Well, I used to hate it - a little bit, at least
But if it doesn't come again - I will certainly miss it. A little bit, at least.
The most difficult yose problem
-
jmeinh
- Dies in gote
- Posts: 35
- Joined: Tue Jul 21, 2020 3:13 am
- GD Posts: 0
- Has thanked: 2 times
- Been thanked: 1 time
Re: The most difficult yose problem
Will we ever get to see Dany's "Not yet" again?
Well, I used to hate it - a little bit, at least
But if it doesn't come again - I will certainly miss it. A little bit, at least.
Well, I used to hate it - a little bit, at least
But if it doesn't come again - I will certainly miss it. A little bit, at least.
-
dany
- Dies with sente
- Posts: 121
- Joined: Fri Dec 20, 2019 2:44 am
- Rank: IGS 4 dan
- GD Posts: 0
- Has thanked: 36 times
- Been thanked: 12 times
Re: The most difficult yose problem
I think there are multi objectives to compare variationsGérard TAILLE wrote:Oops as long as we talk about "preference" we have to define what is the objective.dany wrote: in principle, the solution too
but imho 94 white move is easier for black - black can immediately finish an unclear ko
B+0.5
so I prefer white's other move, which continues the ko
P.S. with ko trick the solution can be extended by 4 moves
For me I consider each player have to respect the best possible result. This best result being a black win by 0.5 point any move that will be a loss regarding this result has to be eliminated.
Here after blackwhite cannot play at "a" because it is great loss isn't it?
black objectives (as winning side):
- max black score
- min variation length
- min complication
white objectives (as losing side):
- min black score
- max variation length
- max complication
that is, there are 3 objectives - score, length, complication
and comparison GT94 with D94 imho looks like:
score, GT94 = D94
length, GT94 > D94
complication, GT94 < D94
-
dany
- Dies with sente
- Posts: 121
- Joined: Fri Dec 20, 2019 2:44 am
- Rank: IGS 4 dan
- GD Posts: 0
- Has thanked: 36 times
- Been thanked: 12 times
Re: The most difficult yose problem
the solution finally foundGérard TAILLE wrote: Black wins by 0.5 points after 105 moves.
What move do you think was the most difficult?
-
Gérard TAILLE
- Gosei
- Posts: 1346
- Joined: Sun Aug 23, 2020 2:47 am
- Rank: 1d
- GD Posts: 0
- Has thanked: 21 times
- Been thanked: 57 times
Re: The most difficult yose problem
The first white move was for me the most difficult white move. Seeing the big point at "a" my intuition as white is to find a sente sequence in the bottom right corner and to play at "b". It is not natural to me to find a seki gote allowing black to play at "a".dany wrote:the solution finally foundGérard TAILLE wrote: Black wins by 0.5 points after 105 moves.
What move do you think was the most difficult?
For black the most difficult move for me was at the very beginning of the second phase. Seeing the two ko threats at "a" and "b" I naturally chose the move at "a" because this ko threat cost nothing while the ko threat at "b" costs 12 points. The point is that I was too lazy to calculte the exact value of the ko and the exact value of the ko threats.
When I discovered that the value of the ko was 30 points and the value of ko theats was 30 points for "b" and "only" 29 points for "a" all become clear. The ko threat at "a" can be used only after a white move move at "c" which loses 1 point.
OC it would be completly wrong to conclude I consider the other moves easy !!!
A beautiful problem which help me a lot to understand basic subtilities for handling a two-step ko.
-
dany
- Dies with sente
- Posts: 121
- Joined: Fri Dec 20, 2019 2:44 am
- Rank: IGS 4 dan
- GD Posts: 0
- Has thanked: 36 times
- Been thanked: 12 times
Re: The most difficult yose problem
I agree. For me "b" the the most difficult move in the problem. But your explanation "The ko threat at "a" can be used only after a white move move at "c" which loses 1 point" is wrong. In the current position "a" is a valid ko threat. Beacause black has not yet added a stone in the left bottom corner. The real reason why "b" is better than "a" is described in Cassandra's diagrams.Gérard TAILLE wrote:For black the most difficult move for me was at the very beginning of the second phase. Seeing the two ko threats at "a" and "b" I naturally chose the move at "a" because this ko threat cost nothing while the ko threat at "b" costs 12 points. The point is that I was too lazy to calculte the exact value of the ko and the exact value of the ko threats.dany wrote:
What move do you think was the most difficult?
When I discovered that the value of the ko was 30 points and the value of ko theats was 30 points for "b" and "only" 29 points for "a" all become clear. The ko threat at "a" can be used only after a white move move at "c" which loses 1 point.
OC it would be completly wrong to conclude I consider the other moves easy !!!
A beautiful problem which help me a lot to understand basic subtilities for handling a two-step ko.
-
Gérard TAILLE
- Gosei
- Posts: 1346
- Joined: Sun Aug 23, 2020 2:47 am
- Rank: 1d
- GD Posts: 0
- Has thanked: 21 times
- Been thanked: 57 times
Re: The most difficult yose problem
You are right Dany. In fact it is difficult to answer the question "What move do you think was the most difficult" because for me the difficulty was not a specific move. The difficulty for me was to compare a ko threat at "a" with a ko threat in the upper right corner. Depending of the exact position different arguments are used and that is a great difficulty.dany wrote:I agree. For me "b" the the most difficult move in the problem. But your explanation "The ko threat at "a" can be used only after a white move move at "c" which loses 1 point" is wrong. In the current position "a" is a valid ko threat. Beacause black has not yet added a stone in the left bottom corner. The real reason why "b" is better than "a" is described in Cassandra's diagrams.Gérard TAILLE wrote:For black the most difficult move for me was at the very beginning of the second phase. Seeing the two ko threats at "a" and "b" I naturally chose the move at "a" because this ko threat cost nothing while the ko threat at "b" costs 12 points. The point is that I was too lazy to calculte the exact value of the ko and the exact value of the ko threats.dany wrote:
What move do you think was the most difficult?
When I discovered that the value of the ko was 30 points and the value of ko theats was 30 points for "b" and "only" 29 points for "a" all become clear. The ko threat at "a" can be used only after a white move move at "c" which loses 1 point.
OC it would be completly wrong to conclude I consider the other moves easy !!!
A beautiful problem which help me a lot to understand basic subtilities for handling a two-step ko.
I showed the diagram after the 40th move where the problem began and it lasted then for several moves. Really difficult indeed.
-
Gérard TAILLE
- Gosei
- Posts: 1346
- Joined: Sun Aug 23, 2020 2:47 am
- Rank: 1d
- GD Posts: 0
- Has thanked: 21 times
- Been thanked: 57 times
Re: The most difficult yose problem
Oops, it is not yet the solution. Strictly speaking white can reach 106 moves !dany wrote:the solution finally foundGérard TAILLE wrote: Black wins by 0.5 points after 105 moves.
What move do you think was the most difficult?
-
Gérard TAILLE
- Gosei
- Posts: 1346
- Joined: Sun Aug 23, 2020 2:47 am
- Rank: 1d
- GD Posts: 0
- Has thanked: 21 times
- Been thanked: 57 times
Re: The most difficult yose problem
I do not see to which post you answer.jmeinh wrote:Yes, Black captures at the top left with move 99. Although at the cost of "complexity" (but that is a matter of taste).
It seems it is not my last one because the purpose of black is to have the shortest sequence isn't it?
-
jmeinh
- Dies in gote
- Posts: 35
- Joined: Tue Jul 21, 2020 3:13 am
- GD Posts: 0
- Has thanked: 2 times
- Been thanked: 1 time
Re: The most difficult yose problem
Well, actually it was your last one, but I had confused the objectives of W and S, so mine was nonsenseGérard TAILLE wrote:I do not see to which post you answer.jmeinh wrote:Yes, Black captures at the top left with move 99. Although at the cost of "complexity" (but that is a matter of taste).
It seems it is not my last one because the purpose of black is to have the shortest sequence isn't it?
-
dany
- Dies with sente
- Posts: 121
- Joined: Fri Dec 20, 2019 2:44 am
- Rank: IGS 4 dan
- GD Posts: 0
- Has thanked: 36 times
- Been thanked: 12 times
Re: The most difficult yose problem
Alexander Dinerstein 3p: The most difficult yose problemdany wrote:Several years ago, in his lecture, Alexander Dinerstein 3p talked about solving this problem.
But imho not all the important moves were explained in his lecture, and some were wrong
great explanation, even though some moves were wrong
-
Gérard TAILLE
- Gosei
- Posts: 1346
- Joined: Sun Aug 23, 2020 2:47 am
- Rank: 1d
- GD Posts: 0
- Has thanked: 21 times
- Been thanked: 57 times
Re: The most difficult yose problem
Just a question Dany. IMO the main idea of the problem is to have a difficult ko fight which have to be prepared correctly, where the ko threats are more or less hidden, and where the different ko threats have to be played in the correct order. For me the length of the sequence is not really the point.dany wrote:Alexander Dinerstein 3p: The most difficult yose problemdany wrote:Several years ago, in his lecture, Alexander Dinerstein 3p talked about solving this problem.
But imho not all the important moves were explained in his lecture, and some were wrong
great explanation, even though some moves were wrong
In this sense isn't it more interesting to limit the number of ininteresting moves (filling dame while the opponent simply pass).
What about this position:
-
jmeinh
- Dies in gote
- Posts: 35
- Joined: Tue Jul 21, 2020 3:13 am
- GD Posts: 0
- Has thanked: 2 times
- Been thanked: 1 time
Re: The most difficult yose problem
After a small change in the order of Black's ko threats (analogous to the order in the video), the White ko trick no longer seems to work, Black seems to refute it. But maybe I'm missing something...
-
dany
- Dies with sente
- Posts: 121
- Joined: Fri Dec 20, 2019 2:44 am
- Rank: IGS 4 dan
- GD Posts: 0
- Has thanked: 36 times
- Been thanked: 12 times
Re: The most difficult yose problem
Yes, if black use top right ko threats before the bottom ko threat, then the trick stops working. And it's even good. There is some artificiality and senselessness in the trick. It doesn't add anything important to the solution.jmeinh wrote:After a small change in the order of Black's ko threats (analogous to the order in the video), the White ko trick no longer seems to work, Black seems to refute it. But maybe I'm missing something...
-
dany
- Dies with sente
- Posts: 121
- Joined: Fri Dec 20, 2019 2:44 am
- Rank: IGS 4 dan
- GD Posts: 0
- Has thanked: 36 times
- Been thanked: 12 times
Re: The most difficult yose problem
with F12 and K13 stones not the same problemGérard TAILLE wrote:Just a question Dany. IMO the main idea of the problem is to have a difficult ko fight which have to be prepared correctly, where the ko threats are more or less hidden, and where the different ko threats have to be played in the correct order. For me the length of the sequence is not really the point.dany wrote:Alexander Dinerstein 3p: The most difficult yose problemdany wrote:Several years ago, in his lecture, Alexander Dinerstein 3p talked about solving this problem.
But imho not all the important moves were explained in his lecture, and some were wrong
great explanation, even though some moves were wrong
In this sense isn't it more interesting to limit the number of ininteresting moves (filling dame while the opponent simply pass).
What about this position: