|
< Earlier Kibitzing · PAGE 4 OF 4 ·
Later Kibitzing> |
Dec-17-09
 | | Jimfromprovidence: <Once> There's actually a final wrinkle to the puzzle. What if white had played 42 Qe8 (with the threat Qxh5+)?
click for larger viewAssuming best play is 42..Rcc2 43 Qxh5+ Kxg3 44 Rb1 Rxh5 45 gxh5, we are left with the folowing position.
click for larger viewIs this a forced win for black? |
 |
Dec-17-09
 | | YouRang: Well, now that I posted the 'easy' answer, it is obvious that I didn't look deep enough to see that this only draws. What was up with 39...Rc5? |
|
Dec-17-09
| | YouRang: Thanks <Once> for the explanation. Add me to the ranks of the confused. Again, today I didn't go deep enough to understand the position because it looked too simple. :-( |
|
Dec-17-09
 | | Once: <Jimfromprovidence> After 41...g5, I think white is thoroughly busted. Fritz rates all of white's choices as -3 or worse. From your diagram, the silicon monster likes 45...Rc6, when the b4 pawn is taboo because of Rc1# and the black king will round up h5. Not exactly what you would call a forced win ("all rook endings are drawn", right?), but in the realm of "best play" I'd say it's enough to chalk up the point for black. |
|
| Dec-17-09 | | Eduardo Leon: <Jimfromprovidence>, after the sequence you gave... 42. e8  cc2 43.xh5+  xg3 44.b1 xh5 45.gxh5
click for larger view... black simply plays...
45...c6!
Black stops the h pawn and defends his own b6 pawn. The b4 pawn is lost anyways. If white captures the b4 pawn immediately...
46.b3+ h4 47.xb4+ xh5
... a tablebase winning position for black is reached. |
|
| Dec-17-09 | | Eduardo Leon: <Once>, YaY! I found a computer move without a computer! (Although, frankly, 45...c6 wasn't that difficult to find.) |
|
Dec-17-09
 | | awfulhangover: 41.-g5?? Why commit suicide when you have an easy perp? |
|
Dec-17-09
| | Jimfromprovidence: <Eduardo Leon> After 45...Rc6, white should not play 46 Rb3+ because it loses a lot quicker to 46...Rc3. Removing the mate threat with 45...Rf1 is better. Now, for simplicity, if 46 Kh4 Rxb4+ Kxh5 black has a tablebase mate in 47 moves! 
click for larger view |
|
Dec-17-09
| | Once: <Eduardo Leon> Good call! A slight improvement would 45... Rc6 46. Rb3+ Rc3 and once again the c4 pawn is immune. |
|
Dec-17-09
 | | chrisowen: An interesting game, looks difficult though escaping a draw. Qc8+ Qg4 knight in iz token play. I lent thoughts for Rxg5 fxg4 then jousting with the idea of Rg2+ Kh1 Rg2+ perpetual but does black have more? Logically it is drawn but after Rc5 Qd8? black is riding on the crest of a wave with g5! A nice composition almost, the standard has picked up. Glancing through that post of <TheaN> he notices the lady is offside handing over a fantastic lot hear of drawing effort: the queen must remain on the light squares d7, e6 to stand any chances covering g4. |
|
| Dec-17-09 | | Eduardo Leon: <Once>, <Jimfromprovidence>, you are right, after 45...c6 46.b3, 46...c3! is stronger. However, in my original analysis, the rook was "stuck" in c6 to prevent the h pawn from advancing. |
|
| Dec-17-09 | | Eduardo Leon: <Jimfromprovidence>, the position you arrived at after 45...c6 46.f1 h4 47.xb4+ xh5 only differs from the one arrived at by the fact the white king is in f1 and not on g1. |
|
| Dec-17-09 | | WhiteRook48: i had Qc8+ |
|
| Dec-17-09 | | CHESSTTCAMPS: A delicious tactical position has been reached: Larsen has advanced his king all the way across the board to participate in a middlegame attack with queens on the board. The winningest tournament player of his era (~1965-1975) is up an exchange plus pawn and threatens to force mate with 36... Qd4+ or Rg2+. But has he paid a price for his aggressive king position? There is only one way to find out, white's only chance in this position: 36.Qc8+!
In view of 36... Rxc8(??) 37.Ng5#, some players might be shocked into resigning - but not very likely in Larsen's case. 37... Qg4!
Forced, but an adequate spoiler. After all, that dominating rook on the 2nd rank can deliver at least a perpetual check - so why worry - be happy. 38.Ng5+!
Necessary, because if 38.fxg4 Rxc8, the mate is gone and black is winning. If 38.Qxg4+ hxg4 39.fxg4 Rg2+ 40.Kh1 Rxh2+ 41.Kg1 Kxg3 also wins for black. Also inadequate is 38.Qxc5 Rg2+ 39.Kh1 Rxh2+ 40.Kg1 Qxg3#. 38... Rxg5 39.fxg4 Rg2+ 40.Kh1 Rxh2+ 41.Kg1 Rg2+ 42.Kh1 Now the essential question is whether or not the enterprising Larsen can still win this with R+3/4 pawns for the Q and white's king trapped in the corner. This is very tricky to visualize and prove a result: A) 42... hxg4 43.Qh8+ Kxg3 44.Qc6 Rh2+ 45.Kg1 and white seems to survive, e.g. 45... g3 (Kxg3?? 46.Qd6+) 46.Rf2 looks good for white. B) 42... Rxg4 43.Rf3 seems to hold.
Therefore, I opt for
C) 42...Rh2+ = a boring GM draw!
But there are plenty of opportunities for error by white - time to check. BTW my record against Larsen is 1-0, disclaimer below. <70+ board simul by Larsen in Teeside, England in 1972 - he had 4 losses in all and a handful of draws. The rest were Larsen wins. I have no knowledge of the overall strength of the opposition by British grade or other measure.> |
|
| Dec-17-09 | | GreenFacedPatzer: Very good puzzle.
At first, I thought white had an easy mate in two, but then I spotted black's counter mate-threats. The games really continues on a knife-edge for some time, with each player _almost_ checkmating the other, but... not... quite. In the end, I didn't see how white could escape.Needless to say, I didn't work through all the complications to safely draw for white. (It's some slight consolation that in this game, neither did the white player.) |
|
| Dec-17-09 | | CHESSTTCAMPS: Obviously I wasn't familiar with the game (or had forgotten it) - so my presumption of "attack" (rather than escape) by black in my post was pretty inept. In any case, a great puzzle! |
|
| Dec-17-09 | | Skakalec: I'm looking at the position after 32... Bd2 and trying to figure out how difficult would it be for ME to find winning move 33.Qh8 |
|
Dec-17-09
| | DarthStapler: I got it but I figured black would take the perpetual check with Rxh2+ |
|
| Dec-17-09 | | Loudpipes: White wins this one at 37. Qxg4+
How could white miss that?
Puzzle solved. |
|
Dec-17-09
| | YouRang: <Loudpipes: White wins this one at 37. Qxg4+ > ?! |
|
| Dec-17-09 | | TheaN: <White wins this one at 37. Qxg4+ > Er... noooh? ^^ |
|
| Dec-17-09 | | ohfluckaduck: Once, ... your right, white erred when he played 40. QD8. QE6 is much better and leads to a draw. |
|
| Dec-18-09 | | muralman: I was rather taken aback by white's divergence from my line where I got a checkmate. I took queen takes queen at move 37. Black has to take my queen with his H pawn. Then I drop down my knight to G5 for checkmate. right? |
|
| Dec-18-09 | | I play the Fred: muralman, your proposal fails to 38...Rxg5. |
|
| Dec-18-09 | | WhiteRook48: 42 Qh8 lasts longest (move 52) |
|
|
|
< Earlier Kibitzing · PAGE 4 OF 4 ·
Later Kibitzing> |
