Incident metering for landscape?

[goob4114]

I have been reading about the different methods of metering a scene. What I can't figure out is what possible purpose could an incident meter reading give you on a landscape that is 5 miles away? I mean, I don't even know how it would work. If I'm taking a picture of a scene - and everything in that scene is far away, wouldn't an incident reading be useless? What difference does it make how the lighting conditions are where my tripod is?! I'm going to be shooting both Ektar 100 and Velvia 50 on a trip coming up, and I need a light meter. I do not plan on taking any shots other than landscapes (I'll have my DSLR for everything else), and I can get a spot meter a lot cheaper than an incident meter that also does spot metering. My plan was to just use the spot meter on the highlights and/or shadows (depending on whether I'm using the negative or slide film) and exposing accordingly. I don't understand what good an incident reading would do.

 

[TCampbell]

Just think of how physically fit you'll be as you hike out 5 miles to take an incident reading! ;-)

Being a bit more serious... no need to hike 5 miles to get an incident reading. There are two reasons for this...

#1... the same sunlight that is landing on that distant scene is also landing right next to you! The only exception might be if your distant scene is in the sun and you're currently standing in shade or under a cloud. Apart from that... the sun is EXTREMELY consistent. No need to go for the hike.

#2... high-end "incident" meters (I have a Sekonic L-758DR) have a reflected light meter as well as an incident meter. My meter has a 1º spot reflected meter (significantly tighter than a camera). This allows me to "spot" the brightest and darkest areas in the landscape. You pick several possible targets and read each of them and the meter will take your targets and find the middle exposure and recommend that. You can also calibrate the L-758 to your camera's specific dynamic range (this requires a special target which you either buy or rent and it works with software that comes with the meter.) Once you calibrate your meter, it knows exactly how much your camera can handle and will warn you if a scene is beyond the dynamic range capabilities of your camera (at which point you reach for your ND Grads or shoot HDR.)

If you're shooting environmental portraiture, then you use the "spot" meter on the background landscape but use the "incident" meter on your foreground subject/model and again... the meter finds the exposure that works for both. Usually this will involve flash since you might not want to shoot the model in "full" sun but the distant background landscape is in full sun. But the meter has flash-contribution metering built in and can tell you how powerful your flash is relative to the ambient light (typically sunlight) so you can adjust flash power accordingly so that your flash is only slightly weaker than the sun -- making your model appear to be in "shade" but the shade won't actually be "dark".

 

[WayneF]

Incident does need to meter the actual light at the subjects position, but this location is really not very important for full direct bright sun, which does not vary. The inverse square law is extremely important, but NOT when the only light source distance is 93,000,000 miles, at any place here on Earth. Even when on the moon, our astronauts could use the same Sunny 16 rule, because the moon is a maximum difference of only about 1/4 of 1% (assuming full direct bright sun here). But when they go to Mars, they will need to open up about one stop.

But the really important thing is that the concept of incident metering is not only about the subjects position, instead it is about metering the actual light source there. Whereas reflective metering (including all Spot metering) instead can only meter the degree of how well the subjects colors reflect that light. And subjects vary.

The white dress will reflect a lot of light, and will read high, and so the reflected meter will underexpose, simply trying to keep it around middle.

The black dress will reflect little light, and will read low, and so the reflected meter will overexpose, simply trying to keep it around middle.

This is just how reflective meters work, including spot meters. Both the white and black will come out middle gray (both are incorrect, without some compensation to correct them). Fortunately, many "average" subjects do contain a wide mix of color and do average about middle tone, so then we imagine our reflective meter got it right. But many exceptions of course.

But the big deal is that the incident meter turns its back to the subject, and meters the actual light (at the subjects location, but INDEPENDENT of the subjects colors). Then when we get the light right, the black dress will come out black, and the white dress will come out white. This is a big plus.

So metering at the source does take into account the inverse square law (distance of light from the subject), which is very necessary. But the mountain at 5 miles and the camera location are both the same 93,000,000 miles from the Sun. If the weather is clear, we can assume the same sun lights both.

But in any case, metering the actual incident light has the plus of ignoring the subjects colors which typically makes reflective metering wrong (in some degree). We don't have to care how well the mountains colors reflect light. The good bet is on the incident meter.

 

[goob4114]

Incident does need to meter the actual light at the subjects position, but this location is really not very important for full direct bright sun, which does not vary. The inverse square law is extremely important, but NOT when the only light source distance is 93,000,000 miles, at any place here on Earth. Even when on the moon, our astronauts could use the same Sunny 16 rule, because the moon is a maximum difference of only about 1/4 of 1% (assuming full direct bright sun here). But when they go to Mars, they will need to open up about one stop.

But the really important thing is that the concept of incident metering is not only about the subjects position, instead it is about metering the actual light source there. Whereas reflective metering (including all Spot metering) instead can only meter the degree of how well the subjects colors reflect that light. And subjects vary.

The white dress will reflect a lot of light, and will read high, and so the reflected meter will underexpose, simply trying to keep it around middle.

The black dress will reflect little light, and will read low, and so the reflected meter will overexpose, simply trying to keep it around middle.

This is just how reflective meters work, including spot meters. Both the white and black will come out middle gray (both are incorrect, without some compensation to correct them). Fortunately, many "average" subjects do contain a wide mix of color and do average about middle tone, so then we imagine our reflective meter got it right. But many exceptions of course.

But the big deal is that the incident meter turns its back to the subject, and meters the actual light (at the subjects location, but INDEPENDENT of the subjects colors). Then when we get the light right, the black dress will come out black, and the white dress will come out white. This is a big plus.

So metering at the source does take into account the inverse square law (distance of light from the subject), which is very necessary. But the mountain at 5 miles and the camera location are both the same 93,000,000 miles from the Sun. If the weather is clear, we can assume the same sun lights both.

But in any case, metering the actual incident light has the plus of ignoring the subjects colors which typically makes reflective metering wrong (in some degree). We don't have to care how well the mountains colors reflect light. The good bet is on the incident meter.

Thank you for the response!

Yes, I understand the inverse square law, but it would be irrelevant in say - this situation :

I am photographing a mountain scene at the "golden hour" from the bottom of a valley (where the sun has already set), but the mountaintops are being lit by golden sun. In that situation, an incident light meter is useless, is it not - that is, if I want to expose for those mountaintops?

All this said, I think I understand how I would go about metering a scene using a spot meter. I would use it to get the dynamic range of the scene and expose accordingly. In other words, if I'm shooting the Ektar color negative film, I would just find the deepest shadows in the scene (where I still want detail) and spot meter there. If the meter suggests a shutter speed of 1/60, then I understand that this will overexpose those shadows to the point that they are middle gray. I get that part. So, I'd just decrease the exposure by 2 or 3 stops to put the shadows back where they belong and shoot at 1/250 or 1/500 and not worry about the highlights so much since this particular film is all but impossible to overexpose (from everything I've read).

On the flip side, if I'm using the slide film, I think the dynamic range of the scene is much more important - since the Velvia film only has room for about 5 stops of range - a spot meter would be absolutely necessary to determine (please correct me if I'm wrong about this). I don't know how an incident meter could determine this. Using the spot meter, if the range is within this 5 stops, I would go with the Velvia and use the spot meter again to determine the brightest areas of the scene (opposite the method of the negative film), and then add 2 stops of exposure (since the spot meter will under-expose this time to the point of bringing the highlights down to middle gray). Being that I've used the spot meter already to determine that the dynamic range of the scene is 5 stops, My highlights will be +2 and my shadows will be -3.

A am very green here, so if I'm way off somewhere, please let me know. This is the method I feel most comfortable trying first. I know I'm going to need a spot meter. My problem is whether or not I also "need" an incident meter for what I'm trying to do - as the incident meters that have the spot meter function as well are very expensive!!

For now, I can use my DSLR for a spot meter - even though it's not "really" a spot meter. It's a partial meter so it covers about 10% of the frame vs a "real" spot meter which I think only meters 1% of the frame.

But I do understand that if I am shooting a landscape in the distance and the only light falling on that landscape is natural light from the sun and its a clear day, the incident meter will give me a reading consistent with the lighting on that distant landscape as well. I perhaps should have clarified.

I guess the modified version of the question should be :

Is there any benefit in investing in an incident meter if I am aware of the necessary adjustments in using a spot meter?

 

[WayneF]

Thank you for the response!

Yes, I understand the inverse square law, but it would be irrelevant in say - this situation :

I am photographing a mountain scene at the "golden hour" from the bottom of a valley (where the sun has already set), but the mountaintops are being lit by golden sun. In that situation, an incident light meter is useless, is it not - that is, if I want to expose for those mountaintops?

If you are at "the bottom of a valley", then it sounds like you are NOT in the same light as the golden peaks. An incident meter would meter for the exposure at the bottom of the valley.

All this said, I think I understand how I would go about metering a scene using a spot meter. I would use it to get the dynamic range of the scene and expose accordingly. In other words, if I'm shooting the Ektar color negative film, I would just find the deepest shadows in the scene (where I still want detail) and spot meter there. If the meter suggests a shutter speed of 1/60, then I understand that this will overexpose those shadows to the point that they are middle gray. I get that part. So, I'd just decrease the exposure by 2 or 3 stops to put the shadows back where they belong and shoot at 1/250 or 1/500 and not worry about the highlights so much since this particular film is all but impossible to overexpose (from everything I've read).

Sort of a hard question. Measuring dynamic range would imply using a spot meter to meter the dark shadows, and then the bright highlights, and doing some calculations and some thought to decide how you want it to come out (using the one single exposure that we can implement).

Yes, it is very important to realize that a spot meter does NOT automatically give any correct exposure. Spot is never about "correct". Spot will simply try to make the selected spot come out middle gray, whatever spot you select. Beginners never realize that some skill is necessary to use Spot metering.

The real advantage of spot is simply to ignore the surroundings. Like a face surrounded by shadows, we want to meter on the face. But we probably don't want that face middle gray, so we have to know to add maybe +1 EV compensation. Experience will be very helpful with Spot metering.

On the flip side, if I'm using the slide film, I think the dynamic range of the scene is much more important - since the Velvia film only has room for about 5 stops of range - a spot meter would be absolutely necessary to determine (please correct me if I'm wrong about this). I don't know how an incident meter could determine this. Using the spot meter, if the range is within this 5 stops, I would go with the Velvia and use the spot meter again to determine the brightest areas of the scene (opposite the method of the negative film), and then add 2 stops of exposure (since the spot meter will under-expose this time to the point of bringing the highlights down to middle gray). Being that I've used the spot meter already to determine that the dynamic range of the scene is 5 stops, My highlights will be +2 and my shadows will be -3.

Sorry, that is not my call, I don't really want to go there, into that special case. Not the numbers, because Sunsets are more difficult than the normal exposure, and depends very much on how we want to make them come out. I think there are no set rules, or standard choices. It requires some thinking, and an opinion, and some trial tries..

In contrast, the incident meter in more normal cases, esp including studio flash photography, makes it all be very easy and precise, if all we want is a good normal proper automatic and correct exposure of a normal case.

A am very green here, so if I'm way off somewhere, please let me know. This is the method I feel most comfortable trying first. I know I'm going to need a spot meter. My problem is whether or not I also "need" an incident meter for what I'm trying to do - as the incident meters that have the spot meter function as well are very expensive!!

For now, I can use my DSLR for a spot meter - even though it's not "really" a spot meter. It's a partial meter so it covers about 10% of the frame vs a "real" spot meter which I think only meters 1% of the frame.

This may depend on the camera. My Nikon D800 says its spot metering is 4mm diameter, or 1.5% of the frame. That is 1/6 of the frame height however, but it is 1.5%. A D7200 says 3.5mm diameter or 2.5% of the smaller frame.

But I do understand that if I am shooting a landscape in the distance and the only light falling on that landscape is natural light from the sun and its a clear day, the incident meter will give me a reading consistent with the lighting on that distant landscape as well. I perhaps should have clarified.

I guess the modified version of the question should be :

Is there any benefit in investing in an incident meter if I am aware of the necessary adjustments in using a spot meter?

The reflective meter, including the spot meter, are all dependent on the way the subjects colors reflect light. We cannot simply meter anything, we have to consider the subjects reflectivity, and how we have to interpret/compensate that reading to get a "correct" exposure. Many average mixed scenes might often come out OK, but very many exceptions certainly do not. For example, if we walk up to a white scene, or with a white background wall, we already KNOW our camera meter is going to underexpose it. Experience gives us a good idea of how much. So we simply correct for it. I think your golden peaks are also not "average".

Incident meters can remove that problem, so I say yes, an overwhelming advantage regarding getting a correct metering automatically. We can ignore the subjects reflectivity, not a factor. We do have to meter at the subjects location, but we get a correct metering of the actual light, suitable for exposure as is. Your sunset might modify that, you likely do not want a "normal" exposure.

I think we would value the incident meter for that reason. But if you're going to piddle with it and set it by trial and error anyway, then the incident meter probably won't help.

However, real world, I have many years of learning how reflective meters work, and so I really never consider anything but the cameras meter outdoors, or any normal snapshots. I have learned to deal with reflected meters. Metering in the camera is less awkward to do, and is convenient and fast, if we know how to deal with it. You speak of film, so proper metering is obviously more important. Whereas digital lets us see the result immediately, so we can always fix it.

But the camera won't meter flash, or other difficult cases are difficult. I consider the incident meter absolutely mandatory for studio flash, where I have to get four lights all adjusted relative to each other (for ratio, etc), and probably repeating the same setup as I used them last time. I simply just set them to meter what they need to do. There is no other way to do it (repeatedly).

Here is an article about that: Why would I need a handheld light meter?

 

[unpopular]

Well. First of all, "real" spot meters have a fixed focal length, so their spot is measured in degrees, not percent. This is where you're getting your number from. Most handheld spot meters are a 1 degree spot. So starting from the meter going out to the subject, you can imagine an a cone with an apex of 1 degree, so the further out you go the the point you're metering, the larger the base's area - the larger the spot.

Camera spot meters are a little different. Because the focal length is variable the geometry changes too. So a 1 degree spot might be 10 percent of the frame using one lens, but a significantly larger spot on another. This is really not that significant in practice.

To use spot metering effectively you're going to want to know your camera pretty well. I would suggest the following test: take a white piece of paper and light it as flat as possible. Set the aperture to something reasonable, f/8 or so and make note of the meter reading at null (+-ev).

Make sure your under/over blinkies are on, and find the point where the sheet of paper clips out. Set the exposure to that point, such that if you increase exposure by 1/3ev, it will no longer be clipped. Confirm minimum exposure in your raw processor. You should have NO useable data in at least some parts of the file (there will likely be some areas that do have exposure, but what you're looking for is any part of the image being clipped).

Now, step up the exposure by 1/3ev all the way until the image starts clipping again. Process the set in your raw processor using asome middle-of-the-road settings.

Now, take a look at the entire set of images. Make note on the lower end where noise starts to become a problem. The point where you can say you can tolerate the noise level for your shadows will be your effective minimum exposure. Now on the upper-end. Take note where things clip out. Pay attention to your histogram or clipping warning. Now, go subtract down 1/3 -2/3ev, depending on how much headroom you are comfortable with. This is your effective maximum exposure.

Now, make note of the distance in value from what the meter was reading at. So if, say, your meter was saying 1/250 and your minimum effective exposure was 1/1000, that means that your shadows can be safely placed at -4ev before getting too noisy.Same with your hilights. If it's that test exposure was at 1/50 with 2/3ev headroom, then your effective maximum exposure will be at +2 1/3ev (if I did my math right... but you should get the idea).

Now, when you go out, find the brightest region and meter from that, you can now increase the exposure to your maximum exposure and nothing will be over-exposed (provided that you metered at the right place). If you're worried about it you can meter the shadows as well to ensure that they will fall within minimum exposure, but be sure that you account for the increase in exposure. In practice, they almost always will be.

My experience has been that when exposing for the hilights you end up with too much shadows and the image will appear over-exposed and washed out. But provided that the scene is not exceptionally flat (such as taken in full shadow) this isn't a problem so long as there are no over-exposed areas (in particular over exposed areas of interest). Just develop out the shadows in raw processing using a curve or gamma/level adjustment.

On one note, obviously don't meter extremely, extremely bright regions without information ... like the sun. In fact. Don't meter the sun, ever.

 

[AlanKlein]

I shoot landscapes with Velvia 50 and always bracket, usually +1 and -1 stops although 1/2 stops might be better, it's just easier with my RB67 MF camera. It's cheap insurance against mis-reading my metering conditions. I would suggest bracketing to start even if you don't want to do that long range. That way you can get to practice and actually see what happens in different conditions. Write down the settings and conditions so you can compare later. Good luck.

 

[goob4114]

Well. First of all, "real" spot meters have a fixed focal length, so their spot is measured in degrees, not percent. This is where you're getting your number from. Most handheld spot meters are a 1 degree spot. So starting from the meter going out to the subject, you can imagine an a cone with an apex of 1 degree, so the further out you go the the point you're metering, the larger the base's area - the larger the spot.

Camera spot meters are a little different. Because the focal length is variable the geometry changes too. So a 1 degree spot might be 10 percent of the frame using one lens, but a significantly larger spot on another. This is really not that significant in practice.

To use spot metering effectively you're going to want to know your camera pretty well. I would suggest the following test: take a white piece of paper and light it as flat as possible. Set the aperture to something reasonable, f/8 or so and make note of the meter reading at null (+-ev).

Make sure your under/over blinkies are on, and find the point where the sheet of paper clips out. Set the exposure to that point, such that if you increase exposure by 1/3ev, it will no longer be clipped. Confirm minimum exposure in your raw processor. You should have NO useable data in at least some parts of the file (there will likely be some areas that do have exposure, but what you're looking for is any part of the image being clipped).

Now, step up the exposure by 1/3ev all the way until the image starts clipping again. Process the set in your raw processor using asome middle-of-the-road settings.

Now, take a look at the entire set of images. Make note on the lower end where noise starts to become a problem. The point where you can say you can tolerate the noise level for your shadows will be your effective minimum exposure. Now on the upper-end. Take note where things clip out. Pay attention to your histogram or clipping warning. Now, go subtract down 1/3 -2/3ev, depending on how much headroom you are comfortable with. This is your effective maximum exposure.

Now, make note of the distance in value from what the meter was reading at. So if, say, your meter was saying 1/250 and your minimum effective exposure was 1/1000, that means that your shadows can be safely placed at -4ev before getting too noisy.Same with your hilights. If it's that test exposure was at 1/50 with 2/3ev headroom, then your effective maximum exposure will be at +2 1/3ev (if I did my math right... but you should get the idea).

Now, when you go out, find the brightest region and meter from that, you can now increase the exposure to your maximum exposure and nothing will be over-exposed (provided that you metered at the right place). If you're worried about it you can meter the shadows as well to ensure that they will fall within minimum exposure, but be sure that you account for the increase in exposure. In practice, they almost always will be.

My experience has been that when exposing for the hilights you end up with too much shadows and the image will appear over-exposed and washed out. But provided that the scene is not exceptionally flat (such as taken in full shadow) this isn't a problem so long as there are no over-exposed areas (in particular over exposed areas of interest). Just develop out the shadows in raw processing using a curve or gamma/level adjustment.

On one note, obviously don't meter extremely, extremely bright regions without information ... like the sun. In fact. Don't meter the sun, ever.

Thanks! Unfortunately, most of your response was geared toward digital, and I am speaking specifically about film here. But, your comment certainly is not wasted. I think it's a great idea to do this on my DSLR to learn the range of my camera. As for the spot metering, what you said makes a lot of sense (about a true spot meter being measured in degrees). That is indeed what I meant. The "partial" meter on my camera claims to cover the middle 10% of the frame - basically the middle focus point and immediate surrounding area. It didn't dawn on me that 10% on a 250mm lens is a much narrower FoV than on an 18mm lens. This leads to another question. If I were to throw my telephoto lens on there and zoom in to an area I'd like to meter for what will ultimately be a much wider shot, will that meter reading translate well onto the ultimate wide shot? If so, that could be the metering technique I use until I invest in a good dedicated light meter.

 

[goob4114]

I shoot landscapes with Velvia 50 and always bracket, usually +1 and -1 stops although 1/2 stops might be better, it's just easier with my RB67 MF camera. It's cheap insurance against mis-reading my metering conditions. I would suggest bracketing to start even if you don't want to do that long range. That way you can get to practice and actually see what happens in different conditions. Write down the settings and conditions so you can compare later. Good luck.

Thanks! I had planned on bracketing the Velvia shots. I'm also bringing some Ektar 100 along, but I wasn't planning on bracketing those - since the film apparently has a very high tolerance to overexposure. I'll just make sure not to underexpose the shadows and let the highlights fall where they may. I am brand new to film, and I've had a Bronica ETRS for a few months. Unfortunately, I only have the one back for it, so I won't be able to switch between the Velvia and Ektar, and it is only capable of 1-stop increments in exposure, so the brackets of +1 and -1 full stop will have to suffice.

Do you use a spot meter for your landscapes?

 

[unpopular]

Typically with slide film you meter the hilights, as I recommend for digtal. However because your dynamic range is more limited and bcause rolloff more forgiving youll want to take better care to meter the shadows to ensure proper placement. You'll only get a couple stops off middle grey at most before things plug up, so in somd cases itll make moe sense to let yor high end blow.

 

[unpopular]

Unless we're talking b/w. That's a whole different kettle of fish.

 

[gsgary]

I shoot landscapes with Velvia 50 and always bracket, usually +1 and -1 stops although 1/2 stops might be better, it's just easier with my RB67 MF camera. It's cheap insurance against mis-reading my metering conditions. I would suggest bracketing to start even if you don't want to do that long range. That way you can get to practice and actually see what happens in different conditions. Write down the settings and conditions so you can compare later. Good luck.

Thanks! I had planned on bracketing the Velvia shots. I'm also bringing some Ektar 100 along, but I wasn't planning on bracketing those - since the film apparently has a very high tolerance to overexposure. I'll just make sure not to underexpose the shadows and let the highlights fall where they may. I am brand new to film, and I've had a Bronica ETRS for a few months. Unfortunately, I only have the one back for it, so I won't be able to switch between the Velvia and Ektar, and it is only capable of 1-stop increments in exposure, so the brackets of +1 and -1 full stop will have to suffice.

Do you use a spot meter for your landscapes?

Why don't you bracket the shots manually if the lens has increments in between f stops or by shutter speed

 

[AlanKlein]

I shoot landscapes with Velvia 50 and always bracket, usually +1 and -1 stops although 1/2 stops might be better, it's just easier with my RB67 MF camera. It's cheap insurance against mis-reading my metering conditions. I would suggest bracketing to start even if you don't want to do that long range. That way you can get to practice and actually see what happens in different conditions. Write down the settings and conditions so you can compare later. Good luck.

Thanks! I had planned on bracketing the Velvia shots. I'm also bringing some Ektar 100 along, but I wasn't planning on bracketing those - since the film apparently has a very high tolerance to overexposure. I'll just make sure not to underexpose the shadows and let the highlights fall where they may. I am brand new to film, and I've had a Bronica ETRS for a few months. Unfortunately, I only have the one back for it, so I won't be able to switch between the Velvia and Ektar, and it is only capable of 1-stop increments in exposure, so the brackets of +1 and -1 full stop will have to suffice.

Do you use a spot meter for your landscapes?

Why don't you bracket the shots manually if the lens has increments in between f stops or by shutter speed

I use to bracket 4 shots at 1/2 stop increments for a total of 5 shots for every scene. But it became too much. I settled on one stop and 2 bracket shots. I only change my shutter speed one stop. There are no half stops with the shutter. This way I can also leave the aperture the same for consistent DOF. Of course, I'd be better off with half stops especially with Velvia and other slide film as positive film changes more rapidly with exposure changes than does negative film.

 

[dennybeall]

These new cameras with their various metering modes pretty much make a meter unnecessary. When using an incident meter "in the Old Days" I would just hold up my hand and meter the hand when I couldn't get to my subject. Worked like a charm.

 

[AlanKlein]

My manual film camera doesn't have a meter.