Chriscrawfordphoto
Real Men Shoot Film.
In my Introduction to Handheld Exposure Meters and How to Use an Incident Meter For Digital Photography tutorials, I explained that incident light meters cannot be used to meter backlit scenes. That isn't entirely true!
When used normally, an incident meter will give incorrect exposure of backlit scenes. This is because an incident meter measures the light that falls on the front of the subject. In a backlit scene, the main light comes from behind the subject, where the incident meter is blind.
In my other tutorial on backlit scenes, I taught you a technique in which you average an incident reading and a reflected light reading of the scene. It works wonderfully, but can only be used with meters that can also do wide-area reflected light metering as well as incident metering.
This tutorial will teach you a technique for determining exposure for backlit subject matter using only an incident meter. This technique is called the Duplex Method. It was invented and named decades ago by Jack Dunn, a British engineer who did pioneering research into exposure metering in the mid 20th Century. Dunn wrote, with George Wakefield, a book called "Exposure Manual." If you're as big of a meter geek as I am, you want to get hold of this book.
The Problem
In a backlit scene, an incident meter used in the normal manner will give readings that overexpose the photograph. This is because the meter's white dome will be in the shade due to the main light source being behind the meter.
Equipment Requirements
Normally, an incident meter has a white dome covering the light sensor. Many meter manufacturers offer a "Flat Diffuser" as an accessory. This flat diffuser is required for the Duplex Method.
Some modern meters feature an incident dome that can be raised and lowered. In the lowered position, it functions like a flat diffuser.
Incident light meters with the standard "Spherical Diffuser" dome.
Incident light meters with the "Flat Diffuser."
Most meters, like the Minolta meter shown here, have a separate flat diffuser that replaces the standard dome.
The Sekonic has an ingenious system that doesn't require the removal of the dome when you need a flat diffuser. Instead of removing the dome, the you lower the dome into the meter head by turning the knurled ring around the base of the dome. When lowered, the dome then functions identically to a meter with a flat diffuser. This is nice because you don't have to carry an accessory, and there's no risk of losing it or the dome when out photographing. Several meters made today have this feature.
How To Use The Duplex Method
1: Mount the flat diffuser on your incident light meter (or lower the dome if your meter has that feature).
2: Take a meter reading from the subject position, with the flat diffuser pointed directly at the camera. This is just like doing a normal incident light reading, except you're using the flat diffuser instead of the dome. Remember the reading it gives. If your meter has a memory function, you can use that.
3: Take a meter reading with the flat diffuser pointed directly at the source of light. Outdoors, this is usually the sun. This reading will be higher than the first reading you took. If it is a sunny day with a clear, cloudless sky then the sun direction reading may be 4 or 5 stops higher than the camera direction reading you took in step two.
Remember this reading, or put it in your meter's memory if it has one.
4: Average the readings. If you're using a modern meter with a memory function, then it probably also has an Averaging button. This makes it super easy...no math!
5: Set your camera to the average reading and take the picture.
Don't Worry If It Doesn't Look Perfect!
The image that results may not look perfect on your camera's screen. That's OK. This method is designed to ensure that the full range of tones are captured by the sensor. To do so, the final image may look underexposed; especially if the difference between the two readings that you took is very large.
In such lighting, it is impossible to get a "perfect" image right from the camera. The goal is to produce a RAW file that can be edited to bring out full detail in the lightest and darkest areas of the image. This technique does that. Notice that I said "RAW File," not JPEG. You must shoot RAW and edit the image using RAW conversion software, like Adobe Lightroom or Capture One.
When used normally, an incident meter will give incorrect exposure of backlit scenes. This is because an incident meter measures the light that falls on the front of the subject. In a backlit scene, the main light comes from behind the subject, where the incident meter is blind.
In my other tutorial on backlit scenes, I taught you a technique in which you average an incident reading and a reflected light reading of the scene. It works wonderfully, but can only be used with meters that can also do wide-area reflected light metering as well as incident metering.
This tutorial will teach you a technique for determining exposure for backlit subject matter using only an incident meter. This technique is called the Duplex Method. It was invented and named decades ago by Jack Dunn, a British engineer who did pioneering research into exposure metering in the mid 20th Century. Dunn wrote, with George Wakefield, a book called "Exposure Manual." If you're as big of a meter geek as I am, you want to get hold of this book.
The Problem
In a backlit scene, an incident meter used in the normal manner will give readings that overexpose the photograph. This is because the meter's white dome will be in the shade due to the main light source being behind the meter.
Equipment Requirements
Normally, an incident meter has a white dome covering the light sensor. Many meter manufacturers offer a "Flat Diffuser" as an accessory. This flat diffuser is required for the Duplex Method.
Some modern meters feature an incident dome that can be raised and lowered. In the lowered position, it functions like a flat diffuser.
Incident light meters with the standard "Spherical Diffuser" dome.
Incident light meters with the "Flat Diffuser."
Most meters, like the Minolta meter shown here, have a separate flat diffuser that replaces the standard dome.
The Sekonic has an ingenious system that doesn't require the removal of the dome when you need a flat diffuser. Instead of removing the dome, the you lower the dome into the meter head by turning the knurled ring around the base of the dome. When lowered, the dome then functions identically to a meter with a flat diffuser. This is nice because you don't have to carry an accessory, and there's no risk of losing it or the dome when out photographing. Several meters made today have this feature.
How To Use The Duplex Method
1: Mount the flat diffuser on your incident light meter (or lower the dome if your meter has that feature).
2: Take a meter reading from the subject position, with the flat diffuser pointed directly at the camera. This is just like doing a normal incident light reading, except you're using the flat diffuser instead of the dome. Remember the reading it gives. If your meter has a memory function, you can use that.
3: Take a meter reading with the flat diffuser pointed directly at the source of light. Outdoors, this is usually the sun. This reading will be higher than the first reading you took. If it is a sunny day with a clear, cloudless sky then the sun direction reading may be 4 or 5 stops higher than the camera direction reading you took in step two.
Remember this reading, or put it in your meter's memory if it has one.
4: Average the readings. If you're using a modern meter with a memory function, then it probably also has an Averaging button. This makes it super easy...no math!
5: Set your camera to the average reading and take the picture.
Don't Worry If It Doesn't Look Perfect!
The image that results may not look perfect on your camera's screen. That's OK. This method is designed to ensure that the full range of tones are captured by the sensor. To do so, the final image may look underexposed; especially if the difference between the two readings that you took is very large.
In such lighting, it is impossible to get a "perfect" image right from the camera. The goal is to produce a RAW file that can be edited to bring out full detail in the lightest and darkest areas of the image. This technique does that. Notice that I said "RAW File," not JPEG. You must shoot RAW and edit the image using RAW conversion software, like Adobe Lightroom or Capture One.
Chriscrawfordphoto
Real Men Shoot Film.
Real World Examples
Here are some real-world examples of photographs made of backlit scenes using the Duplex Method to determine exposure.
Backlit Tree
Here, again, is the example photograph from the beginning of this tutorial.
I determined the exposure using an incident meter in the normal way. The meter had the standard white dome attached, and I metered by pointing the dome back toward the camera from the subject position.
The photograph is overexposed, since the meter could not see the bright light coming from behind the subject.
In this version, I determined exposure using the Duplex Method.
This is a straight rendering, no tonal adjustments were done to the RAW file in Lightroom. It looks great; the brightly lit areas in the background are rendered accurately, and the shadows still have full detail.
If you want even more detail in the shadows, they can be lightened using the Shadow Recovery slider in your RAW conversion software.
Here is the Duplex Method exposure, with the shadows adjusted using the Shadow Recovery slider in Adobe Lightroom.
Backlit Tree #2
Here is another photo of the backlit tree, this time done on a day with hazy sunlight and snow on the ground. The sun was not nearly so harsh as it was in the first example.
I determined the exposure using an incident meter in the normal way. The meter had the standard white dome attached, and I metered by pointing the dome back toward the camera from the subject position.
The photograph is slightly overexposed, since the meter could not see the bright light coming from behind the subject.
In this version, I determined exposure using the Duplex Method.
This is a straight rendering, no tonal adjustments were done to the RAW file in Lightroom. It looks much better; the brightly lit snow in the background has full detail, and so do the shadows.
1958 Chevy Viking Truck
This is kind of a worst case scenario for backlighting. The sun is very harsh and is coming from behind and to the left of the truck.
I determined the exposure using an incident meter in the normal way. The meter had the standard white dome attached, and I metered by pointing the dome back toward the camera from the subject position.
The shadowed areas are rendered nicely; but the sky is way too light and the side of the truck's hood and windshield are blown out, as are the brightly lit patches of snow on the ground.
In this version, I determined exposure using the Duplex Method.
The lighting ratio was very large. The difference between the camera direction reading and the sun direction reading was 5 stops! This kind of lighting is best avoided, as there are always going to be quality compromises. I present this example because there are times when you simply must get the picture now; where you cannot come back later when there is better light.
This is a straight rendering, no tonal adjustments were done to the RAW file in Lightroom. Large adjustments will be required to get a usable image out of this!
The final version, after adjustments in Lightroom.
I raised the shadows up using Lightroom's Shadow Recovery slider pulled all the way up to its highest setting, and I used a little bit of the Highlight Recovery slider to bring back a little more detail in the most brightly lit areas. I also increased the midtone brightness using the Curves adjustment in Lightroom.
This isn't perfect, but perfect is not possible in such harsh, ugly light. I have made beautiful photographs of this truck on overcast days, when the light is soft; and on sunny mornings, when the sun lights the front of the truck. The truck faces east, so in the mornings it is lit by the sun and in the afternoons it is backlit.
Alternatives
I also have a tutorial on metering backlit scenes by averaging a reflected light reading and an incident reading.
Here are some real-world examples of photographs made of backlit scenes using the Duplex Method to determine exposure.
Backlit Tree
Here, again, is the example photograph from the beginning of this tutorial.
I determined the exposure using an incident meter in the normal way. The meter had the standard white dome attached, and I metered by pointing the dome back toward the camera from the subject position.
The photograph is overexposed, since the meter could not see the bright light coming from behind the subject.
In this version, I determined exposure using the Duplex Method.
This is a straight rendering, no tonal adjustments were done to the RAW file in Lightroom. It looks great; the brightly lit areas in the background are rendered accurately, and the shadows still have full detail.
If you want even more detail in the shadows, they can be lightened using the Shadow Recovery slider in your RAW conversion software.
Here is the Duplex Method exposure, with the shadows adjusted using the Shadow Recovery slider in Adobe Lightroom.
Backlit Tree #2
Here is another photo of the backlit tree, this time done on a day with hazy sunlight and snow on the ground. The sun was not nearly so harsh as it was in the first example.
I determined the exposure using an incident meter in the normal way. The meter had the standard white dome attached, and I metered by pointing the dome back toward the camera from the subject position.
The photograph is slightly overexposed, since the meter could not see the bright light coming from behind the subject.
In this version, I determined exposure using the Duplex Method.
This is a straight rendering, no tonal adjustments were done to the RAW file in Lightroom. It looks much better; the brightly lit snow in the background has full detail, and so do the shadows.
1958 Chevy Viking Truck
This is kind of a worst case scenario for backlighting. The sun is very harsh and is coming from behind and to the left of the truck.
I determined the exposure using an incident meter in the normal way. The meter had the standard white dome attached, and I metered by pointing the dome back toward the camera from the subject position.
The shadowed areas are rendered nicely; but the sky is way too light and the side of the truck's hood and windshield are blown out, as are the brightly lit patches of snow on the ground.
In this version, I determined exposure using the Duplex Method.
The lighting ratio was very large. The difference between the camera direction reading and the sun direction reading was 5 stops! This kind of lighting is best avoided, as there are always going to be quality compromises. I present this example because there are times when you simply must get the picture now; where you cannot come back later when there is better light.
This is a straight rendering, no tonal adjustments were done to the RAW file in Lightroom. Large adjustments will be required to get a usable image out of this!
The final version, after adjustments in Lightroom.
I raised the shadows up using Lightroom's Shadow Recovery slider pulled all the way up to its highest setting, and I used a little bit of the Highlight Recovery slider to bring back a little more detail in the most brightly lit areas. I also increased the midtone brightness using the Curves adjustment in Lightroom.
This isn't perfect, but perfect is not possible in such harsh, ugly light. I have made beautiful photographs of this truck on overcast days, when the light is soft; and on sunny mornings, when the sun lights the front of the truck. The truck faces east, so in the mornings it is lit by the sun and in the afternoons it is backlit.
Alternatives
I also have a tutorial on metering backlit scenes by averaging a reflected light reading and an incident reading.
Out to Lunch
Ventor
Thanks Chris. Very informative. In one of your next contributions, I would appreciate if you could share your expertise on how to best deal with chromatic aberration. Not only in Photoshop terms -I don't use it, but more in general terms...i.e. adjusting exposure, contrast etc... Happy Holidays! Peter
Pete B
Well-known
Edit: wrong thread
retinax
Well-known
Chris, your results speak for themselves. But if the lighting was even harsher or the contrast between backlit subject and background higher, this method means risking blown highlights, so with the dynamic range of modern sensors mostly in the shadows, why wouldn't you simply expose for the highlights by using your metering for the sunlit background and do everything else in post? I'm thinking the normal background incident metering should already be the maximum exposure you can give without blowing highlights? Anyway as I wrote, your results seem to prove my thinking wrong, but I don't quite understand why...
charjohncarter
Veteran
Thanks Chris, now I know what the flat diffuser is for, that came with my meter.
Chriscrawfordphoto
Real Men Shoot Film.
There's no way to meter for the bright background with an incident meter. If you stand in the brightly lit background and do a normal incident meter reading (dome pointed at camera), then the dome will still be backlit, producing a badly overexposed image. If you point it at the sun, you'll get an underexposed image since incident meters are not calibrated for reading the light source directly.
That's why we average the light source and camera direction readings. Most of the time, it produces an exposure level that keeps both the darkest parts of the shadowed, backlit subject and the lightest parts of the brightly lit background all within the camera's exposure range.
If the lighting ratio is so big that it is impossible to keep everything within the range of tones the camera can handle, then you do sacrifice the very lightest and darkest tones; the Duplex Method then effectively exposes to keep at least tones in the middle acceptable.
So, there has to be compromise. Such bad lighting really should be avoided if possible. Another alternative is to use fill flash to light the subject, which reduces the brightness range by putting light in the shadows. I don't really like using flash, as it usually looks un-natural even when done right, but for some situations it can be useful.
Chriscrawfordphoto
Real Men Shoot Film.
Its also used for studio work; and if you read the meters instructions, that is probably all they mention for the flat diffuser.
In the studio, it has two uses.
1) Checking lighting ratios. When you use studio lights, you typically have a bright main light and a dimmer fill light. The fill light ensures that the shadows don't go too dark. Contrast in the image is controlled by the difference in brightness between these two lights, and the flat diffuser is used to measure that difference.
Because the flat diffuser is more directional in its sensitivity (it cannot see light coming from an angle as well as the round dome does), you can point it at each light and take a reading, and it will ignore the other light, giving you the brightness of the one light you point it at.
2) Photographing flat objects, like paintings and drawings, and copying photographs.
Just as the dome diffuser simulates the way a three-dimensional subject is lit, the flat diffuser does the same for flat subject matter. It'll give more accurate readings when photographing flat things.
Also, it is used to check even-ness of illumination when copying photos or art. You need the light to be PERFECTLY even over the whole work of art. Even 1/3 stop difference will show up.
To check for even light on a work of art, you use the flat diffuser and take readings with the meter at the center and at all four corners of the piece. All these readings must match. If they don't, you adjust your lights till they do.
CMur12
Veteran
I really appreciate the information you provide here, Chris. It just makes sense that you would have a forum of your own to keep it all together. Thanks to you and Stephen.
I have a flat disc diffuser for my Minolta Autometer III, but I never saw any explanation of its intended purpose. I bought mine when I was doing occasional window-light portraits on Agfachrome 1000 and I used it to take separate readings of lighting from the side (where the window was located) and the front. I would then compare them and decide how to balance them.
I typically got good results from just using the dome diffuser, however.
- Murray
I have a flat disc diffuser for my Minolta Autometer III, but I never saw any explanation of its intended purpose. I bought mine when I was doing occasional window-light portraits on Agfachrome 1000 and I used it to take separate readings of lighting from the side (where the window was located) and the front. I would then compare them and decide how to balance them.
I typically got good results from just using the dome diffuser, however.
- Murray
Chriscrawfordphoto
Real Men Shoot Film.
Peter,
Since this question isn't related to the rest of this thread, I am going to start a new one to answer it. Give me a couple days to dig up some example photos from my archives and write the article, then I'll post it on my forum.
Chriscrawfordphoto
Real Men Shoot Film.
Murray,
I'm glad you're finding my stuff useful.
The dome is best for general purpose work. The flat diffuser is more of a specialized accessory, for backlit scens using the Duplex Method and for the studio tasks I mentioned in my reply to John earlier.
retinax
Well-known
Thanks for your explanations! I still don't completely understand. I don't have an incident meter unfortunately (need to calibrate my old Luna Pro, but I'm afraid the little dome has darkened), so I can't try for myself at the moment.
Then how can I meter at all? Neither away from the sun or toward the sun will work? In a non-backlit situation, the sun would shine on the meter dome as well, no?
I suspect the issue is that I think that a sunlit landscape should get the same exposure no matter from which direction I photograph it, unless I have a foreground subject that's in the shade and that I need texture in (only then would I normally treat it as backlit subject, not because the whole landscape has the sun coming from the side opposite the camera - think of a landscape without much in terms of deep shadow, like flat hills lit by a late afternoon sun).
But using an incident meter will give different readings for a sunlit landscape depending from which direction I want to photograph it, correct?
This makes a lot of sense, thank you.
Then how can I meter at all? Neither away from the sun or toward the sun will work? In a non-backlit situation, the sun would shine on the meter dome as well, no?
I suspect the issue is that I think that a sunlit landscape should get the same exposure no matter from which direction I photograph it, unless I have a foreground subject that's in the shade and that I need texture in (only then would I normally treat it as backlit subject, not because the whole landscape has the sun coming from the side opposite the camera - think of a landscape without much in terms of deep shadow, like flat hills lit by a late afternoon sun).
But using an incident meter will give different readings for a sunlit landscape depending from which direction I want to photograph it, correct?
This makes a lot of sense, thank you.
Chriscrawfordphoto
Real Men Shoot Film.
An incident meter is not usually pointed directly at the sun. Instead, it is pointed at the camera. Even if the dome is brightly lit by the sun when doing the normal camera-direction reading, it will give a lower reading (more exposure) than it would if pointed up at the sun directly.
The only time you point the meter directly at the sun is when using the Duplex Method for backlit scenes. In that case, you must use a flat diffuser, something your Luna Pro doesn't have, and the reading is not used directly; it is averaged with a camera direction reading that is also done with the flat diffuser.
If the sun is in front of the camera, and not behind it, the scene is still backlit even if there is no shadowed foreground object. As such, it will need more exposure than a frontlit scene. The technique in this tutorial accurately gives you the correct exposure for such scenes.
Because you do not have a meter that you can put a flat diffuser on, this technique won't work for you. If you use the dome on the meter to do the camera direction and the sun direction readings needed for the Duplex Method, the exposure will probably not be accurate.
My other backlighting tutorial uses a camera direction reading with the incident dome and a reflected light reading of the scene which are then averaged. It works great for the Luna Pro. See the Metering Backlit Scenes with the Incident-Reflected Average Method tutorial.
Spanik
Well-known
This is a very comprehensive and detailed guide. The added photos are really a great thing. I learened already a couple of things and I'm looking forward to the rest of this series.
Many thanks for this work.
Many thanks for this work.