This comes from the plays inside territory section of the Gokyo Seimyo. It shows the line through . As the variations show, at this point Black cannot avoid ko.
In a real game, the ko threat situation is typically significant. Normally, because the defender has more at stake, assuming that the attacker is komaster gives more realistic picture, even if the defender takes the ko first. (If neither player has a ko threat, who takes the ko is more significant, however.)
If White is komaster the line given is not optimal. Please consider what is best play when White is komaster in a real game, where there are many plays, and plays of comparable size on the board. (I am not presenting this as a problem, because best play depends on a number of factors.)
Enjoy!
Note: The komaster can win kos, but the concept is more specific. You can think of it this way. Black has no ko threat, and White has just enough threats to win any ko. That means that White cannot delay winning a ko, because of the threat that Black will take the ko back and White will have no threat.
Depending on the whole board situation, Black may be able to try another possibility.
[sgf-full](; AW[er][ao][bo][co][cp][dp][dq][ep][fp][gp][gq][hr][iq] AB[br][cq][bp][ap][dr][eq][fq][fr][gr]; W[es];B[ar] C[Black may have the option of keeping sente.]; W[ds];B[cr];W[gs] C[Black may be able to play this variation, e.g., if Black can make 6+ points elsewhere.] )[/sgf-full]
Re: Ko position from Gokyo Seimyo
Posted: Wed May 05, 2010 10:00 am
by Bill Spight
gonut wrote:
Bill Spight wrote:
Harleqin wrote:Black can shed some stones.
Yep.
Now, what about ?
Depending on the whole board situation, Black may be able to try another possibility.
(; AW[er][ao][bo][co][cp][dp][dq][ep][fp][gp][gq][hr][iq] AB[br][cq][bp][ap][dr][eq][fq][fr][gr]; W[es];B[ar] C[Black may have the option of keeping sente.]; W[ds];B[cr];W[gs] C[Black may be able to play this variation, e.g., if Black can make 6+ points elsewhere.] )
Excellent! Except for the last move. This is typically White's sente. (But not by much. )
Re: Ko position from Gokyo Seimyo
Posted: Wed May 05, 2010 11:13 am
by Harleqin
At first glance, I thought at A2 was about 4 zi worse than at C2. However, after A2, D1, C2, there is an 8 zi double gote endgame remaining. If White gets it, the result is in fact 4 zi worse, but if Black gets it, it is 4 zi better, so it seems that on average, it should be about the same. If the game were at an early stage, the tipping factor for me would be that there is still some miniscule amount of aji in the tiger's mouth, so I would like to save those four stones.
Re: Ko position from Gokyo Seimyo
Posted: Wed May 05, 2010 1:24 pm
by Bill Spight
(;GM[1]CA[ISO8859-1]SZ[19]ST[2]AP[GOWrite:2.2.21]FF[4]PM[0]GN[ ]PB[ ]AW[fp][gp][iq][bo][ao][cp][dq][ep][hr][co][dp][er][gq]FG[259:]AB[br][bp][cq][fr][fq][gr][eq][dr][ap]PW[ ] ;W[es] ( ;B[ds] ;W[bq] ;B[cr] ;W[ar] ;B[bs]TB[cs][er:es]TW[ap:bp][is]C[There is 1/3 point for Black around H-01. Net count = 1/3.] ) ( ;B[ar] ;W[ds] ;B[cr]C[The count is the average of -6 and 9 1/3, or 1 2/3.] ( ;W[gs] ;B[cs] ;TB[aq:bq][as:bs]W[fs]TW[eq:fq][fr:gr][hs:is]C[Net result\: -6] ) ( ;B[cs]TB[aq:bq][as:bs][ds:es][er]TW[is]C[There is 1/3 point for Black around H-01. Net result\: 9 1/3] )
)
)
On average, at A-02 is 1 1/3 point better than at D-01, followed by at C-02. OC, sometimes the latter play will be better.
Re: Ko position from Gokyo Seimyo
Posted: Wed May 05, 2010 2:32 pm
by gonut
Bill Spight wrote:
gonut wrote:Depending on the whole board situation, Black may be able to try another possibility.
[sgf-full](; AW[er][ao][bo][co][cp][dp][dq][ep][fp][gp][gq][hr][iq] AB[br][cq][bp][ap][dr][eq][fq][fr][gr]; W[es];B[ar] C[Black may have the option of keeping sente.]; W[ds];B[cr];W[gs] C[Black may be able to play this variation, e.g., if Black can make 6+ points elsewhere.] )[/sgf-full]
Excellent! Except for the last move. This is typically White's sente. (But not by much. )
Thanks for the endgame lesson. Sorry to be dense, but what does "typically White's sente" mean? I.e., why wouldn't White play that last move, which is worth about 15 points?
Re: Ko position from Gokyo Seimyo
Posted: Wed May 05, 2010 4:35 pm
by Bill Spight
gonut wrote:
Bill Spight wrote:Excellent! Except for the last move. This is typically White's sente. (But not by much. )
Thanks for the endgame lesson. Sorry to be dense, but what does "typically White's sente" mean? I.e., why wouldn't White play that last move, which is worth about 15 points?
The last move gains 7 2/3 points, on average. (The swing is 15 1/3 points.) The original position is an 8 1/3 point sente for White. The difference is less than 1 point, so it would not be unusual for White to have nothing better elsewhere than the 7 2/3 point gote after the 8 1/3 point sente. That's why I said that it was "typically" White's sente, even though it is theoretically White's sente.
. As the variations show, at this point Black cannot avoid ko.
?
)
at C2. However, after A2, D1, C2, there is an 8 zi double gote endgame remaining. If White gets it, the result is in fact 4 zi worse, but if Black gets it, it is 4 zi better, so it seems that on average, it should be about the same. If the game were at an early stage, the tipping factor for me would be that there is still some miniscule amount of aji in the tiger's mouth, so I would like to save those four stones.