NavList:

David Fleming,

I may not be reading your results correctly. Are you saying that one method gives you an index correction of 3.6' (rounded to a tenth of a minute of arc), and the other method give an index correction of 0.6' ? If so, something is very wrong. There's no way you would get such a wide range. If not, could you just summarize and tell me what the two I.C. values are that you get by the two methods.

By the way, your mathematical/statistical argument in an earlier post would only apply to identical "visual tasks", but we don't necessarily have that here. The visual process of bringing two Sun images in contact edge to edge is probably a "hyperacuity task" (N.B.: I can't guarantee that, but it's certainly a possibility). Superimposing two Sun images probably is not. Hyperacuity tasks are examples of visual phenomena where the human vision system can resolve differences at a level beyond the normal 1' capability of good (unmagnified) vision. This is a real, well-known property of human vision (Wikipedia article about hyperacuity). This won't really affect the accuracy of the final result which would probably be more than good enough in either case. My point is simply that you can not build an argument on a purely mathematical basis.

Frank Reed
ReedNavigation.com
Conanicut Island USA