[biz-engineer]

Ignore biz-engineer's suggestion. He was trying to be funny and sometimes that doesn't convey well on the internet.

The image with the chart isn't zoomed in enough to see accurately where the plane of focus is, but we see a little. -2 adjust isn't much, maybe is enough. Often the adjustement isn't as much as people think, simply because of the reproducibility of focus conditions. I'd take not only one test shot, but 2 test shots. 1 test shot with pre-focus at closest distance, 1 test shot with pre-focus at infinity.

---------- Post added 22-04-19 at 05:59 ----------

Adjusted exposure a little. Was losing light, outside.

Looks good.Great image.

[clackers]

Focusing is always done wide open even if you set the lens to f16, Terry, so don't do that in testing for front/back adjustment, in fact it makes it harder to judge it.

[pschlute]

Last year, I was told by a fellow member to do these tests with aperture wide open, so that's what I've always done. I realize how shallow the dof is at 1.4. How far do you think I should stop down? Perhaps f/2.8?

The camera will always autofocus at widest aperture. Using a slower aperture to capture the test picture wont alter that. But it will perhaps make it more difficult to assess BF/FF as you will be increasing your DOF.

ps Leaving the affects of focus shift out of this !

[beholder3]

Hello,
I wanted to check for front or back focusing

You should be aware that you are using the focus target completely wrong.

1. The center piece is even folded somewhat backwards, while it actually needs to be folded forward so it is absolutely vertical.
   Once it is vertical you can and must remove the distracting blue thingie.
2. Then you need to shoot from a perspective which is absolutely horizontal to its center (you shot from a higher position).
3. finally these targets only make sense when the focus point you use only covers the vertical center piece from the target, so you have to be quite close to it, because the actuall cross sensor covers quite a big area.
   Once the cross sensor covers parts of the subject in different depths you will start a little lottery, since it is not clear which focus plane position you want.

[Medex]

There is a quite good image store website, you can upload any resolution image you want and then paste the link, that has preview and is visible here on Pentaxforums. But the site is russian and I don't know how to delete images from there at any point of time I want. But there is possibility to remove pictures lets say after 1 hour or 1 day, or even 30 days, if you click the box before uploading

VFL.Ru ???, ??????????? ??? ???????????, ? ??????? ??????? ???????????.

---------- Post added 04-22-19 at 02:40 PM ----------

I do focus adjustment in more practical way, that even don't needs shutter action, especially on cameras, that has focus peaking.
I use tripod and some flat contrast target on 45 degree. I enable focus peaking in Live view, center the point I want to focus at 16x magnification and then perform contrast detection auto focus by half-pressing the shutter. You can see on dysplay the highlighted zone of focus peaking, it is evenly distributed accross the target point. Then I shut down Live view, defocus the lens by hand and perform Phase detection auto focus (via viewfinder) by half-pressing the shutter. After that I return to Live view and chech the focus peaking highlighted zone. You can perform 5-10 repeated actions to check the trend of phase detection autofocusing. In case the highlighted focus peaking zone is more back compared to your target point, you need + adjustment, and vise versa.

[TerryL]

Uploading to the forum results in a reduced resolution image. The forum limits image size. Linking to a different site where full resolution is permitted is what I was trying to suggest.

I see. I completely understand, now. Thank you.
Terry

---------- Post added 04-22-19 at 11:11 AM ----------

Focusing is always done wide open even if you set the lens to f16, Terry, so don't do that in testing for front/back adjustment, in fact it makes it harder to judge it.

Ok, Clackers. I understand.
Terry

---------- Post added 04-22-19 at 11:14 AM ----------

The camera will always autofocus at widest aperture. Using a slower aperture to capture the test picture wont alter that. But it will perhaps make it more difficult to assess BF/FF as you will be increasing your DOF.

ps Leaving the affects of focus shift out of this !

Hi Peter,
That is good to know, and I completely understand, now. Thanks.
Terry

---------- Post added 04-22-19 at 11:18 AM ----------

There is a quite good image store website, you can upload any resolution image you want and then paste the link, that has preview and is visible here on Pentaxforums. But the site is russian and I don't know how to delete images from there at any point of time I want. But there is possibility to remove pictures lets say after 1 hour or 1 day, or even 30 days, if you click the box before uploading

VFL.Ru ???, ??????????? ??? ???????????, ? ??????? ??????? ???????????.

---------- Post added 04-22-19 at 02:40 PM ----------

I do focus adjustment in more practical way, that even don't needs shutter action, especially on cameras, that has focus peaking.
I use tripod and some flat contrast target on 45 degree. I enable focus peaking in Live view, center the point I want to focus at 16x magnification and then perform contrast detection auto focus by half-pressing the shutter. You can see on dysplay the highlighted zone of focus peaking, it is evenly distributed accross the target point. Then I shut down Live view, defocus the lens by hand and perform Phase detection auto focus (via viewfinder) by half-pressing the shutter. After that I return to Live view and chech the focus peaking highlighted zone. You can perform 5-10 repeated actions to check the trend of phase detection autofocusing. In case the highlighted focus peaking zone is more back compared to your target point, you need + adjustment, and vise versa.

What an excellent idea. I'll have to try that. Thank you.
Terry

---------- Post added 04-22-19 at 11:23 AM ----------

You should be aware that you are using the focus target completely wrong.

1. The center piece is even folded somewhat backwards, while it actually needs to be folded forward so it is absolutely vertical.
   Once it is vertical you can and must remove the distracting blue thingie.
2. Then you need to shoot from a perspective which is absolutely horizontal to its center (you shot from a higher position).
3. finally these targets only make sense when the focus point you use only covers the vertical center piece from the target, so you have to be quite close to it, because the actuall cross sensor covers quite a big area.
   Once the cross sensor covers parts of the subject in different depths you will start a little lottery, since it is not clear which focus plane position you want.

Boy, I'm learning a lot about how to do these tests. Can you recommend a better target device to get?
Terry

---------- Post added 04-22-19 at 11:26 AM ----------

The image with the chart isn't zoomed in enough to see accurately where the plane of focus is, but we see a little. -2 adjust isn't much, maybe is enough. Often the adjustement isn't as much as people think, simply because of the reproducibility of focus conditions. I'd take not only one test shot, but 2 test shots. 1 test shot with pre-focus at closest distance, 1 test shot with pre-focus at infinity.

---------- Post added 22-04-19 at 05:59 ----------


Looks good.Great image.

Thank you for the complement. The day didn't turn out so bad, after all.
Terry

[TerryL]

Ignore biz-engineer's suggestion. He was trying to be funny and sometimes that doesn't convey well on the internet.

As far as pixel shift goes, I typically process them using Raw Therapee. I generate a neutral 16 bit TIFF file from there and finish the image in Lightroom as I feel more comfortable with it.

Perfect. Thank you

[monochrome]

And clearly my comments about shooting at multiple apertures are confusing as well. I have done so as a way for me to assess the degree of user error in my tests, but I think I’ve bee fooling myself.

[TerryL]

And clearly my comments about shooting at multiple apertures are confusing as well. I have done so as a way for me to assess the degree of user error in my tests, but I think I’ve bee fooling myself.

I understood you, Paul. I'm the one who's been doing these tests wrong. I need to do a test on my 43 Limited, and everyone's suggestions have been very helpful, and something I can refer back to.
Terry

---------- Post added 04-22-19 at 04:20 PM ----------

And clearly my comments about shooting at multiple apertures are confusing as well. I have done so as a way for me to assess the degree of user error in my tests, but I think I’ve bee fooling myself.

Paul,
I don't see any focusing problems in my azalea picture. I was using the spot setting. And, I was shooting at the closest distance the lens would let me. Also, I need to check my 15-30, too. How in the world do I do that? I have never done an uwa lens, especially a zoom lens.
Terry

[Lev]

Don't forget to set your camera to SPOT focus mode for precise results.

[TerryL]

Don't forget to set your camera to SPOT focus mode for precise results.

I sure will. I've been using select instead.

[stihlmania]

You must use a repeatable camera mount position [like a tripod!] to get any accuracy from the results your "card"!........the images posted so far show to me no real issue with front/back focusing, but the focus test must be set up on a repeatable platform to get accurate results....if you plan to use that test target everything must be aligned to get accurate resultes { I have the same target when I need to test focus and once set up correctly it works great] So far none of my 12 lenses need any adjustment.....