light meter measurement: incident light and the inverse square law
As I understand it the inverse square law states that the intensity of light will diminish in value at a rate equal to the inverse of the square of the distance. My question is this: how can taking an incident meter reading of a subject be accurate without taking into account the distance of the camera from the subject?
Is the inverse square law only applicable to sources of illumination but not reflected light? This seems counter-intuitive but is the only way i can think to reconcile this apparent discrepancy in my head.
Great question, I am trying to get my head around this myself. I do think that the inverse square rule applies to what are effectively point sources. When light goes through a lens, it becomes basically a point source at the middle of the lens (I am sure that there is a correct word, which I do not know). When metering in a normal scene, you are measuring light that is coming from any number of very diffuse sources. I am sure that the inverse square rule applies, but that it is not measurable because there is not a single point to measure from.
As i understand it an incident reading reads the light falling on the subject rather than the light reflected by the subject. You should be as close as possible to the subject and point the incident meter toward the light source. It does work but may not be best for every situation.
Thinking about it, your light source(outside) is the sun. If you take a measurement and go 5, 50 feet or 50 miles the distance hasn't really changed.
And you do not have to be near the subject, just in the same light. So, if you're at the rim of the grand canyon & take a reading, the exposure will be the same on both sides.
Assuming no clouds or shadows or solar eclipse or errant asteroid or something else like falling into a black hole just south of Dallas.( I know it's not nearby!)
When you're using an incident reading you're not reading reflected light, that's what a spot meter or reflected light meter does. Point the meter at the subject.
Incident reads the light falling on the subject.
The inverse square law is pertinent to (generally) a point light source. Flash, street light, spotlight table lamp. etc. Whether you use either kind of meter.
Last edited by John Koehrer; 02-08-2011 at 06:37 PM. Click to view previous post history.
The inverse square law applies to true 'point sources' of light. If you look up illustrations, the light which fans out from a true point source expands outward in all directions, which is why the number of photons falling onto an object decreases by the square of the distance (inverse square).
When the distance to the source is close enough, inverse square does not apply...a softbox used within about 3x largest dimension of the softbox behaves closer to inverse linear. Similarly if you were 2.8 million kilometers from the 1.4 million kilometer diameter sun, the sun is a huge 'softbox'.
Reflected light behaves a bit differently since the object is not a true 'point source', but conceptually is a large collection of points on a surface. So its behavior is not 'inverse square' either.
Sponsored Ad. (Subscribers to APUG have the option to remove this ad.)
The further the subject is away from the camera, the smaller it's image.
A smaller image contains less light than a larger image.
(Throw on a longer lens, and the image gets bigger. But also darker.
Put a faster lens on, and the image will get brighter, but only because the lens has a larger image collecting surface, so captures a larger solid angle cone eminating from the subject.)
Reflected light does behave differently only in as far as the reflection could be (very) directional. A concave mirror, for instance, does not spread light, but does the opposite. So stand in its focus, and you'll find out why it's called focus.
But generally things don't reflect like that, and the inverse square law isn't a bad description of how also reflected light behaves.
Last edited by Q.G.; 02-08-2011 at 07:23 PM. Click to view previous post history.
No, point the incident meter at the camera lens, not the light source.
Originally Posted by jeffreyg
You need to measure the light falling on the subject from the direction of the camera.
Light also 'works' in accordance with the inverse square law when coming from extended light sources. It can't do anything else.
An extended light source is nothing other than a collection of point light sources. The distance from the object to each of those points in the light source will be different, so it gets complicated because of that: the total amount of light an object receives is the sum of what it receives from every point in that extended light source. But no matter from what point it comes, light will follow the inverse square law.
Unless, of course, the distance to the extended light source is large enough to make the differences in distance be so small that they do not matter.
Then you can ignore the complexity and do simple math again. And still that inverse square thing applies.
A good example of such an extended light source is the sun. Even from here, one astronomical unit away, it has an appreciable size. But that matters so little that it can be completely ignored.
Only when light is directional enough will the inverse square law not apply. Lasers, for instance, do spread, but not in such a way that the inverse square law would be applicable.
Last edited by Q.G.; 02-08-2011 at 07:53 PM. Click to view previous post history.
As the distance in question increases, the area decreases proportionately (I THOUGHT cube of distance relative to the inverse cube of the area - but, I don't know - possibly my memory cells are getting rusty). The amount of light given off for each unit of area remains the same. An example: If a measurement limited to a one degree circular area of a gray card from one cm indicates an EV of 10, increasing the distance to the card to 100 meters and limiting the measured area to the same area as before will also indicate 10EV.
Reflected metering will indicate an AVERAGE light output of the entire scene, limited only by the acceptance angle of the meter, which may be one degree, five degrees, ten ... or ?? in "Spot" metering.
Incident metering measures the light falling on the subject - the distance to the CAMERA has no effect. Properly, the meter should be directed, generally, at the camera; it is also useful to point the meter at individual light sources to determine their effect on the entire scene (balance).
Ed Sukach, FFP.
We are indeed talking areas, not volumes, so it is the square, not cube.
Originally Posted by Ed Sukach
The question is: why?
Originally Posted by Ed Sukach
Light has to travel from the subject to the camera to be captured. And along the way, why would it not behave in accordance to the inverse square law?