Reikan FoCal - Automatic AF Micro adjustment software

[neuroanatomist]

@Dr Croubie - the technical answer is that AFMA doesn't adjust depth of field at all. What it adjusts is depth of focus, which is the sensor-side version of DoF, measured in microns. One unit of AFMA is 1/8 of the depth of focus for whatever lens is attached (at max aperture).

[neuroanatomist]

After reading another forum post indicating it works in an XP virtual machine, I think I'll try EOS utility on the VM to see if it connects to the camera (suggested on FoCal's website), and if it does, I'll buy and try.

[HDNitehawk]

After reading another forum post indicating it works in an XP virtual machine, I think I'll try EOS utility on the VM to see if it connects to the camera (suggested on FoCal's website), and if it does, I'll buy and try.

John; I was hoping you would. I am curious how the software results will compare with the settings you arrived at with the lensalign.

@ Dr Croubie, I had the same question when I started using the lensalign. My thinking was that if I new this making adjustments would be easy. What I have found is that the cameras autofocus is not accurate enough for the answer to matter. The DOF and the amount the AF shots vary is more than +/-1 will cover. This is one reason many have problems when trying to find the correct setting for a lens. Wide lenses are harder to pinpoint than long lenses. Zoom lenses can vary over their range. Cameras are only accurate and precise to a certain point. Using the lensalign gave me valuable lessons about what your camera and lens really do and what you should expect.

[bob williams]

Again, I spent the better part of the day working with FoCal and my two long lenses (500 and 100-400) with no success.

Two major issues:

1. Although the software easily and quickly recognizes the smaller target, I couldn't get the software to recognize/validate the larger target provided for use with longer lenses (above 300mm). Tried several reprints, followed the instructions, setup in daylight, etc--but still no luck

2. When set up on the smaller target, the software validates it just fine, but I consistantly rec'd the "inconsistant results" error which is why I thought the larger target might do the trick. I tried my distance to target at 12, 20 and 27 meters but was still unable to get any kind of consistancy with the test. When in live view mode, the focus did slightly drift in and out---like heat waves, but I understand this is normal with long lenses. Oh, and the outside temp was 34 degrees F-- I don't know if that had anything to do with it or not, but thought I would mention it. I also was shooting from my dining room out my back door and my house sits on a slab so no vibration there.

3. I am still happy with the software (at least for the short lenses) and am hoping to get a reply back from Reikan on this issue. If anyone has any suggestions please post them. At present, I am out of ideas.

[HDNitehawk]

Without the software did you have any trouble with the 500mm focusing on the target?
Was the door open? Not through glass I am assuming?

I was hoping you wouldn't have a problem with the 500mm. Let us know if you work it out.

[conropl]

Bob:

I do not have the 500, but I did use this with my 100-400 and the results matched my manual method. A couple of suggestions:

• They suggest printing the target at 300dpi. On my printer that did not look good so I printed at 600dpi which was way better.
• Make sure the paper and ink is not reflective, and make sure your lighting is consistant throughout the test (clouds moving in and out is not good).
• I know you are supposed to be 50 time the focal length for optimal testing, but you would be better off at the distance most shot from. Aslo, when I did my manual testing, I did it at the recomeneded distance and came up with an adjustment of +5. Then using the FoCal at a much closer distance (about 8 meters) I got the same results several times in a row. I guess I would suggest a shorter target distance. That is not normal practice, but it seemed to give me expected results.
• Did you turn off IS? If you did not, then that would make your live view drift in and out (FoCal suggests it be turned off).

Not sure that will solve you problem, but it is what I did.

Also, I was supprised at the results I got with my macro lens, but after checking it out at the setting suggested by the testing, it seemed to be correct.

Pat

[bob williams]

Without the software did you have any trouble with the 500mm focusing on the target?
Was the door open? Not through glass I am assuming?

I was hoping you wouldn't have a problem with the 500mm. Let us know if you work it out.

The camera and lens seemed to work normally without the software and it was focusing fine on the target----I was shooting through the open door and nothing touching the tripod..

[bob williams]

Bob:

I do not have the 500, but I did use this with my 100-400 and the results matched my manual method. A couple of suggestions:

• They suggest printing the target at 300dpi. On my printer that did not look good so I printed at 600dpi which was way better.
• Make sure the paper and ink is not reflective, and make sure your lighting is consistant throughout the test (clouds moving in and out is not good).
• I know you are supposed to be 50 time the focal length for optimal testing, but you would be better off at the distance most shot from. Aslo, when I did my manual testing, I did it at the recomeneded distance and came up with an adjustment of +5. Then using the FoCal at a much closer distance (about 8 meters) I got the same results several times in a row. I guess I would suggest a shorter target distance. That is not normal practice, but it seemed to give me expected results.
• Did you turn off IS? If you did not, then that would make your live view drift in and out (FoCal suggests it be turned off).

Pat

Pat, I will try another reprint as you suggested, I also checked and did everything else you mentioned. I was also haveing the same problem with my 100-400@400. I may also try shooting from the counter top for a little more stability.

Thanks

[bob williams]

I rec'd an Email back from Rich Meston of Reikan Technologies and here are his comments:

Hi Robert

I'm sorry to hear you're having problems with FoCal.

Firstly, the largest and smallest targets are not recognised by the currently released version of FoCal for Target Setup or Target Optimisation. Support for these targets has been added in the next release which is due out at the end of January.

Having said that, you should still be able to run a calibration with the larger target and your lenses. I would suggest you turn Target Validation to off and use the larger target, visually checking through the viewfinder that the centre focal point is aimed at the centre of the target. At the distances you are talking about, slight offsets from parallel will have less of an affect that closer, so it should not be an issue.

Could you let me know if using the larger target in this mode stops the inconsistent behaviour? If not, then it may be good to capture some debug logs, and I can explain how to do that.

Kind regards

So, the problem with the large target explained and I will try re-shooting with it this afternoon (weather permitting). But I think I will try shooting from my cabinet top just to eliminate and vibration.

[bob williams]

Hello, Hello---anyone out there

As suggested by Rich @ Reikan, I went through the procedure again and here is what I reported to Rich.

Rich, I went through the testing routine again with my Canon 7D and 500 F4L---and achieved the same inconsistent results. Here was my setup:

1. Camera and lens sitting on a granite counter top---on a concrete slab (I wanted to remove any chance of movement or vibration from a tripod).
2. I ran the test with the target in both sunlight (ev 11.6) and shade (ev 9.5)
3. I ran the test with the large target at 25 meters and 12 meters

I still had inconsistent runs and always rec'd the "inconsistent behavior" message on every run.

after 20 tests I had two readings that showed up the most often: +6 or 0. Since my own tests with a ruler and target normally indicate that I should be at a +5 to + 8, I decided to stick with the +6 reading.

If you want to see the debugging logs, please let me know how to retrieve and transmit them to you.

On a side note, I think this software you have developed is outstanding---if I can help test and provide you feedback I will gladly do so. Just let me know.

Based on my findings to date I have concluded:

1. There is something wrong with the software when it comes to long lenses; or
2. There is something wrong with my camera or lens.

Your comments are welcome.