Help with lens focus on D7000

[livio]

For a while now I have not been happy with focus autofocus on my D7000 camera so I ran an experiment
I printed out two test patterns, laminated them so they would remain flat and then taped them to a glass
frame on the wall which was sitting flat. I measured 4 meteres from the wall sat the tripod on a solid table
inside and set up the remote trigger. I was watching the camera focal point and it picked the same focus point
iso was set to 100 and set to aperture priority.

at 4 metres at f/1.4 0.8 sec

test-1 by starmedal, on Flickr

at 4m f/2.0

test-2 by starmedal, on Flickr

at 4m f/4.0

test-3 by starmedal, on Flickr

at 4m f/8.0

test-4 by starmedal, on Flickr

at 4m f/11

test-5 by starmedal, on Flickr

at 4m f/16

test-6 by starmedal, on Flickr

Why is the focus so out of whack at f/1.4 and f/2.0 it is not until f/8 that I believe the focus is sharp enough do I have a dud camera or is there
some thing wrong in my technique.

[farmer_rob]

Did you try using live view to get a sharp manual focus at f1.4? This will help prove if the lens is the problem or the autofocus.

It is likely that the autofocus is either forward or back focussing. You can't tell which with a flat target. I'd say that as it is stopped down to f8, the depth of field increases so that your target falls into range.

The D7000 offers a simple solution, so i don't think you have a faulty camera. The D7000 has a microfocus adjust function, so any forward/back focus can be adjusted for easily.

For my d7000, I tested focus wide open on a fixed subject, then used live view to see if I could get a better focus (yes, with one lens). I then used the microfocus adjust in the indicated direction.

Regards,
Rob

[J.davis]

Have you tried this with a different lens to isolate the culprit.

[WhoDo]

The D7000 has a microfocus adjust function, so any forward/back focus can be adjusted for easily.

For my d7000, I tested focus wide open on a fixed subject, then used live view to see if I could get a better focus (yes, with one lens). I then used the microfocus adjust in the indicated direction.

I agree with Rob, that you can't check this properly with a flat subject. Put a set of 3 objects (batteries are good) evenly space and at a small diagonal to the direct line of the camera. Using your middle object as your focal point, autofocus at the same aperture (f/2.8 is shallow enough) and adjust your camera micro focus in between shots one increment at a time until the focal point is sharp in the resulting image. In my case it required a +3 adjustment. The result is that a lens I thought was soft is now tack sharp (well it was all along, I just hadn't adjusted the camera autofocus fine tune properly).

[I @ M]

set up the [#1] remote trigger. I was watching the camera focal point [#2] and it picked the same focus point


Why is the focus so out of whack at f/1.4 and f/2.0 it is not until f/8 that I believe the focus is sharp enough do I have a dud camera or is there [#3]
some thing wrong in my technique.

Did you try using live view to get a [#4] sharp manual focus at f1.4? This will help prove if the lens is the problem or the autofocus.

At this point in the exercise I reckon we should go right back to basics and eliminate a few things. Those things are highlighted in the quotes.

#1 Which remote shutter release are you using? Is it set to actually engage autofocus at half depression of the button? Can your camera be set to and is your camera set to take a picture without focus being achieved?

#2 That very much sounds like you are letting the camera decide where to focus. That is a HUGE mistake and you really need to set your focus to single point AF, especially for testing and also for many situations where you are photographing. The dynamic AF options are handy in some situations but definitely not in this situation.

#3 At this stage, yes, your technique is far from optimal.

[#4] Perfect response and this is exactly where you should restart the experiment. Even with your present chart as it is start afresh by selecting the single centre AF point that you control, ditch the remote control and if the lens is an AFS lens simply put your eye to the viewfinder and rotate the focus ring from one end to the other and find the sharpest / clearest / in focus view of the chart that you can. When you get the best looking focus in the viewfinder simply compare that view with the image that you have posted and if it is not much better than what you have already stop right there because it would appear that you have a hugely defective lens.
Which lens are you using?
If however using manual focus results in a nice clear ( or at least as clear a view as your F/8 sample ) then and only then can you proceed to setting up angled targets and adjusting the AF via the micro focus function in the camera.

[ricktas]

To test correctly, use THIS. Whilst created for a D70, it is still valid for ALL cameras

[livio]

Hi Andrew Thanks for your response it has been valuable and I will re-do the experiment using some of the new tips I have garnered from this forum. When the original test was done I used two lenses a 16-85 zoom and the Nikkor 50mm f/1.4 lens which was the one I posted. At that stage I had no real idea about front focus and back focusing lenses. I didn't even really appreciate that the D7000 has a AF micro adjustment. The 16-85mm lens was good from about f/5.6 and the 50mm lens was good from about f/8 it was really rubbish at f/1.4 in all of the cases I tries the centre focus lock used I wanted to be sure that I had control of the AF that said I would only take the shot when the camera AF locked in to the same reference point I was using the viewfinder at the time I was taking shots. I received a good post from "ricktas" and I am going to try that this evening. What I have learned from this?
1) Even though you pay good money for brand lenses they will behave differently on different camera bodies
2) There is such a thing as forward and backward focus and this build right down to knowing your equipment
3) The test I was initially using is flawed in that it is a flat surface it becomes difficult to gauge correctly how much correction I need to provide the lens
4) I was right to post this question to the forum. I have learned from the experience and knowledge of others to fast track a solution which would have taken me so much longer to find by myself.

Thank you to all who have provided some tips and their knowledge. I really appreciate it

Kind Regards
Livio

[livio]

Thanks Rob, I think you are are correct here I think my test was flawed but that is what learning is all about. I will redo my tests and see what I come up with. Once I know the settings I need I can set up the camera accordingly The live view was definitely sharper photo so I will now start with focus adjustment. Appreciate your post it has helped me heaps

Kind Regards
Livio

[livio]

Rick,

Thanks heaps I have downloaded this document and this looks like a much better way of testing than I was using. I will try this again this evening and see what the results are. I believe that this will amount to being a forward / backward focus issue and it all boils down to knowing your equipment. Thanks to you and the AP forum I have learned some valuable lessons
1) The AP Forums have quite a number of very knowledgeable people who are happy to share.
2) I'm fairly certain this will be a back / front focus issue and there is adjustment on D7000 to allow me to get this right
3) I was right to post this question to the forum. I have learned from the experience and knowledge of others to help me build up my own knowledge in a far faster way than I would have been capable of by myself.

Kind Regards
Livio

[kiwi]

Definately eliminate the auto-focus point selection

[wheellathe]

I had focusing issues with my D7000 after almost a year of convincing my self that it was me I gave up and took it to Nikon. It turns out that I did have a back focus problem.If it's less than a year old you can get it checked under warranty

[reaction]

Focus test charts don't work with cameras that can focus on blank paper . D7000 will have no trouble at all focusing on any blank paper. That's why you have lensalign and other perpendicular test things. Call nikon service and they'll say not to test with a 45deg chart.

Guarantee he'll get 'front focus' with a 45deg chart

[farmer_rob]

Hmm... I have the view, naive though it might be, that the image should be in focus at the same point as the highlighted autofocus sensor point. Whatever is under that point should be in focus, whether it is plain paper or a sharply contrasting object. That is what i set my autofocus adjust for.

Regards,
Rob

[livio]

Thanks to all who have responded to this thread, it has really helped me heaps. I ran the tests that ricktas pointed to me about shooting at 45 degrees to a flat surface. The result was with the D7000 and the f/1.4 50mm lens I had to adjust the AF Mircro adjustment to -20 for the text to be in focus. I could read it at -8 but it was clear that it was still focusing behind the subject. This explains to me why my low light photos of my sons band have been in soft focus. And here I am thinking it was me all along this has been a useful test for me and I'm sure I will pass this on to others as well.

1) Know your equipment, I had a guy tell me this once and for a computer specialist you know it either works or it does not, it is binary, unfortunately cameras don't work the same way. Optics are a variation they fall between a range so it is important to test every new lens and understand how it operates with your body.

2) Trust in the Forum of AP you may not get every answer you want but you certainly get good advice and direction. For me I appreciate the fact I can learn from others. I am currently going doing photography related courses at TAFE but I find that there are gaps and AP provides the perfect forum to fill in those blanks. As I learn more I plan to participate more in AP and provide help to others.

Thanks again to all who have contributed. I now have a solution and something which works okay not fantastic but okay. I also have one of the 70-200mm f/2.8 lenses and that one is brutally sharp I am looking forward to running the same tests to see just how much that needs adjustment down low.

Kind Regards
Livio