[keanex]

Looking for a guide on how to properly adjust the fine tuning on my lens. I just bought a 50/1.8 and noticed some front focus. I know how to adjust it, and have done so, and it focuses beautifully on closer subjects, but it seems to front focus on subjects further away so I'm wondering what is the proper way to do so?

[jatrax]

Here is one article: Fixing Front and Back Focus - Introduction - In-Depth Articles

There are a number of ways to do it actually. A web search will turn up all manner of methods.

However, no matter which method you choose keep in mind that AF fine tuning requires precise (and I really mean precise) setup and testing. A quick setup without precise measurements will generally have a margin of error greater than the original lens error. Meaning any "adjustments" are just random changes that might be for better or for worse.

You need a sturdy tripod, a good target and multiple tests with mirror up and a remote release. The camera and target must be precisely aligned in both vertical and horizontal dimensions. Testing should consist of 5 to 10 images taken and compared for each adjustment.

Blindly adjusting things almost always leaves things worse than you started with. If you knew all that, my apologies, but we often see new comers that find the AF fine tuning and think it is a magic wand. It is definitely a great feature but only if used properly.

[keanex]

Thanks for the tips. I just spent the past 30 minutes doing fine tuning starting at -10 taking a shot of a barcode all the way to +10 at 1.8. I found that my camera was front focusing pretty badly, so -9 is where I ended up. I'll do a more precise tuning tonight with my tripod and a 2 second delay. It won't be the best I suppose but it's definitely looking a little better than it was.

[jatrax]

It won't be the best I suppose but it's definitely looking a little better than it was.

With a 1.8 (or 1.4 or 1.2) lens you are dealing with a error measured in millimeters, or less. So doing it handheld means that the slight movement of the camera caused by pressing the shutter button is enough to cause a focusing error. Just leaning forward a tiny bit as you push the shutter button can move the camera quite a bit. Tripod is essential and so is precise alignment of the target. Use the level feature in the camera to get it perfectly level in both directions. Use a level or plumb bob to get the target perfectly vertical. Good luck!

[keanex]

Thanks, definitely going to set it up on a tripod in a few minutes and create a makeshift depth chart like the ones shown in the guide. I will say that when looking at shots that had front and back object from what I locked in the focus on, made the focus WAY off, like way front. If I focused on an object at a slight angle the object to the right/front would be in focus, and it was definitely noticeable. Thank god for Pentax having fine adjustment easily accessible!

---------- Post added 01-30-16 at 07:35 PM ----------

Okay, so I did a long AB test of tripod mounted focusing and find there's still a slight preference towards front focusing at -10 when 1.8. I doubt I'll notice it when push comes to shove, my question is is it possible for this to get worse over time? If so I'll return it for a less out of adjustment lens.

[CarlJF]

You can also try, or double-check, with the diffraction pattern method: AF microadjustment for the 1Ds mark III, 1D Mk3, 5D Mk2, 7D, 1D X

[zbrueningsen]

Thanks, definitely going to set it up on a tripod in a few minutes and create a makeshift depth chart like the ones shown in the guide. I will say that when looking at shots that had front and back object from what I locked in the focus on, made the focus WAY off, like way front. If I focused on an object at a slight angle the object to the right/front would be in focus, and it was definitely noticeable. Thank god for Pentax having fine adjustment easily accessible!

---------- Post added 01-30-16 at 07:35 PM ----------

Okay, so I did a long AB test of tripod mounted focusing and find there's still a slight preference towards front focusing at -10 when 1.8. I doubt I'll notice it when push comes to shove, my question is is it possible for this to get worse over time? If so I'll return it for a less out of adjustment lens.

If it is a K5 it could get worse over time. Both my K5s did that and I've read some others have also. I think someone called it focus drift one time. But my guess if it isn't the K5 you should not have that worry.

[jatrax]

is it possible for this to get worse over time? If so I'll return it for a less out of adjustment lens.

I've no idea about that, seems unlikely but who knows. I do know that I had to AF tune almost all of my lenses on the k-5. On the k-3 I went through the testing process and finally just gave up, out of 10 tested only one had any noticeable error and it was not quite -1, so I just left it alone.

As to returning it. If I were you I would conduct the test again, in different light preferably daylight. If the results are the same, then yeah I would return it. This does not mean the lens is bad, just that it is 'off' in a direction that makes it not work with your particular camera. The way I understand it is if say the lens is +6 and the camera is -4 then an adjustment of +2 makes it right. But if the lens is +6 and the camera is +5 then you need +11 to make it right, which you cannot do.

You don't say what camera but I believe the k-5 had a particular issue with front focusing in incandescent (yellow) light. So if you have a k-5 you might want to read up on that issue. One of the improvements in the AF of the k-3 was to adjust for this.

[beachgardener]

I had my K-5 calibrated with the DFA 50 macro lens by C.R. Kennedy. It turned out that the sensor distance needed calibrating. Everything is spot on now with out need for fine adjust. Now in the process of checking out lenses with my K-3.

[keanex]

Sorry I have a K-50. I'm going to return it then since it's at -10 and I'll hope for something closer adjusted to my body!

[Ray Cox]

There was a recent article regarding lens fine tuning on Roger Cicala's blog ( Lens Rentals ) that pertains to this issue. It could be of interest to you.

LensRentals.com - How to Use AF Microadjustment on Your Camera

Hope it will be of interest.

[keanex]

Thanks!

So I did some more shooting and am finding -5 works best for me. I'm thinking the poster that said this was a fools errand may have been right! haha. Oh well.

[bschriver11]

I went through several of the focus tests with a a k-30 with a 70mm macro without any luck and then i noticed that when i photographed on some type of fabric i could see the lines of the weave very distinctly so i set up a small target on a cloth table place mat and set the camera at a 45 degree angle to the mat and focused on the target and i could clearly see it was back focusing,from there it was easy to adjust and hits right on now.If the depth of field is shallow enough it creates a distinct line across the frame exactly where it is focusing.

[keanex]

Alright - sorry to keep going with this but it seems there's still a very slight soft front focus at 1.8 when at -10. I could live with it, but I know that in the back of my head I'll always worry about potential "focus drift" or something. Let's see how support is!

[stevebrot]

Blindly adjusting things almost always leaves things worse than you started with. If you knew all that, my apologies, but we often see new comers that find the AF fine tuning and think it is a magic wand.

...and then we see a follow-up post saying that their camera AF sucks. AF fine adjustment is usually not needed with most lenses. What is interpreted as poor calibration is usually poor precision* or the expectation that the camera will read the photographer's mind regarding desired focus point.


Steve

* Precision = the chance of locking the same focus plane twice. When evaluating, use magnified live view with focus peaking as gold standard and determine how many of 10 focus attempts require "+" to adjust and how many require "-". If there is not a clear preponderance one side or the other, there is nothing to be gained by adjustment. Also, use a flat target parallel to the sensor, camera on tripod. The AF system cannot consistently evaluate a slanted, curved, or moving surface.