How consistent are incident light meter readings?

[peterB1966]

I am trying to do a complex indoor / flash outdoor setup and, while my assumption is that whether I meter indoors, outdoors, or form the flash, these are all relative to an absolute EV, i.e. it is not relative to 50% grey or anything else, and if I take an incident meter reading of the same light on different days or in different situations, I should come up with the same values each time.

I decided to test this (staying away from stuff like back-lighting that might mess with the reading vs what the camera captures), and in some instances the exposures from my D610 looked correct, but in others they looked up to two stops too light

Technical: I am using the incident sphere on a Polaris flash meter, which on the Polaris does not recess into the meter (no haters, please: we buy what we can afford). Because I was mainly testing low light situations, I was alternating between ISO 4000 and ISO 1600, don't know if high ISO's would affect accuracy. I have been careful to not shoot too reflective a surface, and have varied between dark and light.

 

[Sirius Glass]

The problem with comparing reflectance readings and incidence readings is that the meters are faced in the opposite directions. Some expect the two to be the same, but the is not always the case. As long as both meters are properly calibrate each will be correct for its perspective and may or may not agree. There are many thread discussing the question why they do not agree, but this should clarify the question.

 

[Chan Tran]

The incident meter reading isn't affected by the subjects and only by the light falling on it. In that scene it's consistent. If you are expecting that it would consistently gives you the good exposure then it won't.

 

[peterB1966]

...If you are expecting that it would consistently gives you the good exposure then it won't.

Could you explain some more? Unless you are referring to e.g. backlit scenes or highly reflective subjects etc that would screw with your exposure, surely the whole idea behind incident metering is meant to be totally consistent?

 

[Paul Howell]

Not sure what OP means by consistent. With a reflective average meter wide view of around 30% meter reading may different depending on how much sky, how close the subject. Incident metering taken in from of the subject facing the camera will spot on every time. Commercial, fashion, industrial photogpghers preferred incident metering for that reason. As a PJ shooing on the fly it was always reflective, only used incident metering on occasion such as an environmental portrait or a fashion shoot.

 

[MattKing]

It is difficult to ensure that your metering technique is exactly identical between two uses at different locations and/or different times.
And unless one is controlling the light, one can't assume that the light is the same.

 

[flavio81]

I am trying to do a complex indoor / flash outdoor setup and, while my assumption is that whether I meter indoors, outdoors, or form the flash, these are all relative to an absolute EV, i.e. it is not relative to 50% grey or anything else, and if I take an incident meter reading of the same light on different days or in different situations, I should come up with the same values each time.

I decided to test this (staying away from stuff like back-lighting that might mess with the reading vs what the camera captures), and in some instances the exposures from my D610 looked correct, but in others they looked up to two stops too light

Technical: I am using the incident sphere on a Polaris flash meter, which on the Polaris does not recess into the meter (no haters, please: we buy what we can afford). Because I was mainly testing low light situations, I was alternating between ISO 4000 and ISO 1600, don't know if high ISO's would affect accuracy. I have been careful to not shoot too reflective a surface, and have varied between dark and light.

I find incident metering the most consistent metering there is. Works for me with 100% rate of success.

If you are shooting a normal scene, such as a portrait with a normal background and normal (not extreme) lighting, and the incident meter is placed on your subject pointing to the most important light source, chances are you'll get a perfect exposure right away.

The polaris flash meters are good machines, no reason to make fun of them.

 

[Chan Tran]

Could you explain some more? Unless you are referring to e.g. backlit scenes or highly reflective subjects etc that would screw with your exposure, surely the whole idea behind incident metering is meant to be totally consistent?

The incident meter measure the light falling on the meter and it's consistent if the light falling on the meter is the same every where in the scene. If you photograph a scene that there is half of the scene in the shade and the other in bright sun then where do you place the meter?

 

[peterB1966]

The incident meter measure the light falling on the meter and it's consistent if the light falling on the meter is the same every where in the scene. If you photograph a scene that there is half of the scene in the shade and the other in bright sun then where do you place the meter?

Thanks for the reply - yes, I was careful to specify metering non-tricky situations such as you have just described. I.e. as described by the poster immediately before your reply.

 

[benjiboy]

With some modern hi-tech light meters from the leading manufacturers, the incidental flash meter reading tells you what proportion of the reading is fash and what proportion is daylight, I find this facility very useful with my Kenko KFM 2100 meter in balancing window light and fill in flash in general.

 

[peterB1966]

With some modern hi-tech light meters from the leading manufacturers, the incidental flash meter reading tells you what proportion of the reading is fash and what proportion is daylight, I find this facility very useful with my Kenko KFM 2100 meter in balancing window light and fill in flash in general.

Yes, I've seen that and drooled in envy.

 

[Paul Howell]

There a few situations that incident meter does not work as well as reflective, best examples that comes to mind is the last edition of AA's book the negative. His example is a distance house with a door in bright light, the foreground an ivy covered gate in shade. If you meter for the gate the door will be too bright, if you meter for the door the gate will be too dark. You can incident meter both and split the difference while bracketing. With the ZS you meter for the gate and develop for the door. Of course that only really works with sheet film.

 

[peterB1966]

There a few situations that incident meter does not work as well as reflective, best examples that comes to mind is the last edition of AA's book the negative. His example is a distance house with a door in bright light, the foreground an ivy covered gate in shade. If you meter for the gate the door will be too bright, if you meter for the door the gate will be too dark. You can incident meter both and split the difference while bracketing. With the ZS you meter for the gate and develop for the door. Of course that only really works with sheet film.

You've just confused the hell out of me: what you describe sounds like reflective metering, as surely the incident meter would have it's back to all of this?

 

[Paul Howell]

The question what is the subject, the gate or the door? If you stand in front of the door the gate will underexposed, if you stand at the gate the door will likely be overexposed. An incident meter is predicated on standing in front of the subject, the meter collecting the light falling on the subject, facing the camera. So what if there are 2 or more points of interests in different lighting, what to meter? A average reflective meter will meter the entire scene and average out the 2 for an average meter. A matrix meter uses a stored program of patterns to try and figure what is the lighting is and calculating the best exposure.

With roll film I would use a reflective spot meter, meter the shadows as zone III then develop the roll as zone Vii and fix as needed when printed.

 

[Sirius Glass]

I find incident metering the most consistent metering there is. Works for me with 100% rate of success.

If you are shooting a normal scene, such as a portrait with a normal background and normal (not extreme) lighting, and the incident meter is placed on your subject pointing to the most important light source, chances are you'll get a perfect exposure right away.

The polaris flash meters are good machines, no reason to make fun of them.

I agree, however reflectance metering is more convenient and is needed for Zone System metering.

 

[Pieter12]

he incident meter is placed on your subject pointing to the most important light source

In my experience, unless you are measuring lighting ratios, the dome of the meter should be pointed to the camera, not the light source.

 

[flavio81]

In my experience, unless you are measuring lighting ratios, the dome of the meter should be pointed to the camera, not the light source.

This is correct too. It depends on what look you want to achieve, really. c.

 

[Paul Howell]

Beyond the Zone System BTZS uses a incident meter, the object is determine a scene brightness range not a shutter speed F stop combo. As with a standard incident metering reading standing at the subject meter the full sun, then a dark shadow and found the brightness range which is input to the magic wheel or phone app which along with all the data collected during testing computes the exposure. When metering a distant object BTZS folks will stand in front of the camera, back to the scene and measure the full light, then using a hat or hands creates the shadows. If you thought the zone was technical read Phil Davis's book.

 

[John Koehrer]

beyond Adams and btzs and compensating for exposure with development,
the simple answer (to me). is point the ball at the camera & go.
If the subject is backlit, and meter like this the side of the subject facing you will be OK and
background over exposed. What's important to you?
Chan has a good point though and it would need a bit of thought or bracketing.
The great thing about incident is the sun is a light source and exposure doesn't change over the
area you're photographing. Taking a pic from one side of the grand canyon? So what?
detail in the shade over there? compensate, put a shadow over the ball probably close 'nuff.

 

[benjiboy]

beyond Adams and btzs and compensating for exposure with development,
the simple answer (to me). is point the ball at the camera & go.
If the subject is backlit, and meter like this the side of the subject facing you will be OK and
background over exposed. What's important to you?
Chan has a good point though and it would need a bit of thought or bracketing.
The great thing about incident is the sun is a light source and exposure doesn't change over the
area you're photographing. Taking a pic from one side of the grand canyon? So what?
detail in the shade over there? compensate, put a shadow over the ball probably close 'nuff.

In backlight, you can use "The Duplex Method" as described in Dunn and Wakefield's Exposure Manual(sadly out of print), in which you point the dome at the Sun, note the reading, then from the subject at the camera, and average the two readings, this works in any light, front side or backlight.

Metering Techniques: How to use an Incident Light Meter for backlit scenes with the Duplex Method: Page 2

crawfordphotoschool.com

 

[JerseyDoug]

The incident meter measure the light falling on the meter and it's consistent if the light falling on the meter is the same every where in the scene. If you photograph a scene that there is half of the scene in the shade and the other in bright sun then where do you place the meter?

I would place the meter back in my pocket and find another subject

Seriously, there is no way to capture both the shadow details of the shaded half of the scene and the highlight details of the sunny half. You will have to decide which matters most and meter accordingly.

 

[Chan Tran]

I would place the meter back in my pocket and find another subject

Seriously, there is no way to capture both the shadow details of the shaded half of the scene and the highlight details of the sunny half. You will have to decide which matters most and meter accordingly.

That is the extrem case where the light is significantly different from one part of the scene to the other part. However, if the light is only lightly different you can can certainly capture the whole range of brightness but in this case you still don't know where to place the meter. The reflective meter when use to measure the light and dark part of the scene you would know how much details you can capture.

 

[Juan Valdenebro]

I am trying to do a complex indoor / flash outdoor setup and, while my assumption is that whether I meter indoors, outdoors, or form the flash, these are all relative to an absolute EV, i.e. it is not relative to 50% grey or anything else, and if I take an incident meter reading of the same light on different days or in different situations, I should come up with the same values each time.

I decided to test this (staying away from stuff like back-lighting that might mess with the reading vs what the camera captures), and in some instances the exposures from my D610 looked correct, but in others they looked up to two stops too light

Technical: I am using the incident sphere on a Polaris flash meter, which on the Polaris does not recess into the meter (no haters, please: we buy what we can afford). Because I was mainly testing low light situations, I was alternating between ISO 4000 and ISO 1600, don't know if high ISO's would affect accuracy. I have been careful to not shoot too reflective a surface, and have varied between dark and light.

Hi Peter,
Beware: there are lots of mistakes around this matter everywhere.
Yes: incident readings are consistent and reliable, if well done, of course, just like reflected readings are.
When I calibrate my films/developers I check both (a well placed gray card for spot, and a well placed hemisphere) and both readings are not just more or less the same: they are identical.
Obviously, if tools are not used correctly, meterings will be wrong.
No problem your meter can't have its hemisphere recessed: just place your hand horizontally two inches above the hemisphere, like a hat, so you'll get rid of the too vertical light that doesn't really illuminate say a vertical face... If your subject is on the grass face up, then it makes sense your hemisphere reads also all that vertical/cenital light without being recessed or hand covered.
Aim your hemisphere exactly from main subject to camera.

 

[Juan Valdenebro]

Apart from that, you'll have to understand the differences between slide film, color negative film, black and white negative film, and digital sensors: in some cases it's blocked highlights what you avoid with your exposure, and in some cases it's rich shadows what you decide with your exposure.

 

[wiltw]

With reference to a known light source intensity (such as a calibration light source at a camera repair shop) all quality meters are 'consistent' (except maybe the cheapo ones)
With reference to 'daylight' the intensity of the light falling on the scene varies from day to day, and even hour to hour...'Sunny 16' is nothing but a rough approximation, ever.
Here is an example of how a meter with Matrix Metering can change its exposure, even as you move the frame around slightly...

lens at a single FL and f/2.8, and the resulting Av mode exposures were

1. 1/4000
2. 1/6400
3. 1/1600
4. 1/6400

Note shots 1 and 4 and framed 'about the same' (they differ vertically) and yet the exposure is different!
...while an incident light meter would give you one reading (at a given angle of incidence)

But even an incident meter will vary in its readings simply due to changes in either/both horizontal and vertical angles, in how you hold it, relative to sun position!

With regard to flash-only, a meter should be consistent unless your flash source itself is not consistent...using a quality studio flash unit, my flash meter will be consistent, not even varying 0.1EV