< Earlier Kibitzing · PAGE 2 OF 2 ·
|Mar-01-09|| ||morfishine: <Once> Thanks for running that out. I chose Bg5; at least it "draws" but like you, I would not want to play this as black|
|Mar-01-09|| ||goodevans: Further to my earlier post, 42 Qf2 is also worth considering.|
I think the winning line is 42 Qf2 Rh5+ 43 Kg1 Rg6+ 44 Kf1 Qh3+ after which white will soon be forced to trade Q for R and black still keeps a strong attack. But I have to admit that I haven't been able to see this through to a definite conclusion.
|Mar-01-09|| ||johnlspouge: Sunday (Insane):
O De la Riva Aguado vs Bologan, 2001 (40…?)
Black to play and win.
Material: B for N+P. White has connected passed Ps on the Q-side and wins if all pieces are off the board. The White Na4 is out of play, so Black should exploit any local superiority quickly. The White Kh2 has 1 legal move, h1. The Black Qf5 pins Pf2 to Qf2 and Re4 to Rd3. The Black Rg6 has an open file, and Rc6 can reload it. The Black Be7 requires activation but does attack Ph4, burdening Re4 and Qf2. The White Rd3 and Qf2 are loose. The Black Kg8 is open to checks on the back rank and a2-g8 diagonal, but is presently secure except for Rd3-d8+.
Candidates (40…): Bxh4, Rcf6, Rg4
Refusal of the sacrifice of Bh4 leaves White faces a strong mating attack against Kh2 with no material compensation. The Black Bh4 still protects d8, removing the zwischenzug Rd3-d8+ as a rescue for Rd3.
41.Qxh4 [Rxh4 Qxd3 wins for Black]
41...Rg5 (threatening 42...Rh5, pinning and winning Qh4, or 42...Rcg6 43..Rh5)
White can save Qh4 only by Qh4-f2 or Qh4-e1. To interpose Re4- h4, White needs to unpin Re4 or else, e.g.,
42.Qf2 Rh5+ 43.Rh5 Rxh5+ 44.Qxh5 Qxe3
and Black is up the exchange. White also needs to protect Pf3. Hence:
42.Rd8+ Kh7 43.Qf2 Rcg6, threatening
(A) 44…Rh5+ 45.Rh5 Qf4+ 46.Kh3 [else, drop Rh4] Rxh4+ 47.Qxh4 Qxf3+ then mate in 1.
44.Rh4 Rh5 (renewing Threat (A))
White has no feasible defense to Threat (A):
(1) 45.Rxh5 Qxh5 46.Qh4 Qxh4#
(2) 45.Kh1 Rxh4+ 46.Qxh4 Qxf3+ 47.Kh2 Qg2#
|Mar-01-09|| ||goodevans: <johnlspouge: ... White can save Qh4 only by Qh4-f2 or Qh4-e1 ... White needs to unpin Re4 ... Hence 42.Rd8+>|
Perhaps white's best plan is to accept that he will have to give up Q for R at some stage, try to do this with as little disadvantage as possible and then try to tough out the ending at a bit of a material disadvantage (see my earlier posts).
To this end 42 Rd8+ may not be such a good idea because it robs white of the chance to get his K off the h-file (e.g. 42 Rd8+ Kh2 43 Qf2 Rh5+ 44 Kg1 Qg5+ picking up the R).
My question is whether the plan outlined above is feasible or whether the material disadvantage is simply too great?
|Mar-01-09|| ||johnlspouge: < <al wazir>: [snip] I didn't look at 42. Qh3 Rh5 43. Rh4 (with or without the Zwischenzug Rd8+), but I think I would have found Qc2+ and Rg6+ [snip] >|
40…Bxh4 41.Qxh4 Rg5
White can leave Rd3 to be snatched, but it is a clear loss. IMHO, the critical variation is therefore not the game variation, but
42.Rd8+ Kh7 43.Qh3 Rh5
Toga gives as best play
44.Rh4 Qc2+ 45.Kg1 Rg6+ 46.Kf1 Qc1+ 47.Ke2 Rxh4
48.Qxh4 Rg2+ 49.Qxf2 Rxf2
The above variation 42.Rd8+ 43.Qh3 (which I missed) requires foresight of the long-distance move …Qc2+ as a resource for Black.
In my practice on http://chess.emrald.net, I found it relatively easy to train myself to foresee long-distance moves that change the direction of an attack. The key point is to make a preliminary scan of the <entire> board, before focusing on a particular theater of action.
|Mar-01-09|| ||johnlspouge: < <goodevans> wrote: <johnlspouge: ... White can save Qh4 only by Qh4-f2 or Qh4-e1 ... White needs to unpin Re4 ... Hence 42.Rd8+> [snip] My question is whether the plan outlined above is feasible or whether the material disadvantage is simply too great? >|
Hi, <goodevans>. The plan is infeasible.
We all have different aims in our posts, e.g., <dzechiel> has the very practical aim of trying to predict best play. Probably because I am trained as a mathematician, my aim is to prove that the key move in a puzzle leads to a win (or draw). To that end, I enumerate <superficially> feasible possibilities, showing that they lead to a loss. Given my time constraints, I have found it impractical to order the possibilities as to feasibility or best play (as <TheaN> does), although I am conscious of the importance of doing so.
|Mar-01-09|| ||goodevans: <johnlspouge: ... IMHO, the critical variation is therefore not the game variation, but|
42.Rd8+ Kh7 43.Qh3 Rh5>
Doesn't this amount to the same thing?
<Toga gives as best play
44.Rh4 Qc2+ 45.Kg1 Rg6+ 46.Kf1 Qc1+>
The first real deviation from the game.
<47.Ke2 Rxh4 48.Qxh4 Rg2+ 49.Qxf2>
Am I missing something here, or did you mean just 49 Qf2?
<49 ... Rxf2>
Isn't 49 ... Qc2+ better?
|Mar-01-09|| ||CHESSTTCAMPS: At first glance, white might appear to have the advantage in this position, with an extra pawn overall and connected passed pawns on the queenside. Black's bishop does not appear particularly effective, with the e-pawn blocking the b8-h2 diagonal, but at least it's aimed at the weak h-pawn. However, black's position does have some strong assets: a very safe king position and more actively placed pieces, i.e. the white rooks are defending laterally and the white queen is placed passively defending pawns on f3 and h4, whereas black has majors very actively placed on f5 and g6. If black does not utilize the advantages of the position by attacking on the K-side, white's pawns will prevail on the Q-side, so the general plan is apparent. |
Typically, strong players search for a candidate move using two crucial criteria: 1. Does it create concrete threats i.e. is it forcing? Dzechiel wisely emphasizes this criterion in his posts - it should always be the primary consideration and gets the best bang for the buck. 2. Does it improve the position of the pieces? This is usually the secondary criterion when there is no obvious move that fits the first criterion. In general, improving mobility is the primary consideration when choosing this criterion. Often, both criteria point in the same direction.
Within the first 5 minutes of looking at this position, I considered one serious candidate move, 40..Rcf6, because it swings the more inactive rook to the K-side and it threatens Qxe4; hence it meets both criteria. At about the 5-minute mark, it occurred to me that 40..Bg5 might be possible, given that the position is rated insane (so look for insane moves!) and it's the only interesting bishop move. Furthermore, there is obvious tactical value in removing the h-pawn to get access to white's king, particularly if black can bring both rooks into the attack. So let's try the second candidate first (at this point I still have no idea) and see where it goes.
40... Bg5 (threatening the deadly Bf4+, very forcing!)
I first looked at hxg5 but quickly rejected it because it shuts the g-file that black now controls. Also it maintains the pawn shelter that black's king may need on h7, now that he has sacrificed the bishop guarding the back rank.
Now this looks dangerous - black can use the rook on c6 to attack on the g-file and the rook on g5 to attack on the h-file. Also, white's rook on e4 is currently pinned, so it makes it difficult for this rook to guard the h-file. Suddenly white's king looks very vulnerable.
42.R3e3 To free this rook from the pin and stay in front of the Pe5
Rh5+ (40...R6g6 41.Rh4 seems to hold, so black must strike fast.)
43.Kg1 (or Rh4 Qf4+ 44.Kh3 Rxh4+ 45.Qxh4 Qxe3 wins) Rg6+ 44.Kf1 Qh3+ 45.Ke1 Qh1+ and it's now clear that white will lose the queen to a pin. Triumph of the active rooks!
Let's look for improvements at move 42:
42.Re8+ Kh7 43.Rh4 R6g6 (threatening Rh5 followed by Qf4+) 44.Nc3 Rh5! (much more convincing than Rg2+ 45.Qxg2 Rxg2+ 46.Kxg2+ Qg5+ 47.Rg4 Qxd8 - too much material reduction and release of tension) 46.Ne2 Rg4!! and it's clear that black can resign because of the pin on the f-pawn.
I think we've struck paydirt. Good, now there's no need to analyze the boring Rcf6.
Time to post and see what happened.
|Mar-01-09|| ||johnlspouge: < <goodevans> wrote: Doesn't this amount to the same thing? >|
Yes, I threw the idea onto the computer before looking closely at the game.
<Am I missing something here, or did you mean just 49 Qf2?>
Correct again, I just meant 49.Qf2.
<Isn't 49 ... Qc2+ better?>
Correct. I usually warn that humans can improve near the end of the computer variations, but I was in too much of a hurry.
I had better get back to making Sunday pancakes for my kids: it is one of the time constraints I mentioned :)
Thanks for keeping me honest, <goodevans>.
|Mar-01-09|| ||Crocomule: Hey, just a kid... very good!|
|Mar-01-09|| ||Jimfromprovidence: The only line I found that was somewhat convoluted was after 43 Qxh5?!|
click for larger view
White is nominally down a queen to a rook and knight.
The Rybka demo says the line continues with 43 Qxh5 Qxh5+ 44 Kg3 Rg6+ 45 Rg4 Qf5 46 Rd8+ Kh7 47 Rxg6 Qxg6+ 48 Kf2 Qc2+ before black finally gets in a position to win more material.
click for larger view
Then, for example, 49 Kg3 e4 50 Nc5 exf3 51 Kxf3 Qf5+ wins.
click for larger view
|Mar-01-09|| ||crwynn: I missed 43.Qh3 as well, didn't see that the Re4 is not really immobile because the Rd3 has a check to save it.|
I figured 43.Qf2 Rcg6 and next move 44...Qe4! would be curtains, but 44.Rd8 Kh7 45.Rh4 was possible, though 45...Qb1 looks pretty fatal in that case.
|Mar-01-09|| ||ChessDaZaster: On Sundays, I just look at the position and make guess based on what just "looks right" - no analysis. I mean, it would be pointless for me to do otherwise, having started playing chess 07/08. 40...Bxh4 looked right to me since it opened up the white king's last wall of defense.|
|Mar-01-09|| ||DarthStapler: I didn't get it|
|Mar-01-09|| ||patzer2: Bologan's decoy sacrifice 40...Bxh4!! forces the opposing Queen into position for a decisive attacking combination on White's helpless King. |
Here's my computer checked (Fritz) breakout:
<40. Bxh4!! 41. Qxh4 Rg5 42. Qh3 Rh5 43. Rd8+ Kh7 44. Rh4 Qc2+ 45. Kg1>
If 45. Kh1, then 45...Rg6! leads to a quick mate.
If 45. Qg2?, then Black gets an amusing composition-like mate-in-three after 45...Rg6! 46. Rd2 Qxd2 47. Qxd2 Rxh4#.
<45... Rg6+ 46. Kf1 Rxh4>
This move wins, but perhaps even stronger here is 46... Qc1+! 47. Ke2 (47. Kf2 Rg1! ) 47... Rxh4 48. Qxh4 Rg2+ 49. Qf2 Qc2+ 50. Rd2 Rxf2+ 51. Kxf2 Qxd2+ .
<47. Qxh4 Qg2+ 48. Ke1 Qxf3 49. Qh2 Rg2> 0-1
|Mar-01-09|| ||patzer2: Although it's certainly a bad move that White would never have made over the board, the position after 45. Qg2? leads to an amusing mate in three composition:|
click for larger view
(Position after 45. Qg2? Find Black's move to mate in three.)
See my previous post for the solution.
|Mar-01-09|| ||al wazir: <patzer2: . . . Find Black's move to mate in three.>|
45...Rg6, threatening 46. Qxg2#. If 46. Qxc2, then 46...Rxh4#. (White can delay the inevitable by one move with a spite check, Rh8+.) If 46. Rd2, then 46...Qxd2 47. Qxd2 Rxh4#
|Mar-01-09|| ||WhiteRook48: 50 Qxg2 Qxg2 is the only way to avoid immediate loss|
|Mar-01-09|| ||johnlspouge: < <crwynn> wrote: I [snip] didn't see that the Re4 is not really immobile because the Rd3 has a check to save it. >|
A zwischenzug to rescue a threatened piece before capturing with another piece is an important staple in the puzzles on http://chess.emrald.net.
|Mar-01-09|| ||CHESSTTCAMPS: <<Once> wrote:
The Bg5 line fizzles out into this...
40... Bg5 41. hxg5 Rxg5 42. Nc5 Rcg6 43. Rh4 Qf6 44. Ne4 Rg2+ 45. Qxg2 Rxg2+ 46. Kxg2 Qxh4 >
Very educational. You, Dzechiel, and I went for this line without seeing 42.Nc5. If black tries 43....Rh5 instead of Qf6, white has 44.f4! which allows the other rook access to h3. Black can try 44...Rxh4+ 45.Qxh4 exf4 46.Qf2 f3 hoping for the blunder 47.Rxf3? Qe5+! 48.Kh3 Qh5+ and black wins. But after 47.Qxf3 Qxc5 48.Qd5+ white has the advantage.
Actually, I won an endgame like this once against a class A player many years ago in a 1st round game at the Maryland Open (R plus 2 connected passed pawns versus R plus 2 connected on the opposite wing) even though my pawns started several moves behind. The result was likely influenced by the late evening hour.
In any case, the game continuation is so much better, I'm kicking myself for not looking into it deeper.
|Mar-01-09|| ||CHESSTTCAMPS: Just to correct/clarify the previous post, the following position might arise after the inferior continuation 40...Bg5:|
click for larger view
After 47.♖xf3, black can try ♕e5+ hoping for the blunder 48.♔h3?? when ♕h5+ wins. However, white is still OK after 48.♖g3 ♖xg3 49.♕xg3 ♕xf5.
From the diagram, white does better with 47.♕xf3 ♕xc5 48.♕d5+ ♕xd5 49.♖xd5, where white has at least a one or two tempo advantage in the endgame.
A sad outcome from a winning position for black.
|Mar-01-09|| ||crwynn: < johnlspouge: < <crwynn> wrote: I [snip] didn't see that the Re4 is not really immobile because the Rd3 has a check to save it. >
A zwischenzug to rescue a threatened piece before capturing with another piece is an important staple in the puzzles on http://chess.emrald.net.>|
Thanks for the link, it seems like a good way to train your ability to see simple tactics quickly and reliably. So far I am only rated 1600 there ;(
|Mar-01-09|| ||ToTheDeath: I calculated to Rg5 and decided white was in a real pickle and stopped, but then I only spend less than a minute on most of these problems and did not consider white's defensive line at all. so i can only claim partial credit i suppose.|
|Mar-01-09|| ||Malfurion: I actually got the first three moves and the winning idea, though I didn't calculate it out as far as the game went..seemed like it was winning though. My first Sunday victory - whoo!|
|Mar-02-09|| ||TheBish: O De la Riva Aguado vs Bologan, 2001|
Black to play (40...?) "Insane"
I'm doing this late in the "game" (already Monday EST), but still Sunday on the west coast. (At least when I started!)
I found 40...Bxh4! yesterday, but didn't have time to do much analysis earlier (moving this weekend). I didn't really find the solution "insane", because there weren't too many candidate moves! The "insane" part would be finding all the variations, I'm sure.
40...Bxh4! 41. Qxh4 (not 41. Rxh4? Qxd3) Rg5!
This is where the real test begins.
42. Qf2 (to avoid the pin after Rh5) Rh5+ 43. Kg1 (or 43. Kg3? Qh3# or 43. Kg2? Qh3+ 44. Kg1 Qh1#) Rg6+ 44. Kf1 Qh3+ 45. Ke1 (or 45. Ke2 Rg2) Qh1+ 46. Ke2 (or 46. Qf1 Rg1) Rg2, winning the queen for rook and bishop.
There are many other lines here, but it's late and it's time to see the game! Having seen this much, my instincts would tell me this sac is sound, and I think I would play this over the board. Since this is a "White to play and win" situation, I'm sure this is the right course!
< Earlier Kibitzing · PAGE 2 OF 2 ·