How effective is the 5D3 Micro Adjust for Zooms?

Anyone try the W/T portion of the Micro Adjust?

I'm mainly curious about the middle portion of the zoom range. I'm unsure how canon "averages" the Micro Adjust in the "middle" of the zoom range. And where is the "middle?" Say, for a 70-200, is the middle 135mm?

Anyone?

Generaly MF is not meant for zooms.

Daship wrote in post #15331242
Generaly MF is not meant for zooms.

Well, that isn't a sensible answer, and of no use whatsoever to the OP. It clearly IS meant for zooms in this case, as it is specifically designed to use different adjustment at different focal lengths of the lens. If that is not meant to be used by zooms, what else is it for?

The 5D3 MFA can be set at the wide end and the long end of a zoom lens individually, the manual then states that it interpolates a third adjustment for the mid-range focal length. Thus giving three different MFA settings as you zoom, reducing the problem of improving one end at the expense of the other when adjusting with a single setting for the whole lens.

To the OP, I'm sorry but I cannot answer your question from any experience. I have the 5D3 but all my zooms are accurate enough that I have felt no need to MFA them. I would assume that the third setting has a set of algorithms which compare the long and short settings you input and come up with a suitable setting for the midpoint. I wouldn't worry where that midpoint is, as the MFA will theoretically be a sliding scale between the amount at each end and the 5D3 is just going to pick the best point for each lens.

However the 5D3 calculates it, it will be a significant improvement on using a system that only allows for a single MFA setting for the whole focal range. If, like myself, you would only MFA a lens that is noticeably out at one end or the other then you can now correct for that without throwing out the other end. If your lenses are not showing front or back focus issues at either end, then I would just leave well alone, which is what I have done so far.

Try it and see. There worse that can happen is you spent time learning how to do it.

It would seem that being able to set 3 data points would be a better approach. I've not ventured into MFA yet, but plan to. I have seen people say they have negative numbers at one end, and positive at the other. That would imply by my simplistic view, that there is probably a focal length that is 0. If you could set a midpoint between the extremes, it would seem that their algorithm would work better. As an alternative, I would think you could sample different focal lengths and adjust the two settings you have to get the best fit across all. That may mean you give up a point or two on either in, to get the best overall numbers.