AFMA Auto Focus Micro Adjustment techniques (Poll)

Just a little background - I've been noticing and have had comments made (in Critique Corner) that maybe I am having some focusing issues, primarily since buying my 6D. Not that my shots were always sharp before, but I seem to be noticing more missed focus shots. I'd like to think it isn't ALWAYS me that's the problem so I thought some AFMA might be needed?

Sometime back (before the 6D) I purchased Reikan's FoCal software to play around with, but I never felt like I got accurate results or maybe it is better to say that I could never really see any change. After some more reading/research I think I can chalk this up to my setup - bad lighting and maybe a less than ideal chart.

So I'm going to seriously sit down here soon and try to do some AFMA on my camera and lenses. However, during some recent research I found out there are a whole bunch of different methods. This led me to look at some of these as well (haven't purchased anything more however) and wonder if one is generally accepted as better than the others. My current plan is to get a better chart and some better lighting and then give FoCal another go. I will then probably test with something like Dot-tune (free) to see if I can validate the FoCal results.

Sorry for the long-winded intro - basically, I'd appreciate if you'd vote for your preferred method and maybe start some conversation on why you prefer it and your experience with other methods.

Thank you for your time!

---------------
Here are a couple of resources that I've come across that helped me better understand two of the most popular methods (so far). If you know of another video, blog, whatever it may be, that will help someone with any of these methods, please share!

Dot-Tune: http://youtu.be/7zE50j​CUPhM (youtube video)
FoCal: http://www.bobkeenanph​oto.com …r-lenses-focal-by-reikan/ (blog w/video)

"Other". I use the ISO 12233 resolution chart. It's available for download from the Cornell University server: http://www.graphics.co​rnell.edu …sc/ISO_12233-reschart.pdf

It's a PDF image, which I converted to a large JPEG (highest quality) and sent to Costco - the result was a 12 X 18 inch glossy print where the smallest divisions are perfectly sharp.

-js

I tried manual. Then I found the dot-tune method and it was so sensible it's all I've ever done. It's pretty easy and quick so I redo it every now and then.

I personally prefer the Dot Tune method but simply because I only have 3 lenses, two of them being fast primes and the other a UWA zoom. If I had a huge amount of lenses I could see FoCal being worth it.

I think lens brand also determines which may be best. If you have a Sigma art lens, then you would be looking at getting the dock as well since you may get different AMFA valued at different distances.

When I had the Sigma 50mm f1.4 EX I got great results using Dot Tune but with my other Canon lenses I only adjust if they seem to need it.

I previously thought I had inconsistent AF problems with my 35mm f2 IS but it was just the picture style making it appear that way. I adjusted my 35 to -1 just by judging with my eye and it's been perfect since.

Thanks for the responses so far! Keep the votes coming! I know there are a lot more than 10 of us doing AFMA

I'm going to cast my vote for Reikan FoCal, mostly because I paid for it and plan to get some good use out of it... However, I will be checking my findings with Dot-tune. Hopefully there isn't much, if any, discrepancy.

I'm assuming that good lighting is a necessity for any of these methods to be accurate? I'm curious to know how you guys are setup for lighting when doing AFMA.

I use Dot-Tune via Magic Lantern. Set up your target, fire up Dot-Tune in Magic Lantern, push your shutter button (I believe it's the shutter button, haven't needed to check focus in awhile), and a few minutes later, your AF is calibrated. Confirmed the results with free standing Dot-Tune, and they're right on.

I chose manual...i.e. real life shooting. I've never used any of the pay systems. I have used Dot Tune on a couple of occasions. Most recently with my 35 mm f/2 IS. I ran it through the method twice. Came up with -2 and -1. Interestingly, it was way off by front focusing. I simply stuck at +3 and it is perfect nailing tack sharp images 99% of the time. The other 1% likely user error/camera body error. I'm not sure why Dot tune wasn't perfect?? In theory, it should work great. In practice, this has not been my experience.

eddieb1 wrote in post #17473169
I use Dot-Tune via Magic Lantern. Set up your target, fire up Dot-Tune in Magic Lantern, push your shutter button (I believe it's the shutter button, haven't needed to check focus in awhile), and a few minutes later, your AF is calibrated. Confirmed the results with free standing Dot-Tune, and they're right on.

eddieb1 - Magic Lantern has had my attention in the past and I'm very curious about it, but I'm kind of hesitant to use it on my 6D only because I've read that they've stopped developing it and don't support it at all. I'm using the 1.1.6 firmware and I think I would have to go to 1.1.3 to be compatible with ML? Not sure how hard it is, or what the potential for breaking my camera is going backwards to an older firmware.

Thanks for your vote!

idsurfer wrote in post #17473180
I chose manual...i.e. real life shooting. I've never used any of the pay systems. I have used Dot Tune on a couple of occasions. Most recently with my 35 mm f/2 IS. I ran it through the method twice. Came up with -2 and -1. Interestingly, it was way off by front focusing. I simply stuck at +3 and it is perfect nailing tack sharp images 99% of the time. The other 1% likely user error/camera body error. I'm not sure why Dot tune wasn't perfect?? In theory, it should work great. In practice, this has not been my experience.

Thanks Cory - I think I saw your post(s) in the 35 f/2 IS thread as well. I don't think I have the experience with any of these methods (especially dot-tune) to help figure out why you results were so off, but I'm curious to know about your setup for testing? Plenty of light and a good target? Not sure how important the target is for dot-tune...

number six wrote in post #17472081
"Other". I use the ISO 12233 resolution chart. It's available for download from the Cornell University server: http://www.graphics.co​rnell.edu …sc/ISO_12233-reschart.pdf

It's a PDF image, which I converted to a large JPEG (highest quality) and sent to Costco - the result was a 12 X 18 inch glossy print where the smallest divisions are perfectly sharp.

-js

Me too, but wouldn't this be manual? I've tried other methods but the ISO12233 chart has worked the best for me. It's so easy to see when you finally hit the spot of maximum resolution.

freestylee30 wrote in post #17473182
eddieb1 - Magic Lantern has had my attention in the past and I'm very curious about it, but I'm kind of hesitant to use it on my 6D only because I've read that they've stopped developing it and don't support it at all. I'm using the 1.1.6 firmware and I think I would have to go to 1.1.3 to be compatible with ML? Not sure how hard it is, or what the potential for breaking my camera is going backwards to an older firmware.

Thanks for your vote!

I am running the latest nightly build in my 6D since it came out, and almost every 6D build before that. Surprisingly no hiccups at all. Reinstalling 1.1.3 is no big deal. You really should give ML a try. It has everything Canon should have included in their firmware, and it doesn't touch the Canon firmware.

gjl711 wrote in post #17473203
Me too, but wouldn't this be manual? I've tried other methods but the ISO12233 chart has worked the best for me. It's so easy to see when you finally hit the spot of maximum resolution.

I agree, I'd consider this manual. I came to the conclusion that these "manual" methods probably took considerably longer than things like dot-tune (manual, but quick?!?!) or paid software. Do you agree? How often do you do it this way and does it take a considerable amount of time?

eddieb1 wrote in post #17473208
I am running the latest nightly build in my 6D since it came out, and almost every 6D build before that. Surprisingly no hiccups at all. Reinstalling 1.1.3 is no big deal. You really should give ML a try. It has everything Canon should have included in their firmware, and it doesn't touch the Canon firmware.

Good to know, thanks. I'll have to do some more research!

freestylee30 wrote in post #17473216
I agree, I'd consider this manual. I came to the conclusion that these "manual" methods probably took considerably longer than things like dot-tune (manual, but quick?!?!) or paid software. Do you agree? How often do you do it this way and does it take a considerable amount of time?

It doesn't take long and I do it once thoroughly anytime I get a new lens or body. It probably takes 15 minutes or so a lens/body. I might check it sometime later if I have the chart out but so far, I have never had a lens that has changes properties. Once dialed in, it seems to stay that way.

The process is quite simple. I set up the chart outside in a nice sunny spot Like the poster above, I have it printed and then mounted to a piece of large foam board. Set up the camera on a tripod at whatever distance I need to have to (50x FL) un-focus manually and let the AF do it's thing. Then check the focus with 10x liveview to see if it needs adjustments. This dials it in pretty quick without need of taking a picture. Once I have it dialed in I take a couple pictures and adjust the MFA +/- a few ticks and viewing in PS look for the setting that gives me the best resolution. That's about it.

I used both FoCal and Dot-Tune in the past, and I liked Dot-Tune much better - better results, and much faster (when using under Magic Lantern).

freestylee30 wrote in post #17473193
Thanks Cory - I think I saw your post(s) in the 35 f/2 IS thread as well. I don't think I have the experience with any of these methods (especially dot-tune) to help figure out why you results were so off, but I'm curious to know about your setup for testing? Plenty of light and a good target? Not sure how important the target is for dot-tune...

I just printed a back and white focus chart from the internet. It looked like any other focus chart with criss cross patterns on it. I taped it to the wall in my garage when there was plenty of light flooding in from an open garage door. I put the camera on a tripod and set it back 6 feet or so and used the Dot tune method. Again, in theory it should give you the MA # that would yield the most accurate focus. This was not the case with my lens. In the end, I don't care at all about MA #'s. All I want is consisntanly sharp/acuratley focused images. I got there with the lens, albeit by simply shooting a bunch of real life images at a variety of distances, in a variety of lighting conditions. This said, I'd probably be a little disheartened if I was sporting a +/-15-20 or so