spot vs. incident meter
Posted 27 February 2007 - 01:11 PM
My problem is that I'm confused as to which reading to go by when measuring the same subject, or piece of the subject. I may be overly obsessive about this but it's been bugging me for a while.
There have been times when I'll take an incident reading from right under a light at height X, and then a spot reading of a person's face (say the person's illuminated cheek), also at height X, and the spot reading will be 1 or even 2 stops above the incident reading.
People have always told me to go with the incident reading, since it better approximates the human head, but I'm wondering if the spot reading is more accurate since it measures the precise area I'm looking at (i.e. if I wanted to expose for the highlight on the cheek) and it also takes in to account the distance of the camera from the illuminated object.
Would the answer be different if we weren't talking about a subject's head, say a chair or something less round and dome-like? If for example, if someone is wearing a dark shirt but has pale skin. An incident reading would give you an overall exposure but a spot reading would give you a more accurate idea of the difference between the shirt and skin. (This could be the same for a light picture hung on a dark wall)
And if the incident reading is best, should I just use the spot reading to get an idea of the relative ratios that different areas of the frame have as compared to other areas, and not for the actual f-stop readings? Or should I just not use the spot meter option at all unless I'm only measuring a perfectly flat surface.
Anyway, these questions keep bugging me. I don't want to rely on my meter as a crutch, but I don't feel like I can let it go until I've understood this. Any clarification would be much appreciated.
Posted 27 February 2007 - 04:24 PM
Posted 27 February 2007 - 05:41 PM
A reflectance meter tells you an f-stop that will render the subject a medium gray. Caucasian skin tones are approximately one stop brighter than medium gray under flat light (which you've witnessed). A white peice of paper is about 2-1/2 stops brighter than medium gray. Setting you lens by the spot-meter's reading would render those subjects at a medium gray brightness, which would appear underexposed. So for a "normal" exposure under flat light you might spot-measure Caucasian skin tones and open the lens one stop from the reading. Or read a piece of white paper and open up 2-1/2 stops.
An incident meter measures the light falling on a subject; with the dome taking into account (and averaging) light from multiple angles, such as key and fill. Aiming the meter's dome straight at the camera lens produces a reading that will render a "normal" or average exposure for all the light hitting the subject (at least in the position where you measured it). Of course there are nuances to doing this, but this is the general idea.
But what if you don't want your subject to appear at a "normal" exposure? What if you want your subject to appear as if it's in a shaft of bright sunlight coming through a window? Or in shadow? You wouldn't set your lens to the incident meter's reading then -- you'd use the reading as a basis for deviation, choosing how much above or below that "normal" brightness you want it to appear.
And spot meters do not take into account the distance of the subject to the lens. What they take into account is obstructions between the lens and subject, like smoke and atmospheric haze (when looking at distant subjects). If the angle of measurement is too large (say, 10 degrees instead of 1 degree), then a spot reading may also be capturing lighter or darker areas surrounding your subject, and averaging them in to the reading.
Posted 27 February 2007 - 06:40 PM
This should clear up all of your confusion.
Posted 27 February 2007 - 07:12 PM
For instance, if I have a kicker or backlight that's shining light on the rear half of the side of someone's head, I can usually assume THAT reflected highlight is roughly 1 to 2 stops over my aperture setting, even though my incident reading from behind my actor's head is probably the same as my key.
What you need to do is get familiar with your light meter, shoot some tests, and see how various lighting setups correlate with the readings you got while shooting. Be sure to take notes on every take, this can be a quick reference that you should keep plugged into your brain for future usage.
Posted 27 February 2007 - 09:23 PM
He justified metering this way by saying that if he meters the incident light he feels he is just getting an approximation of light values. If he spot meters from the camera he feels like he is reading light the same way the film will since the film is exposing what is reflected at it. He also said that by no means was this a scientific justification but it works for him.
Of course you still have to pick an exposure based on your spot readings.
"The Negative" by Ansel Adams has great info on this as well as some info on the Zone system that may help round out your choices on exposure.
Posted 27 February 2007 - 10:19 PM
But sometimes, like in many daylight situations, incident is faster, I don't like being an obsessive meter reader.
I'm not sure this helps you but....
Posted 28 February 2007 - 07:34 AM
if it works for him, fine, but isn't it just a question of how you've learned it and what you draw from experience? i know extremely well how a face looks on neg, reversal, video, hd whatever on this and that lighting setup with this ratio. do i know how a face will look if one side of the face spots two stops under the other, and how this will look in front of a green wall another stop further down? not really. i'm not lost of course but it doesn't come naturally. i'll still use my spot meter to check so that wall doesn't become too dark though.
He justified metering this way by saying that if he meters the incident light he feels he is just getting an approximation of light values. If he spot meters from the camera he feels like he is reading light the same way the film will since the film is exposing what is reflected at it.
Posted 28 February 2007 - 08:49 AM
A clarification, regarding this...
"saw a lighting demo by Christian Sebalt ASC once and he said that he never uses his incident meter. He prefers to be by the camera so that he can evaluate the scene as a whole and fire off spot meter readings from there"
I have used a contrast filter, determined roughly my mid gray in the frame, took reading from spot meter and exposed, it was fine. Can u please comment. I will definitely try more number of times, but pl do share ur thoughts. Thank you.
- Karthik Ganesh
Posted 28 February 2007 - 09:19 AM
I experienced the advantages of spot reading in close-up shots. But it's difficult to maintain the continuity with spot reading.