|
< Earlier Kibitzing · PAGE 1 OF 2 ·
Later Kibitzing> |
| Oct-14-07 | | dzechiel: White to move. Material decidedly uneven (white has two bishops and a pawn for his queen). "Insane" <sigh> Wow. Right away I see an idea that keys on the open g-file. It goes: 28 Kh1 Rxd7 29 Rg1+ Kh7 30 Bf6 Rd4 (to stop 31 Be4#) 31 Rg7+ Kh8 and white has at least a draw. At this point white might be able to play 32 e3, which attacks the rook and covers d1 with the bishop. If 32...Rb4 then 33 a3 and the rook can no longer guard e4. Now there are a number of flies in the ointment:
- black doesn't have to take the rook on d7, he can take the pawn on e7 instead, giving back an exchange. - black can offer up the queen on move 30 with 30...c5 to stop the bishop from delivering mate. White could respond with 31 Bxb7 Rxb7 and we still have our draw in hand. Seeing as how this is "insane", I'm not going to spend all evening on this position. I feel strongly that the first move is 28 Kh1 trying to weave a mating net against the black king, but I can't see it all. Time to check. |
|
| Oct-14-07 | | dzechiel: Hey! Pretty close. I should have seen that the rook could stop the check by interposing on f5 as well as capturing on e4. Looking forward to Monday. |
|
| Oct-14-07 | | beginner64: I found all the correct moves, until 30..Rd5. I did not anticipate that defense, that is a key move due to which this problem is "Insane". I think if that move is not considered, problem drops down to a Friday level. Also, it is important to note that even after Rxf7, king does not get a chance to escape the files. |
|
| Oct-14-07 | | psmith: <dzechiel> Nice analysis. I got to pretty much the same place as you and decided to check as well. |
|
| Oct-14-07 | | psmith: Fritz 5.32 points out that 36. ♖g1+ is an inaccuracy. The very pleasing quiet move 36. ♔g2! declining the discovered check leads to a forced mate: 36... ♕d7 (36...♖g1+ just puts it off by a move) 37. ♖g4+ ♔h7 38. ♗g6+ ♔g8 39. ♗xe8+ ♔h7 40. ♖g7+ ♔h8 41. ♖f7+ ♔g8 42. ♖f8+ ♔h7 43. ♖h8#. |
|
| Oct-14-07 | | psmith: Nonetheless the whole game is quite amazing. |
|
| Oct-14-07 | | King mega: I am too lazy to check on this one :) |
|
Oct-14-07
 | | An Englishman: Good Evening & Good Heavens: I got this one. Not just the first move, but all of the main variations including the game through move 32. I'm really shocked. The rest of the game is pretty shocking, too. |
|
| Oct-14-07 | | MostlyAverageJoe: My brain overheated.
For a couple of minutes, I kept thinking: if only white pawn was on h3, not h2, then Kh2 followed by Rg1 and the rest should play itself. Now I need to write 100 times:
Kh1 if g-file is open...
Kh1 if g-file is open...
Kh1 if g-file is open...
...
What really annoys me is that a while ago I had to scramble hard to survive my opponent's Kh8 followed by Rg8, in a similar configuration. AAARGH. |
|
| Oct-14-07 | | MostlyAverageJoe: Okay, I thought white had some magnificent vision when the Q was sacrificed on move 23, but I see no win for white if black had only played: 27. ... Rxd7
Now, 28. Rxd7 is pretty much forced, but 28 ... Qc5 and the e-pawn is no more a threat. Am I missing something here?
|
|
| Oct-14-07 | | RandomVisitor: After 22...g5:
click for larger view
(22-ply)
1. (0.10): 23.Bxg5 Rxd1+ 24.Rxd1 Nd5 25.Rxd5 hxg5 26.Qd4 c5 27.Qg4 Qc7 28.Rd1 Qxe5 29.Bb7 Rd8 30.Rxd8+ Bxd8 31.Bxa6 Qa1+ 32.Kg2 Qxa2 33.Bxb5 Qxb3 34.Bc4 Qb7+ 35.Qf3 Qxf3+ 36.Kxf3 Kf8 37.h3 Ke7 38.h4 gxh4 39.gxh4 Kf6 2. (0.00): 23.Rxd8+ Rxd8 24.exf6 gxf4 25.fxe7 Re8 26.Rd1 f6 27.Rd7 Qc5 28.Bxf6 Kf7 29.Bh4 fxg3 30.hxg3 a5 31.Rc7 Qe5 32.Ra7 Qc5 33.Rc7 Qe5 34.Ra7 Qc5 35.Rc7 Qe5 36.Ra7 Qc5 37.Rc7 Qe5 38.Ra7 Qc5 39.Rc7 Qe5 40.Ra7 Qc5 41.Rc7 Qe5 42.Ra7 Qc5 43.Rc7 Qe5 44.Ra7 Qc5 45.Rc7 Qe5 <3. (-0.81): 23.exf6> gxf4 24.fxe7 Re8 25.Rd7 Rc7 26.Rad1 Kg7 27.R7d4 Qc5 28.Rxf4 f5 29.Rfd4 Rexe7 30.Bxe7 Rxe7 31.Rd6 Qa3 32.Rxc6 Qxa2 33.b4 Qa4 34.Rdd6 Qxb4 35.Rxa6 |
|
| Oct-14-07 | | beginner64: <psmith: Fritz 5.32 points out that 36. Rg1+ is an inaccuracy.> Absolutely, I thought so too. Kg2 is much more natural. Thanks for presenting Fritz' confirmation on this. |
|
| Oct-14-07 | | PositionalTactician: Quote: <MostlyAverageJoe> Okay, I thought white had some magnificent vision when the Q was sacrificed on move 23, but I see no win for white if black had only played: 27. ... Rxd7
Now, 28. Rxd7 is pretty much forced, but 28 ... Qc5 and the e-pawn is no more a threat. Am I missing something here?
Reply: Hey Joe, I think that after 28...Qc5 White can play 29.Kg2 and win. The point of this move is to play 30.e4. 29.e4 is not possible because of 29...Qc1+. Here is my analysis 29.Kg2 Qa3 (I think that black has to leave the queen on this diagonal, else Rd8, and if the rook takes the pawn on e7, I think that the two bishops and rook are too strong for the queen.) 30.e4 Qc5
31.e5(threatening Bf6) Qb4(if 31...Qa3 then 32.Bf6 Kg8(32...Kh7 33.Be4 is about the same position) 33.Rd8 Note that black cannot player Rxe7)
32.Bf6 Kh7
33.a3! and the black queen is overloaded. White will play either 34.Rd8 or 34.Be4 on the next move, depending on where black move. This is my own analysis. Can someone verify whether this is correct? |
|
| Oct-14-07 | | Alphastar: Since there is nothing really forcing, what I'd like to play here is 28. Kh1! with the idea Rg1+ and some mating threats.
Black has a lot of tries here, but lets first look at the most obvious:
28. Kh1 Rxd7
29. Rg1+ Kh7 (Kh8? Bf6+ Kh7 Be4#)
30. Bf6! and now it becomes very hard for black to prevent Be4#. Some tries:30. ..h5 31. Be4+ Kh6 32. Bg7#
30. ..c5 returning the queen seems best. 31. Rg7+ Kh8 32. Rxf7+ Kg8 33. Rg7+ Kh8 34. Bxb7 Rxb7 35. Rg5+ Kh7 36. Rxc5 and I think white has a won game. Maybe black has some better 1st move tries: 28. ..f5 29. Rg1+ Kh7 (Kf7 Bh5#) 30. Bf6! and black is in huge trouble again. The first thing he has to deal with is Rg7+ Kh8 Bh5! and Rg3+ Kh7 Bg6+ Kg8 Bxe8+ etc is unstoppable. Right, I think I've seen enough now, time to check. |
|
| Oct-14-07 | | cheski: <MostlyAverageJoe: .... I see no win for white if black had only played:
27. ... Rxd7 >
Or just about anything else but 27...Qb7.
Looking at the board without the engine's help, I cannot see why that Queen move was so disastrous. BTW, this particular Weinstein wasn't a Garry then :-) |
|
| Oct-14-07 | | goodevans: I didn't even get close! Very nice finish. |
|
| Oct-14-07 | | willyfly: Material is certainly not equal. White is down a ♕ and has only two ♗s and a ♙ (passed) for compensation. Because it is Sunday I'm not going to spend much time on this but I noticed one interesting line which I am submitting 27 ♔h1 ♖xd7 28 ♖g1+
if 28...♔h7 29 ♗e4+ ♔h8 30. ♗f6#
if 28...♔h8 29 ♗f6+ ♔h7 30 ♗e4#
of course this is totally speculative dependent on Black's 27th move. ok let's look and then on to Monday
-----
how about that?! I almost got it. I missed that 29 ♗e4+ could be met with 29...f5 |
|
| Oct-14-07 | | znprdx: Well this position calls for desperate measures. f5>f6 is pretty but useless Rc1 is equally unconvincing which leaves R[7]d6 as a definite maybe. Clearly some kind of combination of these ideas is the only hope. I] 28.f5 e6xf5 Aha 29.R[1]d6 presuming Rxd7[R] 30. Bf6+ Kg6 31. Rxc6 Qa7 and nothing. OK Enough wishful thinking here comes II] 28. Rxc7[[R] Qxc7[R]29. f5 e6xf5 30.Kh1 Na this is crazy. Back to square one. III] 28.Rd8 R[8]xe7 29.f5 e6xf5 Could it be possible? 30.Bh5 and after Kh1 White attacks on the g-file as Rg1+ threatens to connect the rooks. There is a fascinating unexpected potential here, but it is unclear as to what (if any) sequence is truly forced, however a draw seems increasingly plausible. I am in way over my head – unfortunately this requires setting up the pieces while looking forward today’s commentary. The computers will be overheating, but of course we will maintain our collected calm and cool – it has been a rather feisty week :) |
|
| Oct-14-07 | | znprdx: Truly delightful - 28.Kh1 takes nerves of steel to leave the rook >en prise< but can't Black do better with an immediate Kh7? (or perhaps Qb8 or Qb6) This slow motion attack is as good as it gets 34.e4 was pure serendipidity. ...35.Rd1+ was desperately pathetic, while 36.Rg1+(note it is a check <CG>) was perfectly natural and effective, notwithstanding Fritz's Kg2 which seems absurd. What would be the follow up? A basic principle of chess is to never allow for any resourceful counter-sacrifices. |
|
| Oct-14-07 | | ConstantImprovement: <First ideas:
White is too much down in material for something slow, radical measures are necessary: A mating attack.The only way to reach the black king is the g-file, with the two bishops assisting at the mate. Initial check: Against Be4(+) could come f5, so probably there needs to come Bf6 (blocking the f-pawn) somewhere before Be4(+). After these starting thoughts, let us try to bring that into existence: 28. Kh1 (the only move to assist the rook). Now it is getting really interesting: Is there really a threat? And if it is: What can black do against it? (I am deliberately extensive about
it to illustrate something for beginners. I can be wrong in my variations, nonetheless :) ). Should black not move at all, white could play:
29. Rg1+ Kh7 (See: 29. ... Kh8 is bad because of 30. Bf6+ Kh7 31. Be4#) 30. Bf6! (See: Be4+ f5) threatening Be4#. This can not be parried adequately, for instance 30. ... Rg8 31. Be4#.> I. After looking at White's threat, now our first thwarting try:
28. ... Rd7: 29. Rg1+ Kh7 30. Bf6! (As above)
1. 30. ...Rd1 is simply followed by 31. Be4#
2. 30. ... c6 (To guard e4) 31. Bb7:, winning the queen
3. 30. ... Rd4! (The best try: Guarding e4 with the rook) 31. Rg7+ (31. Bd4:? f5 with no immediate mate; 31. e3 Rb4 32. a3 is interesting, but slow, see: 31. ... Re7:) 31. ... Kh8 32. Rf7:+ Kg8
33. Rg7+ Kh8 34. Bd4: (Now, because f5
is not possible anymore); White is winning, because black, since white is
threatening discovered check, can not do anything against the maneuver Be4-Bg6 with Rh7++, Kg8, Rh8# to follow. II. Black could play f5 at once:
28. ... f5 Now it is getting difficult, there will be at least one quiet move necessary, and then black can for instance play Re7: to give the king the flight square f8.
1. 29. Rg1+
a. 29. ... Kf7 30. Bh5#
b. 29. ... Kh8 30. Bf6+ Kh7 31. Rg7+ Kh8 32. Rc7: Qc7: 33. Bh5, and white is winning because of the threat Bg6 and Rh7++, Kg8, Rh8#.
c. 29. ... Kh7
c1. 30. Bh5 Rf7:, see 2.
c2. 30. Bf6 Re7: 31. Rc7: Qc7:
c21. 32. Bh5 and after Bg6+ Kg8 it is difficult to continue.
c22. 32. Be5
2. 29. Bh5
a. 29. ... Rd7:
30. Rg1+ Kh7 (30. ... Kh8 31. Bf6+ Kh7 32. Rg7+ Kh8 33. Bg6 and white is winning) 31. Bg6+ Kg8 (Kg7 is the same) 32. Be8:+ Kh7 (32. ... Kh8 33. Bf6+ Kh7 34. Bg6+ Kg8 35. e8Q#) 33. Bg6+ Kg7 34. e8Q, and white is easily winning
b. 29. ... Re7:
30. Be7: Rd7: 31. Rg1+ Kh7 32. Bg6+ with perpetual check after 32. ... Kg8 33. Be8+ Kh7 etc., but 32. ... Kh8 33. Bf6+ Kg8 is critical. <Conclusion:
The critical line is 28. Kh1 f5! 29. Bh5 Rd7:! 30. Be7: Rd7: 31. Rg1+ Kh7 32. Bg6+ Kh8 33. Bf6+ Kg8, and it is critical for white.> |
|
| Oct-14-07 | | officeplayer: 23.exf6. First of all, the queen sacrifice for two minor pieces was NOT sound in this position. As <RandomVisitor> noted, white was able to maintain equality with 23.Bxg5, then the game(with perfect play from both sides)continues: 23. ... Rxd1+ 24. Rxd1 Nd5 25.Rxd5 hxg5 26. Qd4 c5 27. Qg4 Qc7 28. Rd1 Qxe5 29. Bb7 Rd8 30. Rxd8+ Bxd8 31.Bxa6 Qa1+ 32. Kg2 Qxa2 33. Bxb5 Qxb3 34. Bc4 Qb7+ 35. e4 Qb2 36. Qe2, resulting in this position: 
click for larger view
which is a dead draw. |
|
| Oct-14-07 | | aragorn69: <Random Visitor> <<3. (-0.81): 23.exf6> gxf4 24.fxe7 Re8 25.Rd7 Rc7 26.Rad1 Kg7 27.R7d4 Qc5 28.Rxf4 f5 29.Rfd4 Rexe7 30.Bxe7 Rxe7 31.Rd6 Qa3 32.Rxc6 Qxa2 33.b4 Qa4 34.Rdd6 Qxb4 35.Rxa6> Intersting that Rybka doesn't play 27.gxf4 as in the game. It must have a better move than 27.-Qb7 in store, probably 27.-Rxd7. And then? |
|
| Oct-14-07 | | ConstantImprovement: After computer analysis it shows that 28. ... Rd7: was a mistake. 28. ... f5 makes it much more difficult for white, who has to find some only moves. 29. Rg1+ (29. ... Bh5 transposes after the only black defense Re7:!.) Kh7 (Mate after anything else) 30. Bh5 (the only move) Re7:! 31. Bg6+ (the only move) Kg7 (fast mate against everything else) 32.
Rd8 (the only move, after Bf5:+ and Bg6 white still has to find Rd8) and white will win after 32. ... Rc8 33. Rc8: Qc8: 34. Be7:. The difficult thing, in my opinion, was
in the 29. Rg1  line to keep the possibility of Rd8 in mind. In the 29. Bh5-line it was necessary to see that after 29. ... Re7: 30. Rg1+ the move Kf8 does not work because of 31. Be7:#. After 28. ... f5 white will only get an endgame with R-B-B against Q, which is still about 3.75 according to Fritz, but will be still not so easy to win. |
|
| Oct-14-07 | | officeplayer: So after 23.exf6, black was able to win with accurate play and this goes: 23. ... gxf4 24.fxe7 Re8. Here white made another mistake with 25.Rd7, as this line (which was also played in the game) is certainly much worse than the best: 25. gxf4 Qc7 26. Kh1 Kh7 27. Rg1 f5 28. Bh5 Rxe7 29. Bg6+ Kg7 30. Bxe7 Qxe7 31. Bxf5+ Kf8 32. Bg4 c5, reaching this position: 
click for larger view
which is won for black, but with a lot more resistance from white. If we continue the analysis of 25.Rd7, we will find out that with 25. ... Rc7! 26.Rad1 Kg7! black is still winning, while the immediate 25. ... Kg7? allows 26.Kh1! and now 26. ... Rc7 is met by 27.exf4! (or 27.Rg1!) 27. ...Rxd7 28.Rg1+ (or 28.exf4+) 28. ... Kh7 29.Bf6 
click for larger view
and now black must return the queen: 29. ... Qd4 30. Rg7+ Kh8 31. Rxf7+
Qxf6 32. Rxf6 Rexe7 33. f5 exf5 34. Rxh6+ Kg8 35. Rxc6 Rd1+ 36. Kg2 Rg7+ 37.Kh3 Rdg1 38. Bd5+ Kh8 39. Rc8+ Kh7 40. a4 and it is not too clear if black could actually win (see diagram): 
click for larger view |
|
| Oct-14-07 | | playground player: You'd think a move like White's 28 Kh1 should have set off Black's alarm bells. "Why is he doing that? Why is he letting me pick off his Rook? What's he got in mind?" It makes me wonder if Black was under time pressure. |
|
 |
|
< Earlier Kibitzing · PAGE 1 OF 2 ·
Later Kibitzing> |
Kibitzing
sac: 23.exf6