FoCal AFMA Software: What are your techniques and tricks?

[RunAndGun]

Lately a few threads have popped up about AFMA and FoCal software and since I DL'd it a few days ago, as well, I thought I'd start a thread for those that use it to post what techniques/procedures they are really using and how well they think the results are.

I assume everyone that has it read the documentation and their guidelines/recommendations, etc. So... Are you doing it "by the book" (50x focal length, FoCal targets, etc.) or other ways such as different target distances(closer/farther away or trying to approximate what distance you use that particular lens the most), non-FoCal target(like a high quality Siemens Star or other back-focus chart/target because the FoCal targets printed w/an inkjet look soft) and for zooms, if you can only set AFMA at one offset(like on a 5DMKII) are you setting it for the wide or long end, or the focal length that you use the most? And any other tips, tricks, quirks or something I may not have thought of that you've found are welcomed, as well. Hopefully this is a thread that can help us all.

I've only had it a few days, so I stuck pretty close to "the book". I calibrated a 5DMKII and MKIII with my 8-15mm f/4L, 15mm f/2.8, 24-105mm f/4L, 100mm f/2.8 macro and 70-200mm f/2.8L IS v2. I'm waiting for my other MKII body and 16-35mm f/2.8L v2 and 85mm f/1.8 to come back that a friend borrowed for some timelapses. I tried to get in the 50x focal length range when I could. Space constraints in my office prohibited that at 200mm on my 70-200mm and on my 100mm macro I calibrated it pretty close to the FoCal target since I'm usually using that pretty darn close to whatever I'm shooting with it. And my W/A's were right on top of the target as well, otherwise the software wasn't giving me an "A-OK" for the target recognition.

One thing that I did find a little surprising, and this was only ONE test with ONE lens after calibration so I"m not going to read too much into it, was that on the multi-point focus test with the 70-200 on the MKIII, the center point was NOT the most precise. It indicated that the top-dead-center point and the second down from it were the two best. Like I said, ONE test with ONE lens, but I figured the dead center point would have easily "won".

So anyway, let's rock 'n roll...

 

[neuroanatomist]

I test at both 25x and 50x focal length. For zooms I test at several points in the range, approximately every 1x of zoom range, e.g. the 24-105mm is a 4x zoom, so I test at 4 focal lengths - 24mm, 50mm, 85mm, and 105mm (I will admit to being a little lazy with the 28-300L, and only testing at 6 points in the 11x zoom range). That was even with bodies allowing only one AFMA value per lens, but holds true with two values - more data on both distance and focal length results in better compromise values, biased by my usually-used distances and focal lengths and an appreciation that longer focal length at a given distance means shallower DoF and more critical need for AFMA.

From the setup side, best advice is lots of light - I use three 150 W-equivalent lamps.

 

[drolo61]

Where do I find the targets? Have bought the license and programm works fine (dust recognition) but now I would need the targets and feel stupid for not finding them. Can anyone help?

 

[JEAraman]

Where do I find the targets? Have bought the license and programm works fine (dust recognition) but now I would need the targets and feel stupid for not finding them. Can anyone help?

They come with the setup files. once you unzip the original zip file you'll see 'Target Images'.

 

[drolo61]

JEAraman, thanks a lot, was so perocupied and solely looked in the folder created by the software installation process, just fouund them in the original download.

 

[thebowtie]

I've used the software on my kit (50D, some EF-S 18-200 IS and some EF non-L and L lenses) - and love it.
What I did find is that my EF 50mm f/1.4 was the hardest to get consistent AFMA for.
I agree with the OP's remarks about getting a high level of illumination - I just used available light and found that I got inconsistent results (I have a glass-roofed conservatory where I used natural light).

What I did do (that I think helped) was to print out the targets on high-quality matte paper (I used EPSON paper) and mounted them on foam-core, and then I can use Blue-tack (removable adhesive) to mount the targets and ensure a perpendicular mount to the lens.

 

[mirekti]

What I did do (that I think helped) was to print out the targets on high-quality matte paper (I used EPSON paper) and mounted them on foam-core, and then I can use Blue-tack (removable adhesive) to mount the targets and ensure a perpendicular mount to the lens.

Where would be the best to go to and print these out?
Can I come to Walmart and ask for matte paper?

The other thing is I set 85mm at 425cm and the target covered some 12 AF points.
How should I test those on the left and right side, should I just print out two additional sheets and place them into that area?
As I'll be checking cross type only it seems to me that I should place the paper in vertical orientation no landsacape. Does this sound good?

 

[PeterJ]

Does anyone know why the 50x figure is often recommended? Personally I use the distance I normally use the lens at most. For example 50x for a 100 macro is 5 meters and on the odd occasion I use it at that distance f/2.8 isn't that focus critical, for true macro I usually manually focus so I set mine at about 1 meter which tends to be about the range I use most for quick handheld snaps where AF is important.

 

[Mt Spokane Photography]

Does anyone know why the 50x figure is often recommended? Personally I use the distance I normally use the lens at most. For example 50x for a 100 macro is 5 meters and on the odd occasion I use it at that distance f/2.8 isn't that focus critical, for true macro I usually manually focus so I set mine at about 1 meter which tends to be about the range I use most for quick handheld snaps where AF is important.

Its fine to use it at your commonly used distances. 25-50X is recommended because that provides a wide range of correction. If you move much closer than 25X, many lenses will start to front focus, so if you always shoot close, thats fine, but longer distances may not be focused as accurately.

Also note that it is not really crucial to have a lens perfectly focused, most of the time, a small error of a few points AFMA is not noticible.
The actual curve below shows this, there is little difference in sharpness at the peak of the curve, even if AF is off by +/- 3 points.