Fujifilm AF Inconsistency Issue; My Final Word—For Now!

This is one of the scenes I shot to further test for AF inconsistency in my Fujifilm X-T4.  Focus
was with a single point, AF-S on the John Deere logo and headlights.  I then made two additional 
exposures re-autofocusing in-between.  See below for close crops and showing the autofocusing
inconsistency. As an aside, this was a very high contrast subject.  The X-T4 handled it with no problem.
X-T4; 16-55mm f/2.8 lens @ 35mm; 1/250th sec. @ f/5.6; ISO 200
(click to enlarge) 

Unless some new information comes to light, this is my final word about the autofocusing inconsistency issue I, as well as many others, have found with our Fujifilm X cameras.  For much more comprehensive information you can find the three posts I wrote on this issue here, here and here.  You can also find additional information about this on the Internet and in camera forums.  I'll make this short and sweet.

As a follow-up, for me to better understand the practical effect versus theoretical effect of the autofocus inconsistency issue, I set my X-T4 in various ways to test the autofocus accuracy.  Over several days, I randomly photographed many different types of subjects—the kind I normally would photograph.  In other words, the autofocus inconsistency issue is real but will it really affect my day-to-day photography?

I photographed wide angle, normal and telephoto.  I photographed wide open and stopped down.  I photographed up very close, at infinity and many distances in-between.  I set the camera for shutter activated autofocus, back-button autofocus, both on AF-S, AF-C and manual focus with 'Instant AF.'  I used three different sizes of single focus points, starting from the smallest.  I tried to tailor the focus point to the size of the item on which I was focusing so there would be no error with the camera grabbing the wrong subject.  Through all of it, I was cognizant of my shutter speed.  I kept the shutter speed high so camera movement would not be a factor.  Also, image stabilization was active on the 'Continuous' mode for all photographs.

Additionally, instead of just making one exposure of a subject then moving to another, I also repeatedly focused on a single subject to see if there was an inconsistency in the autofocus.  I did that in two ways.  I activated autofocus on a subject, then made an exposure.  Keeping the camera on the same subject, I activated the autofocus again and made another exposure, then another.  

The other method was to activate autofocus on a subject and make an exposure.  Then I activated the autofocus on a subject that was at a different distance, then back to my subject with AF activated again.  I wanted to see if there was a difference between repeatedly autofocusing on the same spot and if there was a variation if I autofocused on a spot, then focused at a different distance, then back to AF on the original spot.  Clear as mud?  I hope I explained it well.

After importing the images into Lightroom Classic, making 1:1 previews and editing each photo as I normally would (including adding some Texture and my normal Sharpening), I looked at each image for sharpness.  For ease of remembering where I focused, I made sure that my single focus point was in the center of the frame so it would later be easy to find if whatever was under that focus point was in sharp focus.  In other words, taking depth of field out of the equation.

What did I find?  

I shot a total of 114 images in the same manner as how I would normally work.  All were made on bright sunny days.  Of the 114 images I made, 12.2% of the images were out of focus.  Two were significantly out of focus (not even close), ten slightly out.  That is not good news.

Of the 19 separate sequences of three consecutive shots I made where I repeated autofocus on the same subject without moving the  camera,  one of the three images was slightly out of focus in four of the sequences.  The other two images in those sequences were in sharp focus.  Again, not moving the camera, just re-autofocusing between images.

This is crop of the above image.  The John Deere logo is sharp.  This was the first in a three
image sequence in which I re-autofocused between each image. (click to enlarge) 

This is the same crop from the second image in the sequence.  It is clear the John Deere log is not
accurately focused and is not sharp.  The third image in the sequence was also sharp.
(click to enlarge) 

Of the two sequences when I focused on the subject, made the exposure, then focused at a different distance before autofocusing again on the subject, all were in sharp focus. 

From that very limited but unscientific test, it appears that autofocusing at a different distance between exposures improved accuracy.

In conclusion, from a practical point of view, the autofocus inconsistency issue can impact my day-to-day photography.  For whatever reason, with the autofocus set in several different ways, the autofocus was inconsistent for some technical reason.  As I find things now, I would never just make one exposure of a subject.  I would make at least three just to be sure at least one is in sharp focus.  Again, that is not how it should be.

All of this said, I don't plan on selling my Fujifilm gear.  When I return from the cross-country road trip I may send my X-T4 to Fujifilm USA to have the camera thoroughly checked out and the autofocus system checked and calibrated to ensure it still meets factory specs.  After that, if the issue persists, I'll then make a decision as to whether or not I'll hold onto it.  I really don't like the idea of selling my Fujifilm gear as it brings me the most pleasure of all the digital cameras I've owned.

Again, as before, my tests are largely unscientific and for my own knowledge.  Take them for what you think they are worth.  I do this to better understand my gear so I can continue to improve my photography.

Join me over at my website, https://www.dennismook.com.  

Thanks for looking. Enjoy!  

Dennis A. Mook  

All content on this blog is © 2013-2022 Dennis A. Mook. All Rights Reserved. Feel free to point to this blog from your website with full attribution. Permission may be granted for commercial use. Please contact Mr. Mook to discuss permission to reproduce the blog posts and/or images.