb3+). Hence his choice of move...
55.Kh2 Kxh4 56.Rxa2 ...
At last falls the paladin of the a-file, but others remain to carry on the fight.
56...Qf3
All's set for the final assault upon the White King. White could have resigned here but played it out until the result was well beyond doubt...
57.R2d2 Qf1!
With the idea 58.Rxf1 gxf1=N+ 59.Kg2 Nxe2 and wins. That would have been a more fitting conclusion, one feels...
58.Rxg2 Qxd1 59.Rf2 Qg4
60.Rg2 Qxg2+ 61.Kxg2 Kg4
1-0 - Finally tossing in the sponge.
Over to you...
|
|
Big endgame cinema! :)
Wow, what a battle! In the first diagram, my impression at once was to fear for
the white king: all batteries of black are aimed at his home and it was very
consequent to do the attack as you did.
In the second diagram, I think the queen is in advantage. The two rooks have
to remain on the d-file. They need to protect each other in order not to fall for
a fork of the agile queen. As the black king is snug in his fianchetto castle, there
is little danger of a rook check, the only chance to rearrange the rooks. And black
can move freely with the queen, attack the white king and forward the a-pawn.
36.h3 seems to be a big blunder. White should have played 36.Kc1 in order to
approach the king to the pawn, going for move repitition. Maybe white still was
in the illudion he was winning??! Or he was enjoying to get crushed in such a
beautiful way...
A great game, ionadowman, very exciting and inspiring endgame. I hope I can
make my addition soon, I have to look through my games for a nice endgame,
although it for sure will not reach that level!
|
|
Thanks, brulla -
... for your comments.
Quite a few of my annotated games feature endgames that I found interesting or exciting or challenging at the time I played them, but I don't think I will post any of those here. There are other endings though that I think I will show later...
But I will post the occasional Master endgame (I have one or two at least in mind), and the odd endgame study. If there is one thing I would like GK to add to this site it is the endgame study, separate from but associated with the mating puzzles. I can see, however, that such a thing would be no simple matter to administer on a site like this.
I would be very interested in seeing any endgame that you felt proud of, even if it were just one single move that made it special.
Cheers,
Ion
|
|
Even the simplest ...
... of endgame positions can lead to intriguing play. Consider this study by Richard Reti.
The position id one that could have arisen from actual play.
w
There are just 4 men on the board: K and P vs K and P. However, whereas the White pawn can be caught easily by the Black King, even with White to move, Black's pawn, it would appear, can just run on home to queen. Surely then White is busted?
Well, no. White can draw this, amazing as it seems. He begins, as you would expect with a King move:
1.Kg7 ...
One of the features of this endgame is was it tells us about the geography of the board. As will be inferred by what happens, neither of the alternative K moves are any good...
1... h4 (Natch!)
2.Kf6! ...
Here we reach - perhaps more literally than usual - the crossroads. Suddenly the WK is threatening to link up with and guide to promotion his c-pawn. For instance:
[A] 2...h3
3.Ke6 h2
4.c7 and then 4...h1=Q 5.c8=Q+ or 4...Kb7 5.Kd7 etc;
[B] 2...Kb6
3.Ke5!
and now if Black captures the c-pawn, White runs down the h-pawn ( 3...Kxc6 4.Kf4)
or else:
3... h3
4.Kd6 h2
5.c7 etc.
White draws the game!
Neat eh?
|
|
Very famous...
...but always surprising over again. Thanks, ionadowman!
|
|
The Art of Analysis...
It was in a chapter under that title that Paul Keres gave a fascinating account of exploring what had seemed to be an advantageous endgame until it reached a point at which his (Black's) winning chances seemed to vanish away. The game Szapiel-Keres had reached the time control with the following position and White to play:
w
Black's far advanced passed pawn is a definite asset, but his rook protects it from in front - not the ideal place to do so. Meanwhile, White has some nasty threats on the K-side.
Without going into a blow-by-blow account of Keres's search for the win, suffice to say that after lengthy and frustrating analysis he could find nothing better than the following continuation after:
41.Rh8+ Kg6 42.Ne5+ Kf5 (decided on after careful consideration of the alternatives)
43.f3 Kf4 44.Ra8! ...
At this point in his analysis, Keres hit what seemed to be an insurmountable wall. His planned 44...Ke3 foundered upon 45.Rxa3+! Rxa3 46.Nc4+ Kxd4 47.Nxa3 and kiss my hand to any winning chances. Nor did advancing the pawn do any good: 44...a2 45.Ra3 Nc3 46.g3+! Ke3 47.Rxc3+ Kxd4 48.Ra3 Kxe5 and the pawn plus in this rook ending is probably, according to Keres, insufficient.
With everything clear cut, and with his winning hopes crumbling to dust, Keres "suddenly hit upon a paradoxical thought that would never have come to anybody at first glance, namely, the sacrifice of the one pride and joy of the Black position, the pawn on a3".
44... Ke3!!
Here we enter what Jimmy Hatlo used to call the "How-cum Dept". "How-cum" Black's only chance to win came by voluntarily entering a knight and pawn ending a pawn down, and (get this) all the pawns on one side (the Kings') of the board? Let the sequel show;
45.Rxa3+ ...
White had no choice but to go into the line Black had in mind. After 45.Nc4+ Kf2! 46.Nxa3 Ne3 Black wins in short order. In this line, 46.Rxa3 transposes into the game continuation.
45...Rxa3 46.Nc4+ Kf2!
47.Nxa3 Ne6!
The point. Black will smash up White's K-side pawns whilst his opponent struggles to reactivate the distant knight and whilst the enemy king remains imprisoned in the corner.
48.f4 Nxg2 49.f5 exf5
50.d5 ...
White's best chance lay in freeing up his d-pawn to advance but it's not quite enough to save the game, owing to his King's unfavourable placement:
50...Nf4 51.d6 g5...
According to his own account, the GM ended his analysis here, as he considered the position winning for Black. For example, 52.d7 g4! 53.hxg4 fxg4 54.d8=Q g3+ 55.Kh1 g2+ 56.Kh2 g1=Q# - a very neat finish!
In the event, the actual game continued until Black's 51st, whereat White tried :
52.Nc2 g4 53.hxg4 fxg4 54.Ne1 ...
Hoping for 54...Kxe1? 55.Kg3 with a draw. Black wasn't having any...
54...g3! 55.Kh1 Kxe1 0-1.
Game over: 56.d7 Kf2, with the familiar finale, or 56.Kg1 h5 57.d7 Ne6 58.Kg2 h4 and the win is straightforward enough, e.g. 59.Kh1 Kf2 60.d8=Q g2+ (Not 60...Nxd8?? Stalemate) 61.Kh2 g1=Q+ and mate next.
I don't know about you, but I find this sort of thing fascinating.
Cheers,
Ion
|
|
Karsten Müller has a monthly column on chess cafe called "Endgame Corner". In this column he buries the readers in variations with very little verbal explanation of what is going on. Perhaps we could ask questions about the positions in his columns. Mark Dvoretsky also has a column called the Instructor, when Dvoretsky features an endgame, I always learn something because of his verbal skills. Both have endgame books out, if their endgame books are anything like their columns; I know which one I would buy.
|
ericaalders
21-Jul-09, 12:21
|
This post was deleted by ericaalders on 04-Sep-09, 14:06.
|
|
Very informative on ...
...Knight vs "bad" Bishop. Actually, one could learn a lot of general priniciples from that one endgame.
I see that there are some useful elementary endgame lessons for beginners on the site recommended by ericaalders. In the two years "coaching" 10-12 years olds at a primary school several years ago, I was struck by how few knew how to checkmate with K and R vs lone K, or, having 4 Queens vs lone King how to avoid Stalemate...
When one of the stronger players had extreme difficulty finding a win with a Knight and 2 rooks against a lone King, I began to wonder if I was doing any good... Of course, the right plan was to move the knight out from underfoot... :)) Hey, ho...
Cheers,
Ion
|
|
Here is an exercise from chess cafe
How to demonstrate the bishop’s superiority despite the wrong rook’s
pawn?
E99.02 Choudhary,J (2187) - Nitin,S (2365)
7th Parsvnath Open New Delhi IND (5), 14.01.2009
the answer can be found at:
www.chesscafe.com
|
|
Chess ending
I recently played an endgame where an advanced pawn and bishop were matched against a rook.
I was playing Black and I had the difficult task of winning such an endgame. Black's prospects don't look very great, he has to supervise b8 at all times, he also has to be very careful with trading pawns because losing the a5 pawn in a wrong circumstance can mean certain loss.
42.Kc4 Ke8 43.Kc5 Kd8 44.Kd6 Rb6+ 45.Kc5 Kc7
The first step of the plan is complete. Now the King guards b8 and the Rook is free to roam the board.
46.a4 Rb2 47.h3 Re2 48.Kb5 Rxe5+ 49.Ka6
White threatens to take on a5 and start the steamroll. Black has to be very cautious if he wants to win.
49...Kb8 50.Bd7 Rd5 (better would have been Rf5 with a faster win) 51.Be8 Rf5 52.Bc6 g5 53.Be4 Re5 54.Bc6 g4 55.hxg4 hxg4 56.Be8
The constant threat to exchange pawns is very real and provides good practical drawing chances for White.
56...Rf5 (finally, deciding to liquidate things) 57.Bc6 (the a5 pawn is White's real target) Rxf6 58.Kxa5 e5 59.Kb6
White was successful in creating the steamroll but the rook manages to hold it back.
59...Rf2 60.Bb5 (preventing Rb2+ and threatening a mating net with a4-a5-a6-a7) Rxg2 61.a5 Rb2 62.a6 Rxb5+ (after paving the way for the g-pawn to safely queen, the rooks exchanges itself for the bishop to remove the mating threats.) 63.Kxb5
The last chance to go wrong. After 63...g3 or 63...Ka7, White wins. 63...Kc7 is the only move for Black. After this, White resigned.
|
|
Sweet ending...
... nyctalop. At several points Black's path to the win was very narrow - at several points, just one move.
I also liked the exercise mentioned by blake78613. The first move is easy to find; the second Bc6 no so simple. It has to be said, though, that despite the bishop being the "wrong" bishop, were there no Black knight in the position, White would still win. The reason is that Black has no access, or can be deprived of access, to the h-pawn's queening square.
For interested beginners who might be wondering what we're talking about, the type of ending in question is something like this:
w
If, say, 1.Kg5 Kh8 and White has no way of digging the BK out of the corner. With a bishop that could reach h8, White's win would be easy enough, but now:
2.Kh6 Kg8 3.Bd5+ Kh8 4.Bc6 Kg8 5.h5 Kh8 etc.
A rather surprising draw if you haven't seen this sort of thing before: A whole bishop and pawn ahead, and no win.
You would be surprised how much tactics can be found in "pure" pawn endgames. Here is a situation that arose in one of the last OTB games I ever played. The ending, in which I had the Black pieces, came down to this:
w
My opponent was a strong, well known player in Canterbury, but it is clear here that Black has a big advantage. So I was surprised at his next move:
52.Kc1? ...
I can only surmise that White thought I would play 52...Kd3? 53.Kd1 after which Black has all sorts of problems to overcome if he wants to win: 53...Ke3 54.Ke1 d3 55.Kd1 d2 56.b4! cxb4 57.b3 and if 57...Kd3 the position is Stalemate. Once you know the concept of "The Opposition", though, the right move is easy to come by:
52... Ke3!
53.Kd1 Kd3
54.Kc1 Ke2
... and White resigned, as the d-pawn is home free.
As the query attached to White's 52nd indicates, he had a much stronger defence: to take up the Opposition himself:
52.Ke2! ...
Now, since the Black King can not improve his position, a pawn move must be indicated:\
52... d3!
53.Kd2 Kd5
54.Ke1! ...
Amazingly enough, this allows Black to take up the Opposition! Why? It's a tempo thing. Had White played 54.Kd1 Ke3 55.Ke1 then 55...d2+ 56.Kd1 Kd3 Zugzwang! 57.b4 cxb4 58.b3 Kc3 and Black wins easily. So, to proceed:
54... Ke3
55.Kd1 d2
56.b4! cxb4
57.b3 ...
Haven't we seen this before? Yes. Recall the possible line (52.Kc1? Kd3?) in the actual game. So, how do we avoid the stalemate then?
57... Ke4!!
Only thus. Everything else draws. Yet it is not immediately clear how giving up the passed pawn - apparently for nothing, be it noted - will win!
58.Kxd2 Kd4
59.Kc2 Ke3
and now it is obvious that the White King will soon be driven away from the b-pawn, whence Black will win, something like:
60.Kc1 Kd3 61.Kb2 Kd2 62.Kb1 Kc3 63.Ka2 Kc2 64.Ka1 Kxb3 and wins.
I've always rather regretted that Dave didn't find this defence. Would I have found the winning move 57...Ke5!, then? I don't know.
Cheers,
Ion
|
|
Even grandmasters get it wrong...
The following diagram is from a game GM L. Ljubojevic vs GM W. Browne, Amsterdam, 1972.
b
Black to play: what should the result be?
Browne played 1...f5, and his opponent successfully reeled in the pawn:
2.Kb4! ...
And now if 2...f4 3.Kc3; or 2...Kd5 3.Kc3 Ke4 4.Kd2 Kf3 5.b4 and White will queen first and draw. So, after Ljubojevic's response, both sides agreed the draw.
Browne missed the win! Maybe he thought 1...f5 was good enough - a miscalculation; or possibly he thought the position offered no more than a draw anyway. What that shows is that even in the simplest of positions, GMs can, every now and then, go wrong. Correct was:
1...Kd5!
Now White can choose [a] to bring his K across (2.Kb4), or [b] advance his own pawn (2.b4)
[a] 2.Kb4 Kd4!
stops the WK cold. After:
3.Ka5 f5 4.b4 f4 5.b5 Kd5!!
A crucial finesse! Instead, 5...f3 only draws (6.b6 f2 7.b7 f1=Q 8.b8=Q Qa1+ 9.Kb6 Qb2+ 10.Kc7 =!) But now it will be White who queens first. How will that benefit Black? Let's see, then.
6. b6 Kc6! 7.Ka6 f3 8.b7 f2 9.b8=Q ...
It's no good promoting to a knight, even with check. That won't stop Black promoting to queen.
9...f1=Q+
Check! White must move his K, and, as 10.Ka7 gets mated instantly -
10.Ka5 Qa1+ 11.Kb4 Qb1+ 12.Kc4 Qxb8 and wins easily.
[b] Now, what happens if White advances the pawn at once?
2...b4 f5 3.b5 f4 4.b6 Kc6!
transposes into line [a], with the same result:
5.Ka6 f3 6.b7 f2 7.b8=Q f1=Qch 8.Ka5 Qa1+ etc.
It is the hidden resources like ...Kc6! and especially 5...Kd5!! in line [a] that makes endings like this so special.
Cheers,
Ion
|
|
I have a knight endgame for you against Coach Ion!
Here is a game between me and Coach Ion, one of my better efforts against him. It is a game that finished last night of which I was guilty of premature resignation. Here is the game.
gameknot.com
Hope you have a good read.
Bye for now.
Joanne
|
|
Check out the final position...
... Unfortunately, Joanne resigned that knight endgame just when it was getting interesting. But some of the analysis offered by nyctalop and myself by way of comment reveals some of the fascinating possibilities that might have been realised had the game continued. At the time, I thought my winning chances only fair, but I hadn't then thought very far ahead, but nyctalop's commenntary has persuaded me that they were rather better than 'fair'.
Meanwhile, something a little different. Material becomes generally a big issue in endgames, such that whoever is the material ahead is usually the one with the initiative and the winning chances. It is the exceptions that can make the whole deal much more intriguing. A pawn sac in the endgame - given the promotion prospects foregone - might well be the equivalent of a piece - even a queen - sac in the middle game.
Here's an obscure ending from a game played in Czechoslovakia, 1972. Black has one extra pawn for White's bishop. Can Black achieve anything with his advanced pair of pawns?
Piklni vs Serbrnik
b
Doesn't seem like it, does it? Especially as White has a passer of his own...
In fact, Black went for broke:
1...Kc3
2.Bb5 a2+
3.Ka1 Kc2
4.Ba4! ...
And Black found himself with nothing more to do on the Q-side. Whatever he tried, White takes the b-pawn:
4...Kc3 5.Bxb3 Kxb3 6.d6 and White wins easily.
Was Black lost all along? No. His attacking idea was really the only one to save the game, and in fact it ought to have won!
1...Kc3!
2.Bb5 a2+
3.Ka1 Kb4!!
This was the move! 3...Kc2?? was a losing blunder (overlooking the retrograde bishop move - the most common oversight).
Now, White can try:
[a] 4.d6 Ka3 and Black mates next move (5.d7 b2#), one of those rare K and P mates.
[b] 4.Bc4 Kxc4 5.d6 Kc3 6.d7 b2+ 7.Kxa2 Kc2 8.d8=Q b1=Q 9.Ka3 Qb3#.
[c] 4.Ba4 Kxa4 5.Kb2 Kb5 whereupon Black reels in the d-pawn, whilst the White king remains entangled in the two Black passed pawns.
Any other try by White is just Variation [a].
Cheers,
Ion
|
|
A nice adapted study by Grigoriev
White to play and win.
The first move is virtually forced, anything else loses. 1.a3! The point behind this move is to prevent Kb4, capturing the white pawns. Black has to try 1...h5. Now after 2.Kh4 Black has to capture the a-pawn with 2...Kxa3. Play continues 3.c5 Kb2 4.c6 a4 5.c7 a3 6.c8Q a2 (This would be a draw if Black didn't have the h-pawn, not that factor works against him) 7.Qb7+ Kc2 8.Qa6 Kb1 9.Qb5+ Kc2 10.Qa4 Kb2 11.Qb4+ Kc2 12.Qa3 Kb1 13.Qb3+ (Now Black can either give up the pawn with 13...Kc1 or allow the following mate) 13...Ka1 14.Kg4 h3 15. Qc2 h2 16.Qc1#
|
|
Fantastic key move...
... and the closing mate is equally piquant (though I think the sequence should read 14.Kg5 h4 15.Qc2 h4 16.Qc1#). Nor can Black try the 'lets hide in the corner' trick after 7.Qb7+, as White still brings the Q gradually down to b3: 7...Ka1 8.Qe4! (cutting corners - the Q could zigzag more slowly but equally surely with 8.Qc6) 8...Kb2 (only move available) 9.Qb4+ Ka1 10.Qa3 Kb1 11.Qb3 Ka1 12.Kg5 etc.
I really enjoy endgame studies - let's have more of these!
|
|
Another endgame study by Grigoriev
White to play and win. Things look pretty bad for White. Black has 3 passed connected pawns on the Q-side and it doesn't seem White can stop them from queening.
1.g4. Preparing to create a passed pawn on the K-side. 1.h4 losses to 1...c4 2.Ke4 h5.
Black answers with 1...Kb7. If 1...a5, then 2.Ke4 a4 3.Kd3 Kb7 4.h4 Kc7 5.h5! Kd6 6.g5 a2 7.Kb2
With an inferior position than the main line.
2.h4 a5 3.Ke4 Kc7 4.h5! The key winning move. Kd7 5.g5 Ke6 6.gxh6 Kxf6 7.Kd3 a4 8.Kc2 c4 9.Kb1 b5 10.Kb2 b4 11.Kb1!
Black is in zugzwang. He can't move his King or else White queens and if he advance his pawns, White will block and capture all of them.
|
|
Remarkable!
I've not seen that one before. I had a few tries with Black's trying to ram home the pawns really quickly, without move his K at all - but the Wk intervenes just enough, and the old g4-h4-h5!-g5! sequence just carries on as and when it can.
That K vs treble-pawn confontation is, of course, fairly elementary though interesting endgame motif. (I do hope nyctalop that you and I aren't the only ones reading this thread!) Recall this from Chess Primers?
w
White to play; what result?
White wins with, yep, you guessed it
1.Kc1!
Now, the Black King dare not move (e.g. 1...Kf7 2.h7 and variations thereof), a piece of knowledge to store away as part of your technique. Also to be stored in the memory is what happens to the pawns.
[a] 1...b3 2.Kb2 d3 (2...c3 3.Kxb3 etc) 3.Kc3 d2 4.Kxd2 b2 5.Kc2 c3 6.Kb1 and Black's remaining pawns throw themselves upon the sword.
[b] 1....c3 2.Kc2 b3 3.Kxb3 etc
Once the Black pawns have been eliminated (continuing from [a] 6...c2 7.Kxc2 b1=Q+ 8.Kxb1 ...), Black is in Zugzwang. The King must move, the unblocked pawn advances, the King is unable to reach it.
One of those "essential" pieces of endgame knowledge that you never know when will come to your aid.
|
|
no no
you're not the only two in this thread!
Cheers and thanks for great endgame material!
Peter
|
|
It's really disappointing...
... that no one else seems to be interested in contributing to this thread. I would have thought there would be something for everyone here, if they were willing to add something of their own.
Here's something that beginners might like to add to their repertoire:
w
In this diagram, white is to play. What result? It looks as though it ought to be a draw, doesn't it? Material level, position sort of symmetrical, the lone pawns blocking each other.
But it's not so simple.
Suppose White were to play
[a] 1.Kd5 ...
then after Black's instant rejoinder:
1...Kb4!
White finds he must lose touch with his own pawn, and hence lose the game:
2.Ke4 Kxc4
and Black wins. After 1...Kb4, White is in Zugzwang, and must lose ground.
Does that mean he is in Zugzwang in the diagram position? By no means. White has a better move:
[b] 1.Ke5!! ...
Now you can see White's cunning plan - if Black plays 1...Kb4, then 2.Kd5! places Black in the same Zugzwang White faced in line [a]. So, let Black have an equal share of low cunning and play:
1...Ka4
'Come on then, White: attack my pawn!' says Black. White does.
2.Kd6!!
Of course, having read so far, you will have seen that 2.Kd5?? loses, but White has now a choice. Black is already lost:
2...Kb4
3.Kd5! Zugzwang!
3...Ka4
4.Kxc5 and wins.
OK, then, does that mean the the diagram position is winning for White? Actually.... not yet. There is one line we haven't tried for Black:
[c] 1.Ke5 Kb6
2.Kd5 Kb7!
3.Kxc5 Kc7
Now, White can grovel around trying to pick up a tempo, but Black, with care, can (and must) always ensure that the moment White takes the c-pawn, Black can occupy c7 with the King.
This position is a known draw -
w
White (to move) can make no progress from here. Example:
4.Kd5 Kd7 ('Opposition')
5.c5 Kc7
6.c6 Kc8
7.Kd6 Kd8
8.c7+ Kc8
9.Kc6 Stalemate.
Let's look back at the original diagram position again, and imagine the play went:
1.Ke5! Kb6! (1...Ka6! is also playable for Black)
2.Kd5 Kc7? (the second diagram position has been reached with Black to play)
Now Black loses:
3.Kxc5 Kd7
4.Kb6 Kc8
5.Kc6 Kd8
6.Kb7 Kd7
7.c5 ...
and White will shepherd the pawn home.
In conclusion: the diagram position is with best play a draw, but there are plenty of ways for either side to go wrong. The position is by no means dead!
Now, suppose the whole position were moved a rank towards the black side of the board:
w
White to play: what result?
All considerations remain the same, except that the line analogous to [c] proves to be no defence for Black. The position reached -
w
...is a win whoever has the move. Of course, Black to move is Zugzwang, but White to move goes (if you haven't seen this before)
1.Kd6 ... (1.Kb6 is just as good)
1...Kd8
2.c6 Kc8
3.c7 Kb7 (forced)
4.Kd7 and wins.
Here endeth the lecture. I do hope somebody gets some benefit from it...
Cheers,
Ion
|
|
Ion,
I think the great challenge to master chess is the ability to have a good three dimensional thinking. What I mean here is to clearly see the future board situation before pieces are moved to its location.
If the readers were chess beginners I believe they would have difficulty in following this thread. Take me for example, when I see the last position you posted, I can see the solution in less than 2 seconds (knowing that there is a zugzwang potential in the position), but to follow the notation (without a board) that’s a real challenge to me (Note the situation here where advanced players will have no need to follow the notation while the beginners will have no talent to follow the notation). May be the number of the diagram should be increased, but that is too much work.
Also, I guess it is only advanced players can contribute to post their endgame as players like me do not have any good endgame to share. But if it were the endgame blunders, I believe we would have plenty to share…
|
|
I would contribute but don't have the time to do all the necessary graphics and notations. Instead I need to concentrate upon my games. I no longer have the time to do annotations either. I have to work.
Yusuf, you mention 3d thinking. Really, it should be identified as 4d thinking. The likely position in a certain number of turns away is seen clearly. This allows evaluation of the possible position 'in the mind's eye'. There is a time element involved that needs to be accounted for. Thus, you have four dimensions instead of just three. Good 3d thinking is a prerequisite but is not enough in itself. The 4th dimensional thinking aspect is learned through experience and practice.
Also, certain positions follow known patterns. Masters know these patterns and do not have to think, merely identify. It is hard for lower rated players to understand that they are often required to think more than the higher rated player to get similar results in any particular position. But it is true. Higher rated players generally don't waste their time thinking about unpromising lines. They simply have a much better feel for the game.
People often talk about the incredible calculating ability of a Master or Grandmaster. While that may be true, most lesser players overlook the higher level player's ability to play a game accurately largely through intuition. It is an important factor that only enters the game through previous experience. At the same time, anyone playing a chess game through intuition alone is likely to get slaughtered by an equally talented peer.
One model does not accurately describe the thinking ability of talented players. Since models are simplifications by definition, they only help us understand certain aspects of the whole system. There are many factors involved; and each individual uses more or less of these various factors and abilities that other players may or may not consider important, all depending upon personality and experience.
I could easily write an entire book on the subject. Unfortunately, I neither have the time nor the space to do so here.
|
maca
30-Aug-09, 04:16
|
ionadowman
What yusuf_prasojo said about weaker players not feeling having anything to share is probably the major reason why only stronger players are participating here. I'm a 1800+ -rated player, but still, I feel that only recently I've been able to play games that would be truly worth showcasing. Still, significant of my games end into rough tactical blunders by either player. Only lately I have felt that some games of I won had no obvious tactical blunders made by my opponent.
To that end, I think I have an interesting rook-ender that I would appreciate if someone could look:  game . In the middle game, I was able to secure a more active position due to what I think was a slight mistake by my opponent, and although my intent of winning a pawn fell through when the pawn that I had won quickly fell, I decided to reject my opponent's draw offer and play for a win, using my superior rooks. At one point, when my own rook became passive I though I had pushed too hard and I was going to lose, but in the end my passed pawns turned out to be winning. I am not sure if either one of us played optimally in the end game, but at least there were no obvious tactical blunders.
Regards,
MaCa.
|
|
I have criticized Karsten Muller's column in chesscafe in the past because he doesn't explain things. But this month he has done really well.
His analysis of the following endgame is really well done:
101.01 Gustafsson,J (2634)-Morozevich,A (2787)
The meat of the analysis is in the variation 70..Rg3! I would work through and understand that variation before looking at the main game.
www.chesscafe.com
|
|
the second game in Muller's column has me stumped. The game being analyzed is
Ivanchuk,V - Anand,V (2791), XXVI SuperGM Linares ESP (5), 24.02.2009
One of the subvariations ends in the following position (White to Move)
Muller says this is a draw. Would someone explain how this is a draw.
|
|
yusuf_prasojo...
I think you raise points worth thinking about. It is true that in starting this thread, I was making certain assumptions. First of all, I discovered as a near-beginner some fascinating aspects of the endgame that now I (and you too) would regard as elementary. There was something intriguing about the ability of pawns on certain squares of the 7th rank (e.g. a white pawn on a7, c7, f7, h7 with a king on the 7th or 8th rank protecting it) being able to stand off a queen, yet on other squares (b7, d7, e7, g7) could not. It was great to learn the technique of winning with K and P vs lone King - and to find out that K and P did not always win. That in certain special circumstances, a bishop plus pawn could nor defeat a lone King (see earlier posting).
It was this discovery, that things aren't always as they seem, that has led me to a lifelong fascination with the endgame. I guess part of my motivation was perhaps to correct an impression many beginners and intermediate players that the endgame was a chore and a bore. Even the great Janowski allowed his disdain for that aspect of the game, much to his cost.
Another assumption was that anyone playing though these endings would follow them through on a chess-board. Without readers are willing to do that, it seems to me unlikely they will benefit very much thereby. True, being able to "read" the game from the diagram is a useful discipline for OTB play, but by the time one becomes proficient in following lines "in one's mind's eye", one will usually have progressed beyond the more elementary endgame techniques.
But I also had in mind sharing something of the wonder (or wonders) of the chessboard. This is why I would welcome any endgame by any player, at whatever level, who felt he or she had achieved something - had made some kind of discovery, or had any reason to feel proud of even just one move in the game.
In that regard, thanks to maca for that rook ending. He certainly made the most of his advantage in that game! It was somewhat unusual in that neither side seemed interested in bringing the rook behind Black's Q-side majority, but maybe such a plan was less than ideal anyhow! (I'll comment briefly on that game in another thread).
|
|
Draw
This looks like a draw. E.g. 56. a8=Q Rxa8 After 57. Kxa8 the white king is too far away to assist the rook in winning the pawns. E.g. 57. .. g5 58. Kb7 g3 59. Ra3 Kg6 60. Kc6 Kf5 61. Rxg3 Kf4 62. Ra3 g4 63. Kd5 g3 draw. Better seems 57. Rxa8 but black will again shoulder the white king with 57. .. Kg6 58. Kc6 Kf5. See for instance first chapter of Practical Rook Endings by Mednis for this manoeuvre.
|
|
maca...
... A very nicely played ending. I did think maybe in the early stages that you allowed White into the game, but he never managed to pull himself close enough to make a real contest of it.
I haven't gone into the game in any real detail, but here are a couple of comments -
[a]
b
25...Ra5!?
This seems OK - there are certainly worse moves you could plausibly have chosen. But I rather like this line:
25...Rc2, with the idea
26.Rxc7 b5! 27.Rc6 b4!? 28.Rxa6 b3!? 29.Kd3 Rxf2
30.Rb6 Rxh2 31.Rxb3 Rg2 etc
This is, to transform a Q-side advantage into a K-side one. White can not prevent the loss of his g-pawn. From this point, Black can contest the promotion of White's sole remaining pawn, even giving up the rook for it eventually, in the hope and expectation that the 3 K-side passers will defeat White's rook.
It is by no means certain that this idea is really better than the game continuation, though.
[b] Three moves later, I did wonder if Black might have found more counterattack buy playing 28.Rc8+ instead of the more passive-looking 28.Rc5.
[c]
w
Here, White played
42.Rf4? and lost quickly: 42...Kc5 43.Rf5+ Kc4 and White resigned in a hopeless situation (e.g. 44.Rf3 Rd3 etc)
It is clear that in the diagram position, White is in a bad way anyhow, but there was more resistance to be had in
42.Rf5, holding back the enemy King.
Now, suppose Black decides to go after the c-pawn with the rook - a plausible line, methinks:
42...Rd3 43.Rf6+ Kc4 44.Rxh6 Rxc3 45.Rh8 Kb4
46.h6 a4 47.Ra8 ...
Black would still have a fight on his hands (though Black still should win from here).
Of course, even better for Black would have been 42...a5, but White would still have had reason to continue playing for a little while longer.
Sometimes it is the saving of the draw that gives reason for a certain pride. Here is a position from the game that ended my starting winning streak on GK. We'll take up the story before we get to the rook ending, after Black's 43rd move:
w
From a Sicilian Dragon, Black had blunted White's attack by giving up a knight for 3 pawns and the pawn roller that you see in the diagram about to squash White like a bug. Fine for Black - but I had the White pieces. The hope I had placed in my Q-side majority was fading every move, but it did lead to a different plan from the obvious promotion motif:
44.b5 axb5 45.Nxb5+ Kc5
This was what I was hoping for: to decoy, beguile and lure the Black King to the Q-side...
46.Nc7 Kb4
Black sees nothing untoward...
47.Nxe6! fxe6 48.Rxe6 Kxa4
This was my plan: to defend a rook ending two pawns down(!) However, it is not so easy for Black to defend both his pawns...
49.Rg6 Rf4
No good was 49...Rg1 50.Kc3 Kb5 51.Kd4. With the move played, Black hoped to cut White off along the 4th rank.
50.Rg5 Kb4
Black gives up one of the pawns in favour of pushing through the other. I don't think a better plan was available to him at this point.
51.Rxd5 Kc4 52.Rd8 ...
Probably simpler was 52.Rg5 at once. Rooks go behind the passed pawn (most of the time).
52...Rd4
The exchange of rooks would have been hopeless for White: 53.Rxd4+?? Kxd4 54.Kd2 g3! 55.Ke2 Ke4 56.Kf1 Kf3 57.Kg1 g2 58.Kh2 Kf2 Black wins.
53.Rg8 Kd5 54,Kc3 Ra4
Black can not maintain the 'cut-off' of White's King from approaching the K-side.
55.Rg5+ Ke6
In turn, White establishes a 'cut-off' of his own. It is very temporary, but lasts long enough to gain a tempo.
56.Kd3 Kf6 57.Rg8 Kf5 58.Ke3 Ra3+ 59.Kf2 Ra2+
60.Kg3.
At this point, Black acknowledged that he wasn't going to make any further progress and the game was drawn. Whew!
|
|
Blake
Averbach (in Turmendspiele Band1, Sportverlag Berlin 1988) gives the following position by Maiselis, 1956 where white has the move and black only draws if the white king is on either a8, b8 or c8:
From the position you give in your last post, we can have the plausible moves 56. a8=Q Rxa8 57. Rxa8 Kg6 58. Ra5 g3 59. Ra3 Kf5 60. Rxg3 g5 to reach
White has gained a tempo for the white king compared to the Maiselis position, which would be enough to win were it not for the slightly worse position of the rook which will again lose that tempo. This illustrates how close this endgame is. The end could be 61. Kc6 Kf4 62. Ra3 (62. Rg1 is essentially Maiselis position, i.e. drawn) g4 63. Kd5 g3 drawn.
|
| |
|
Please log in to your account to play chess online!
