< Earlier Kibitzing · PAGE 2 OF 2 ·
|Sep-07-04|| ||tamar: Calli's improvement 29 Nc4 is very strong. If Rubinstein has a win after that, it is extremely well hidden. But does that mean the whole sacrifice was unsound? Or merely trappy?|
Using backward logic, knowing how active
White's pieces could have become, especially the rook that landed on h3, I think that was a key mistake for Rubinstein to play 22...Qxg3. He wins a pawn, but White profits later from mobility on the third rank.
Instead Akiba had an enormously strong continuation based on suppressing the rook's activity and blasting open the center at the same time.
22...Bb7 23 Rh3 f4!! This move leaves White with no defense. The strongest line according to Shredder 8 on Deep Analysis is now -4.28 after the following line
24 Ke1 Rad8 25 Bd3 Rf5!! another stunning move. 26 gxf4 Qxf4 27 Kd1 Ne3+ 28 Rxe3 Qxe3 29 Bxf5 and now Black wins by checking with the queen and keeping the King on the Kingside subject to mating patterns.
What do you folks think? Would 22...Bb7 23 Rh3 f4 lead to a crushing win for Black?
|Sep-08-04|| ||beatgiant: Nice analysis, <bunti>. That looks good for Black.|
But what if White instead sticks to <Calli>'s original idea of counter-attacking in the center? For example, 29. ♘c4 ♗d5 30. ♕d4 ♕g5 31. cb f3 32. ♘e3 f2+ 33. ♗xf2 ♘xf2 34. ♕xd5+ ♕xd5 35. ♘xd5 ♘xh3 36. ♗xh3 , and White has survived to a not bad endgame.
As <tamar> pointed out above, there are lots of tough branches, so we need some kind of guiding idea to study this.
|Sep-08-04|| ||Calli: <Tamar> Yes, I think your 22...Bb7 23 Rh3 f4!! is devastating. It proves the soundness of the sacrifice. However, since Rubinstein did not calculate it that way, he *was* being bit speculative. Still, even the way he played, I could not label the sacrifice as unsound because he could still probably draw even with White's best defense.|
<beatgiant> Yes, 29...Bd5 30.Qd5, attacking that bishop is the way to go. White has sufficient counterplay, IMHO.
|Sep-08-04|| ||Chessical: <Calli Tamar et al> So where does Euwe's position start to deteriorate? It must be before move 22. Perhaps, Euwe should have tried the sharp:|
13.Bxg7 Nf4 14.Qd2 Rg8 15.Nc3 (and if 15...Rxg7?! 16.Nb5)?
|Sep-08-04|| ||tamar: <So where does Euwe's position start to deteriorate?> Hard to say. The whole idea of 4 d4 cxd4 5 Qxd4 e6 6 c4 Nc6 seems like a dead end to me. Looking over the B29 lines in Explorer, Black won a huge majority in the early days. |
|Sep-08-04|| ||beatgiant: I would be interested in some more explanation of the ideas behind <tamar>'s surprising line with 25...♖f5 . It doesn't look like the rook has much additional scope for activity on the 5th rank, and besides it is exposed to White's bishop. Even after being shown, it is far from obvious to me what motivates this sequence of moves.|
Since it looks like <tamar> and <Calli> understand it, could one of you give a little bit of annotation for that line?
|Sep-08-04|| ||tamar: <beatgiant> <I would be interested in some more explanation of the ideas behind <tamar>'s surprising line with 25...Rf5> |
Okay. I thought Rubinstein's 22...Qxg3 wasted time, and proposed 22...Bb7
23 Rh3 f4!! 24 Ke1 Rad8 25 Bd3 Rf5!! as leading to an overwhelming attack. Here is what I was thinking, although I never imagined it would be so crushing.
22...Bb7 23 Rh3 f4 --The best part of this sequence is that it aids Black's attack without giving White any way to unravel his awkward pieces.
The King on e2 blocks the bishop and can only go to e1 to gain cover; d1 is out because of the knight fork on e3. Staying on e2 gives Black a free tempo with the f3+ push. Taking gxf4 is almost unthinkable; it brings the Queen in powerfully. On the other hand, not taking on f4 leaves the rook on h3 powerless to defend e3. So in some variations, White plays gxf4 though it looks suicidal!
24 Ke1 Rad8--Another free developing move, hemming in the King, and making a King flight to d1 even less achievable.
25 Bd3-- Attacking h7 and eyeing e4 for either the knight or bishop
25...Rf5!!-- Threatening Re5+, stopping Bh7+, and even preparing to pin along the e file if the pieces come to e4. What a move! Taking on e5 always exposes the rook on a1 to attack, gaining another tempo.
Here are some attempts to defend after
25...Rf5. I lost the original analysis on Shredder 8 Deep Analysis, which had incredible depth, and is perhaps worth duplicating if I can get enough time, but these give the ideas during a briefer look.
26.Bxf5 exf5 27.gxf4 Qxf4 28.Qb3 Qf2+ 29.Kd1 Qf1+ 30.Kc2 Qxh3 31.Bd4 Qxh5 32.Qc3 Qf7 33.Rf1 Qd7 -2.72/13
26.gxf4 Qxf4 27.Kd1 Bg2 28.Kc1 Bxh3 29.Bxf5 exf5 30.Qa4 Qe3 31.Qc2 Qf2 32.Kb1 Ne3 33.Qc1 Qg3 -2.89/13 (Here the original line given in the earlier post--27...Ne3+ is better, with a King hunt, although this line wins also.)
26.Rh4 Re5+ 27.Ne4 Ne3 28.Qb2 Rxd3 29.Bxe5 Qxc4 30.Rxf4 Nc2+ 31.Kf2 Bxe4 -3.16/13
26.c5 fxg3 27.Bxf5 exf5 28.Bd4 Rxd4 29.Qb3+ Bd5 30.Qxg3 Rf4 31.Kd1 bxc5 32.bxc5 -3.53/13
26.Nb3 f3 27.Kd2 Nf2 28.Nd4 Rf6 29.Bh7+ Kh8 30.Rh2 Qxg3 31.Rxf2 Qxf2+ 32.Kc1 Qe3+ 33.Kb2 Rxd4! 34.Re1 Qf4 35.Qf2 Rxc4 -4.23/14
26.Be4 fxg3 27.Qd1 Nf2 28.Bxf5 exf5 29.Qe2 Nxh3 30.Nf3 Qc6 31.Rd1 Rxd1+ 32.Qxd1 Qe4+ 33.Qe2 Qb1+ -6.14/13
|Sep-08-04|| ||bunti: Good job tamar!! if you have the time, i would love to see the deep analysis from shredder 8 on how to defend ...25. Rf5. Submit this to the editor for the next edition of art of attack.(lol) |
|Sep-09-04|| ||Honza Cervenka: 29.Nc4 would be better than 29.Rh4 but black's position is still clearly superior thanks to mighty passed pawns e+f. Simple 29...e3 looks quite well. I think that Rubinstein's sac of Bishop was based rather on evaluating of positional factors than on any concrete winning line and that it was almost certainly sound. On the other hand, to analyze possible lines after the 20th move of this game is great fun and above posted analyses are very interesting. Especially Tamar's analysis of 22...Bb7!? is excellent. |
|Sep-09-04|| ||Calli: <Honza> Not fair! We all looked at e3 and f3 for days and couldn't come up with anything definitive. You can't just come in here at the end without giving a line for us to rip up. :-) |
|Sep-09-04|| ||Honza Cervenka: <Calli> Well, I don't see any forced winning line there either but also I don't see any way to consolidate permanently white's position. For example, 30.Nd6 Bd5 31.Rd1 Be6 32.Bc4 Kh8 33.Bxe6 Qxe6 34.Qc4 Qf6 35.Qb3 bxc5 36.bxc5 f3! 37.Bxe3 Rab8 38.Qd3 Qe6 doesn't look like a satisfactory solution for white. Of course, this line is not forced but is there anything significantly better for white to play? |
|Sep-09-04|| ||Calli: IIRC, 29.Nc4 e3 30.Rd1 is better and then Nd6 later. If 30...Kh8 31.Rd6 for instance. |
|Sep-09-04|| ||Honza Cervenka: <Calli> If 29.Nc4 e3 30.Rd1, then what about 30...Qg5? For example, 31.Nd6 Ne5 threatening Qxg1 or Nf3+ is bad for white. |
|Sep-09-04|| ||Honza Cervenka: And if 29.Nc4 e3 30.Rd1 Qg5 31.cxb6, then 31...axb6 32.Nxb6 Ne5! |
|Sep-09-04|| ||beatgiant: The reason I rejected 29. ♘c4 e3 is lines like 30. ♘d6 ♗d5 31. ♗c4 ♕e6 32. ♖d1 ♗xc4 33. ♕xc4 ♕xc4 34. ♘xc4 and White seems to be surviving. In that line 31...♕g5 32. ♘e4 doesn't improve anything. |
A possible continuation in that ending is 34...b5 35. ♖h4 ♘f2 36. ♗xf2 ef+ 37. ♔xf2 bc 38. ♖c1 .
29. ♘c4 was a very good find by <Calli> that is holding up very well. There's a lot to learn by studying the possible attacks and defenses in a position like this.
|Sep-10-04|| ||Honza Cervenka: <beatgiant> Let's try to prolong a little bit your line 29.Nc4 e3 30.Nd6 Bd5 31.Bc4 Qe6 32.Rd1 Bxc4 33.Qxc4 Qxc4 34.Nxc4. What about 34...f3? If 35.Bxe3, then 35...Rae8 36.Kd2 (36.Rd3 Rf4 37.Rc3 b5 38.Nd6 Rxe3+ ) 36...Re4 37.Rc1 (37.Kd3? Rxc4 ) 37...b5 38.Nd6 Rxe3 and black wins. If 35.Nxe3, then 35...f2+ 36.Ke2 Rae8 37.Bh2 Rxe3+ 38.Rxe3 Nxe3 39.Kxe3 f1=Q 40.Rxf1 Rxf1 etc. |
|Sep-10-04|| ||beatgiant: Thanks for that analysis, <Honza Cervenka>. I did not analyze that ending in depth, because I was proposing 29. ♘c4 ♗a6 to avoid such an ending and thereby keep one of Black's main advantages, the attack on the exposed White king.|
The best defense I have been able to find starting from <Honza Cervenka>'s 29.♘c4 e3 30.♘d6 ♗d5 31.♗c4 ♕e6 32.♖d1 ♗xc4 33.♕xc4 ♕xc4 34.♘xc4 f3 is 35. ♗xe3 ♖ae8 36. ♔d2 ♖e4 37. ♖f1 ♖xc4 38. ♖1xf3 ♖d8+ 39. ♔e1 ♖xb4 40. cb ab , and this position still looks easily won for Black.
I haven't managed to find convincing improvements anywhere else in the line either, so 29. ♘c4 e3 is looking good! You master annotators never cease to amaze me!
|Sep-10-04|| ||tamar: <bunti> Here is the Shredder 8 Deep Position Analysis on the position after
22.Ke2 Bb7 23.Rh3 f4 24.Ke1 Rad8 25.Bd3 Rf5
I let it run for about a day and a half, just to be sure :)
22... Bb7 23.Rh3 f4 24.Ke1 Rad8 25.Bd3 Rf5!! Start Position
A) 27.Kd2 Nf2 28.Bd4 Nxh3 29.Bxf5 exf5 30.Qxf5 Qxc4 31.Qe5 Qxb4+ 32.Kc2 Qc4+ 33.Kb2 Qf7 34.h6 Ba6 35.Rc1 f2 36.Nd2 Qd7 37.Be3 Re8 38.Qc3 Qf7 39.Nf1 Ng1 -4.08/19
B) 27.Bd4 Rxd4 28.Nxd4 Qe5+ 29.Kd2 Qxd4 30.Qc3 Qe3+ 31.Kc2 (I was rooting for this to be the main line, as it features some acrobatic minor piece play in one variation, and an unlikely King march ending in consecutive discovered checks in the other!) Be4 32. Rh4 f2 33.Rxg4 ( or 33.h6 Qxd3+ 34.Qxd3 Ne3+! with the win of a piece after queening on f1) Qe2+ 34.Kb3 Qxg4 35.Qd4 Rd3 36 Qd8+ (Now the fun starts) Kf7 37.Qc7+ Kf6 38.Qd8+ (white must keep checking, but where can the black king find safety?) Kf5 39 Qf8+ Ke5 40.Qb8+ Kd4 41. Qd6+ Bd5 42 cxd5 Kxd3! 43 dxe6 discovered + Ke2 discovered+ wins (as does 43...Qd4 less dramatically)
|Sep-14-04|| ||Calli: Just getting back. <Tamar> Haven't had time to go over your lines but looking forward to doing so.|
<BeatGiant> I think it is key to leave the Knight at d6. After 29.Nc4 e3 30.Nd6 Bd5 31.Bc4 Qe6 then 32.Bxd5 Qxd5 33.Qb3 Qxb3 34.axb3 bxc5 35.bxc5 f3 36.h6 What do you think?
|Sep-21-04|| ||beatgiant: <I think it is key to leave the Knight at d6. After 29.Nc4 e3 30.Nd6 Bd5 31.Bc4 Qe6 then 32.Bxd5 Qxd5 33.Qb3 Qxb3 34.axb3 bxc5 35.bxc5 f3 36.h6 What do you think?>|
You are right, this is a better defense. It's not easy, although Black is still barely staying ahead in all the lines I've explored after this. The reply is probably 36...gh , and then perhaps 37. ♗xe3 ♘xe3 38. ♖g3+ ♔h8 39. ♔f2 ♘c2 40. ♖a4 a6! 41. ♖xf3 b5 42. ♖a2 ♘xb4 43. ♖d2 with good drawing chances for White. Black's extra material consists only of the doubled h-pawns, and White still has the advanced knight post. Is this the kind of line you had in mind?
|Sep-22-04|| ||Calli: ,beatgiant> I was mainly looking at 36...f2+ alsong the lines of |
32.Bxd5 Qxd5 33.Qb3 Qxb3 34.axb3 bxc5 35.bxc5 f3 36.h6 f2+ 37.Ke2 Rab8! and now if 38.hxg7? fxg1N!+ 39.Rxg1 Rf2+ 40.Kd3 e2 looks like a win
Instead White has 38.Rg3! Rf4 39.Bxf2 Rxf2+ 40.Ke1 Nxh6 41.Rxa7 looks OK.
Your 36...gxh6 line looks fine for White also. Why not 40.Rag1 with threats Rg7, Nf7.
|Sep-22-04|| ||tamar: <Calli, beatgiant> I've been away, but am interested if you have settled whether the 29 Nc4 e3 line is good for White or Black. My gut feeling is that White should have a saving or winning resource... |
|Sep-22-04|| ||beatgiant: <Calli>, I just noticed that I have been analyzing without Black's 34...bxc5, while you have this move. Sorry about that oversight. I have some more lines, but nothing conclusive. Black usually ends with a small material advantage which is hard to convert to a win. Maybe I will publish a summary later if I have time. |
|Sep-26-04|| ||beatgiant: <tamar><My gut feeling is that White should have a saving or winning resource...> I believe it is either a draw or a win for Black, but I haven't seen any lines where White has winning chances. I doubt I will ever "settle" it... although maybe the stronger players like <Calli>, <Honza> and <tamar> have a chance of doing so. I will give a few of the many lines I looked at.|
I start from 29. Nc4! e3 30. Nd6 Bd5 31. Bc4 Qe6 32. Bxd5 Qxd5 33. Qb3! Qxb3 34. axb3 f3 (sorry folks, I don't have time to also analyze <Calli>'s 34...bc here) and now the pawn break idea 35. h6 (trust me, 35. cb f2+ is no help).
The best I found for Black so far is then 35...f2+, with the idea of capturing the h-pawn with the knight instead of pawn to keep the pawn structure intact. If now 36. Bxf2 ef+ 37. Ke2 Nxh6 38. Rf1 Ng4 39. Rh4 Rf4 40. c6 Rd4 looks good enough for Black. So the focus turns to 36. Ke2 Nxh6. Now White has either 37. Rxe3 or 37. Bh2.
On 37. Rxe3 follows 37...fg(Q) 38. Rxg1 Nf5 39. Nxf5 Rxf5 40. c6 b5 41. Rc1 Rc8, and I think Black will win with careful play.
So the right defense is 37. Bh2 Ng4 38. Rh4 Nxh2 39. Rxh2 Rf4 40. cb Raf8 41. Rf1 ab 42. Kxe3 Rxb4 43. Rfxf2 Rxb3+, and it looks like a draw.
At this moment I'm back to the conclusion we had before: if there is a win after 29. Nc4, it is well hidden.
|May-25-06|| ||TheoreticalNovelty: I really like this game. Rubinstein sac's the Bishop to get two connected passed pawns as compensation.|
< Earlier Kibitzing · PAGE 2 OF 2 ·