|
< Earlier Kibitzing · PAGE 2 OF 3 ·
Later Kibitzing> |
| Sep-27-09 | | patzer2: Wow! Today's Sunday puzzle illustrates the insane difficulty of the endgame, with Alekhine's idea of 36...Rxc7 37. Bxc7 38. b4! axb4 39. Bb6 = as described in his game note to Black's 36...Kf7 "mistake." However, after <BeatGiant>'s suggestion of 36...Rxc7 37. Bxc7 38. b4! Kf2! 39. bxa3 Be5!  the outcome appears to be in doubt as White still has winning chances. See <OhioChessFan>'s post. According to http://en.wikipedia.org/wiki/Endgam..., endgame tablebases are limited to six pieces or fewer. Since the position in <OhioChessFan>'s post is a ten piece position, we might have to wait for some advances in programming and technology before we'll know the real solution to today's puzzle. P.S: Does anyone know how much hard drive memory it takes to install and fully utilize a six piece Nalimov endgame tablebase? I've heard estimates anywhere from 1TB to 5 TB. For now, I'll limit my searches in this area (six piece tablebases) to online database querys. |
|
| Sep-27-09 | | TrollKing: Way above my pay grade :-) |
|
Sep-27-09
 | | chrisowen: I thought caution rook takes or Kf7 splits it down the line. I went with Kf7, it was the wrong option eventually. |
|
| Sep-27-09 | | kb2ct: <Does anyone know how much hard drive memory it takes to install and fully utilize a six piece Nalimov endgame tablebase? I've heard estimates anywhere from 1TB to 5 TB.> 1.2 terabytes.
You can get them from me if you send me an external 1.5 terabyte drive. :0) |
|
Sep-27-09
 | | Open Defence: <<Does anyone know how much hard drive memory it takes to install and fully utilize a six piece Nalimov endgame tablebase? > I used to know this but I cant remember... |
|
| Sep-27-09 | | MiCrooks: Is there an online database that does 6 piece endgames? I thought the Nalimov one only did 5. Is there another? |
|
| Sep-27-09 | | Marmot PFL: Spent little time on this but probably would have played it the same as Lasker. The most interesting thing to me is whether white can improve in Alekhine's drawing variation and I will try to look at this later. |
|
| Sep-27-09 | | patzer2: Correction: It would take an 11 piece tablebase to solve the position (White to move) in <OhioChessFan>'s post -- unless a program or someone can simplify it to a 6 piece position. |
|
| Sep-27-09 | | Eisenheim: There's only two real choices here, but candidly I NEVER would have figured out 37 ...b4!! I thought through both kxf7 and rxr, but couldn't devise a plan to smash white's center. I considered a bishop sac, but that got me nowhere. still its been a good week and we have a new Russian friend from our GOTD yesterday and I look forward to her commentary |
|
Sep-27-09
 | | Jimfromprovidence: <MiCrooks>
<Is there an online database that does 6 piece endgames? I thought the Nalimov one only did 5. Is there another?>They are up to 6 now.
http://www.k4it.de/index.php?topic=...
"Bishops of opposite colors"? I already knew that; they are either black or white. |
|
Sep-27-09
 | | OBIT: Looking at this again the next day, I noticed the same idea that <beatgiant> also points out: after 37...b4, instead of taking the pawn, White can play 38. Kf2! With this move, White's center pawns stay mobile. Black's a-pawn obviously can't promote, but it does limit the activity of the White bishop, and *maybe* that's enough to draw. Various move orders from here are possible, of course, but White's basic idea has to be to push the center pawns. So, let's say 37...b4 38. Kf2! bxa3 (38...b3 39. a4! looks very promising) 39. Be5+ Kf7 40. Bc3 Ke6 41. Ke3 a2 42. d4 Kd6 43. d5 Kd7 44. Kd4 Kd6 45. e5+ Kd7. It doesn't take much more for White to win this. For example, if White can get a pawn to g5, then Kc5 followed by e6 forces the Black king to e7, and then Bf6+ wins. So, if 46. h4!?, Black has to make sure White cannot continue g4 followed by g5 to retain the blockade. It's tough, for sure. |
|
| Sep-27-09 | | sfm: Hmmm. In a way it is very simple - afterwards.... |
|
| Sep-27-09 | | BOSTER: After 36...Rxc7 37. Bxc7 the move 37...b4 looks really very attractive only because black has nothing to lose more.
But chess is not a checkers, it is not necessary to take a pawn. This is why all analysis by <johnispouge> and Co. is not very correct.( my opinion).
I can not accept such solution, when <Athamas> said: <My first instinct is Rxc7, Bxc7 b4...
I am not sure why>
But I agreee with <BeatGiant> that after 38. Kf2 white has more chance to win that black to draw. 38...bxa3 39. Be5+ Kf7 40.Bc3 black can not force the centre pawns to move onto dark squares and block them .
Certainly,<Patzer2> is right:< The outcome appears to be in doubt>. |
|
| Sep-27-09 | | zb2cr: Hi <znspdrx>,
The references to Alekhine are because the game annotations are by Alekhine, as noted at the bottom of the game viewer window. |
|
| Sep-27-09 | | goodevans: Just as 38 axb4 isn't forced, if white plays 38 Kf2 instead then 38 ... bxa3 isn't forced either. To my mind 38 ... b3 (threatening 39 ... Kf6) has a lot more going for it. |
|
Sep-27-09
| | pittpanther: An interesting endgame. I did get the correct idea due to the opposite colored bishops. b4 follwed by Ba6 is the only way to get the white pawns fixed on dark squares. There are plenty of these hard puzzles I miss but fortunately I had a good day today. |
|
| Sep-27-09 | | MathMage: <OBIT (38...b3 39. a4! looks very promising)> 39. a4?? Kf6 and wins. If 36. ...Rxc7 37. Rxc7 b4 38. Kf2, then 38. ...b3 39. Be5+ Kf7 seems to leave Black with good chances because White's bishop cannot leave the long diagonal. Black can connect his queenside pawns whenever he wants, and the King and Bishop may be enough to stop the center pawns with the restriction on White's bishop. A possible continuation would be 40. Ke3 Ke6 41. Bc3 a5 42. d4 (42. Bxa5?? b2; 42. a4?? Bc6) a4. Does Black survive after trading his Bishop for the passed pawns? |
|
Sep-27-09
| | OBIT: <MathMage>After 37...b4 38. Kf2 b3 39. a4 Kf6 40. Ba5 keeps the pawn from going anywhere (40...b2? Bc3+). The plan from here is Bc3 followed by a5 and Bb2. On b2, the bishop can go to a3 to chase the Black king off d6, and I think that should be enough to win. |
|
Sep-27-09
| | OBIT: Looking at 37...b4 38. Kf2 b3 39. a4! Kf6 40. Ba5 a little more, I'm convinced White wins. From here, 40...Ke6 41. Bc3 a5! prevents White from playing a5 himself, but even this positional advantage is not enough to hold the draw: 42. Ke3 Bc6 43. Bb2 Bxa4 44. d4 Bb5 45. d5+ Kd6 46. Kd4 a4 47. e5+ Kd7 48. e6+ Kd6 49. Ba3+ Kc7 50. Ke5 wins. After 37...b4 38. Kf2, Black's only move is 38...bxa3. The point is that the bishop has to stay on the a1-h8 diagonal to watch a1, and from there it is harder for the bishop to support the advance of the center pawns. |
|
| Sep-27-09 | | TheBish: Bogoljubov vs Ed Lasker, 1924 Black to play (36...?) "Insane"
From what I know about opposite-colored bishop endgames, they are often drawn if the pawns can be blockaded. Thus... 36...Rxc7 37. Rxc7 b4! 38. axb4 Ba6
If the White king were closer to the d-pawn (like on f2 or f1), then White would win. 39. d4 Bd3! 40. e5 Bc4 and Black can blockade the center pawns, i.e. 41. Kf2 Kf7 42. Bb8 a6 43. Ke3 Ke6 and Black needs only to safeguard his kingside pawns and the draw is secure. |
|
| Sep-27-09 | | WhiteRook48: I got the first two half moves |
|
| Sep-27-09 | | Athamas: Reading the comments and then looking at the position again... My initial reaction was similarly b3 or possibly a5 or Kf6, but they all seem sub-par to just accepting the pawn. Ke2 easily squelches black's chances if you push to b3. However, I think black still has decent drawing chances with his doubled passed pawns. 38. Kf2 bxa3 39. Be5+ Kf7 40. Ke3 Ke7 41. Ba1 a2 42. d4 Kd6 43. d5 Kc5 This is about as much as I can see... still looks like black has some decent drawing chances |
|
| Sep-27-09 | | MathMage: <OBIT> Oh, I see. Thanks! But what about 44. ...Bc6 anticipating 45. d5+ Bxd5 46. exd5 Kxd5? It seems like Black has a drawn endgame from there, at worst. |
|
| Sep-27-09 | | Old Wolf: after ..b4; Kf2 bxa3; Be5+ Kf7; Ke3 Ke6!; Bc3 a2; d4 Kd6. Now if White ever plays d5 then Black has Kc5, which looks like a draw because White can't get e5 in now, and the WK can't leave the defense of the pawns. |
|
| Sep-27-09 | | DarthStapler: I didn't get it |
|
 |
|
|
< Earlier Kibitzing · PAGE 2 OF 3 ·
Later Kibitzing> |
Kibitzing
Annotations by 
sac: 27.Rf1