Yes....Adjustments have to be made anytime the cb is within about 8 to 10 inches or closer to the ob.
Maybe this method Oikawa has would work nice when the balls are close. That would be sweet.
To clarify, when I said the GB guess doesn't matter, I meant typical shots where the CB and OB aren't too close. Indeed, the further away they are, the smaller the error from misjudging the GB is. However, if it's a close cut, you can use the self-correcting principle I mentioned in the last page, to make it as accurate as you want, even when OB and CB are close.
So, without the self-correction, it's just as bad at close shots, but if you are willing to spend a quite long time iterating a few times, it works well.
There is actually maybe an even more accurate way of doing it for shots that are not thinner than a half ball hit (although even slower, so I'm not saying it's practical), which I have been tinkering with, which would remove any need of guesswork, and give an accurate result for any CB/OB distance. I just never got to testing it enough to be able to say how well it works, but if my current assumptions are correct, it should work, just requires good mental arithmetic to be accurate.
I'll try to explain the basic idea, maybe someone can spot a flaw.
If you use the same angle measurement system as mentioned by me, but instead of getting CB-GB and OB-target values and subtracting their difference, you measure the following:
CB - center OB
CB - edge OB (edge towards the cut direction)
OB - target
Subtract the difference between the first two, and replace the second value with that. Now you have:
CB - center OB
Difference between CB-center & CB-edge
OB - Target
Now, subtract the difference between the first and last values, and forget both of original values afterwards, only keeping the differences. Now you have:
Difference between CB - center OB & OB - target
and
Difference between CB - center & CB - edge
Now we, sort of, deduct the real GB location mathematically based only on those values.
We need to know two things:
Real angle value of a full ball hit is 0
Real angle value of a half ball hit is 2.5.
If any of our values if above 2.5, this means our shot isn't half ball hit or thicker, and the method doesn't work.
Now, if we say, for example, the first value is 1.5, and the second value is 0.5, it might look like this:
The first value (1.5, red line relative to the upper yellow line) tells the angle value to pot the ball if it was a full ball hit (which we know it's not, since the angle value is much larger than 0), and the second value (0.5, difference between yellow lines) added to the first value tells the angle value if it was a half ball hit (which we know it's not, since half ball hit is 2.5). So 1.5 = full hit assumption, (1.5 + 0.5) 2.0 = half ball hit assumption.
So, we know that the real angle value for our GB target is somewhere between 1.5 and 2.0.
With those values, 1.5 and 2.0, we can imagine/visualize two number lines:
0_________1_________2____2.5
_______________1.5__2_______
0 to 2.5
1.5 to 2.0
To get the answer, find the point where those two values meet, when both values are gradually going from one end to the other at a linear speed, such that they both start and end at the same point in time. To do this, i think of a table of the values, in some amount of divisions. In this case, I observe that the line can be neatly divided into 5 parts, and get an accurate answer by thinking of a table like this:
0 - 1.5
0.5 - 1.6
1 - 1.7
1.5 - 1.8
2 - 1.9
2.5 - 2
The crossing point is between 1.5 - 1.8 and 2 - 1.9, closer to the latter one, so the value is probably like 1.88'ish. But .1 precision is already 1/80th of a cut angle, you don't need 1/800th precision. So 1.9 is fine as an answer.
For example, with 0->1 and 0->2 values "racing", this crossing point would be at 0, since the only point they have the same value at the same point in time is 0.
Or, another example, with 0->2.5 and 2->2.4, it would be near 2.4.
To convert that angle value into an aimable overlap, you can, for example, use the table I mentioned on the last page.
This is very hard to explain neatly in words, I hope it's understandable enough.