|
< Earlier Kibitzing · PAGE 1 OF 2 ·
Later Kibitzing> |
| Feb-02-09 | | whiteshark: The Big Beliavsky |
|
| Feb-03-09 | | esticles: From the Official Tournament Site:
<41 b6!
At first sight this looks crazy as it allows Black to exchange into a favourable king and pawn endgame. But it is not so. 41...Rd8
Black would dearly like to play 41...Rd4+ but he realises that he can’t. White wouldn’t reply with the awful 42 Kb5?? Rxa4 43 Kxa4 e3 44 b7 e2 45 b8Q e1Q leaving him a pawn down in a queen endgame, but the amazing 42 Kc5!!. This move and the following sequence is analogous to the Saavedra study (except that White is not obliged to underpromote at the end of it). If 42...Rxa4 43 b7! Ra5+ 44 Kc4! Ra4+ 45 Kc3! Ra3+ 46 Kb2! and White will promote his queen*. * I'm since told that the resultant queen versus rook endgame might still not be winnable. Maybe Big Al was bluffing!?> |
|
Jan-01-11
 | | Phony Benoni: 
click for larger viewLooks simple. 41.b6 and what does Black do? 41...Rd7 42.Ra7 Ke6 43.b7. 41...Rd1 42.b7 and with the Rb4 block available White should win. Other moves look passive. But that Black e-pawn must not be ignored. So 41.b6 Rd4+ 42.Kb5/b3 Rxa4 43.Kxa4 e3, and Black promotes as well. Mulled over that for a few seconds, and a light hit. <Saavedra!> 
click for larger viewIf you don't know the study, White wins with 1.c7, even though he's a rook down, since the rook can't stop the pawn. The whole solution goes <1.c7 Rd6+ 2.Kb5> (2.Kc5? Rd1!) <2...Rd5+ 3.Kb4 Rd4+ 4.Kb3 Rd3+ 5.Kc2 Rd4! 6.c8R!!> <6.c8Q? Rc4+! draws by stalemate; instead, White threatens Ra8+ with mate to follow) <6...Ra4 7.Kb3>, and wins with the threats of 8.Kxa4 and 8.Rc1#. So, in today's game, 41.b6 Rd4+ 42.Kc5! Rxa4 43.b7!, with the same idea, minus the underpromotion shtick. That's not all, of course, and Black did hold on for a while. And I see from the kibitzing that there might have been a tablebase ♕ vs. ♖ draw. But I think the point today is to see the Saavedra idea of giving up the rook to push the pawn. |
|
| Jan-01-11 | | Sparta: Another game proving you need to know your endgames! White may not have gotten the point had he not known the Saavedra position as Phony Benoni nicely wrote out. On the game, it seemed that white was playing solely to reach a favorable endgame. I particularly liked 30. Rd1 as it forced an exchange of rooks and allowed white to attack black's a-pawn before black attacked white's. |
|
| Jan-01-11 | | Brandon plays: I got b6. However, I missed the fact that Kb5 was in fact losing whereas Kc5!! was winning. I did not consider the fact that the rook could drop from a4 down to 1 to check from behind. Drat. |
|
| Jan-01-11 | | SufferingBruin: I've been working my a** off to learn endgames. I was set to go all professorial and show step-by-step how white wins even a rook down. Instead, I'll type this: What <Phony Benoni> said. Happy 2011, everybody. |
|
| Jan-01-11 | | M.Hassan: "Very Difficult"
White to play 41.?
White is a pawn down.
It is not difficult to see that White should push a pawn and that is the b pawn: 41.b6
I spent quite some time after this move to go for the best Black response. At first I thought:
<41...........Rd4+
42.Kc5(not b5) Rxa4
43.b7 and Black can not stop promotion of the b pawn>
So, I chose the following line:
41............Rd8
42.Kc5 Rb8
43.Kc6 Rc8+
44.Kd7 Re8
45.b7 e3
46.Re4 e2
47.Rxe2 Re7+
48.Kc6 Re6+
49.Kc7 Re7+
50.Kb6 Re8
51.Rxe5 Rxe5
52.b8=Q
And White wins. Let's see how the game continued |
|
| Jan-01-11 | | SamAtoms1980: Got the first 5 moves.
After a look of several minutes, the most attractive general plan was to push the pawn, and guard it with the King while our rook picked off Black's pawns. |
|
| Jan-01-11 | | dzechiel: White to move (41?). Black is up a pawn. "Very Difficult." New player Brandon is visiting this evening and this position is one of the first times he has been presented with this type of problem. We looked at the obvious pawn push
41 b6
at first, and while white does queen first, the lines still seem to favor black, eg: 41...Rd4+ 42 Kb3 Rxa4 43 Kxa4 e3 44 b7 e2 45 b8=Q e1=Q OK, both sides have promoted, and white does get the first check here, but black's extra passed pawn is protected and the white queen will only get a couple of checks. While this line is tempting, I'm not thinking that it's even good enough for a draw. So, back to the drawing board. How about...
41 Ra7+
instead. By giving this check, we move the rook out of the potential skewer that the black rook might give by checking on d4. The black king will move 41...Kf6
and now we can try the pawn push.
42 b6
I'm not sure what black should try here. But the rook check followed by the pawn race 42...Rd4+ 43 Kc5 e3 44 b7 e2 45 b8=Q e1=Q
does not favor black. Now that the rooks are still on the board, I think white will checkmate black in short order. Instead, black will probably try to block the pawn back on move 42 with something like 42...Rd8
and now it gets a little messy. Maybe
43 b7 Rb8 44 Kc5
and it looks like white will win.
Time to check and see how this played out. |
|
| Jan-01-11 | | dzechiel: Wow, that is a wonderful move (after 41...Rd4+ Kc5!!). Wish I had seen it. |
|
| Jan-01-11 | | Fezzik: What a brilliant way to start the New Year!
Almost any fish would at least consider 41...b6, but most would stop when they saw that Black's R can get back to the 8th rank in time. Beliavsky saw farther, thanks to his intimate knowledge of the classics. Great puzzle, CG.com! |
|
| Jan-01-11 | | thegoldenband: I saw the key idea (not recapturing the rook) but went with the wrong move, missing the skewer after 41. b6 Rd4+ 42. Kb5?? Rxa4 43. b7 Ra1 44. b8=Q Rb1+. 42. Kc5 is lovely. |
|
| Jan-01-11 | | Edoneill: Nice end to the game. You really need to know your endgames though otherwise 41. b6 seems a little futile. It's interesting to see that with a bit of endgame knowledge you can convert a seemingly hopeless position into a winning one. |
|
| Jan-01-11 | | mig55: White wins a tempo by giving up his rook after Rd4.... |
|
| Jan-01-11 | | UnsoundHero: I can't yet see how White wins after 41 b6 Rd4+ 42 Kc5 Rxa4 43 b7 e3 44 b8(Q) e2 45 Qb3+ Kf6 46 Qe3 Ra2. |
|
| Jan-01-11 | | Dr. J: After 41 b6 Rd4+ the continuation 42 Kc3 also seems to win - at least, I have little trouble beating the Crafty EGT. Am I missing something? |
|
Jan-01-11
 | | Sastre: <UnsoundHero: I can't yet see how White wins after 41 b6 Rd4+ 42 Kc5 Rxa4 43 b7 e3 44 b8(Q) e2 45 Qb3+ Kf6> 46.Qf3+ Rf4 47.Qxe2 is winning for White. |
|
| Jan-01-11 | | David2009: Beliavsky vs Kotronias, 2009 White 41? White has an important choice: push the b Pawn immediately with 41 b6 or prepare with 41 Kc5 expecting 41...d3 42 Re4 Ke6 and
Black is poised to stop the b Pawn. This so line is unattractive.
41 b6 looks good. Defences: (A) 41...Rd4+ 42 Kc5! Rxa4 43 b7 and the Pawn promotes; (B) 41...Rd8 42 Kc5 and if d3 43 Re4 etc.
Analysing further is beyond my skill.
Before consulting the game I will set the position up on Crafty End Game Trainer link:
http://www.chessvideos.tv/endgame-t... 
click for larger view [Beliavsky vs Kotronias 41? (POTD 01/01/11)] The EGT defends with line (B) no real surprise there and I win an interesting ending starting 42...Rb8 43.Rxe4 Ke6 44.Re2 Rc8+ 45.Kb5 Rb8 46.Kc6 Rc8+ 47.Kb7 Rc3.
Time to check:
=====
I got the main idea. The EGT's defence seems to be neither better nor worse than the game line. I'm called to lunch: review other comments later. |
|
| Jan-01-11 | | Marmot PFL: The trick to promote the b pawn works here 41 b6 Rd4+ 42 Kc5 Rxa4 43 b7 Ra5+ 44 Kc4 etc, but black might still be able to capture g2 then make a fortress for his king via g6, Rf5-h5-f5 to cut off white's king. |
|
| Jan-01-11 | | Marmot PFL: For instance 44...Ra4+ 45 Kc3 Ra3+ 46 Kb2 Rd3 47 b8(Q) Rd2+ 48 Kc8 Rxg2 and I don't see how white can win. |
|
| Jan-01-11 | | Isbjorn: <David2009: The EGT defends with line (B) no real surprise there and I win an interesting ending starting 42...Rb8 43.Rxe4 Ke6 44.Re2 Rc8+ 45.Kb5 Rb8 46.Kc6 Rc8+ 47.Kb7 Rc3.> How do you win from there? I've tried several times against EGT. In some lines black is forced to sacrifice a rook against the b-pawn, but because EGT makes a king move instead of the gratuitous rook check, I'm never able to capture any pawn with my king, and end up with a rook (the white is king too far away) against a king and 1-2 advanced pawns. The other line is essentially yours above, but there I don't see how to make progress either. |
|
| Jan-01-11 | | Isbjorn: Basically, I find that EGT advances the black king to a central position on the d-file, preventing the white king from doing anything useful after the black rook is sacrificed for the white b-pawn. |
|
| Jan-01-11 | | Charles E: Isaborn, the line I took against the EGT was:
41 b6 Rd8 42 Kc5 Rb8 43 Rxe4 Ke6
44 Rc4 Rc8+
and with the rook exchange the b pawn promotes |
|
Jan-01-11
 | | Penguincw: b6 is so obvious following the rule of " Passed ♙awns must be pushed" .One quick note is that after 41. b6 black has to play 41. ♖d4+ first and after 42.♔b5 or ♔b3 then 42.e3 loses to 43.b7 and if 43.♖d8 fails to 44.♖a8 and that's game over. 
click for larger view |
|
| Jan-01-11 | | Eyal: I think <Marmot PFL> is right: after 41.b6 Rd4+ 42.Kc5 Rxa4 White may queen the b-pawn, but it's not the end - it seems that Black can still set up a fortress and draw. So, as the original report from the tournament site says, Beliavsky may have been bluffing. Btw, regarding <dzechiel>'s line starting with <41.Ra7+>: here Black plays <41...Ke6!> (rather than 41...Kf6?) and it should be a draw. The point is that in case of 42.b6 Black can defend by 42...Rd1, and 43.Ra3 Rb1 44.Rb3 doesn't work, since after a rook exchange the black king is just in time to catch the b-pawn - which wouldn't be the case if it's on f6. |
|
 |
|
< Earlier Kibitzing · PAGE 1 OF 2 ·
Later Kibitzing> |
Kibitzing
