I am writing a code that will calculate the solid angle subtended by an off axis disk. I am using the table of answers in this paper on page 257 to check if my code is giving correct answers.
The table says that for a disk with radius 1, a source with distance 0.8 from the disk center and height 1 above the disk center will subtend an angle of 1.3488367 steradians. It also has a reference to a book (which possibly used a different method of calculating the angles) which says the solid angle is 1.34883 steradians.
My code gives the answer 1.348829928531692 steradians. When rounded to the same number of sig figs as the paper this is 1.3488299.
This agrees with the book, but not the paper. The error is only small but it is enough to cause problems for the application I am using.
There are two possible reasons for this error:
1) My code is wrong. I could have a mistake somewhere, it could be a result of rounding done during the calculations, it could be the fact that I use a numerical method to calculate the elliptic integrals used in calculating the solid angle etc.
2) The paper is wrong. I doubt this a little bit as this is a published paper, but I have checked my code meticulously and I can't find any errors.
So my question is this, where papers written around 1959 prone to errors in accuracy due to less powerful/less access to computers? I am hoping that my code gives the more accurate answer, as if it is not I could face problems later on.
Thanks!
note: In my code I have used the same method as the method discussed in the paper. This method provides an exact solution in terms of elliptic integrals.