Overexposed

[ashokgoyal42]

Hi,

Not realizing my light meter was set at incident i/o reflective I ended up overexposing 2 rolls of 120mm Panf.

Should I underdevelop in terms of time and agitate less? (I will be developing these one roll at a time so that at least one of them turns out right.) I will be using LC29 diluted 1:29 because I need to develop at room temp 27C.

Ashok

 

[Poohblah]

Depends on how much you overexposed. One roll at a time is good too; I've used that trick on rolls when I was unsure of the condition of the shutter. But unless the material at which you pointed the light meter was quite black or in the shadows, I wouldn't worry about it. Don't be too liberal in your underdeveloping. Also, I would make sure that you are developing rolls of 120 film and not 120mm film, the latter of which is very difficult to find in rolls.

 

[ashokgoyal42]

Depends on how much you overexposed. One roll at a time is good too; I've used that trick on rolls when I was unsure of the condition of the shutter. But unless the material at which you pointed the light meter was quite black or in the shadows, I wouldn't worry about it. Don't be too liberal in your underdeveloping. Also, I would make sure that you are developing rolls of 120 film and not 120mm film, the latter of which is very difficult to find in rolls.

Hey Poohblah,

So 120 film isn't 120mm - WOW! Now that I reflect on this it was stupid of me to assume 120 referred to 120mm - so what does it refer to?

I was fortunate in one way that I was shooting in extremely poor light, but I think instead of 30 seconds (accounting for reciprocity)I exposed for 60 seconds in quite a few shots. I'm planning to reduce development time by 10% in the first roll and then see what that leaves me for the second.

Ashok

 

[MattKing]

Hey Poohblah,

So 120 film isn't 120mm - WOW! Now that I reflect on this it was stupid of me to assume 120 referred to 120mm - so what does it refer to?

Ashok

120 is just an arbitrary number that was assigned to the film size when it was originally developed/marketed.

Most film size numbers are the same, although some might have some connection with one or more dimensions. Think, for example, of 828, or 616, or ....

EDIT: here is a link to an interesting chart: http://www.camerapedia.org/wiki/Film_sizes_and_designations

 

[Poohblah]

Hey Poohblah,

So 120 film isn't 120mm - WOW! Now that I reflect on this it was stupid of me to assume 120 referred to 120mm - so what does it refer to?

I was fortunate in one way that I was shooting in extremely poor light, but I think instead of 30 seconds (accounting for reciprocity)I exposed for 60 seconds in quite a few shots. I'm planning to reduce development time by 10% in the first roll and then see what that leaves me for the second.

Ashok

Long exposures are tricky to expose and develop properly, so I would do one roll at a time regardless of whether or not you overexposed.

 

[2F/2F]

Setting your light meter at incident instead of reflective would give you a correct average exposure for the light at which you pointed the dome, which you would not get with a reflected reading, so everything is fine. Develop as normal.

Why do you think that your film is overexposed? Did you do something other than just setting the meter to incident instead of reflective?

 

[ashokgoyal42]

Setting your light meter at incident instead of reflective would give you a correct average exposure for the light at which you pointed the dome, which you would not get with a reflected reading, so everything is fine. Develop as normal.

Why do you think that your film is overexposed? Did you do something other than just setting the meter to incident instead of reflective?

When just the last shot was left on the last of the two rolls I saw the dome over the reading hole, so when I pushed it back to reflective the light meter went from suggesting 5.6 for 30 seconds to 5.6 for less than half of 30. Also the incident light was the light falling on my meter not on the subject, so how would that be correct? Wouldnt I have to reflect of the subject to the camera for the incident light reading to show an incident light reading correctly? The scene was poorly lit so I might get away with long exposure, but I think I should underdevelop one roll so I know how to handle the second. I'm even tempted to cut the rolls in half and test one portion.

And by the way I instinctively kept feeling that I'm overexposing ... the readings did seem poor but I followed the meter after I checked it's battery level.

Ashok

 

[tkamiya]

Hi,

Not realizing my light meter was set at incident i/o reflective I ended up overexposing 2 rolls of 120mm Panf.

Should I underdevelop in terms of time and agitate less? (I will be developing these one roll at a time so that at least one of them turns out right.) I will be using LC29 diluted 1:29 because I need to develop at room temp 27C.

Ashok

I think the answer really depends on the reflectivity of the subject. If you had the meter set to incident and it was pointed to the subject as you would normally do for reflective, your meter will read lower. Therefore, the film will be over-exposed. If your subject was 100% reflective, then the reading will be accurate. If your subject was 0% reflective, then it will be infinitely overexposed. I'd say recreate a similar scene with similar object and figure out the difference before you can do anything...

 

[ashokgoyal42]

I think the answer really depends on the reflectivity of the subject. If you had the meter set to incident and it was pointed to the subject as you would normally do for reflective, your meter will read lower. Therefore, the film will be over-exposed. If your subject was 100% reflective, then the reading will be accurate. If your subject was 0% reflective, then it will be infinitely overexposed. I'd say recreate a similar scene with similar object and figure out the difference before you can do anything...

Good thought that one ... I can easily test the situation today as the photography was done in my dining room. My subject was 0% reflective.

Ashok

 

[tkamiya]

That can't be true... if the subject was 0% reflective... your meter would have had any reading.... Assuming you mean it was very dark and less reflective, I think the margin of error is much greater... good luck.

 

[ashokgoyal42]

That can't be true... if the subject was 0% reflective... your meter would have had any reading.... Assuming you mean it was very dark and less reflective, I think the margin of error is much greater... good luck.

My subjects were a group of dried up Bonsai and cactus - in very poor light purposely against a white wall. The meter was dead. Let's see ... I'll know today. I'll post the results.

 

[2F/2F]

When just the last shot was left on the last of the two rolls I saw the dome over the reading hole, so when I pushed it back to reflective the light meter went from suggesting 5.6 for 30 seconds to 5.6 for less than half of 30. Also the incident light was the light falling on my meter not on the subject, so how would that be correct? Wouldnt I have to reflect of the subject to the camera for the incident light reading to show an incident light reading correctly? The scene was poorly lit so I might get away with long exposure, but I think I should underdevelop one roll so I know how to handle the second. I'm even tempted to cut the rolls in half and test one portion.

And by the way I instinctively kept feeling that I'm overexposing ... the readings did seem poor but I followed the meter after I checked it's battery level.

Ashok

Reflected meters are affected by subject brightness, which is why they are such miserable contraptions if not used with skill and understanding. Your meter suggested less exposure than when in incident mode because it was telling you something different. A reflected meter tells you how to make anything within its field of view appear half a stop below middle grey in a normal print, no matter how light or dark it is in reality. An incident meter tells you how to make a middle tonality in reality appear as a middle tonality in a normal print.

Reflected meters do the same thing as incident meters if you fill the reflected meter's field of view with a grey card, and then add 50% more light than the meter recommends (open up half a stop, or slow down 1/2 shutter speed).

As long as the incident meter (or grey card) is in roughly the same light as your subject when you metered, your exposures will be near to textbook perfect (barring technical failure/inaccuracy of any of the parts of your "image chain" that affect negative exposure, such as shutters, f stops, etc.).

You should point incident meters at the light you want to measure; at that for which you want to expose. So, if you point it at light that is different from the light that your subject is in, you will get an "incorrect" exposure for the subject.

Using a directly-read averaging reflected meter reading to judge whether or not your incident meter is giving you an ideal exposure is near futile, because there will nearly always be a mismatch between the two, and often a significant one.

At any rate, negative film can survive some overexposure quite well. I would develop more based on the contrast of the light in which you shot than on possible exposure mistakes in the direction of overexposure. If you shot in contrasty light and want a normal-looking negative, I'd underdevelop. If you shot in contrasty light and want a contrasty negative, I'd develop normally, or even push if you want extreme contrast. If you shot in average-contrast light, and want an average contrast negative, I'd develop normally. See what I mean? There are three steps in what determines the contrast on your print: 1. The contrast of the brightnesses within your composition (A.K.A. luminance range or subject brightness range), 2. How that compositional contrast is rendered on film by your combination of film emulsion, exposure, and development, and 3. How the contrast on the negative is rendered on photo paper by your combination of paper, enlarger, developer, and what have you.

 

[JBrunner]

5.6 for 30 seconds to 5.6 for less than half of 30

That's only about a stop or so. Develop normal. A 1 stop thicker neg isn't much of an issue, in fact I and a few others prefer it. And you'll only get that if the reflectance of the subject was equal to gray. If the subject was above average reflectance, the suggested exposures about match. In any case you are well within the latitude of negative film, it takes much more than a stop over to really screw it up. Under developing it however, won't be as forgiving.

 

[2F/2F]

If you had the meter set to incident and it was pointed to the subject as you would normally do for reflective, your meter will read lower.

There is only a fraction of chance of this statement being correct for any random photo. You have named one possibility, but there are three. Your meter is either in the same light as the subject, in less light than the subject, or in more light than the subject.

Your meter would read lower (as in telling you to give more exposure than you need for the subject) if the meter was in less light than the subject.

Your meter would read higher (as in telling you to give less exposure than you need for the subject) if the meter was in more light than the subject.

Your meter would read the same as the subject if it was in the same light as the subject.

In other words, if you want the "correct" exposure, meter the light your subject is in, or light just like it...and in even more plain words, if you do not meter the light your subject is in, or light just like it, you will not have the "correct" exposure.

 

[tkamiya]

Why....?

Let's say I have my meter set to incident. That means the white dome "thingy" is out. I measure incident light falling on the subject. I measure a value.
Now, I point the meter in the same setting the other way, pointing to the subject. Now, the light it recieve is the one that was reflected back from the subject. This light can never be more than the light falling on the subject.

Where's the error in my logic?

I just did an experiment... using my desk lamp as a source. My 'subject' is a white board. Facing the light, it reads 1 second and f/4. Facing the white board, it reads 1 second and f/2.8 and 6/10th.

 

[2F/2F]

Why....?

Let's say I have my meter set to incident. That means the white dome "thingy" is out. I measure incident light falling on the subject. I measure a value.
Now, I point the meter in the same setting the other way, pointing to the subject. Now, the light it recieve is the one that was reflected back from the subject. This light can never be more than the light falling on the subject.

Where's the error in my logic?

The error in not in your logic, but in the assumptions you are using to apply it. It is in the thinking that the light falling on the meter when it is pointed back at the subject cannot be greater than or equal to the light falling on the subject; in thinking that it must be less.

Incident meters measure the light that is falling on the dome, independent of the subject. When the dome is pointed at the subject, the same amount of light that is falling on the subject can be falling on the dome, or more light than is falling on the subject can be falling on the dome, or less light than is falling on the subject can be falling on the dome, because the light that is falling on the dome is not all coming from the subject, but from the light that is illuminating the dome of the meter.

 

[tkamiya]

How, how, how??? Incident meter is nothing more than a light meter calibrated to show 18%, right? Assuming the subject itself does not emit light, like light bulb, how can it illuminate the sensor at higher value than the light it's falling on? (of course, I am also assuming the subject is not a concave mirror)

The only way I can think this may not be true is more of a back-lit condition - what am I missing?? We are talking about the type of incident meter where the dome only faces 180 degrees at a time, right?

 

[2F/2F]

How, how, how??? Incident meter is nothing more than a light meter calibrated to show 18%, right?

No, no, no!!! An incident meter is nothing more than a light meter designed such that an object that reflects 18% (some brands use different values) of the light that is falling on it in the real world is rendered as middle grey on a normal print.

What you have described is a reflected meter: a meter that tells one how to render that which it is pointed at as a middle grey value.

Assuming the subject itself does not emit light, like light bulb, how can it illuminate the sensor at higher value than the light it's falling on?

Who said that the subject is illuminating the meter? I said exactly the opposite. It was my whole point. The light meter measures incident light, independent of the subject. As such, it can be in the same light as the subject, less light than the subject, or more light than the subject, and not only in less light than the subject, as you are stating.

At its most simple (already stated in this thread): Reflected meters tell you how to make what they are pointed at into a middle grey. Incident meters tell you how to make a middle grey into a middle grey.

 

[2F/2F]

My subjects were a group of dried up Bonsai and cactus - in very poor light purposely against a white wall. The meter was dead. Let's see ... I'll know today. I'll post the results.

What does a white wall in the background tell a reflected meter compared to what it tells an incident meter?

If anything, this "accident" will show you that you ought to be measuring incident light for most things instead of reflected light when you want to use direct readings from a light meter. Yes: It was pointed the wrong way, and in this particular case, the white wall will influence the reading (tell you to underexpose) if it was close to the meter. However, unless you were going for a silhouette, I can assure you that the incident metered shots will be "better" exposures than the reflected metered shots.

Also, practically speaking, with a white wall behind dark bonsai plants, and an incident reading with the dome pointed at the subject, I would underdevelop your negs. You can always send that wall back to white in printing from an underdeveloped neg, but you will have to go to much more extreme levels in printing to bring a white wall behind plants down to "less white" if you do not underdevelop.

 

[Galah]

When just the last shot was left on the last of the two rolls I saw the dome over the reading hole, so when I pushed it back to reflective the light meter went from suggesting 5.6 for 30 seconds to 5.6 for less than half of 30...Ashok

This suggests that you may have overexposed by approximately one stop: The film can handle that. Just proceed as usual , or halve development time (?)

 

[ashokgoyal42]

As long as the incident meter (or grey card) is in roughly the same light as your subject when you metered, your exposures will be near to textbook perfect (barring technical failure/inaccuracy of any of the parts of your "image chain" that affect negative exposure, such as shutters, f stops, etc.).

.

The incident meter was in the same light as the subject. I was about 2 feet from the plants, the light was coming in from the left - it was very weak. I think I'm going to go for normal development for one roll.

Thanks for your notes on the evaluation of the luminance of the scene. Very informative.

 

[ashokgoyal42]

5.6 for 30 seconds to 5.6 for less than half of 30

That's only about a stop or so. Develop normal. A 1 stop thicker neg isn't much of an issue, in fact I and a few others prefer it. And you'll only get that if the reflectance of the subject was equal to gray. If the subject was above average reflectance, the suggested exposures about match. In any case you are well within the latitude of negative film, it takes much more than a stop over to really screw it up. Under developing it however, won't be as forgiving.

Awesome flickr stream !

Yes, I will develop one roll normally.

 

[JBrunner]

Why....?

Let's say I have my meter set to incident. That means the white dome "thingy" is out. I measure incident light falling on the subject. I measure a value.
Now, I point the meter in the same setting the other way, pointing to the subject. Now, the light it recieve is the one that was reflected back from the subject. This light can never be more than the light falling on the subject.

Where's the error in my logic?

I just did an experiment... using my desk lamp as a source. My 'subject' is a white board. Facing the light, it reads 1 second and f/4. Facing the white board, it reads 1 second and f/2.8 and 6/10th.

There are a few things at work here, but it helps to realize that a reflected reading gives the luminance of the subject (doesn't read the light directly, it reads the luminance of the subject, bright things have more luminance.) and will expose that subject for middle grey. Now for instance, in the case of a caucasian person, you probably don't wan't to expose their skin as middle grey, so in that case the exposure needs to be adjusted. In other words a reflected meter pointed at a caucasion gives a reading that is "wrong" in regards to what an incident reading would give (this will vary according to how the meter's angle of acceptance, a spot meter will for instance give exact middle gray exposure recommendations for particular tones within a composition, while a meter with a wide angle of acceptance will give a reading closer to the incident reading, but is easily fooled in high contrast situations, or situations where the subject being exposed for doesn't predominate the composition.) It isn't really wrong of course, it's just telling you the exposure for the middle, and also averaging the luminance it sees. You have to extrapolate the correct exposure for where you want the tones to fall. If you meter something pure white with a reflected meter and expose according to the meter, it will expose as grey, see? If you want it to expose as near bullet proof on the neg "white", you'll need to open up 4 or more stops. Caucasian skin meters about a stop and a half bright if you have a reflected reading that is reading predominantly the skin tone. Following the meters exact recommendation in that case would get you an underexposure.

In the case of flipping an incident meter around the reading will be what is falling on the dome. If the light is coming from the side, overhead, or all around it will read the same or nearly the same. If from behind or in front, the readings will of course be different.

Also, if you are going to make a mistake, always error in the favor of overexposure (fortuitously, as you may have done), because when erred to under exposure, well, you sure as hell can't print what ain't there.

 

[tkamiya]

In the case of flipping an incident meter around the reading will be what is falling on the dome. If the light is coming from the side it will read the same or nearly the same. If from behind or in front, they will of course be different.

Thank you JBrunner. That was my whole point. If I flip the incident meter, and assuming the typical arrangement of a light source, subject, and meter, the light falling on the incident meter will be different.

Is it wrong to think (and I tested it here) if the light source is on camera side (front lit) and meter using incident meter facing the camera, then flip it, in former case, it will (of course) read correctly, and on the latter case, it will read lower based on reflectivity of the subject?

As I understand it, the incident light meter measures light falling on the dome. It has no capability to distinguish light from source or light reflected from the subject. Light is light.

 

[Thomas Bertilsson]

If you make an incident reading you point the meter toward the camera. You will be measuring the light that the camera sees, and that's what you want. It doesn't matter what angle the light is coming from, because it is in fact the light that the camera sees, at the very same angle. No compensation required.
But, you have to be very exact about WHERE you take the incident reading. If it's a big and vast landscape, it's easy. If you're indoors or in partial shade, where the lighting is mixed, with possibly very dark and very bright illumination in direct adjacency, you have to be very careful.
For example, if you are photographing a person's face, and take an incident reading, part of the face can be in shade. You have to decide how you want those shaded tones to show - do you want to expose normally and just let the camera record it as it is, potentially with shaded parts not having tone, or is too deep on the toe of the film curve? Or do you want to overexpose by shading the meter and just rely on the ambient light coming from the direction of where the camera is? Or something in between?

I've always relied on incident metering, and I never get unprintable negatives; in fact I almost always get great negatives that print very easily, unless I just guess exposures, in which case I have myself to blame. It takes a bit of practice to get it right.


I don't know too much about reflective metering, honestly. I've never had to use it.