Choosing the Right f-stop for Night Shots with Long Exposure

Tip711-Img1 1600

Learn how to choose the right f-stop for long exposures during night shots, and get the desired expression for your images. E.g. a smooth silk-looking water surface.

By using long exposure times for exposing water in movement, its possible to achieve a very nice silky effect. Typically an exposure time of 15-60 sec is enough to turn a moving water surface into a flat silky looking surface.  Depending on the conditions you may need to expose up to a couple of minutes.

To achieve this long exposure time for a night shot, simply by selecting a smaller aperture like f22 and f32, may not be the best solution, as lens diffraction at these apertures have a significant negative influence on the contrast and the sharpness of the final image.

You will achieve a much sharper image, if you use a lower f-stop in the range f8  f11. To ensure the right exposure time, try using high quality Neutral Density (ND) filters instead.

Tip711-Img2A full - thisThe image above shows a zoom into the central part of the image without any image adjustments added.  The same image is shot at both f=8 and f=22. It is obvious using f=22 decrease the sharpness and the local contrast significantly. You can hardly read the Gulf logo.

Sometimes its a good idea to make an aperture sweep to learn more about the behavior of a specific lens. It is necessary to use a very sturdy tripod and manual focus, in order not to be influenced by variations in the autofocus system.

Tip711-Img3 full - thisThe 4 images above are 200% zooms into to central part of an image.  At this zoom level we can see how the lens diffraction effect gradually decreases the sharpness of the image.   At f11 we still have great sharpness, but at f16 an obvious loss of sharpness appears. At f22 we really lose a significant amount of sharpness and local contrast.

Lets take a look at the surface of the water and observe the effect of using long exposure time for the shoot.
Tip711-Img5 full - thisIn this case the 15 sec exposure gives the wanted silky look of the water surface. For the best result, I should use a neutral density filter giving a 15 sec exposure at f8 or f11. The image is shot using a wide-angle lens so there will be no problem with depth of field using f8 or even f5.6.

If I havent got my ND filters with me, and I still want the silky water look from the f22 shot, then I can use the Clarity tool in Capture One 7 to improve on the lost local and also improve quite a bit on the loss of sharpness.

Tip711-Img6 full - thisA 100% zoom into the central part of the image.  The Clarity tool from Capture One 7 has been used in the image to the right.  I use Clarity=40 and Structure=20 for this image.  These settings bring back the loss of local contrast and improve the perceived sharpness of the image.

With the Clarity tool we have been able to dramatically improve the loss of quality due to lens diffraction at f22. A better solution to achieve the 15sec exposure time would have been to use a high quality ND filter and an aperture between f8 and f11.

All the best,

Niels

The Image Quality Professor

Niels V. Knudsen

Niels V. Knudsen is Phase One’s Image Quality Professor. He is responsible for breakthrough advancements in image quality both in Phase One’s medium format camera systems and in its digital imaging software.

15 thoughts on “Choosing the Right f-stop for Night Shots with Long Exposure

  1. Peter Moloney

    This is one of the things you learn in college, and take for granted! In the film days it was almost impossible to see this, but with digital it is obvious and can be viewed instantly in Capture One. I would urge every photographer to test this for themselves – you will be as amazed as I was when I photographed a ruler with lens tilt at various apertures – do test it!
    Peter

    Reply
  2. Ian

    I have a basic Canon EOS50D and Kit Zoom lens. If my lens gave crappy results at F/22 I would be looking to change my lens. I photograph many items where I need the extra depth of field and do not want to deal with focus stacking. I made many comparison images at a four stop range and the difference is barely perceptable. The lens you are using IMHO is not a very good one.!

    Reply
    1. Tan68

      Very late response but I am looking through old posts and thought I would add this…

      I hear that lenses affect when diffraction is visible. I believe it is also a thing that happens anyway and so isn’t just that some lenses show it yet some lenses don’t…

      I will make up a unit of blur and say the lens and camera show 12 units at f/11 as diffraction starts to become visible. The exact aperture when diffraction becomes visible at a pixel level varies by pixel size.. With good technique, a really good lens might show only 4 units of blur at f/5.6. An average lens might be at its best at f/8 with 10 units of blur.

      Looking at the average lens only, I might think it is a better lens because I will see little difference between its sharpest point at f/8 (10 units) and when diffraction becomes noticeable (12 units). The lens appears to show no real decrease in quality but this is because the lens, at its best, was already providing a level of quality similar to that when diffraction first occurs… A poor lens may, at its best, be so not sharp that diffraction effect is masked until very small apertures…

      The good lens, on the other hand, will appear to suffer horribly. There will be a noticeable loss. The loss stands out because the good lens was a lot better at f/5.6… By time diffraction arrives at f/11, both the good and not so good lens will show similar levels of sharpness…

      There are things other than sharpness and the good lens may still offer some benefit even as diffraction becomes visible.

      I believe this to be generally accurate.

      Reply
  3. Robert Ezergailis

    Rather than neutral density filters, change the ISO setting on the camera to a lower ISO to get the f stop and exposure duration that is optimal. Neutral density filters always reduce the optical quality of a lens.

    Reply
  4. Werner Bonitz

    Unfortunately the sharpness in the buildings is counterproductive to the “silky” appearance of the water. If the best appearance for the water is (in your example) f/22 at 15 sec, then it will probably need an even longer exposure than 15 sec at f/8 – f/11in order to counteract the increased sharpness in the water. This doesn’t apply any more after the exposure time is long enough to thoroughly merge the water movement. With my particular lenses I am happy with my results up to f/16 but avoid going for a smaller aperture.

    As to the comment about lowering the ISO setting on the camera: If you are trying to get maximum sharpness then I assume you are already shooting at the lowest ISO possible. So ND filters are the only alternative.

    Reply
    1. Werner Bonitz

      How did you know that I use an extremely good lens? Ah, I guess you actually meant the opposite of what you said.

      “Pure”, according to the dictionary is (among other things): a (1) : unmixed with any other matter (2) : free from dust, dirt, or taint (3) : spotless, stainless
      b : free from harshness or roughness and being in tune —used of a musical tone.
      In other words: “the best”. Your command of the language seems to be on a par with your “critical eye”.

      I have no quarrel with your being happy with what you get from your camera but there are some guys (and gals) out there, myself included, that aim for something better. The loss of quality at both ends of the aperture range (large and small) is well documented and is the result of “diffraction” (law of physics). It applies to all lenses, although it is of course less noticeable with better lenses.

      Last, not least, assuming that the Image Quality Professor and everybody else are using inferior lenses (your original and your last post) is quite presumptuous; especially considering the lens that you use (a “Kit Zoom lens”, is it?).

      Reply
  5. Werner Bonitz

    Oops, made a mistake.

    The first post with reference to an inferior lens was by Ian, not by Dragan Mikki. So, my apology to Dragan Mikki for my last paragraph: the “Image Quality Professor” and the reference to the “Kit zoom lens” should be excluded from it (the rest is still applicable). I guess my last paragraph can still stand for Ian though (“with everybody else” excluded) along with my third paragraph.

    I better stop before I stumble over myself.

    Reply
    1. Dragan Mikki

      Thank you for pointing out at my obvious mistake. Worth noting that you have made a career of stating obvious! Isn’t that why you call yourself a Professor?
      Instead of vesting your time why don’t you try and help integrate Wacom Pen and Tablet with Windows 8 OS? Instead of skirting this issue it is about time you do something about it. Windows 8 OS have been around for a year!

      Reply
      1. Werner Bonitz

        You think that I am the Image Quality Professor? A great honor, but I can’t claim that title. You are right though that I am vesting my time in a hopeless discussion. I hope that you will find your peace.

        Reply
  6. Dragan Mikki

    Deep stuff Professor! Please give us more.
    Interestingly, you appear to be undermining the legitimacy of Phase One as well! Who are you? Please don’t tell us you are Pippa Middleton disguising as an Image Quality Professor.
    And you won’t waste my time if you tell me that one day Capture One will work with Wacom in Windows 8.

    Reply
  7. Tan68

    One thing to consider for night pictures are stars formed by light sources.

    In the samples, the stars are more visible at smaller apertures. A decision can be made to use an aperture for maximum sharpness or use a smaller aperture for more star effect.

    How stars look vary by lens. Some lenses don’t need to be stopped down as much for stars to form. Some lenses have well defined stars while others have less sharp stars and some have stars with more points/arms. Number of arms varies with aperture blades.

    I think I remember odd number of blades are good but that would have to be confirmed if anyone is interested in lens star characteristics…

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Heads up! You are attempting to upload an invalid image. If saved, this image will not display with your comment.