I’ll answer the second question first, right now: You need a light meter for video. The question is: do you want it as part of your camera or as a separate tool?
What is a light meter?
A light meter reads (measures) light, and gives you the results in a format that you understand. To know more about measuring light, read How to measure light.
Most light meters will give you measurements in terms of:
However, because it is a ‘light’ meter, it might also tell you a little bit about light in terms of Luminance or Illuminance.
Light meters take the guesswork out of judging exposure. You are free to agree or deviate from the results (called ‘readings’) but at least you have a valid starting point.
Let me rephrase that last sentence for emphasis: One of the biggest blunders in image-making is when the image-maker assumes the light meter reading is to be followed religiously. It would be better if you thought of a light meter reading as an extremely accurate weather forecast. If rain is forecast but you feel like playing football outside, don’t let that stop you.
Types of light meters
Actually there is only one type of light meter, and that is the incident light meter. Light falls on a sensor, and the sensor measures it.
However, people are more interested in knowing how the light gets to the sensor. If it stopped on the way at the grocery store for a candy bar they want to know about it. For this reason, you have the following two classifications:
- Incident light meter – light falls directly on the sensor.
- Reflective light meter – light bounces off some place (or places) and then arrives at the meter.
Why would anyone care where the light has been? For one reason they want to make sure their light is a good boy, and is behaving as he ought to.
E.g., if you have a light bulb in the scene and you want to know how much light it is giving out, you can measure the light coming off it directly. On the other hand, if you don’t care much for the number, quality and quantity of light sources in your scene, you need only worry about reflected light. For this reason, traditionally, light meters are used in this fashion:
- Incident meters – when you want to measure the sources of light directly, and in relation to one another. There’s always a comparison happening, even if there’s only one light (the other is shadow).
- Reflective meters – when you only care about the end result (how light from various sources bounces off an object).
Can you make one of the above do the other’s job? Yes and maybe yes. Incident meters when pointed to the scene will read whatever light falls on it. Reflective meters cannot measure light directly without a standard reflective surface. And, you’d have to switch each light on and off to take independent readings…meh.
From the above scenario, you might visualize some practical ‘problems’:
- To measure incident light at a point, you have to be at that point. In a studio, you can hover around the actor or model and take readings from all sides. However, if you’re shooting a tree on a mountain far away, an incident meter is dead weight.
- To measure both incident and reflective light, you’ll need some kind of ‘averaging’ system, simply because there are many kinds of light falling on the meter sensor.
To overcome these practical problems, we have two kinds of light meters:
- Incident light meter – a separate tool that can be taken to the point of measurement and is light enough to carry around.
- Reflective light meter – usually found inside a camera because a camera manufacturer knows its imaging sensor better than anybody else. A third-party reflective light meter might be used with multiple camera sensors but will deliver different results.
The ‘averaging’ system can be divided into three groups:
- Complex algorithm for the entire scene (most used, and actually very accurate, big secret sauce stuff).
- Center-weighted (Only looks at the center portion of the image).
- Spot meter (Only looks at a small portion of the scene, somewhere between 1-5 degrees). E.g., at a distance of 10 feet, 1 degree spot is a circle with a diameter of 2 inches. At 100 feet, the circle has a diameter of almost 2 feet. For a 5 degree spot, the equivalent values are 8 inches and 9 feet.
Mind you, there is a complex algorithm for all three of the above reflective meter types – because even center-weighted and spot readings must combine light of various qualities collected over an area.
To recap, modern light meters are produced for four kinds of readings:
- Complex algorithm
When to use an incident meter or reflective meter?
Here’s a twister: Can a spot meter be used as an incident meter? The answer is No and Yes.
- No – you cannot compare light bulbs with a spot meter. You could point the spot at two points (say the left and right sides of a face) to compare light levels between the two, but they must absolutely be of the same material. E.g., if one side of the face is covered by shiny hair you cannot compare it to the side that doesn’t have hair. If you’re photographing Harvey Two-Face, then you might get shot for doing something this stupid.
- Yes – you can compare light as it affects the camera sensor. The camera sensor is made up of pixels that read light, and it doesn’t care where the light comes from.
Bottom line, the two things you can do with a light meter is: compare lights, and find the exposure of a scene as it affects your camera sensor. Therefore, you have these rules of thumb:
- Without light bulbs (strobes or continuous sources) to measure and compare, an incident meter is pointless.
- Without a reflective meter, everything is pointless.
Luckily, modern digital cameras have in-built reflective meters that are calibrated with the camera sensor and are highly reliable. The simplest way to get the best out of your camera sensor is via this meter. Don’t leave home without it.
Support tools for video
Video has help from many quarters. Four critical tools are:
It’s not enough to judge the exposure, the image must conform to a predetermined standard, usually Rec. 709. Therefore, you’ll find talk of placing skin tones at a certain IRE, ETTR, never over-exposing your shot, blah blah – all of which were rules created not for image quality reasons but for technical delivery. Though, to be absolutely fair, digital sensors back then didn’t have the latitude to go beyond many of these parameters. These rules became rules of survival. E.g., if you were a documentary filmmaker armed with a 1/5″ camera, you never worried about artistic interpretation because you always relied on one or more of the four tools above. This is why most video-originated material look alike.
Today, sensors are capable of latitudes that go beyond what you’d find in a scene. This gives you the freedom to move around, and not be rigid in your exposure. How do you deal with this freedom? There are two fundamental ways. Let’s understand them.
Do you need an incident light meter for video?
Ever since the DSLR video revolution, everyone thinks in-camera reflective meters are good enough, and the days of the incident meters are numbered. That’s baloney.
I can understand the ‘want to make a quick buck’ filmmaker completely relying on the in-camera meter to get the shot. Now that we have external monitors on set it’s easy for any idiot to turn up or turn down various light units individually until they have stumbled on to a ‘look’. What’s missing here?
Previsualization. If you come to depend on an external monitor, you’ll always need it. Here are two scenarios for a newbie cinematographer:
You carry around an incident light meter and compare light falling on different points in the scene. Using your imagination (or the Zone System) you place your greys according to your vision. Initially, you’ll make terrible mistakes, but you still have an LCD screen or external monitor that’ll prevent you from screwing up big time.
Then, you’ll call on your crew to tune the light until it’s perfect. Ten years later, you will see light like a master. Even without an external monitor, armed with just your light meter, you’ll be able to walk into any set and create the vision you have in mind. The in-camera meter will still help you stay within the confines of the sensor’s ability, but you’re no longer a slave to it.
Changing technology will not affect you.
You use your in-camera meter exclusively, and rely heavily on your external monitor. Every time you set up a scene, you must recheck on your external monitor because you don’t have a mental image of the light levels. You create your vision as you move along. You ask your crew to tune the lights up or down till you’re in the ballpark of your vision. Obviously, this process takes more time. Eventually, you too, will start seeing light (albeit in a different way) and will be able to craft a look based on your experiences.
What you can never do, though, is after ten years you’ll still not have the confidence to walk blind onto a set and light a scene to get a predetermined vision (Any fool can look at the monitor and say ‘I told you so’ – try reproducing what’s in your head). Without an external monitor, you’re dead.
Before you go off a wrong tangent, neither of the two scenarios is good or bad. In both cases you need a crutch – in the former it’s an incident light meter and in the latter it’s an external monitor or LCD. It would be unfair to yank away an external monitor but not a light meter! Choose your poison.
One way to mix things up a bit is to give each cinematographer the other’s tool. The professional who no longer has a light meter can still light with an external monitor. The professional without his or her dependable external monitor will be fired.
It all depends on what kind of cinematographer you want to be.
Which light meters should you buy?
Many older analog meters still work, but you never know how accurate they are until you’ve tested them. On the whole, I strongly advise against used light meters if you don’t have the means to test them thoroughly. If possible, you might want to avoid photography-centric light meters too (though they are perfectly usable and if you can’t afford anything else go for it).
Three manufacturers known for video or cinema light meters are Gossen, Spectra Cine and Sekonic. Here are some models:
- Sekonic L-398A (about $200)
- Sekonic L-308DC (about $270)
- Gossen Digipro F2 (about $300)
- Spectra Cine IV-A (about $340)
- Sekonic L-758Cine (about $805)
Update 2015: Though I prefer the above meters, eventually I purchased a Sekonic 478D (Amazon, B&H) because it is the only meter that will take high frame rates and high ISOs, both of which are important with the cameras I use (Sony a7S and a7S II).
Do you use a light meter? Why or why not?