Behind the scenes of some endgame tesuji
-
Kirby
- Honinbo
- Posts: 9553
- Joined: Wed Feb 24, 2010 6:04 pm
- GD Posts: 0
- KGS: Kirby
- Tygem: 커비라고해
- Has thanked: 1583 times
- Been thanked: 1707 times
-
Uberdude
- Judan
- Posts: 6727
- Joined: Thu Nov 24, 2011 11:35 am
- Rank: UK 4 dan
- GD Posts: 0
- KGS: Uberdude 4d
- OGS: Uberdude 7d
- Location: Cambridge, UK
- Has thanked: 436 times
- Been thanked: 3718 times
Re: Behind the scenes of some endgame tesuji
Let's try to quantify how big l5 is then. Locally the game went as follows,
And then after black had got s3 (so lower right super alive) he played to increase territory on left and damage right in sente, which did actually surprise me as I thought m2/m3 was likely best local yose for him, but lets count this instead of the ambiguous position above. So here white spent 2 moves to get 2 points at k6 and reduced black to the triangled territory.
Compared to black first at l6: the same triangles are marked and squares are the extra black territory. There's 20 of them. So that's swing of 22 for 3 moves difference or 7 1/3 points per move. (I ignore slight change at h5 because when black defends white gains by saving that in sente, but black gains by getting h8 peep to help with centre thinness.)
Now for the top, if black spends 1 gote and gets sente monkey jump:
If white spends 2 gotes then I don't know how to count top right for black, but white has gained 8 squares compared to before.
If black gote, he has 8 points (squares) in corner beyond triangles.
If white gote, then white is creating followup at a which actually makes ~4 points in the corner, and black's move there only makes ~3, so let's say current value is 2 for white, 1.5 for black so net -0.5 for black which I will round to 0. So if black gote was 8 and white gote was 0 then current value is black expects 4 points.
Going back to black getting n18 and monkey jump black got 17 points beyond the triangles. So with these 3 gotes it went from black 17 to white 8 and black 4 ie black 13 to white 8 so difference of 21 or 7 points per move. 21 is close to 22. So seems they are about the same size
(though that was with black possibly making a mistake to make L5 bigger than it really was to settle the ambiguous area).
I suppose there's also the meta-gaming aspect that anyidiot 15 kyu up can see that n18 is a very big move, but l6 area is the kind of move you might miss the true size of, so if I want to bet on my opponent's lack of strength I could take n18 for myself and hope he doesn't see how valuable l6 is and then I can grab it later than I should if I assumed I was playing someone who will punish my mistakes.
And then after black had got s3 (so lower right super alive) he played to increase territory on left and damage right in sente, which did actually surprise me as I thought m2/m3 was likely best local yose for him, but lets count this instead of the ambiguous position above. So here white spent 2 moves to get 2 points at k6 and reduced black to the triangled territory.
Compared to black first at l6: the same triangles are marked and squares are the extra black territory. There's 20 of them. So that's swing of 22 for 3 moves difference or 7 1/3 points per move. (I ignore slight change at h5 because when black defends white gains by saving that in sente, but black gains by getting h8 peep to help with centre thinness.)
Now for the top, if black spends 1 gote and gets sente monkey jump:
If white spends 2 gotes then I don't know how to count top right for black, but white has gained 8 squares compared to before.
If black gote, he has 8 points (squares) in corner beyond triangles.
If white gote, then white is creating followup at a which actually makes ~4 points in the corner, and black's move there only makes ~3, so let's say current value is 2 for white, 1.5 for black so net -0.5 for black which I will round to 0. So if black gote was 8 and white gote was 0 then current value is black expects 4 points.
Going back to black getting n18 and monkey jump black got 17 points beyond the triangles. So with these 3 gotes it went from black 17 to white 8 and black 4 ie black 13 to white 8 so difference of 21 or 7 points per move. 21 is close to 22. So seems they are about the same size
(though that was with black possibly making a mistake to make L5 bigger than it really was to settle the ambiguous area).I suppose there's also the meta-gaming aspect that any
-
Knotwilg
- Oza
- Posts: 2432
- Joined: Fri Jan 14, 2011 6:53 am
- Rank: KGS 2d OGS 1d Fox 4d
- GD Posts: 0
- KGS: Artevelde
- OGS: Knotwilg
- Online playing schedule: UTC 18:00 - 22:00
- Location: Ghent, Belgium
- Has thanked: 360 times
- Been thanked: 1021 times
- Contact:
Re: Behind the scenes of some endgame tesuji
While Uberdude was posting his analysis, I did mine: two difference games (attempts, I'm 15 kyu at it) and one Kata move analysis in 1 sgf.
Edit: I now realize they aren't really difference games
Edit: I now realize they aren't really difference games
Last edited by Knotwilg on Fri Jan 03, 2020 1:42 am, edited 1 time in total.
-
xela
- Lives in gote
- Posts: 652
- Joined: Sun Feb 09, 2014 4:46 am
- Rank: Australian 3 dan
- GD Posts: 200
- Location: Adelaide, South Australia
- Has thanked: 219 times
- Been thanked: 281 times
Re: Behind the scenes of some endgame tesuji
While people are still thinking about Uberdude's and Knotwilg's analysis, here's another position from Shirae's book. This one is for reading, not counting, and I'm asking because I actually don't know the answer...
The book solution is a move that wasn't on my radar:
But my question is: if black tries 
at one of these two alternative points, I can see that black is thin and there's more than one way for white to punish black, but what is white's best continuation?
The book solution is a move that wasn't on my radar:
at one of these two alternative points, I can see that black is thin and there's more than one way for white to punish black, but what is white's best continuation?
Last edited by xela on Fri Jan 03, 2020 4:03 am, edited 1 time in total.
-
Uberdude
- Judan
- Posts: 6727
- Joined: Thu Nov 24, 2011 11:35 am
- Rank: UK 4 dan
- GD Posts: 0
- KGS: Uberdude 4d
- OGS: Uberdude 7d
- Location: Cambridge, UK
- Has thanked: 436 times
- Been thanked: 3718 times
Re: Behind the scenes of some endgame tesuji
A complication in this one if the white group has spare eyes in the centre such that white can ignore black's move in the corner as a pure gote yose, then bad move 1 actually has a bigger followup than the 'correct' move, in that s18 next actually makes a few 1st line points for black and white can't make 1 eye in sente at q17.
-
jlt
- Gosei
- Posts: 1786
- Joined: Wed Dec 14, 2016 3:59 am
- GD Posts: 0
- Has thanked: 185 times
- Been thanked: 495 times
Re: Behind the scenes of some endgame tesuji
There are so many variations, I don't know what the best response is but here is an attempt. If someone finds a mistake or a better response I would be interested.
at 
. If Black can't fight the ko, he will play 
differently:-
Bill Spight
- Honinbo
- Posts: 10905
- Joined: Wed Apr 21, 2010 1:24 pm
- Has thanked: 3651 times
- Been thanked: 3373 times
Re: Behind the scenes of some endgame tesuji
xela wrote:The book solution is a move that wasn't on my radar:
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins
Visualize whirled peas.
Everything with love. Stay safe.
At some point, doesn't thinking have to go on?
— Winona Adkins
Visualize whirled peas.
Everything with love. Stay safe.
-
xela
- Lives in gote
- Posts: 652
- Joined: Sun Feb 09, 2014 4:46 am
- Rank: Australian 3 dan
- GD Posts: 200
- Location: Adelaide, South Australia
- Has thanked: 219 times
- Been thanked: 281 times
Re: Behind the scenes of some endgame tesuji
I had different move orders which I think come to the same result!jlt wrote:There are so many variations, I don't know what the best response is but here is an attempt. If someone finds a mistake or a better response I would be interested.
If black plays the one-space jump:
before 
is better move order, because if black wants to try the ko variation with black stones at a and b, then black has to make the ko shape at an earlier stage, allowing 
? The gap at a means that later, b and c will be sente for white. I think it leads to the same count as your variation: black has one less prisoner but one more point of territory. White has a 9-point corner, rather than 7 points after the book solution and follow-up moves.