Author: Albert Silver
Date: 13:48:50 01/07/06
Go up one level in this thread
On January 07, 2006 at 14:38:07, John Merlino wrote:
>On January 07, 2006 at 13:09:42, Albert Silver wrote:
>
>>On January 07, 2006 at 12:40:09, John Merlino wrote:
>>
>>>On January 07, 2006 at 11:16:04, Albert Silver wrote:
>>>
>>>>In some testing against Toga in 20 min. games, Rybka played a strong maneuver,
>>>>and temporary pawn sac that preliminary analysis suggests may have been the
>>>>winning one.
>>>>
>>>>The following position arose after Toga's 36th move:
>>>>
>>>>[D]8/1r3q2/2p1p1k1/p1PpBppp/P1nP4/2P2PP1/4R1KP/4Q3 w - - 0 37
>>>>
>>>>Here Rybka played 37.h4! gxh4 (37...g4 will need to be better investigated)
>>>>38.Kh3!! hxg3 39.Qxg3+ Kh7 40.Bf4 Qf6 (I couldn't find any real improvement over
>>>>this BTW) 41.Qh4 Qxh4+ 42.Kxh4 Re7
>>>
>>>What about 38...Rb2?
>>>
>>>If 39.Rxb2 Nxb2 40.gxh4 f4! Draw
>>>If 39.Rxb2 Nxb2 40.Kxh4 Nxa4 41.Qa1 Qb7! Draw {42.Qxa4?? Qb1 Black wins}
>>
>>How about 41.Qc1 Qe7+ 42.Kh3 Qb7 43.Bd6! (43.Bf6 looks fancy but runs into Qf7)
>>Nb2 44.Bf8 and Black is in trouble. 44.f4 for example, 45.Qb1+ Kf6 46.Bd6
>>
>
>Your line looks good, but The King is not convinced that "Black is in trouble".
>After 46.Bd6 fxg3
>
>47.Be5+ Kg5 48.Qc1+ Kg6 etc.... Draw
Rybka views this line a bit differently. It gives 47.Be5+ Kg5 and here it
initially gave 48.Bxg3 and if 48...Kf6 49.Be5+ Kg5 50.Qg1+ winning.
However I could not find a win after 48...a4 where despite winning a piece with
49.Qe1 Qf7 50.Qc1+ Kg6 51.Qxb2, white appears to have no way to win due to the
combination of the a-pawn and perpetual threats after 51...Qxf3.
After some more time though, Rybka switches to 48.f4+ Kh6 49.f5 Qf7 50.f6 Qg6
51.Qc1+ Kh7 52.Qxb2 and now the line after 52...Qg4+ 53.Kg2 h4 54.Qb1+ Kh6
55.Qd3 h3+ 56.Kh1 Qg6 57.Qe3+ Kh7 58.Bxg3 Qxf6 59.Bf4 seems better for white,
though there is still a lot of play before declaring it won, if indeed it can be
won.
It is worth pointing out that black is clearly trying hard to save the game in
all this, which is sufficient to declare the initial pawn sac by Rybka as being
correct and best, unless one can show that there were better options.
Albert
>or
>47.Bxg3 Kg7 48.Qc1 Nc4 {not 48...Kg8? 49.Be5 Kh7 50.Qb1+ Kh6 51.Bf4 Kg7 52.Bc1
>Qf7 53.Qxb2 Qxf3+ and White should be able to convert} 49.Qg5+ Kh7 50.Qxh5+ Kg7
>etc... Draw
>
>jm
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.