Incident Metering

[Digital]

When using an incident meter it helps if the light your subject is in is uniform. I encountered a situation where my subject (log structure) was lit in direct sunlight, and there were shadows were the roof over hung.
I metered for the hghlights since I figured I could rescue the detail in the shadows in post processing. If I metered the shadow area, I would have blown the highlights. I read recently that if you blow the highlights it is practically impossible to retrieve them in your image editor.
Sure enough (how is that for a "Southern" term), I was able to retrieve the details in the shadow areas in post processing.
This has been my observation.
Comments welcomed.

Bruce

[Manfred M]

Bruce - I primarily use my incident meter for flash photography, usually using a model but sometimes I work with still life work. I tend to not use dramatic lighting that results in either crushed shadow details or clipped highlights.

My normal approach is to meter the side that the light is falling on, with the measuring dome fully extended, pointing at the light source, when shooting in daylight. This will provide an average reading based on a reading of all light coming in from the hemisphere you are measuring from. If shooting a person I will meter close to their face and in a still life, I would meter where the light hits the main subject.

If the light is very extreme (split lighting) I might meter the dark side just to ensure that I stay within the dynamic range of my sensor.

That being said, I rarely use an incident meter in outdoor work. I will use the camera's meter and I will adjust my exposure, based on what I am seeing on the camera's histogram (I check all three channels). Blowing out the highlights is not good and in those situations I might let some of the shadows slip into black. With the camera's meter, I generally have at least an extra stop to play with in the highlights.

[DanK]

This sounds like exactly the situation for which your camera's spot meter would be ideal.

[Digital]

Ted, Manfred, and Dan, thank you for your comments.


Bruce

[Digital]

Manfred, I was playing with my light meter when I photographed the log building. I also set my camera to aperture preferred which indicated an exposure setting 1 stop from the light meter reading. Basically, I was happy with the results.


Bruce

[xpatUSA]

Manfred, I was playing with my light meter when I photographed the log building. I also set my camera to aperture preferred which indicated an exposure setting 1 stop from the light meter reading.
Bruce

Sounds about right, Bruce. The difference of one stop being proportional to the reflectance of the logs, meaning that the logs reflected only half of the light falling upon them.

This old chart may help with understanding that - or it might confuse ***:



It doesn't fully agree with your specific experience. You'll notice that "Weathered Wood" is only 1/3 stop different to mid-gray, indicating that your logs are more reflective than Kodak thinks they should be.

*** I'll be happy to answer any questions about the chart ...

[William W]

. . . I encountered a situation where my subject (log structure) was lit in direct sunlight, and there were shadows were the roof over hung. I metered for the hghlights since I figured I could rescue the detail in the shadows in post processing. If I metered the shadow area, I would have blown the highlights. I read recently that if you blow the highlights it is practically impossible to retrieve them in your image editor. Sure enough (how is that for a "Southern" term), I was able to retrieve the details in the shadow areas in post processing. This has been my observation. Comments welcomed.

Comments:
1. When using an Incident Light Meter, in that scenario, metering the highlight area is the first step in correct protocol: the (typical) next step is to understand and then apply any latitude you have in the Film or the Sensor. What that means is, in simple (digital) terms, you push the exposure as far to the right as possible without blowing the highlights or limiting what highlights you do blow. For example there are some situations where you might choose to have spectral highlights blown. The rationale for (typically) pushing the exposure to the right as far as possible is to ensure the greatest amount of exposure possible I the shadow areas, ergo, ensuring the least amount of increase in the shadow areas in Post Production (or Wet Printing if using Film). If using Digital medium, then use the lowest possible ISO, preferably then native ISO also will assure the least amount of visible noise infiltrating the shadow area when that area is brought up in Post Production. (If using Film, there are development techniques available to increase shadow detail when one is developing the Negative or Positive as well as dodging and burning techniques when Wet Printing).

2. Re using a TTL Spot Meter: Firstly (pedantically few if and cameras provide a ‘true’ Spot Meter, but that is by the by). This method is one that I have used. If you ‘Spot Meter’ the highlight area and ‘Spot Meter’ the Shadow area you can firstly get a range, however it is important to remember that any REFLECTIVE Light Meter Reading is affected by the Colour and Saturation of the Subject being metered, so, as per Ted’s chart, the Colour of the Log does need to be taken into account.

3. Another method: The scenario seems a non action shot, that is, you have time to set it up. In this case consider framing the shot and then use whatever TTL ‘Averaging’ metering system your camera provides. For example ‘Matrix’ in Nikon, or ‘Evaluative’ in Canon. These metering systems are quite smart and rarely blow any highlights. Then Bracket the shot. I know that using my Canon DSLRs, I would typically pull ±⅔ Stop for the scene you describe, based upon my previous testing the headroom that I have in my cameras.

4. You also have the Histogram to interrogate; some Cameras also provide ‘Blinkies’ both are good tools aid pushing to the right.

All that said - practicing and experimenting with different methods to get to the same best practice answer is fantastic and I encourage that.

WW

[Digital]

To be more specific, with the incident meter the reading was: f/16 @ 1/125, ISO 200. The camera meter (aperture preferred) gave me a reading of f/16 @ 1/100, ISO 200.
I shot a total of 10 shots (5 using the reading of the incident meter, 5 using the reading of the camera's meter).
I had to increase the shadow detail more in the shots that I used the incident meter as a guide than I did using the camera's meter as a guide. This makes sense since the incident meter reading did not take into account the reflectance of the logs, which were in the shade.
Comments welcomed.

Bruce

[William W]

To be more specific, with the incident meter the reading was: f/16 @ 1/125, ISO 200. The camera meter (aperture preferred) gave me a reading of f/16 @ 1/100, ISO 200.
I shot a total of 10 shots (5 using the reading of the incident meter, 5 using the reading of the camera's meter).
I had to increase the shadow detail more in the shots that I used the incident meter as a guide than I did using the camera's meter as a guide. This makes sense since the incident meter reading did not take into account the reflectance of the logs, which were in the shade.
Comments welcomed.

(technical) comments not very possible unless you provide two more details:

1. The Metering Mode (and the camera)
2. The range of the Scene that was in shot when you metered the scene

Meaning
1. Matrix/Evaluative, Center Weighted Average, Spot, Partial etc.
2. A picture would be good, yet a description would suffice

- eg - 'I used Center Weighted Average and the scene (LtoR) in the viewfinder extended from the log in bright front lit sunlight to shadow area from the building overhang and (TtoB) clear blue sky to bright green grass on the left and grass in shadow on the right'

***

Educated guess -

As I understand the meaning of your conclusion, I think it is incorrect.

IF you used the incident meter at the sunny end of the scene and pointed the dome towards the camera then f/16 @ 1/125, ISO 200 would be an understandable reading i.e. overexposing 2/3 Stop referencing the F/16 Rule for Front-Lit Sunlight, Carrollton, Georgia being 33N that seems about correct for this time of year assuming. Also it might have been a tad late in the afternoon or a few hours after sunrise. In any case it's not educational to quibble over 2/3Stop.

The Camera's meter reading is reckoning 1/3Stop more overexposed than the Incident reading: this could be for many reasons, which is why we need the answers to the above.

And also to confirm you used the Incident Meter as I described.

In any case the 'reflectance of the logs' matters not to any Incident Light Meter Reading. That's the point of using and Incident Reading, doing so doesn't need to take into account the colour, hue, or saturation of the Subject/Object being photographed.

WW

[xpatUSA]

To be more specific, with the incident meter the reading was: f/16 @ 1/125, ISO 200. The camera meter (aperture preferred) gave me a reading of f/16 @ 1/100, ISO 200.

Bruce,

The difference between those meterings is 0.32EV, as Bill just said - very close to the Kodak chart I posted earlier. With everything else being equal, the ratio of shutter times is good enough: log of (1/100 divided by 1/125) divided by log of (2) equals 0.32EV.

But in post#6, you said:

"I also set my camera to aperture preferred which indicated an exposure setting 1 stop from the light meter reading."

Which was your result? One-third stop difference or one whole stop?

Did "1 stop"in post#6 perhaps mean 1 click on the setting wheel?

[William W]

... in post#6, you said:
"I also set my camera to aperture preferred which indicated an exposure setting 1 stop from the light meter reading."
Which was your result? One-third stop difference or one whole stop?
Did "1 stop"in post#6 perhaps mean 1 click on the setting wheel?

I didn't pick up on that. Thanks Ted.

[xpatUSA]

I didn't pick up on that. Thanks Ted.

No problem, Bill.

Over to Bruce ...

[Digital]

The camera was set for Matrix metering. The scene was lit by very bright sunlight up to the overhang which placed the logs in shade. The grass in front of the structure was in bright sunlight as well. The photos were taken at 3:13 p.m. (EST).
By one stop I mean a whole click from 1/125 (incident meter) to 1/100 camera's meter that the camera selected when set to aperture preferred.
I know that the incident meter reading did not take into account the reflectance of the logs. I apologize if I did not make myself clear.

Bruce

P.S. The incident meter reading was taken in the sunlight pointed toward the camera.

[pschlute]

By one stop I mean a whole click from 1/125 (incident meter) to 1/100 camera's meter that the camera selected when set to aperture preferred.

"Stop" means a halving or doubling of the exposure. So 1/125 to 1/60 would be one stop. 1/125 to 1/100 would be one third of a stop.

[DanK]

Bruce,

At the risk of being repetitive: it seems to me that you are making this far more complicated than it need be.

The key to this, as Bill explained, is to expose to the right. If the dynamic range is so large that your camera crushes the blacks when you expose the right, then you need to bracket, but it sounds like that wasn't the case here.

Cameras include very good tools for doing this, and IMHO, all you got for bringing an incident meter into the situation was needless complexity. It will not give you better results than your camera's reflective metering system will.

Do what Bill has in his #2 in post 8. More specifically, this is how I would handle it:

1. Spot meter off the brightest areas. These are the ones you need to protect from clipping.

2. Because the meter will try to make whatever you point it at neutral, you will have to open the aperture to brighten those areas. I would probably start by opening up two stops (see Peter's comment about what a stop is), although that may well be too little. Take a shot and look at the histogram. If there is still empty space on the right, open the aperture a little more, the amount depending on how much space you have.

All of this can be done in a very short time. Even if looking at the histogram shows that you have to shoot again, that won't take as long as walking over the the subject, figuring out where to place the incident meter, etc.

I don't think I have used an incident meter since the 1970s. If I did studio portraits, I probably would, but for the work I do, there is no value in carrying one. Unlike film cameras, digital cameras provide a more accurate indication of how well you have set exposure than any meter: the histogram. And digital cameras make it trivial to correct if you metered in a less than ideal way, given that additional shots are completely free.

Dan

[Digital]

Peter, I stand corrected. I should have thought this out better.

Dan, the whole exercise was to see how my incident meter stacked up against the light meter in my camera.

Bruce

[xpatUSA]

The camera was set for Matrix metering. The scene was lit by very bright sunlight up to the overhang which placed the logs in shade. The grass in front of the structure was in bright sunlight as well. The photos were taken at 3:13 p.m. (EST).

Personally, Bruce, I don't use Matrix or Evaluative - preferring instead to make my own decisions as to what exposure to set, based on Spot Metering.

Dan's advice about metering for the brightest parts of a scene is good enough to start with - but with the caveat that those parts may not necessarily be desirable - for example, harsh bright reflections from bright metal parts (cars, motor-cycles) or from glass (white clouds in window-panes). You could just let those blow ...

By one stop I mean a whole click from 1/125 (incident meter) to 1/100 camera's meter that the camera selected when set to aperture preferred.

Thanks for the clarification - the which Peter has already addressed in post #14. Remember, too, that "one click" can mean a different exposure step on different cameras - for example, a half-stop or a third-stop.

I tried incident-metering for a while and it does work but I am too lazy to carry more than just a camera. I'm even too lazy to carry a gray card, lens filters, or anything else but that's just me ...

Please do join the "I always shoot Center-Spot" gang. Take the camera somewhere and wave it around in various areas to get used to the metering numbers given (leave EC at zero and ISO at 100).

Worth a few reads if you haven't already:

https://en.wikipedia.org/wiki/Exposure_value

In your scene you had sunlit grass. Around here, that is about a half-stop greater than neutral so I would apply one or two clicks-worth of positive EC (1/3 or 2/3 EV) to what it meters. By so doing, I've turned the camera into effectively an incident light-meter!

To me, the more important part of exposure is learning how known objects are likely to meter and what the ambient lighting represents - such that you could take a well-exposed shot without any metering at all just based on experience ...

"I am a Matrix-oholic ... been using Spot for over five years now" - welcome, Brother.

[Digital]

Ted, Thanks for the suggestions. I will definitely give spot metering a try.

Bruce

[xpatUSA]

Ted, Thanks for the suggestions. I will definitely give spot metering a try.

Bruce

Good to hear. It will help if you get a feel for how big your camera's spot size is in the LCD/EVF. A common value is about 2 percent of the area, for what that is worth; which means that spot-metering the moon ain't gonna work ...

[Manfred M]

Peter, I stand corrected. I should have thought this out better.

Dan, the whole exercise was to see how my incident meter stacked up against the light meter in my camera.

Bruce

The two are not really comparable, as they are using two completely different approaches.

Reflective metering attempts to emulate what incident metering does, but assumes an average scene (much like you would get when light bounces off an 18% gray card that fills your viewfinder).

In incident light meter measures the amount of light falling on a subject, so if your subject is going to impact the reading. In an "average" scene with a good mix of dark and light tones the exposure should be very close, but in isolation a dark area might appear to be a touch too dark and a light area might appear too bright. In cases like that, I tend to either bracket or rely on the camera's histogram to ensure I have enough good data (ETTR) to work with.