|
< Earlier Kibitzing · PAGE 2 OF 2 ·
Later Kibitzing> |
Jan-20-13
 | | FSR: It's too hard for me to calculate to the end, but 42.Kg3! is clearly forced. Then if 42...a5? 43.bxa5 Kd5 44.Kf2 or 42...h6? 43.g5 hxg5 44.hxg5 must be  . What White wants to do is liquidate the king-side pawns, then run over and eat the queen-side pawns. |
|
Jan-20-13
| | FSR: <rilkefan: The pawns can't move first, or the black king gets in. And it's no good running to the queenside as the black h pawn is faster. So play Kg3 and worry about what to do next after black's move.> Precisely. |
|
| Jan-20-13 | | awfulhangover: Comp proves that 51. - Kd7 would have drawn! |
|
Jan-20-13
| | FSR: Not surprising at all. If you go to the Nalimov tablebases and put in positions with Q + g-pawn + h-pawn versus queen, without Black even having a pawn, many positions are drawn. For example, with pawns on h4 and g5, W queen on g4, W king on h3, and B queen on d2, Black has 15 different drawing moves.
click for larger viewhttp://www.k4it.de/index.php?topic=... |
|
| Jan-20-13 | | vinidivici: Queen WITH PAWN(s) v queen (NO PAWN) ending could be tricky. I have posted somewhere about the basic principles of this ending... For stronger side
1. Don't place the king IN FRONT OF the pawn(s).
2. Stronger side SHOULD place the king on the same file/rank where the defender's king is standing, or an adjacent file/rank. For weaker side
1. Facing knight/rook pawn: Best place for king is in the corner that diametrically opposite to the promotion square. When this is come to the application, weaker side could hope that a counter-check could be less probable. 2. Facing bishop/central pawn: Sorry to say this...there is no way out for this kind of endgame, theres no drawing zone. Your opponent play rightly, you WILL BE LOSE. So you can pray for your opponent to play less precise. For both sides:
1. The queen is better placed in the center of the board. I can add that these particular theories for this particular endgame more exact in term of application than the other particular theories for the other particular endgames. |
|
Jan-20-13
 | | chrisowen: At get go dine ogle 42.Kg3 gadzooks in both again,
kings have ground, to cover ace in b1 versus a8 line low freeing maneovre seek on a cycle one g3 black king has nowhere to go both in h6 looses, to h5 in whilst e5 plead for gamble f3 feel the noose tighten in every it is angle covered by white in creak again tardy it odd in yeah 40.Kd5 in ty it dog in dig 40...Kb6 the duke why it roving along in low mind black king bucket in stopping a5 live on h8? ha gen das move booting b1 both again seat ooh utility seems it d5 allows right off in kingg3 cadence in rising e5 whizz back f3 looks tempo kg3 bind aint it over in bat I'm clinching a8 before b1 us inclined to believe in together on black eyes penetrate racing back took it down in a8 grey goose, her in e8 checksee tease again difficlut change light open in shute eeking g4 as h5 throttle see bod a teteatete in good recind hind wag one
edict free 54.Kh5 d3 queen among time hodge bodge 55.Kh6 a lump facing a 61.Kg7 registrate defeat for black eg success nugget in titter riff i see kingd5 live on g3! back some gallantry in b6 you ride a swindle apt? |
|
| Jan-20-13 | | prbprb2: Does Moro have better chances with 35 .. Rb5? It seems to me he can pick up both Q side white pawns. Seems like he could get a draw then. No? Could someone comment for me, edify me? |
|
Jan-20-13
 | | Jimfromprovidence: Based on the comments of <awfulhangover> and <FSR> on the characteristics of Queen + 2 pawns vs. queen endgames, I'm amending my earlier comments. Nunn, in his book, “Nunn’s Chess Endings Volume 1” states that “Queen and two pawns generally win against a queen, but one of the most astounding discoveries to emerge from the 6-man datababses was the finding that queen + h pawn + g pawn vs. queen is generally drawn if the defending king is in front of the pawns”. Now I’m thinking that it’s not good enough the general observation that white can queen first; it’s the specifics related to the uniqueness of the position that matter. With that in mind now I’m thinking that 44 g5 was either a mistake or inferior, and that 44 h5 was a much clearer winning move.
click for larger viewAssuming 44…Ke5, then 45 h6.
click for larger viewBlack still cannot play …a5 to promote because white’s king can still get there in time. And the further up the board black’s king gets, the more likely he is subject to mating traps. So black has to retreat his king while white pushes his king up the f file and pushes his g pawn as well. (The white king can get as far as f5 and still catch the passed b pawn after ....a5). I think that this is the way to go. |
|
Jan-20-13
| | FSR: Hmm, I think you may be on to something, <Jim>. I also remember being struck by that Nunn quote, which is what motivated me to look at the Nalimov tablebase for the ending where Black had no pawn. |
|
| Jan-20-13 | | Marmot PFL: White has to try 42 Kg3, and if 42...a5 43 ba Kd5 (43...b4 44 a6 and white queens with check) 44 Kf4 and if 44...b4 45 Ke3, otherwise g5, h5 etc. What else? If 42...Ke5 43 Kf3, now 43...Ke6 44 Ke4 or Kf6 44 Kf4 are just slow ways to lose, so 42...Kd4 43 Kf4 a5 (43...Kc4 44 g5 is too slow) 44 ba b4 and when both sides queen it is white's move with an extra pawn and better placed king. Not sure I could win that, but for Shirov it's another matter. |
|
| Jan-20-13 | | Patriot: Material is even. Black threatens 42...Kf4, winning. The only candidates that prevent this are 42.Kg3 and 42.Kg2. 42.Kg3 is the candidate that looks strongest. 42.Kg3
42...h6 43.h5 Ke5 44.Kf3 a5 45.bxa5 b4 46.a6 b3 47.a7 b2 48.a8=Q b1=Q This could easily draw, because I don't see a winning skewer. 42...a5 43.bxa5 b4 44.a6 b3 45.a7 b2 46.a8=Q+ wins.
Does 42.Kg2 make any difference? There was nothing to suggest this. The immediate 42.g5 Kf4 looks lost and 42.h5 Kf4 looks no better. This position really depends on where the kings are when the pawn break ...a5 occurs. I would play 42.Kg3 but it's really not clear. |
|
| Jan-20-13 | | Marmot PFL: < Jimfromprovidence> I wasn't sure which pawn to push but I wanted the black king out so went for 44 Kf4 instead of 44 g5 Ke5 45 Kg4. Seems like less calculating to just take opposition here. |
|
| Jan-20-13 | | Marmot PFL: <42...h6 43.h5 Ke5 44.Kf3 a5 45.bxa5 b4 46.a6 b3 47.a7 b2 48.a8=Q b1=Q This could easily draw, because I don't see a winning skewer.> 45...b4 46 Ke3 and white catches the pawn. |
|
| Jan-20-13 | | Patriot: Wow! 42.Kg3 h6 43.h5 Ke5 44.Kf3 a5 45.bxa5 b4 46.Ke3! or 46.Ke2! wins. 46.a6 is a big mistake--just move the king toward the b-pawn and stop it. When black tries to stop the a-pawn, white has g5! at his disposal. This is really something good to keep in mind! |
|
| Jan-20-13 | | Patriot: <Marmot PFL> Absolutely correct! Thanks! |
|
Jan-20-13
| | Jimfromprovidence: <Marmot PFL> <I wasn't sure which pawn to push but I wanted the black king out so went for 44 Kf4 instead of 44 g5 Ke5 45 Kg4. Seems like less calculating to just take opposition here.> I wish I had thought of that. That move basically means that if both sides promote and white is able to force a queen exchange, he will surely win, kind of like in this position below, with black to move. 
click for larger viewSo Nunn's observations are right on. |
|
| Jan-20-13 | | Marmot PFL: <Jimfromprovidence> White has to be able to win the Q vs 2P ending after 44 Kf4 Kc4 45 g5 Kxb4 46 h5 a5 47 g6 hg 48 hg a4 49 g7 a3 50 g8(Q) and stops a2, then brings the king over 50...Ka4 51 Ke4 b4 52 Kc4 b3 53 Kb4 and wins by a tempo. that I wasn't sure of at first but didn't realize 44 h5 was winning either. |
|
| Jan-20-13 | | Marmot PFL: White can also torture black and reach this kind of position with white to move and win, a pawn down-
click for larger view1 Qxb5+ Ka3 2 Qa5+ Kb2 3 Qb4+ Ka2 4 Kc2 and wins |
|
| Jan-20-13 | | Kikoman: <Kg3> that's my pick. ^^ |
|
| Jan-20-13 | | agb2002: The material is identical.
At the moment, 42... a5 is not a threat because the white pawn would promote with check. If the black king tries to eliminate White's b-pawn then White would promote in 5-6 moves while Black would need 8. The real threat is 42... Kf4, capturing White's g- and h-pawns. The white king must protect the king side pawns. Therefore, 42.Kg3: A) 42... Ke5 43.Kf3 (gaining space and reentering the square b5-f1 to stop Black's b-pawn) A.1) 43... Kd4 44.Kf4
A.1.a) 44... a5 45.bxa5 b4 (45... Kc5 46.g5 b4 47.Ke3 wins) 46.a6 b3 47.a7 b2 48.a8=Q b1=Q 49.Qa7+ and it seems that White can force the exchange of queens either on the b-file, the square e4 or the square f5. A.1.b) 44... Kd5 45.h5 Ke6 46.g5 Kf7 (46... a5 47.bxa5 b4 48.Ke3 + -) 47.Kf5 Ke7 (47... Kg7 48.Ke6 Kf8 49.Kf6 Kg8 50.g6 hxg6 51.hxg6 Kf8 52.g7+ Kg8 53.Kg6 and mate in four) 48.g6 with the same idea as the previous variation. A.1.c) 44... Kc3 45.g5 a5 (45... Kxb4 46.h5 promotes two moves before Black) 46.bxa5 looks similar to A.1.a. A.2) 43... Ke6 44.Ke4 looks similar to A.1.b.
B) 42... Kd4 43.Kf4 transposes to A.1. |
|
| Jan-20-13 | | James D Flynn: I did not see a win after 42.Kd5 instead of Ke5.
Material is equal with White having a pawn plus on the K-side and Black a pawn plus on the Q-side. Both ks are closer to the K-side but this does not confer an advantage to the Q-side majority here because the a pawn is backward and would be taken if advance, moreover if Black plays a5 with his K still on the long white square diagonal and no closer to the Q-side the white pawn would Q first with check and win. The current placement of the Ks forces White to play Kg3 to avoid Kf4 winning both Whites K-side pawns : the alternative 14.Ke2 Kf4 loses both K-side pawns and it would take 6 moves to take off Blacks Q-side pawns and only 3 to remove Whites K-side pawns and unblock the h pawn from queening.
42. Kd5(close enough to stop the White a pawn) 43.g5 a5 44.h5(if bxa5 b4 45.a6 Kc3 46.a7 Kb7 47.a8=Q+ Kxa8 49.h5 b3 50.g6 hxg6 51.hxg6 b2 52.g7 b1=Q 53.g8+ The game is drawn) axb4(if a4 45.g6 hxg6 46.h6(not hxg6 a3 47.g7 a2 and both pawns Q) 45.g6 hxg6 46.h6 (if hxg6 Ke3 wins)b3 47.h7 b2 48.h8=Q b1=Q probably draws although Black is 2 pawns up. |
|
| Jan-20-13 | | James D Flynn: After perusing the other Kibitzers analysis I find nobody considered 42.Kd5. While i understand this looks an unlikely square since if White promotes on either a8 or g8 it comes with check, but it seems to draw. |
|
| Jan-20-13 | | beenthere240: WWMD. What would Magnus do? |
|
| Jan-20-13 | | M.Hassan: "Insane"
White to play 42.?
Each side has 3 pawns
42.Kg3 Obvious Kd5
43.Kf4 a5!
44.bxa5 b4
45.Ke3 Kc6
46.g5 b3
47.Kd3 b2
48.Kc2 Kb5
49.Kxb2 Kxa5
50.h5 Kb5
51.g6
Black King can not reach one of the advancing pawns
1-0
There was not any so called "insane" move
Time to check |
|
| Apr-30-21 | | g15713: E. 1.0
Black to move at 37
click for larger viewFinalGen has Black drawing after 37...axb5
<"Capture towards the center is often valid when the choice stands between the a-pawn and c-pawn (as well as f-pawn and h-pawn).">
<"It should also be ignored when appropriate."> 'https://chess.stackexchange.com/que.... FinalGen analysis with White to move at 38 after 37...axb5 g4 draws
h3 draws
b3 draws
Kf2 draws
Kg2 draws
Kf1 draws
h4 Black wins or Draw
Kh1 Black wins or Draw
b4 Black wins in 18
37...axb5 38. b4? 38...c5! wins
E. 1.1
click for larger viewF. 1.0
White to move at 50 in the actual game
click for larger viewSyzygy says White is winning with 50. Qe8+ or Qf3 or Qh8+
'https://syzygy-tables.info/?fen=Q7/... Chess 'User: Jimfromprovidence quote: <Nunn, in his book, “Nunn’s Chess Endings Volume 1” states that “Queen and two pawns generally win against a queen, but one of the most astounding discoveries to emerge from the 6-man databases was the finding that queen + h pawn + g pawn vs. queen is generally drawn if the defending king is in front of the pawns”.> That's the rub if the defender's king is cutoff from blocking passed h-pawn and his queen has no perpetual check then he loses. Understanding Chess Endings by John Nunn, my favorite chess writer, is a great place to start, then his Nunn's Chess Endings Volumes I and II - my two cents on great chess books on endgames. Of course there are many others -
<"Part of the problem is there are too many good endgames books and not enough people read them!"> aansel |
|
 |
|
< Earlier Kibitzing · PAGE 2 OF 2 ·
Later Kibitzing> |
Kibitzing
icon.