Damn you Zenbeam!
The sheer amount of mathematics it takes to actually refute such a proposition is way beyond my . . . level of industry.
Not that you have convinced me, mind you!
The number of orientations is very high, as opposed to one, in the naturally occurring case. So the first question which occurs to me is just how “washed out” a rainbow we are talking about? Rainbows are spectacular, but they are hardly bright. I am sure that a random sample of orientations would be far less than ten percent likely to be within the narrow range you described.
Also, in those cases where the axis of the cylinder is directly perpendicular to the axis bisecting the source/viewer angle it seems to me that rotation within the plane perpendicular to the bisecting angle would allow different frequencies to be sent to the viewer. I don’t have a clue what the overall statistical picture would be, but it falls very far below the 100 percent that would be the same color in the natural model.
Oh, yeah and what about the RRR paths from all those cylinders outside of your “rainbow” area which happen to send single color spectra to the viewer because of their orientations? Perhaps your cylindrical rainstorm would just be a sparkling field of colors.
Sigh. I will have to think a long time about this proposed oddity. I won’t even venture into wondering who started making fiberrain particles to replace spherical raindrops, or why. Or what keeps them rigid along their axes.
Tris