There are a lot of articles about right exposure, various tricks to improve the picture quality, like export to the right, exposure to the left, as wall as a technical features to protect photographers from errors, like highlight correction and shadow correction.
Let's start from exposure. I hope you know enough about Aperture, Shutter speed and ISO, so we will not waist time by describing trivial things. However, we still need to define what is our purpose. Let's say we want to have well-exposed photography. But what is it?
Seems we can divide all photography by two main arias: photography of the scene (when we don't have the main subject or object), something like landscape and photography of subject, or object (the photography when we want to depict the person or object properly). Some people can argue that even in landscape photography we usually have the main object. This is true. If we have the object it means that the photo with this particular object might be represented in the second group. However, pretty often we merely want to represent the nice scene that we see and in this case - we need to move it to the first group. But why we need to divide our photos to this two groups? - because we will use two different approach to set up exposure for these groups.
The photography of the scene suppose we don't have a main object in the scene and in this situation the more broad dynamic range our photography has the better. How we can achieve the most broad dynamic range? - The easiest way to achieve it is to use Multi-zone (or Matrix) metering. Yes, you can use spot mitering, check the light in darkest and brightest sides of the scene and try to calculate the exposure, but Multi-zone metering will do it much faster and precise. Spot metering still good for experiments and for learning purpose, but not for real life.
The photography of the subject is entirely opposite case! If we do the photo of the person we mast have the right exposure. Here we can use Spot metering mode to set up the right exposure. For example you need to take a photo of the person on the strait sunlight (not the best case, but why not), you can set the metering point to the lighted part of the face (not the brightest, but lighted). In this case you'll get the picture where the sun-lighted skin properly exposed the shadow side in the shadow and the reflected light very bright but usually not overexposed (of course it depends on the light that you have, but most of modern cameras have enough dynamic range to to it properly in the daylight). What will happen with brightest part of the picture (not with the person's face)? Probably it will be overexposed. Is it good or bad? - Usually it doesn't matter, because you have the subject (the person) and it is exposed properly. If you decide the image should be brighter or darker you can do it later (whe you shoot to the RAW you usually have expose reserve in one or two steps, so usually it isn't a problem.
It is also possible to use Center -weighted metering mode for photography of the subject but it is wary from camera to camera, the algorithms of this mode and center zones are different between cameras and manufactures so it is pretty impossible to prognoses the result and no sense to use this mode.
Sometimes we have some special view of course and can use exposure compensation to correct exposure that we have from lightmeter. In any case to do the test shot and check the result visually and by histogram is always a good idea.
However, when we use a metering mode (Multi-zone or Spot metering - doesn't meter) in automated shooting modes (Aperture priority, Shutterspeed priority, ISO priority, and Program mode) we usualy have a problem: camera automatically change the exposure when we re-frame the image. For example we use Spot metering and check the light on the skin, then we move a camera to have a good composition and now the central point measure the exposure in the dark area behind the person, camera automatically change the exposure and we will got wrong exposure of the person. The same situation with Multizone metering mode, if some light or dark object get to the frame, camera will change the exposure. We have three options to resolve this problem:
- Use exposure lock (most modern cameras has it)
- Use exposure compensation
- Use manual mode
What is advantages and disadvantages of these methods?
For exposure lock method the main advantages is we can take right exposure, recompose the frame and have good composition with the properly exposed objects. The disadvantage of this method is you need to remember that you have to press lock button every time after you get exposure, if you forget about it - you get randomly exposed shot (exposed as your camera decided). Of course to keep it in mind during all session isn't easy, you might be distracted, think about composition or communicate with model. Using this method you have a high risk of have a wrong exposed shots.
Exposure compensation not too good when you need to correct the light for one shot. Here is a question how you should correct it (the usual answer is you'll get some experience) but the idea is you do it manually and you need evaluate the scene manually to understand do you need to move it to half, one, or more steps. Yes, it is possible, you can do it base on your intuition, or by memorizing parameters of the proper exposed subject (if we use Spot metering for example), and set up the compensation to have the same parameters (shutter speed, aperture, ISO) after you recompose the frame. Again, it is possible, but it will take a lot of affords, seems that much more than using exposure lock, so hard to recommend exposure compensation for this purpose. However exposure compensation seems might be extremely useful when you need to change the tone for all session, for example when you need to have all series over or underexposed.
The third method is to use Manual camera mode. This method is a little bit tricky, but unexpectedly looks as a most efficient and fast. If you set up your camera in manual mode with no auto ISO, your camera can't change the exposure, but it still can measure it. You can use Multizone metering or Spot metering to set up the exposure, and do one or two test shots to check the histogram and correct the exposure, if necessary. After that you have you exposure, all done. You can change the frame as you want and as you need, exposure will be the same. You don't need to change any settings between shots if the light conditions are the same, it looks perfect! But the main disadvantage of this method - if the light conditions change, your camera will not change the exposure. Usually it leads to using Manual mode in studia, because it is obvious, the main question is: is it possible to use advantages of this method in other situations? And the main question is what approach: Exposure Lock, Exposure Compensation or Manual camera mode to choose to eliminate the problem in real life?
Lets consider the Manual mode a little bit more carefully. If the light condition is the same it has the huge advantage in compare with auto modes: you set it up and can do the series (five, may be ten or more) perfectly exposed shots without any changes. If you do the wedding - doesn't matter what is on the background or what color suit have the person you currently shoot. If you do the street photography - the tones or the colors - doesn't matter if you have the same sun.
But light outside is always changed, it's never the same! - Are you sure?
Let's say we have a sunny day. If you have no clouds in the sky - the light the same. Let's consider the whole sky in the clouds - actually the light pretty same, it might vary a little bit, but if you'll see that it's become darker or lighter you simply to meter your subject again and change your exposure. This is might be a good idea to check the exposure and your results from time to time.
Let's think about cloudy weather when we have shades and sunlight and there changes fastly. In this situation it might happen that you measure exposure in sun-lighted moment and after that cloud come and now our object in the shade, and after that again in the sunlight. What should we do in this situation?
Let's consider the Ansel Adams's Zone System . We can see that the average Caucasian (European) skin tone placed in zone 6 we also can see the light stones in this zone. We can see the dark stones and landscape shadows in zone 4. It means that the difference between sun-lighted and shadow subjects near two stops. For clouds it might vary a little bit, because clouds are never as solid as rocks so the difference might be near one stop. You can check it with your camera and decide what is the difference for you.
The main idea hear is two stops isn't too much especially if you shoot in RAW. So if cloud comes and your image become one stop darker - it isn't problem at all you can easily change it later in LightRoom or any other RAW converter. We remember that we are try to properly expose the skin and it is in zone 6, it means that if it becomes one or two stops darker it will be in zone 4 and you can easily to fix it in LightRoom. Even if it becomes a little bit brighter you still have enough reserve to fix it. This is a theory of course, bat this theory can give us a strong base to understanding how we can use the camera.
Of course light changes not just because of clouds, it might happen that we need to shoot in the shadow, in the fores or in the building. If it happens we should adjust the exposure of course. Let's say we do a street photography, so we go in the street and shoot randomly to sun and dark side of the street. Of course we can use a matrix mode to find the average exposure and after that shoot all session with one exposure. It is possible, but in this case the dark side will be a little bit too dark and lighted side will be a little too bright. It happens because when we see, our eyes instantly adjust the amount of light and we see the dark scene bright and vise versa. When we will see the photo we will compare them with what we saw and it will be too dark and too bright for us. What can we do? - we can use the Matrix mitering mode and take a two exposure on sun-lighter and on the dark side of the street. Let's say we use ISO 100 and aperture 5,6 it means that we will have two numbers of shutterspeed, one for light and one for dark, let's say it will be 500 and 250. This is one stop or two click of shutterspeed adjustment wheel and simple rule dark side - to click slow, bright side - two click faster.
You can say that it might be more than one stop and this is true, it might be two or three or in contrast sunlight even 3 stops. But, we don't need to 100% compensate the light changes, it should be clear, that here we have the sunlight and here we have the shadow. Actually we see the light and shadows with our eyes, so it should be on the photo. The question is how bright and dark it should be, but this is a question of your test. We should remember that we still correct it later in LightRoom. The average correction might be near one stop.
If however we get from the sunlight to the building to the forest or to the deep shadow it might be a good idea to correct the exposure, again, it shouldn't be so big difference. Again, you can do it once and after that do all shorts in this circumstances without any additional adjustments.
Now we can compare three options that we considered: Exposure Lock, Exposure compensation and Manual camera mode. The main disadvantages of the first two methods are: if we want to have proper exposure we need to measure exposure every time before the short and use exposure lock or exposure compensation if we want to recompose our shot. If we use Exposure Lock this method allow us to have right exposure, but if we use exposure compensation we will have arbitrary results. On the other hand Manual mode allow us to do the adjustments not for each short, but for series. In the other words we do the adjustments only when we think we need to do it, and we don't need to fight with camera to overpower the algorithms. In the manual mode camera give us advice what exposure it's consider as a right expose and it is only our solution do we want to follow this advice or not.
Manual mode also seems much more fast approach (you don't need to do adjustment for each shot).
The automated mode might be faster than manual mode if we don't do any adjustments at all. But in this case we will have only some kind of picture, it will be impossible to say about right exposed subject or about any consistency from picture to picture. The exposure will vary and it will take long time to align it in the RAW editor or Photoshop. The worse thing here - your will not have any sign should you do the picture brighter or darker, the color density also will vary from picture to picture because it depends from the exposure.
Okay, we decided what is better for the right exposure of our images. What about the corrections? The Idea of Highlight correction is to protect the image from overexposure, how it is possible to do? - There no magic here, your camera do underexposures images (usually near the half of the stop) and after that push the tone curve to do it brighter. The advantage of this feature - you have half of the stop reserve for your highlights, the disadvantage - the quality of the picture is worse. This feature might be useful if you shoot in automated mode, because your camera might decided to do you image brighter - so your image will be overexposed and reserve in half of stop might save your image. So it might save several of your shots but the price of it - the quality of all your shots become worse. In manual mode we don't have arbitrary changes of exposure so we don't need the exposure corrections.
What about exposure to the right and exposure to the left?
Exposure to the left do the same thing as an highlight correction it gives us the reserve in highlight. In the past it had some sense because DSLRs had a very narrow dynamic range and it was very easy to overexpose the image. Nowadays, it isn't so important so we don't need to use it.
The idea to expose to the right is to have the better quality image. It might be useful, but if we expose to the right, we need to understand that the reserve in lights our camera give us become smaller, so we have higher risks to overexpose the image without possibility to fix the problem in the RAW editor. It might give the perfect results, but easily might brake the image. The main rule here might be to take the perfectly exposed shot first and after that if we want some experiment - to take image exposed to the right.
The only one thing we have to consider now is a Manual mode with the auto ISO. The problem that we have here is exactly the same as the problems for other auto modes - the camera choose exposure itself. So we will have the same problems with Exposure compensation or Exposure Lock as with Aperture priority mode for example. Furthermore, automatic ISO even worse than Aperture priority, because in Aperture Priority camera automatically change the shutter speed and it has the broad range in shutterspeeds. The range of ISO is much narrow and it's affect the image quality. If we are talking about ISO, it might be a good idea to consider it as a film quality. We shoot in the good daylight conditions we can use ISO 100, we have to shoot with bad light - we simply set up ISO 800 or 1600 or even 3200. Go back to the daylight - install new ISO 100 film and continue shooting.
So we see the conditions, set up the ISO, choose aperture and adjusted the shutterspeed. If we don't have enough shutterspeed - we can consider to choose higher ISO or open the diaphragm. During the session we can adjust the shutterspeed a little bit up and down to meet the current light conditions. It should be enough to have perfectly exposed and consistent images for all session.