Illumination fall-off effect tends to be most a problem for what type of shots? Indoor where shots involve lots of objects close together, or objects at a distance, or outdoor shots on a sunny or cloudy day?


Fall-off depends on the distance from the source of light.

For relatively small light sources, the amount of illumination an object receives follows the inverse-square law. That means that an object that is twice as far away from the light source as another object receives only one-quarter of the amount of light.

Outdoors, in bright sunlight, the difference in distance between any two things on Earth and the sun is negligible. The farthest two things on Earth can be from one another is only about 13 thousand kilometers (and no two things in a picture taken on the Earth can be that far apart), and the distance to the sun is about 150 million kilometers. You would hardly be able to measure, let alone see, the difference in the brightness of the light falling on objects whose difference in distance from the light source is less than one tenth of one percent. On overcast days the light source is the clouds (the sun still provides the light, of course), and they're still too far away from your subjects for fall-off to be a concern.†

Indoors, with the source of light much closer to one subject than to another, you will be able to see (and photograph) a difference. That makes lighting indoor group portraits tricky -- you either have to use very powerful lights from far away (if you have the space) or find a way to make the distances between the lights and each of the subjects more-or-less the same. (Or, with a bit of artistic storytelling, you can "feature" some subjects at the expense of others.)

Using very large light sources can help; within a certain distance of a large source, light will fall off in a linear fashion (things twice as far away get half as much light) or even more gradually.

On the other hand, sometimes fall-off is exactly what you want, and taking advantage of it by putting the light source very close to your subject will enhance the effect.

† For landscapes at dawn or dusk, you may find that you can see fall-off if clouds are the main source of illumination, but that's likely to be the reason you're taking the picture and not a technical problem to deal with.

| improve this answer | |
  • 1
    Just to add some theory for the mathematically inclined: Infinity small point source fall-off is proportional to 1/(distance)^2, infinity long, thin linear source (think of a never-ending florescent tube) fall-off is proportional to 1/(distance), and infinity large planar sources (flat and extending out forever in two dimensions, kind of like on an overcast day or a giant illuminated ceiling) in theory have no fall-off; the light level is the same at the surface as it is far away. How useful this theory is depends on how well a given real-life source matches one of the above. – Sean Sep 23 '11 at 21:14
  • @Sean: Interesting comment, and it brings to light an intriguing question: How does one compute falloff for light sources that are neither point sources nor infinitely large planar sources? Large lighting apparatuses of varying size from a little larger than a light bulb to many feet in width and height are commonly utilized in photography... – jrista Sep 24 '11 at 4:22
  • 1
    @jrista: That's why God (or Gossen, or somebody) invented light meters. There are too many independent variables (when you include the environment, subject reflectance, and so on) for calculation to be practical. A reasonable rule of thumb is that light sources begin to act "pointy" when you're twice the diameter/diagonal away. The transition is difficult enough to map in a theoretically perfect black room with theoretically perfect sources, but it gets really hairy in practice: very close to a very large softbox, the "fall-off" can be actually be negative because the diffuser isn't perfect. – user2719 Sep 24 '11 at 4:48
  • @jrista - I actually don't think Stan's comment is a good answer to your question. First, the direct answer is - you do spatial (2D or 3D) integrals on your sources and add the attributes of each infinitesimally small point sources (and the thanks probably go to Newton, not God). Also, Stan mentions reflections and environment, but this is not different than the effect these have on point sources. To sum up - you asked how to compute and not how to measure, hence my clarification. – ysap Sep 27 '11 at 14:30
  • @ysap: (Newton vs. God comments aside...) I think Stan got the baser meaning of my question, and given that you've noted integration is necessary to arrive at a valid accurate result, I think Stan's answer is exactly what is necessary in the real world: use a light meter. ;) – jrista Sep 27 '11 at 16:28

Illumination fall off will be a problem where you have point light sources which tend to be indoor or outdoor at night. The further you get away from the light source the less light will fall on the object and the less it can reflect. This reduction in intensity happens as a square of the distance so its not linear. To experiment with this try shining a torch at an angle against a wall and study how the light changes with distance. The way the light from the beam fades further away from the torch is illumination fall off. Outdoors during the day a scene will be flooded with light so you won't see a similar effect as the sun will give light of equal intensity in all parts of the scene that are exposed to it even though it is a point source its a long way away and very bright so you wont see changes in intensity across what is in comparison a very small field of view. On cloudy days you won't see fall off either as the clouds act like a giant soft box in front of the sun diffusing and softening its light. This is why on cloudy days you don't get shadows either.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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