How do You Interpret Incident Meter Readings When Filming Outdoors?

[Karl Lee]

A while back, I ran my first 800' of film through my new SR3. As I'm really just a hobbyist, my goal was to not only make sure that my camera, mags, and lens were all working well optically and mechanically, but also to see how I would fare with my first solo attempt at manual metering. While I was a freelance AC in college and had the opportunity to work on quite a few film projects, I never delved into the finer points of metering and having the final call on exposure, so from a metering standpoint, this was a great learning experience for me.

 

I have a Sekonic L-508 meter which has integrated spot and incident metering. While I'd love to learn proper spot metering techniques, I'm not really comfortable with spot metering at this point, so for this batch of filming I stuck to incident metering. As I was filming 50D, all of my filming was done outdoors. As such, my overall lighting conditions fell into two basic categories - direct sun with highlights and shadows, or overcast, diffused daylight. As I wanted this to be a good learning experience, I kept notes and recorded the incident meter readings for each shot, including readings in both highlights and shadows when filming in direct sunlight (when possible...it's not the easiest thing to do when filming skylines and distand subjects). I also kept notes of the exposure at which I filmed each scene, and in addition to filming most scenes at what I believed to be proper exposure, I intentionally underexposed and overexposed some of the scenes to experiment a bit with the film's latitude.

 

In the end, I was quite pleased with the results from my test rolls and the film transfer...kudos to Rob and the crew at Cinelab for such a nice job! As for the quality of my metering, I was pretty happy considering it was my first major attempt at manual metering for cinematography, but I did notice that some of my shots, particularly those in direct sunlight with highlights and shadows, and a few shots I tried during magic hour, looked a little underexposed in the darker areas. In these particular shots, there really wasn't much technique to my metering...I'd take an incident reading in the direct sunlight, then film at or perhaps 1/3 stop above the incident reading. In retrospect, I think had I opened up a complete stop or stop and a half above the incident reading in these situations, the resulting image would have looked better.

 

If nothing else, I'm learning first-hand something I've heard repeated over and over...meter readings, even when taken properly, are a good point of reference, but ultimately proper interpretation of the readings and experience yield the best results. That said, are there any good rules of thumb when it comes to interpreting incident readings outdoors? When interpreting incident meter readings taken in direct sunlight or during magic hour, for example, is it customary to overexpose by a stop or so? On the flip side, when filming in overcast conditions with uniform, diffused daylight, do you usually go with the incident reading or under/overexpose a bit? I realize that these kinds of generalizations may not really be applicable in practice, since the lighting conditions and highlight/shadow composition is always going to vary from one shot to another, but I thought it would be worth asking.

 

In terms of interpreting meter readings, I think David Mullen phrased it best in a recent post..."Metering [a grey card] is the beginning of deciding what the exposure should be, not the end." While metering itself isn't terribly difficult, the key is knowing how to interpret the meter readings and apply them to each unique scene and lighting composition...a skill that's probably best learned from experience, and hopefully a skill that I'll be able to perform better as I continue experimenting with shooting film.

 

[David Mullen ASC]

Some of this is helped by some simple testing, shoot a person and a grey scale in frontal light in 1-stop exposure increments, over and under, so you have an idea of what 1-stop over looks like, etc. Of course, in real life it is a bit more complex when the subject is in mixed light, shadowy light, etc.

 

But keep in mind that if you are shooting negative, you have some flexibility if you mis-expose and generally, overexposure is more correctable than underexposure, within reasonable limits. This is one reason why I tend to rate a stock slower, just to give myself some extra density. Rating a stock 1/3 of a stop slower, for example, isn't so much about consistently overexposing because 1/3 of a stop is within a margin of human error of exposing... but it does slightly tilt your exposures away from underexposing as often.

 

I was recently watching some movies shot by David Watkin and Conrad Hall on blu-ray, where you are see the grain structure more clearly, so it's easier to spot an underexposed shot that is printed up... what I've noticed is that they both had a tendency to shoot near wide-open all the time for a well-exposed negative and then print down when they wanted more mood. So with negative film, if you are going to err, it's good to err on the side of more exposure.

 

Unless maybe if you are shooting skiers dressed in white on a snowy slope that is backlit by the sun...

[Karl Lee]

Some of this is helped by some simple testing, shoot a person and a grey scale in frontal light in 1-stop exposure increments, over and under, so you have an idea of what 1-stop over looks like, etc. Of course, in real life it is a bit more complex when the subject is in mixed light, shadowy light, etc.

 

Good suggestion, David. Actually, I performed a similar test on a wide landscape shot which had both bright, direct sunlight and shade. I did a complete aperture sweep starting at T2.5 (wide open on my Canon zoom), then closed down to T2.8 and worked my way all the way to T22 in full stop increments, stopping the camera between aperture changes to make the different aperture increments easily identifiable. My spot metering indicated an f/4 - 5.6 split in the shade and an f/11- 16 split in the direct sunlight. When I checked out the shot on the film transfer, it looked like the sweet spot for this particular scene was right around T8, which happens to be the midpoint between the two incident readings. T11, while acceptable, looked a little underexposed in the shade compared to the T8 shot. The T5.6 shot was also acceptable, but compared to the T8 shot, the areas in direct sunlight were starting to look very slightly overexposed.

 

I've read previously that when shooting a scene which has both direct sunlight and shade/shadows that taking the average of the readings in the shade and in the sunlight is a good starting point, but just because that happened to be the optimal aperture setting in this particular test, I'm not sure that it's really a proven, reliable metering method. Every scene is different, and my guess is that in many cases, skewing the lens aperture more toward one reading over the other will yield better results.