What is Focus Shift?

Focus Shift is an optical problem that occurs due to Spherical Aberration, when an object is brought into focus at maximum aperture and captured with the lens stopped down. Focus shift can lead to blurry images and focus errors, when working with subjects at close distances and using fast aperture lenses. With the lens aperture fully open or “wide open”, incoming rays of light converge at different focal points due to spherical aberration along the optical axis, as shown in the top illustration below:

Focus Shift

When the lens is stopped down (the size of the aperture is decreased), light rays no longer reach the edge or the “periphery” of the lens and only the ones close to the optical axis make it through. As a result, the point of best focus with the circle of least confusion is moved to the right, as shown in the second illustration above. If focus is not re-adjusted after this change of aperture, it will shift the sharpest focus plane (hence the name “focus shift”) away from the lens, essentially moving it slightly behind the focused area. Imagine focusing on an eye, only to find out later that you ended up with a nose in focus instead, just because you changed camera aperture.

Focus Shift and Fast Aperture Lenses

Most lenses prone to focus shift problems have very fast maximum apertures of f/1.0, f/1.2 and f/1.4 for one major reason – a big portion of the lens surface is used to transmit the light. Fast prime lenses with uncorrected spherical aberration will always have focus shift problems for this reason. In fact, any fast lens with spherical aberration problems will exhibit focus shift at different apertures. Slower lenses with maximum aperture of f/1.8 and smaller that also suffer from spherical aberration problems will have a much less pronounced focus shift, so it will not be as noticeable in images.

Phase Detect AF Problem

What many photographers do not know, is that their AF lenses are always set to the maximum aperture when phase detection autofocus is used. The reason for this, is that phase detection uses two small sensors to measure AF accuracy (depending on the angle of light) and they require a lot of light to work effectively, which is why lens aperture stays wide open even when the lens is stopped down on the camera. The aperture changes only when you finally press the shutter button and opens back up at the end of the exposure. At smaller apertures, the light comes at a much smaller angle, making it harder for the AF sensors to see if the subject is in focus or not. On top of that, there is much less incoming light for the sensors to see much contrast, which also causes autofocus accuracy issues. These are the main reasons why camera manufacturers claim that autofocus does not work on lenses slower than f/5.6-f/8 (on the latest Nikon DSLRs). Interestingly, the most optimal lenses for the phase detect AF system are the ones that have a maximum aperture between f/2.0 and f/4.0. Large-aperture prime lenses with maximum aperture larger than f/2.0 have a very small depth of field, so they just take longer to obtain correct focus.

Why is this all important and what does phase detect AF have to do with the focus shift? As shown in the illustration above, focus shifts to the right (or away from the camera) when aperture is changed. Since phase detect AF leaves the lens aperture wide open while autofocusing, the sharpest focus plane is going to shift when you take a picture with the lens stopped down. This is potentially a big problem, because it means that you might not be able to achieve correct focus, unless you stop down enough to compensate the shift (by increasing depth of field). Sadly, even the AF Fine Tune function in some advanced cameras like Nikon D7000 is not going to help, because it does not allow micro-adjusting focus for different apertures. If you adjust focus for a lens at f/2.8, focus will certainly shift at f/1.4 and vice-versa.

Focus Shift: Phase Detect vs Contrast Detect

Contrast Detect (Live View) AF System, on the other hand, does not have these problems on advanced DSLRs that allow changing aperture in Live View mode. Since Live View shows everything through the lens and adjusts the lens focus based on image contrast, changing aperture and then refocusing will provide correct focus – that’s because aperture will actually change on the lens in contrast detect mode. Some entry-level DSLRs do not provide the ability to change aperture in Live View mode, which would again result in focus shift when changing lens aperture.

Reducing Focus Shift

As you can see from this article, focus shift can be a very problematic and annoying issue on fast aperture prime lenses. While there are some tricks you can employ to reduce or get rid of focus shift, most of them are unpractical. Unless camera manufacturers provide the ability to fine tune autofocus on each aperture, there is no easy way to deal with this problem. Unless, of course, manufacturers design lenses with aspherical elements to reduce spherical aberration and focus shift problems; or incorporate focus shift data into camera firmware, which will automatically adjust focus based on a lens and its aperture. Anyway, here is the list of tricks or workarounds to reduce focus shift:

  1. Use maximum aperture – take pictures at the maximum aperture and you won’t have to worry about focus shift. Might not be practical for most lenses, because they are soft wide open. See the next bullet point for an alternative solution.
  2. AF Fine Tune optimal aperture – if your camera has the ability to fine tune autofocus, set your lens to its optimal aperture that you will be primarily using, then fine tune autofocus. You will then have to shoot at this optimized aperture all the time and stop down when needed. Using larger apertures will result in focus errors after this type of calibration.
  3. Use a slower lens – if you want to avoid focus shift problems, use slower f/1.8-f/2.8 lenses that have much less issues with focus shift.
  4. Stop down the lens – usually stopping down the lens to apertures smaller than f/2.8 will take care of the focus shift problem due to increased depth of field. Not very practical for fast aperture lenses, but will certainly take care of the problem.
  5. Use Contrast Detect AF – not practical for most situations, because Contrast Detect AF is slow and requires the mirror to be raised up, blocking the viewfinder.
  6. Use Manual Focus Lenses with Aperture Rings – a manual focus lens with an aperture ring will allow you to control the aperture from the lens, so you can stop it down before manually acquiring focus. You will have to reacquire focus every time you change aperture though.

Comments

  1. 1
    ) Marian
    November 8, 2011 at 1:16 am

    Very well written! Thanks!
    This is a problem that spoils a lot of images but many people don’t know about it and blame the lenses for this.
    I have 50/1.2 Ai-s that has a sever focus shift at close focusing distances and pictures should be taken very carefully using Live View.
    The viewfinder of the D7000 reacts on brightness changes up to about f/2.5. I suppose that the electronic confirmation dot has a similar “max. aperture” and cannot guess the right focus. There is suppose that the focus shift also messes up but I am not sure.

    Nasim, do you have an idea about the electronic confirmation dot light sensitivity? Is it really that limited?

    • November 13, 2011 at 11:16 pm

      Marian, as far as I know, the electronic confirmation dot you are referring to uses phase detect to determine focus in normal mode. In live view mode, you should only see a red/green square. Since you are using a manual focus lens, you should rely on sharpness you see on the LCD only.

  2. 2
    ) Peter
    November 8, 2011 at 7:34 am

    Excellent. Great illustrations, too. Will read a few times.

    One thing, I’ve always been told there is a “sweet spot” in a lens, usually midpoint on the aperture scale. Also, if you stop down too much, like f/16 or f/22, you will lose sharpness. Now focus shift. There seems to be a mess of overlapping lens issues which can get confusing. Con you enlighten me?

    Also, all this seems to be a good reason to” buy the best lens you can afford.”

    • November 13, 2011 at 11:07 pm

      Peter, yes, every lens has its own sweet spot. For some it is f/4-f/5.6 and others are sharpest at f/8-f/11. And yes, stopping down to very small apertures like f/16 will decrease image quality due to diffraction (on any lens) – I will write an article on diffraction soon. And buying the most expensive lenses won’t help, since even the most expensive lenses have issues with chromatic aberration, distortion, etc. You just need to know which lens has which problems and try to work around those :) More details to come!

  3. 3
    ) Rod
    November 8, 2011 at 8:34 am

    Hi Nasim,

    Thanks for an very informative article, I didn’t know this problem exists!
    I would explain some of my ‘not so sharp’ images and wondering what went wrong!!

    Regards,
    Rod

  4. 4
    ) Muhammad Iqbal
    November 9, 2011 at 7:57 am

    Ah… thanks for the explanation. I think this explains why my AF-S DX 35mm f/1.8 Nikkor suffers from focus shift and severe chromatic aberration in its widest aperture.
    I wonder if any AF-S DX 35mm owner has the same experience.

    Regards,
    iqbal

    • November 13, 2011 at 10:42 pm

      Iqbal, yes, the 35mm f/1.8G DX is known to have issues with heavy CA, especially in the corners.

  5. 5
    ) Ondrej Marsalek
    November 9, 2011 at 10:23 am

    Thank you, well written. It seems that being able to autofocus while DOF preview is active would help with this. Let’s say you have a fast lens, stop it down to around 2.8, use DOF preview, focus while it is still active, then expose. Unfortunately, at least my D7000 does not autofocus while the DOF preview button is pressed.

    • November 13, 2011 at 10:38 pm

      Ondrej, it would be difficult to focus accurately while looking into the viewfinder…but I will give it a shot with the DOF preview button – that seems to be a good idea.

  6. 6
    ) Paweł Kowalczuk
    November 9, 2011 at 4:45 pm

    Thank You. It was very interesting. :) I mostly use 3 buttons on my d700. First one I press is the function button closing the aperture to desired stop. Next one is AF-on to start af in continous mode and try to keep focus on the corner of the af field if I’m composing this way. Then the shuter relese button to measure light and release the shuter. Have You tested focus shift in this config? I believe aperture is set this way to it’s position before AF start to mess things up the way You described it. I believe this function was designed for testing bookeh or subject isolation before taking a picture, but still there…. it seems You have Your problem solved?

    Or am I mistaken?

    I do so because at some point I noticed that having focus set in the center of the frame and after blocking AF and recomposing to avoid focus change and taking the picture…. still ends up in a bit out of focus subjects in the corners. ;) Then I developed this way and I can see AF working while I’m recomposing so there must be a difference between focus in the corners measured by AF and focus taken in center and then recomposed with AF blocked.

    Many guys recomend this way of making AF more user friendly but it’s just not enough.
    Ohhh I use mostly 50mm 1.4g version (a bit soft but not a bad lens), 24-70 2.8, 70-200 2.8 on d700.

    What do You think?

    • November 13, 2011 at 10:32 pm

      Pawel, interesting, I have never tried to focus in a depth of field preview mode (I am assuming that’s the button you are talking about). I should try it out and see if it works.

  7. 12
    ) Raghav
    November 24, 2011 at 10:11 pm

    Yet another well written article.. I always had trouble
    Understanding focus shift, until i read this..

    Though i don’t reply much, i ve been following your
    Site for over a year.. It’s so helpful and thanks for providing
    The entire artice on feeds.. Am a surgeon & i read
    Most articles in my android feed reader, during my breaks :)

  8. 13
    ) Steve
    October 17, 2012 at 9:29 am

    Great article Nasim, a lot of valuable info there for any DSLR shooter. I’m glad there’s a post regarding this on such a well known Nikon/DSLR/Photography site.

    Cheers,
    Steve

  9. February 7, 2013 at 5:12 pm

    Hi Nasim,
    This article has really helped me understand the problem regarding focus shift, as I own the Canon 50mm 1.4 and it isn’t great at Af, wether focus shift is the problem or just poor Af. I would love to know if there are any fast canon lenses out there that have had their software inside tweaked to combat the problem.
    I hear sigma are making a devise for the public to buy that enables us to fine tune the lens to the body,will this tool be able to program in anti focus shift issues on the lens?
    Regards,
    Andy.

  10. 15
    ) nicolae
    April 29, 2013 at 5:44 pm

    thanks for article Just found my 50mm 1.2 ais exibits focus shift :(

  11. 16
    ) Eddie
    May 5, 2013 at 10:54 pm

    Nasim,

    I bought a Nikon 24 1.4G recently. I found the focus at 1.4 is
    inaccurate and exhibit shift which required fine tuning in my d600
    body (~ -12).

    I bring this to the Nikon cs to fix it, after checking, both of my d
    600 and 24 1.4G have shift in focus.

    The CS suggested to tune my 24 1.4G at 1.4 (if i am using 1.4
    frequenetly), but he says this tune would affect across all aperture
    as step down (ie may have shift when step down from 1.4). The CS says
    it is a “normal” optical phenomenon for a fast lens to exhibit “focus
    shift” when step down from big aperture like 1.4. I asked them whether
    it is a defect of my lens, he insist not a defect but rather a
    “normal” optical phenomenon.

    As I read through your article on focus shift, focus shift is due to
    spherical aberration. So, I want to ask:

    1)    will lens like 24 1.4G (with aspherical element) will also suffer
    from “irreversible” or “intrinsic” foucs shift that cannot be amended?
    (for which they called it normal optical phenomenon)

    2)    So, for me, is a tuning of the lens at 1.4 for my 24 1.4G is a good
    solution to me? Or should I insisted on changing the lens if this is a
    real defect?

    Thanks for your expert advice

    Lam

  12. July 29, 2013 at 11:26 am

    In your opinion, does the Sony translucent mirror and hybrid phase/contrast detect focus alleviate these problems? BTW, great articles on here… very informative!

  13. 18
    ) Koen
    August 30, 2013 at 10:11 pm

    Nasim,
    That certain cameras cannot focus with f8 maximum aperture lenses has nothing to do with the amount of light falling on the focus sensor but has all to do with the separation distance of the AF pairs.
    At f2.8 the camera uses light from the extreme edge of the lens which give a large separation and therefore a better focus accuracy. At f8 maximum aperture the separation is a lot smaller and it will be hard for the camera to accurately determine focus. Most cameras don’t even have AF pixels where the light hits the AF sensor at f8. The light intensity doesn’t matter much. AF at f8 doesn’t even work in bright light.
    This is a common misunderstanding.

    • September 3, 2013 at 1:22 pm

      Koen, I think that you are partially right – the separation of AF pairs on phase detect sensors is what is used for calculating AF (as I pointed out in this article). However, you cannot neglect the intensity of light as well. If light was not important , then you should be able to acquire AF accurately in low light situations at say f/2.8. There should be plenty of separation right? But that’s not the case – AF accuracy goes down the drain as the light conditions worsen. Also, some lenses can autofocus very well past f/5.6 (the new 800mm f/5.6 is a good example of that), which proves that phase detect AF is much more complex than just simple separation of points between two sensors. I think you are wrong in stating that most cameras don’t have AF pixels where the light hits the AF sensor at f/8 – by the time the light reaches the mirror from the diaphragm, it is spread evenly across the mirror/secondary mirror, no matter what aperture is used. Where you are right, is that at smaller apertures, the degree of separation on the two AF sensors is smaller.

    • September 3, 2013 at 1:38 pm

      Koen, I have modified some of the language in the above article – please give it a read and let me know what you think :)

      Thank you for your feedback, I really appreciate it!

  14. 21
    ) Koen
    September 7, 2013 at 4:00 am

    Hi Nasim,

    Light does play a role in AF but it is not as important as the aperture. Most modern cameras have a focus sensitivity of less than EV 0. The 6D even does EV-3. Still it can’t focus a f5.6 lens with extender resulting in f8 maximum aperture in daylight. The reason is that the light passes a slit and a prism or lens before hitting the focus sensor. All the other light is blocked. At f8 maximum aperture the light doesn’t fall on a AF line pair and no focus can be determined. A camera like a 1D has extra line pairs very close to the centre that can work at f8 but focus is not great because of the small separation.
    Your corrected text is much better but don’t forget that the main reason a lens is wide open when you’re not taking a photo is to allow you to see through the viewfinder. Imagine having to use the viewfinder at f16. You won’t see much.

    The two pictures at the top nicely shows why large aperture lenses can be soft wide open. It is very difficult to make large optics that produce a sharp focus point AND do that for the whole visible spectrum.

  15. 22
    ) Suresh
    September 10, 2013 at 11:28 am

    Sir,
    I have a Nikon 28mm f2.8 lens ,if the distance of the subject is more than 2 meeter, usually it shifts focus (in AF mode) to INFINITY ! ,now myself I find out a simple solution (other than Manual Focusing !), AF-On mode switch activation with this lens .its works well

    The lens is in warranty period ,is it possible to fix the problem ?

    Suresh

  16. 23
    ) Cesar
    September 19, 2013 at 6:40 am

    Great article, that clears up a lot of things for me. Thanks Nasim! Thankfully I mostly use lenses like this in largest aperture (that’s what they’re designed for, right? :))

  17. 24
    ) Mark
    December 28, 2013 at 7:20 pm

    Wish I would have known about spherical aberration and focus shift earlier. Would have saved many hours trying to figure out why my Nikkor 80-200 f/2.8 has focus shift at close range (3 meters) and apertures other than wide open. Nikon service thinks this is normal behaviour. It would be great if camera manufacturers would have an option to focus at f/4 instead of wide open for these problem lenses.

  18. 25
    ) Victor
    February 2, 2014 at 10:59 am

    Remember the old Rolleiflex?… zwei augen… that would be a solution for focus shift… one lens for focus and one for taken the picture…
    BTW I’m 67…

  19. 26
    ) Victor
    February 2, 2014 at 11:39 am

    Sorry for the confusion, thinking a little more, I realize that “zwei augen” would not solve the focus shift problem… probably a software connecting the distance information with the aperture, and changing the focus in accordance with the aperture would fix it.
    But I still love the old Rolleiflex…

  20. 27
    ) Ryan
    April 15, 2014 at 1:55 pm

    I think you have phase detect and contrast detect backwards. Live view uses phase detection.

    • April 15, 2014 at 3:16 pm

      Nope, please check your sources – live view is 100% contrast detect on DSLR.

  21. 29
    ) Lee
    April 23, 2014 at 12:31 am

    Great article. However if one were to use only the center AF point for all fast lenses (f2.0 or faster), which is *always* the sharpest without the usual anomalies like CA, coma, will focus shift occur, since light that runs exactly on the optical axis is unaffected by aperture size?

Comment Policy: Although our team at Photography Life encourages all readers to actively participate in discussions, we reserve the right to delete / modify any content that does not comply with our Code of Conduct, or do not meet the high editorial standards of the published material.

Leave a Comment