I finally dialed in the last of my Canon lens today, a 400MM F5.6, and it was the only one that required a major compensation. By itself -8 was needed and with the 1.4X converter in place -1 was correct. Most of my lenses were on with little or no correction, for example my 70-200MM F2.8 was -2 by itself and 0 with the 1.4X in place.
I was a little surprised with the 400MM to see how much difference there was in required compensation between the bare lens and with the 1.4X in use.
Rather than follow the exact Canon procedure I shot a series of 5 shots each with the compensation at -10, -5, 0, +5, and +10 with and without the converter in the first iteration of testing to get a basic idea of where the compensation needed to be set and then fine tuned it in the second go round. The EXIF data nicely shows the AF microadjust compensation level so you know exactly where it was set for each shot.
I am planning to try the 400MM for some of my daughter's daytime soccer games this spring. It was a little too long for a lot of the work on my older 1.3X crop 1D Mark II but seems to be a good match for the full frame 1DX Mark II. But I expect the 70-200MM 2.8, often with the 1.4X converter in place, will get a lot of use. I plan to pick up a 300MM F2.8 before the Fall indoor season if I like shooting the outdoor games with the 400MM prime. The indoor dome is pretty tight and I know a 400MM would be too long for many of the shots there. In any case going from shooting primarily with a zoom to a prime will take some adjustment on my part but it will be a good learning experience.
I do not use the Canon technique for AFMA. I have found the “DOT TUNE” method on YouTube to be very effective.
I have found doing AFMA to be deceptively difficult. Your technique and skill at taking the test shots is really put to the test. Taking multiple shots, and averaging the results is highly advisable.
The type of lighting that you are using can make significant difference, too. Canon points out that you should make AFMA adjustments under the anticipated lighting conditions. Using indoor artificial lighting will yield measurable differences compared to test shots taken outdoors in bright sunlight.
After spending countless taking test shots, under varying lighting conditions, both indoors and outdoors, I have concluded that my “L” lenses are pretty much in the ballpark, arguably spot on target. The only lenses I have that seem to need AFMA are consumer grade, wide aperture primes, which do not focus consistently enough to make AFMA useful.
For example, the EF 85mm f/1.8 USM is a good example of inconsistent focusing. The common complaint is that the lens is soft wide open. It is not. The amount of slop in its’ ability to focus exceeds the DOF when the lens is wide open.
The 100-400 F5.6 is the only lens I own that required a significant change and although it is an L series lens it is probably less "L like" than my other L series lens since at F5.6 it is slower than my other L series.
I agree with you about the importance of simulating shooting conditions to get the most out of the fine tuning process. The good news with my 400MM 5.6 is that it was very consistent with its focus discrepancy and easy to dial in and definitely was much improved after the process.
The 70-200 F2.8 and 400 F5.6 were the two I was most concerned about but the way I use them they are probably the easiest to consistently dial in because both are primarily used with a single selected AF point. I suspect the field results from this process are less impressive when the lens is subsequently used with a wider array of AF points.
In any case the 400MM was back focusing prior to adjustment. When I bought this lens and my 1DM2 back in 2005 I sent the pair back to Canon for adjustment and that is probably why the lens now needs an offset dialed in when used with the 1DX M2 body. Focusing is still spot on with the 1DM2 so whatever Canon did 12 years ago held calibration very well.
" But I expect the 70-200MM 2.8, often with the 1.4X converter in place, will get a lot of use."
The Canon EF 70-200mm f/2.8L IS II USM Lens is one of the very few lenses that responds well to the use of a tel-con.
I want to suggest another option for you. The Canon EF 300mm f/4L IS USM Lens with the 1.4x tel-con. Why, yo uask? It works well and you get a 420mm f5.6 lens with IS.
This past weekend I have been comparing the Canon EF 300mm f/4L IS USM Lens with the tel-con vs the Canon EF 400mm f/5.6L USM Lens.
As most already know I consider the Canon EF 400mm f/5.6L USM Lens a must have lens for everybody. It is a unique lens for sure.
" I want to suggest another option for you. The Canon EF 300mm f/4L IS USM Lens with the 1.4x tel-con. "
I like your logic and I am planning on a similar approach this Fall except going with the 300MM F2.8 instead of the F4.0 version because I will be doing a lot of shooting under poor lighting so the extra stop really helps with focus speed and noise. (and it is a VERY expensive extra stop given the price difference between the two 300MM choices).
What would be perfect for me would be the 300MM F2.8 with a 1.4X extender like Canon builds into the 200-400MM F4.0 that allows the extender to instantly be switched in or out via a lever rather than physically removing/installing it. If they could create a standalone version of this type of 1.4X lever switched converter with good optics I would be an early adopter.