2
\$\begingroup\$

Equipment : Nikon D7000, Godox Flash on camera with soft box diffuser.

Background: I have been assigned the task by my family to make portrait albums of the identical twins born to my close relative. The pics I have taken are all indoor. flash pointing towards baby straight (power of flash (1/128).

Issues: The AF shown via the view-finder is shown on the baby and when I check on my laptop the sharpness shows a bit earlier from the baby. and the second issue is noise even when using flash.

Edited to keep the issue to AF

  1. is this issue of camera not able to tell exactly where the AF is or its getting changed during the click ?

some of the stats from the pics taken

  1. ISO 1000, f5, 1/50 sec, 28mm - here the focus was shown on baby but turned out to be before on the mat when viewed in laptop. Noise visible. flash fired.
  2. other pics ISO and speed is same, focal length keeps varying.

facts on equipment: Godox TT685n flash is brand new. DSLR is old 5 years old. the stock lens unfortunately has fungus\fungi (whatever you call it).

\$\endgroup\$
5
  • \$\begingroup\$ Please limit your question to one topic. Are you more concerned with the autofocus issue, as the title of the question suggests? Or are you more concerned with exposure/ISO/noise, as the body of the question suggests? It's perfectly acceptable to ask two different questions about the same photos, but asking about two unrelated issues in the same question is not a good fit for the way the Stack Exchange system is designed to work. Thanks. \$\endgroup\$
    – Michael C
    Commented Oct 11, 2020 at 14:42
  • \$\begingroup\$ thanks Michael C, I have edited the question to AF only. \$\endgroup\$ Commented Oct 11, 2020 at 15:02
  • \$\begingroup\$ What AF mode (AF-C, AF-S, etc.) were you using? What AF point selection method were you using? \$\endgroup\$
    – Michael C
    Commented Oct 11, 2020 at 15:06
  • \$\begingroup\$ I am using AF-A and single point selection method is being used. \$\endgroup\$ Commented Oct 11, 2020 at 15:12
  • \$\begingroup\$ One additional observation. AF-S will lock focus at half press allowing you to recompose without changing focus. AF-C will refocus if the camera or subject is moved. AF-A will automatically select S or C depending upon whether the subject is stationary or moving. I have no idea the crteria the camera uses to make the selection of single or continuous in auto mode so personally I don't use it. \$\endgroup\$ Commented Oct 13, 2020 at 16:11

3 Answers 3

3
\$\begingroup\$

Some possible reasons the focus distance in the image was closer to the camera than where the autofocus point used was superimposed on the scene at the time the image was taken:

  • You locked focus and then recomposed. If you are using AF-S focus mode once focus is confirmed the camera will hold the focus at that distance as long as the shutter button is half pressed (or any AF back button is pressed).
  • You zoomed in or out after locking focus. Most lenses do not hold focus when the focal length is changed. Those that do are called parfocal lenses and true parfocal lenses are very expensive, usually in the five figure (USD) range. They're mostly used by cinematographers. Some cheaper zoom lenses with fairly narrow maximum apertures can appear to be effectively parfocal because the extended depth of field due to the smaller aperture masks the change in focus distance as the lens is zoomed from longer focal lengths to shorter focal lengths (but doesn't work as well if the lens is zoomed from shorter to longer focal lengths).
  • The area of sensitivity for each AF "point" is usually much larger than the square you see in the viewfinder. The camera will AF on whatever it detects with the greatest contrast anywhere within the active AF area.

If those don't solve your issue, please see the more general question How do I diagnose the source of focus problem in a camera?

This answer to the above linked questions covers a wide variety of possible causes of focusing issues and provides a plethora of links to other, more specific questions here at Photography SE about each possible cause.

\$\endgroup\$
3
\$\begingroup\$

Focus (both auto and manual) can be surprisingly complex and errors are more noticeable at short distances and narrow depth of fields.

In addition to possible user error, there are a number of mechanical factors to be aware of. The one I want to concentrate on is Calibration.

Focusing is performed on a sensor that is not the same as the main image sensor that actually captures the picture. (Note that mirrorless cameras may differ). What this means, among other things, is that the focal distance used to focus may differ from that of the actual picture, resulting in a positive or negative offset from intended focus.

Your camera has an Auto Focus Fine Tuning adjustment that will allow you to add a small focus offset per lens to your camera. Your camera will remember by lens serial number as well as a possible general offset for all lenses if desired. There are many articles on do it yourself fine tuning such as Auto Focus Fine Tuning.

Alas, while useful, nothing is ever as simple as we wish. Ideally a single general calibration would correct the difference between the focus sensor and the image sensor, but the reason a per lens setting is provided is because of the mechanical errors added by the lens. Even worse is that the ideal lens offset value varies with lens settings and you can only enter a single value.

As an aside, Canon has supported 2 fine tuning values per lens for a couple of years now and Nikon just provided that capability in the new D750.

To give you an idea, here is a calibration run of a Nikon 18-55 kit lens at both 18mm and 55mm on a D7200.

18-55 at 18mm 18-55 at 18

18-55 at 55mm 18-55 at 55mm

As for the noise you see, ISO 1000 is probably a lot higher than you need with flash.

My general advice would be:

  • Focus on the eyes
  • Get the flash off of the camera
\$\endgroup\$
1
  • \$\begingroup\$ I agree that the camera is front focusing... very common at short focal distances. This is easily confirmed/rectified by switching to live view photography (contrast detection autofocus). The AFMA setting is also usually different for different subject distances as well as zoom position and aperture setting.. \$\endgroup\$ Commented Oct 11, 2020 at 19:28
0
\$\begingroup\$

I would suggest you consider the fact that focus "points" are not really points, but areas. They are often significantly larger than the viewfinder might suggest. Anything in that focus area could be what is focused on, not just what you (as a human) expect.

On some system it also helps if you use the contrast detect focus system in certain circumstances. Phase detect focus systems are TTL and will operate with the aperture wide open (regardless of what shot-time setting you use) with the aperture closed down automatically at shot time. Contrast detect systems usually operate at the aperture setting to be used at shot time, which means there is no danger of focus shift - a possible issue with some lenses, although I'm dubious this is your problem.

Flash technique leaves something to be desired and is not helping.

... are all indoor ... flash pointing towards baby straight (power of flash (1/128).

Never do this indoors unless you've a huge ceiling height.

Instead bounce the light off the ceiling or "white card". Point the towards the ceiling - i.e. not directly at the subject. This gives two huge benefits. Firstly it lets the light bounce off the ceiling in a diffuse way and spread out, getting rid of all that nasty direct flash effect. Secondly you can use more power (not 1/128th flash !) and hence a lower ISO.

This technique gives far more natural looking lighting.

Sometimes either the ceiling is too high or too far from white to be useful. For this reason I normally use a reflective gadget that can attach directly to my flash and act as a large diffuse reflective surface. The one I use is a Rogue FlashBender 2, but there are many variations on this. It looks awkward, but it works really well.

This is a more sophisticated version of the old "bounce card" trick used since flashes came about. You can literally rig something with a white card and some elastic bands - and I've done this myself. This basically directs the flash forward instead of half going backward when the flash is pointed upward towards the ceiling. The advantage of the flashbender is that the surface is much more reflective and can be bent (it has some stiffness to maintain shape) allowing better of control.

Careful choice of bounce angle and using a bounce card helps a lot.

ISO 1000, f5, 1/50 sec, 28mm - here the focus was shown on baby but turned out to be before on the mat when viewed in laptop. Noise visible. flash fired.

Probably worth reading up on the technique of dragging the shutter in this context and bounce cards.

The D7000 apparently has some issues with using high ISO when not needed with flash. This blog post from Francois Malan discusses it and offers some advice. The blog gives two potential workarounds :

  1. Set the built-in flash (setting e3) to “Commander” mode, then in the accompanying sub-menu to TTL mode. This prevents auto-iso from raising the ISO, but you have to live with the longer preflash sequence (since the camera thinks that it needs to control an external flash)

OR

  1. Disable auto-ISO whenever you want to use the flash. This can be made more convenient by putting auto-ISO settings at the top of “my menu” and assigning the customizable fn-button to the top of “my menu”. But this remains a cumbersome solution and forces you to remember changing the iso setting dependent on flash usage – exactly the kind of task loading auto-iso is supposed to solve.

If all else fails remember that people shot completely manually for decades and it is not as difficult as it sounds, although it scares people when they start.

If you use these techniques you should be able to avoid the high ISO, the direct lighting effects and it might even help focus, as some systems use a "pre-flash" low intensity burst to help lock focus.

\$\endgroup\$
2
  • \$\begingroup\$ If you're going to insist on concentrating on the other question the OP edited out of this one about autofocus, you should probably mention that bouncing off the ceiling requires a white or other neutrally gray colored ceiling unless one can live with the color cast that a non-neutral ceiling will add. \$\endgroup\$
    – Michael C
    Commented Oct 12, 2020 at 17:51
  • 1
    \$\begingroup\$ @MichaelC I think I did mention about color casting by off-white ceilings. I haven't read anything that convinces me the OP has a focus issue that isn't down to lighting and the true size of focus areas - who hasn't this caught out over the years ? \$\endgroup\$ Commented Oct 12, 2020 at 18:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.