On Sun, Jul 17, 2005 at 05:03:16PM -0700, Godfrey DiGiorgi wrote: > > On Jul 17, 2005, at 2:06 PM, John Francis wrote: > > >>Perhaps you're seeing evidence of the fact that the "1.5x crop > >>factor" is a simple *linear* approximation of the field of view > >>difference. Field of View in fact does not change linearly with > >>respect to focal length on a format, a true factor for an 18mm lens > >>is probably a little closer to 1.4-1.3x to give a 100% accurate > >>calculation. I'll make up a chart demonstrating this for you in a > >>little bit. > > > >... and you'll find the factor is 1.5, not 1.4 or 1.3x. > > > >True, angle of view doesn't change linearly with focal length, > >or crop factor, or whatever. But if you reduce the (linear) > >width of the sensor by the 1.5 crop factor, you need to reduce > >the focal length by exactly that same factor of 1.5 to maintain > >the same angle of view. > > How do you explain the difference in the results from comparing the > angle of views as I did with Rui's FoV calculator, John? Just round > off error? There seems to be a predictable pattern to the error.
I don't know - I don't use any FoV calculator except the one in my head. But to me it seems painfully obvious that if you take a full-size diagram of the FoV of a 35mm frame, and a diagram of something with exactly the same FoV, but with a baseline smaller by the factor of 1.5 (or whatever crop factor you choose), the focal length will be reduced by exactly that same linear factor. Clearly this isn't obvious to everybody, but I don't know how to explain it better.
