Photo Taco – AFMA Explained!

Jeff Harmon answers year old listener question from Improve Photography writer Aaron Taylor:

Here's a Photo Taco episode if there ever was one: AF Microadjust/Fine Tune. Why do it? When to do it? How important? Is it obsessive or necessary? Use a Lenscal device or just something you print yourself?

Jeff talks about how autofocus micro adjustment (AFMA) is something specific to DSLR cameras, not needed and therefore not offered in mirrorless cameras. Not all DSLR cameras have the feature to do AFMA, but if it does then it is something worth doing – especially if you shoot portraits using narrow depth of field.

Jeff talked about an 11 point setup needed to do AFMA:

  1. Set your shooting mode to aperture priority. You can do this in manual mode but since you will be putting your camera on a tripod we don’t really care what the shutter speed is so that is one less thing to worry about.
  2. Set your aperture to the very widest setting, which means the very smallest number possible. If you have an f/2.8 lens set it to 2.8. If you have one that opens up to 1.8 or 1.4 then set it there. If your lens doesn’t go any wider (again the smallest aperture number) than 5.6 then you really don’t need to do this AFMA process.
  3. Set ISO to 100. You need to get images as clean as possible so you don’t want to increase the ISO.
  4. Set the focus point to use the single, center focus point. With some cameras you could get away with other setups but this is the safest way to get the most accuracy.
  5. Set your focus mode to single shot on Canon, af-s on Nikon. Again, this is a setting that will get the most accuracy out of phase detect autofocus that you are testing.
  6. Turn off image stabilization if it is available on the lens. We are setting up something very controlled here and in this controlled situation image stabilization could cause image problems making you think it is an autofocus problem.
  7. Good amount of light. Phase detection autofocus is most accurate with plenty of light so it is better to either do AFMA outdoors or add a lot of artificial light indoors.
  8. Use a tripod. You will be comparing images to one another in order to figure out what the AFMA value should be set to and this is much easier when they are all exactly the same.
  9. The camera needs to be as level with the target as you can make it. Don’t have the target significantly above or below the camera.
  10. The camera needs to be a specific distance from the target. As a general rule of thumb you multiple the focal length of the lens by 50. For example, if the focal length is 24mm you multiply that by 50 to get 1200 millimeters which is 1.2m or roughly 4 feet. There are a few online calculators you can use to get more precise measurements that I will put links to in the show notes.
  11. If you have done AFMA with this lens previously, set it to 0.

Then Jeff outlines 3 methods or processes to doing AFMA that seem to have almost equal support in the photography community with fanboys for each one:

  1. Ruler method – setting up a rulertarget on a 45 degree angle and then shooting shots with different AFMA values set through the camera menu to find which one most consistently results with sharp focus right where the focus point was placed in the scene. You can use DIY targets here with a ruler or tape measure, but paid for targets simplify things and are nicer to use.
  2. Software calibration. Tethering your computer to your computer you can have software figure out what AFMA value most consistently has sharp focus where the focus point is. The most popular of that software by what seemed on overwhelming majority is FoCal from Reiken.
  3. Dot Tune method invented by Fred Miranda. Ingenious out of the box thinking that leverages a comparison between contrast detection autofocus and phase detection autofocus plus the focus confirmation lights that come on when a lens is in manual focus mode.

After using all three methods Jeff preferred the Dot Tune as a process with FoCal software coming in as a very close second. A distant third was the ruler method. He did the tests many times across his four lenses capable of wide open apertures and couldn't find that one method was more accurate than another and unfortunately each method has fans of that method dead convinced it is the most accurate so it is hard to say if there is a meaningful difference in accuracy.

Links mentioned on the episode:

  • Photo Taco – Hyperfocal Distance Explained (Depth of Field explanation): https://improvephotography.com/44559/photo-taco-hyperfocal-distance-explained/
  • Photo Taco – Focus Points Explained: https://improvephotography.com/36979/photo-taco-focus-points-explained/
  • LensAlign MkII target: https://www.amazon.com/dp/B004G3PANY/ref=psdc_3109910011_t1_B0041L3JS4
  • Datacolor SpyderLensCal SLC100: https://www.amazon.com/gp/product/B0041L3JS4/ref=oh_aui_detailpage_o03_s00?ie=UTF8&psc=1
  • FoCal Target Distance Tool: https://www.reikan.co.uk/focalweb/index.php/online-tools/test-distance-target-size-calculator/
    Michael Tapes Design LensAlign Distance Tool: https://michaeltapesdesign.com/lensalign.html
  • Fred Miranda Dot Tune: https://www.fredmiranda.com/forum/topic/1187247

Other Photo Taco Resources:

  • Vote for Jeff’s “Cull” module idea to be added to Lightroom: https://feedback.photoshop.com/photoshop_family/topics/less_redundancy_actions_during_import_into_lr_instant_rating_during_import
  • Photo Taco Archive: https://improvephotography.com/category/taco/
  • iTunes review link: https://itunes.apple.com/us/podcast/photo-taco-quick-photography/id980781096?mt=2#
  • Facebook Photo Taco Group: https://www.facebook.com/groups/phototaco/
  • Photo Taco Instagram: https://www.instagram.com/phototacopodcast/
  • Photo Taco email: [email protected]
  • Jeff’s Twitter: https://twitter.com/harmon_jeff
  • Jeff’s Facebook Page: https://www.facebook.com/jsharmonphotos/
  • Jeff’s Portfolio: https://jsharmonphotos.com
  • Jeff’s Instagram: https://www.instagram.com/harmonjeff/
  • MacPhun Luminar: https://macphun.evyy.net/c/362006/185399/3255
Scroll to Top