stillstonevisions wrote:If you are correct, then perhaps it is less of an issue than I was making of it. I just did a test measurement from the front of the lens to a macro subject at a typical range. I don't know exactly how to measure to the entrance pupil, but I suspect the front of the lens housing is a good approximation. Anyway, distance to the center of the subject was 240mm. Change in position of the lens focusing from the front to the back of the subject was about 3mm, or just over 1%. There could be instances when I might be a bit closer than that, but probably not more than 2%, based on this test.
I'll keep testing.
The position of the entrance pupil can be estimated by looking in your lens through the front element when using an aperture preview (or actual long exposure time). The apparent position of the aperture blades is where the entrance pupil is. If you can estimate how much (if any) it moves when focusing to the extremes of your required focus range, you can also judge if it is less than the physical travel distance of the focus rail. The smaller of the two will change perspective the least.
It is hard to predict how much the entrance pupil moves, because it is influenced by the optical elements between the subject and the aperture blades. With internal focusing it may be hard to find along which path the apparent aperture position moves (if any) even if the lens barrel itself extends/shortens. It may also vary at different magnification settings (presumably more when closer to 1:1 because the extension is largest).
My guess is that, besides the preferred stationary lens and moving film plane setup, in general the (internal) lens focusing will move the entrance pupil less than moving it with a focus rail.
P.S. Because it may be hard to see, you could also judge from the amount of magnification difference between the front and back one of your original focus range images. The settings with the least magnification difference apparently changes perspective position the least.