Chriscrawfordphoto
Real Men Shoot Film.
In my Intro To Handheld Meters Tutorial, I explained that incident light meters are often the best choice because, unlike reflected light meters, they cannot be fooled by unusually bright or dark subject matter. The incident meter avoids that problem by simply ignoring the subject altogether!
Instead of being pointed at the subject, the incident meter is pointed away from the subject, toward the camera. This allows it to measure how much light is illuminating the scene. The meter's light sensor is covered by a white translucent dome, which simulates a three-dimensional subject. In the real world, objects receive light from all directions. The incident meter's dome is also lighted from all directions, or most of them. It does not see light coming from behind the subject.
Some incident light meters from my collection.
The Sekonic L-758DR is a modern multipurpose meter that includes an incident meter and a one degree spotmeter in one unit.
The Gossen Ultra-Pro (Mastersix outside the USA) is basically a reflected light meter, but it has a small incident light dome that slides in place over the meter cell. Because a larger dome is theoretically more accurate than a smaller one at measuring light that comes from the sides, Gossen sold the "Luna-Sphere" attachment shown on the Ultra-Pro in the middle. In practice, the small built-in dome works fine.
The Minolta Flash Meter IV is a system meter whose capabilities could be expanded by a multitude of accessories that added reflected light metering, spot metering, and even the ability to meter through microscopes. The "Flat Diffuser" shown above the meter is one of the most useful accessories. It is used for incident metering when photographing flat objects, like artwork, and for measuring lighting ratios with studio lighting.
The big spherical diffuser at the top is one of the more unusual items from my collection. It is an accessory made by Topcon, a now-defunct Japanese camera manufacturer, back in the 1970s. When fitted to a standard 50mm lens, like a filter, it allowed you to take incident light readings using the camera's built-in through-the-lens exposure meter.
How To Use An Incident Light Meter
At its most basic, an incident meter is incredibly simple to use. You simply stand near your subject and point the meter's white dome directly back toward the camera (or the place you'll be standing when you take the picture if you don't have the camera on a tripod). Make sure the meter is in the same light as the subject, then push the meter's measuring button and set your camera with the settings the meter gives.
What if you cannot get close to the subject? You can take the meter reading somewhere else, as long as the light is the same as it is at the subject position. Often, when I am doing landscape work, I'll just stand in front of the camera to do the incident reading rather than walk about a long distance into the scene. In cases where you cannot get to the subject, and the light is completely different than where you are located (a lighted stage during a concert, for example), you cannot use an incident meter. A reflected light meter would be necessary for that.
In the real world, there are situations where it isn't quite so simple. What if the subject has both bright sunlit areas and deep shadows? Where do you place the meter? What about backlit scenes?
We'll begin with the easiest lighting situations.
In normal lighting, the subject is pretty evenly illuminated. Examples include subjects lit by overcast or cloudy sky, subjects in the shade, and subjects lit directly from the front, or from an angle that is not too large, by the sun without very deep, dark shadows.
The metering technique with the incident meter in these types of lighting is super easy. Stand by your subject, point the white dome directly back toward the camera position, and take the reading. Set your camera to the aperture and shutter speed the meter indicates, and take the picture!
Instead of being pointed at the subject, the incident meter is pointed away from the subject, toward the camera. This allows it to measure how much light is illuminating the scene. The meter's light sensor is covered by a white translucent dome, which simulates a three-dimensional subject. In the real world, objects receive light from all directions. The incident meter's dome is also lighted from all directions, or most of them. It does not see light coming from behind the subject.
Some incident light meters from my collection.
The Sekonic L-758DR is a modern multipurpose meter that includes an incident meter and a one degree spotmeter in one unit.
The Gossen Ultra-Pro (Mastersix outside the USA) is basically a reflected light meter, but it has a small incident light dome that slides in place over the meter cell. Because a larger dome is theoretically more accurate than a smaller one at measuring light that comes from the sides, Gossen sold the "Luna-Sphere" attachment shown on the Ultra-Pro in the middle. In practice, the small built-in dome works fine.
The Minolta Flash Meter IV is a system meter whose capabilities could be expanded by a multitude of accessories that added reflected light metering, spot metering, and even the ability to meter through microscopes. The "Flat Diffuser" shown above the meter is one of the most useful accessories. It is used for incident metering when photographing flat objects, like artwork, and for measuring lighting ratios with studio lighting.
The big spherical diffuser at the top is one of the more unusual items from my collection. It is an accessory made by Topcon, a now-defunct Japanese camera manufacturer, back in the 1970s. When fitted to a standard 50mm lens, like a filter, it allowed you to take incident light readings using the camera's built-in through-the-lens exposure meter.
How To Use An Incident Light Meter
At its most basic, an incident meter is incredibly simple to use. You simply stand near your subject and point the meter's white dome directly back toward the camera (or the place you'll be standing when you take the picture if you don't have the camera on a tripod). Make sure the meter is in the same light as the subject, then push the meter's measuring button and set your camera with the settings the meter gives.
What if you cannot get close to the subject? You can take the meter reading somewhere else, as long as the light is the same as it is at the subject position. Often, when I am doing landscape work, I'll just stand in front of the camera to do the incident reading rather than walk about a long distance into the scene. In cases where you cannot get to the subject, and the light is completely different than where you are located (a lighted stage during a concert, for example), you cannot use an incident meter. A reflected light meter would be necessary for that.
In the real world, there are situations where it isn't quite so simple. What if the subject has both bright sunlit areas and deep shadows? Where do you place the meter? What about backlit scenes?
We'll begin with the easiest lighting situations.
In normal lighting, the subject is pretty evenly illuminated. Examples include subjects lit by overcast or cloudy sky, subjects in the shade, and subjects lit directly from the front, or from an angle that is not too large, by the sun without very deep, dark shadows.
The metering technique with the incident meter in these types of lighting is super easy. Stand by your subject, point the white dome directly back toward the camera position, and take the reading. Set your camera to the aperture and shutter speed the meter indicates, and take the picture!
Chriscrawfordphoto
Real Men Shoot Film.
Below are some real-world examples:
Stop Sign
This stop sign is lit evenly by the sun. I held the incident meter in front of the sign, dome pointed at the camera. The resulting exposure is perfect. The white parts of the sign have full detail, as does the house in the background. At the same time, the shadowed areas are not too dark.
This is a straight rendering, no tonal adjustments were done to the RAW file in Lightroom.
Red Truck and Pink House
This photograph was a fairly typical sunny-day scene. This straight rendering, with no tonal adjustments, shows good midtones and full detail in whites. I held the meter in the sunlit part of the scene, not the shadowed part.
Shadows are a little dark, and can be lightened if you want. The main objective was to make sure the whites rendered right. Dark tones that go too dark can be recovered when processing the RAW file much more easily than can whites that go too light.
Dead Sunflower In The Snow
This dead sunflower was photographed in soft overcast light. Everything rendered beautifully, from the bright snow to the darker parts of the dead leaves.
This is a straight rendering, no tonal adjustments were done to the RAW file in Lightroom.
Voors Jewelry
This jewelry store was lit by soft evening light late in the day. I stood in front of the building and pointed the incident meter's dome back toward the camera. The exposure the meter recommended was perfect.
This is a straight rendering, no tonal adjustments were done to the RAW file in Lightroom.
Stop Sign
This stop sign is lit evenly by the sun. I held the incident meter in front of the sign, dome pointed at the camera. The resulting exposure is perfect. The white parts of the sign have full detail, as does the house in the background. At the same time, the shadowed areas are not too dark.
This is a straight rendering, no tonal adjustments were done to the RAW file in Lightroom.
Red Truck and Pink House
This photograph was a fairly typical sunny-day scene. This straight rendering, with no tonal adjustments, shows good midtones and full detail in whites. I held the meter in the sunlit part of the scene, not the shadowed part.
Shadows are a little dark, and can be lightened if you want. The main objective was to make sure the whites rendered right. Dark tones that go too dark can be recovered when processing the RAW file much more easily than can whites that go too light.
Dead Sunflower In The Snow
This dead sunflower was photographed in soft overcast light. Everything rendered beautifully, from the bright snow to the darker parts of the dead leaves.
This is a straight rendering, no tonal adjustments were done to the RAW file in Lightroom.
Voors Jewelry
This jewelry store was lit by soft evening light late in the day. I stood in front of the building and pointed the incident meter's dome back toward the camera. The exposure the meter recommended was perfect.
This is a straight rendering, no tonal adjustments were done to the RAW file in Lightroom.
Chriscrawfordphoto
Real Men Shoot Film.
We have looked at scenes that were pretty evenly illuminated. Now, let's consider scenes with high contrast light. In bright sunlit scenes, there are often very deep shadows cast by the sun. If the scene has both brightly lit areas and deep shadows, where do you place the meter?
In the Intro To Handheld Meters Tutorial, I wrote that for digital photography the most important thing is getting correct exposure for the lightest tones in the image. Digital cameras have little overexposure lattitude. If you overexpose, whites will be completely blown out with no detail. You can recover overexposed light tones to some extent in RAW conversion software, like Lightroom or Capture One, using the Highlight Recovery sliders; but, the ability to do this is limited.
With dark tones, there is a lot more lattitude. Underexposed images can be recovered more easily using the Shadow Recovery sliders in post processing.
Because of those characteristics of digital sensors, it is most important to expose for the brightly lit areas. This may push the shadows down to where they're too dark, but you can recover them when you process the image. What I am teaching you assumes that you shoot RAW, not JPEG. The ability to recover detail in deep shadows and in overexposed light tones is severely limited with a JPEG file.
So, the answer is to place the meter in the sunlit area, not the shadowed area.
Below are some real-world examples:
Garage Light
I held the incident light meter in front of the light fixture so it would receive the full sunlight that illuminates the brightest part of this scene.
The light tones are rendered correctly, but the shadows are too dark, especially next to the garage, where the shadows are much darker than the shadow on the garage door.
This is a straight rendering, no tonal adjustments were done to the RAW file in Lightroom.
In this version, I held the incident light meter in front of the garage door in the shadowed area on the front of the garage.
As you can see, the image is overexposed. The shadowed area renders as white, which is really too light since it is in shadow. The deep shadow next to the garage, with the garbage cans looks great, and if I were just photographing them it would be the right exposure for them.
The big problem here is the sunlit areas on the front of the garage and in the background landscape. They're way too light, and the whites are completely washed out.
The version where I metered in the brightly lit area is much better. But what about those shadows? Aren't they too dark?
Here is the version where I used the incident light meter in the sunlit area.
This time, I used the Shadow Recovery slider in Lightroom to open up the shadows, which are too dark in the straight, unedited version.
They could probably have been lightened even more, but I wanted to preserve the feeling of deep shadows from the very harsh light.
What About Backlit Scenes?
Backlit scenes are the hardest to meter, and for that reason, I am creating separate tutorials on metering for backlighting. There are several possible approaches.
The most accurate, but most difficult, is to use a handheld spotmeter to find the important light-tone areas in the scene and adjust the reading the meter gives to place them at the correct values on the tonal scale. I will have a tutorial for spotmetering soon.
There are two simpler methods for metering backlit scenes.
Here's a short description of each:
The Duplex Method: This technique uses an incident light meter with a flat diffuser. Two readings are taken, one with the meter pointed at the camera and the other with it pointing at the sun or other main light source; the readings are then averaged. Tutorial for this is coming soon.
Incident-Reflected Average: With this technique, you take a reflected light meter reading of the scene and an incident light reading of the scene, then average them.
In the Intro To Handheld Meters Tutorial, I wrote that for digital photography the most important thing is getting correct exposure for the lightest tones in the image. Digital cameras have little overexposure lattitude. If you overexpose, whites will be completely blown out with no detail. You can recover overexposed light tones to some extent in RAW conversion software, like Lightroom or Capture One, using the Highlight Recovery sliders; but, the ability to do this is limited.
With dark tones, there is a lot more lattitude. Underexposed images can be recovered more easily using the Shadow Recovery sliders in post processing.
Because of those characteristics of digital sensors, it is most important to expose for the brightly lit areas. This may push the shadows down to where they're too dark, but you can recover them when you process the image. What I am teaching you assumes that you shoot RAW, not JPEG. The ability to recover detail in deep shadows and in overexposed light tones is severely limited with a JPEG file.
So, the answer is to place the meter in the sunlit area, not the shadowed area.
Below are some real-world examples:
Garage Light
I held the incident light meter in front of the light fixture so it would receive the full sunlight that illuminates the brightest part of this scene.
The light tones are rendered correctly, but the shadows are too dark, especially next to the garage, where the shadows are much darker than the shadow on the garage door.
This is a straight rendering, no tonal adjustments were done to the RAW file in Lightroom.
In this version, I held the incident light meter in front of the garage door in the shadowed area on the front of the garage.
As you can see, the image is overexposed. The shadowed area renders as white, which is really too light since it is in shadow. The deep shadow next to the garage, with the garbage cans looks great, and if I were just photographing them it would be the right exposure for them.
The big problem here is the sunlit areas on the front of the garage and in the background landscape. They're way too light, and the whites are completely washed out.
The version where I metered in the brightly lit area is much better. But what about those shadows? Aren't they too dark?
Here is the version where I used the incident light meter in the sunlit area.
This time, I used the Shadow Recovery slider in Lightroom to open up the shadows, which are too dark in the straight, unedited version.
They could probably have been lightened even more, but I wanted to preserve the feeling of deep shadows from the very harsh light.
What About Backlit Scenes?
Backlit scenes are the hardest to meter, and for that reason, I am creating separate tutorials on metering for backlighting. There are several possible approaches.
The most accurate, but most difficult, is to use a handheld spotmeter to find the important light-tone areas in the scene and adjust the reading the meter gives to place them at the correct values on the tonal scale. I will have a tutorial for spotmetering soon.
There are two simpler methods for metering backlit scenes.
Here's a short description of each:
The Duplex Method: This technique uses an incident light meter with a flat diffuser. Two readings are taken, one with the meter pointed at the camera and the other with it pointing at the sun or other main light source; the readings are then averaged. Tutorial for this is coming soon.
Incident-Reflected Average: With this technique, you take a reflected light meter reading of the scene and an incident light reading of the scene, then average them.
Ronald M
Veteran
Nice work Chris. I got some friends started with incidents 30 plus years ago. They all loved it.
Nikons metering is more able to cope with scenes like the stop sign, but Leica is same old average. What Nikon can not do is meter cloudy overcast correctly. Not my Nikons fault as I have many and they all do it wrong, pro models to prosumer.
Sekonic 398 needs no battery. Love it.
When things get down and dirty, Pentax digital spot never misses, but is far more tricky to use than sekonic.
I read sometime back the movie people all use incident because there is no way to bracket a movie and reshoots cost big $. You can also just walk across the stage and make sure light is even. Portrait studio is the same. Get things right the first time.
Nikons metering is more able to cope with scenes like the stop sign, but Leica is same old average. What Nikon can not do is meter cloudy overcast correctly. Not my Nikons fault as I have many and they all do it wrong, pro models to prosumer.
Sekonic 398 needs no battery. Love it.
When things get down and dirty, Pentax digital spot never misses, but is far more tricky to use than sekonic.
I read sometime back the movie people all use incident because there is no way to bracket a movie and reshoots cost big $. You can also just walk across the stage and make sure light is even. Portrait studio is the same. Get things right the first time.
agfa100
Well-known
Merry Christmas Sir! I still use my Sekonic L28c2 I picked up in the PX in Saigon in 1969, it was the best 25.00 I ever spent and it came with the full slide set.
wbill
wbill
Chriscrawfordphoto
Real Men Shoot Film.
You're right, most Hollywood cinematographers use incident meters. The Sekonic L-398 you mentioned was originally called the Norwood Director; it was designed for movie work, and later on people began using them for still photography too.
The director was designed by an American named Don Norwood, and originally was an American-made product. Sekonic bought the design later on and has improved upon the original while keeping the same basic look.
helen.HH
To Light & Love ...
Brilliant Chris !
Well thought out, Concise & Easy Read...
Thanks for putting it All at my fingertips
Well thought out, Concise & Easy Read...
Thanks for putting it All at my fingertips
Ko.Fe.
Lenses 35/21 Gears 46/20
If you are standing where subject of photography is and pointing it to the camera, it doesn't matter where light is. Back or front. Isn't it?
Good topic anyway, my M-E metering fails miserably under difficult light.
Good topic anyway, my M-E metering fails miserably under difficult light.
charjohncarter
Veteran
Well done Chris! Really good examples for incident use. I'm a small meter fan so I use a Sekonic L308s (which also has a couple of flash metering modes). Before that I used, for over 30 years, a Sekonic-Master. The Master has an inverted cone. I used to know the reason for the inversion, I think it had to do with side lighting. Excellent tutorial Chris.
Chriscrawfordphoto
Real Men Shoot Film.
John,
Your Sekonic Master was actually the Weston Master IV. Sekonic paid Weston for a license to manufacture the meter in Japan.
Many of Sekonic's early meters were foreign designs that Sekonic bought. The most famous is the L-398 Studio Deluxe, which was originally the American-made Norwood Director. Sekonic bought the design when the last American manufacturer of it, Brockway, stopped making it.
The inverted cone incident diffuser was created by Weston to work like a hemispherical diffuser, without infringing on Norwood's patent for the round diffuser. The Norwood Director was the first incident meter to have a round diffuser; all previous incident meters had flat ones, which are less accurate for three dimensional subjects. The invention of the round diffuser is what made incident meters go mainstream!
The original version of the Weston Invercone was smaller and flatter than the later version, and it was not as good as the round diffusers.
In the 1960s, Weston redesigned the Invercone to be very large and with a different shape. The "Improved Invercone," which was first sold with the Master IV and continued to be included with the Master V, Euro-Master, and Euro-Master II meters.
The Improved Invercone was actually the best incident metering system ever designed; it is more accurate in extreme side-light than round diffusers, and can even compensate to some degree for backlighting! I wish someone made a modern meter with the later invercone design.
I have a Master V that was rebuilt by Ian Patridge in the UK with a new Selenium Cell (virtually all old Weston IV and V meters now have dead cells). It is a very accurate system, better in many ways than modern incident meters, except for the crappy low-light sensitivity.
charjohncarter
Veteran
Thanks for that write up. You are right about the low light problem. Still, for a very long time (and still does) it provided incident readings for me. Even as little as fifteen years ago people didn't understand the incident reading. It is probably going in that direction again with Digital camera users.
Godfrey
somewhat colored
Nice write up, Chris. It echoes what I have in my Focal Press book "Basic Photography" published circa 1958. I studied that book fifty years ago and committed it to memory.
To me, the critical thing to understand for proper exposure metering is the difference in metering for a digital sensor vs film. It has to do with the difference in the recording mediums' response curves and how you render an image from them. Metering for a digital sensor is more akin to metering for narrow-latitude positive transparencies except that the blackpoint baseline is a matter of a judgement call, essentially answering the question "how much noise is acceptable vs how much dynamic range can the medium achieve?" given that the saturation limit is an impenetrable and unforgiving flat line. Metering for film can accommodate more variability since both the toe and shoulder portions of the response curve are soft.
G
To me, the critical thing to understand for proper exposure metering is the difference in metering for a digital sensor vs film. It has to do with the difference in the recording mediums' response curves and how you render an image from them. Metering for a digital sensor is more akin to metering for narrow-latitude positive transparencies except that the blackpoint baseline is a matter of a judgement call, essentially answering the question "how much noise is acceptable vs how much dynamic range can the medium achieve?" given that the saturation limit is an impenetrable and unforgiving flat line. Metering for film can accommodate more variability since both the toe and shoulder portions of the response curve are soft.
G
Chriscrawfordphoto
Real Men Shoot Film.
Yeah, I see a lot of digital users claiming that meters are obsolete because you can look at the photo on the screen, or can look at the histogram.
Problem is, both of those are not very accurate on most cameras. I've shot hundreds of photos where the histogram said I'd have blown highlights, but when I opened the RAW file in Lightroom, the highlights had full detail with no adjustment. The meter was right, the camera wrong!
Chriscrawfordphoto
Real Men Shoot Film.
Thanks, Godfrey.
One nice thing about digital is the ability to open up the shadows after exposing for the highlights. With slide film, people had to settle for high-contrast images that often had blocked up shadows in order to keep the highlights.