Manual for the warrior of available light

As promised a couple of times earlier, today I want to talk about one of my favorite types of shots: available light portraits. Using natural light you can get stunning portraits and today I want to give you a step-by-step guide.

As usual many routes lead to Rome and I want to give you one that can serve as a starting point for experimenting. Also, it is important to keep in mind that different people have different opinions. From a learning perspective though it is typically more effective to first explore one route and then compare to other routes, adapt, and modify. Here Bruce Lee’s principle is a powerful guide: Adapt what is useful, reject what is useless, and add what is specifically your own.

Before we go into the actual process, let’s quickly go over a few technical things that will make our lives much easier. Once out of the way, we can focus on the images.

Technical requirements.

1. Shoot in RAW format. This will allow for easy white balance adjustments in post processing.
2. Put your camera into manual mode (M). Then you can control ISO, shutter speed, and aperture separately. Set your ISO to 100 (or whatever your lowest ISO is). I assume basic familiarity with this mode.
3. Put your camera into spot metering, typically indicated by a small dot in the metering menu. Not the one with the brackets around it – that’s center-weighted average.

We are good to go now. So let’s talk about the actual process – I assume that you have a model…

The process.

Finding a location that works. Although it sounds obvious it is actually a bit more subtle. There are many things to keep in mind when choosing a location in view of the harshness of light etc. I will go with one of the most basic, yet one of the most powerful setups: open shade. For this find a wall or a background that is in the shade. This will give some very soft light. Typically walls that face north work well: this means the wall is in the shade and is itself casting a long shadow.

Direct your model. Giving directions is key to get the result you want. If your model is not super-experienced he or she has no idea how things look. In fact, often what feels great looks weird and vice versa. You are the one looking through the viewfinder – you have to say what needs to be changed. I typically walk around my model for some time and direct without looking through the viewfinder. I carefully check the light and for things that do not belong in the frame. Only when I have the pose that I want, I take my camera and take a few shots – I tend to be a slow shooter and take only very few shoots.

Composition. This brings us to the composition aspect. Basically, anything goes. There are so many rules but in the end, depending on the situation many things can look great. So do not go by the rules but rather take your time compose and recompose until you have something you like. As a starting point, the rule-of-thirds often works. Visually think of your frame as being cut into thirds in horizontal and vertical direction and align your model (or your model’s eyes) on one of those lines.

Metering. This part is every important. We are spot metering and what we really want is to expose for the face, i.e., we want to make sure that the light in the face is right. The spot meter in your camera meters what is called “18% grey”. That’s about ‘middle-dark’ grey – not so important. Important is that skin color is not 18% grey but lighter or darker depending on skin type and you need to correct for this. That’s what the exposure compensation is for: exposure is measured in stops (i will go into these details elsewhere) and you dial in a positive or negative compensation depending on the skin lightness. Rules of thumb:

1. very light skin: +1.5 stops
2. light skin to normal skin: +1.0 stop
3. milky dark skin: -1.0 stop
4. very dark skin: -1.5 stops

So point the metering patch in the viewfinder (typically the thing right in the middle) at your model’s face and since you are in manual your camera will show on a scale from say -3 to 3 where your exposure lies. Make sure that the reading of the face is roughly consistent with the number from above – you can adjust this reading by changing the aperture and the shutter speed (I assume that you are familiar with this). Then take a shot. Check your exposure of the skin on the screen. If it is too dark, increase exposure compensation, if it is too light decrease. For example the picture below was more on the challenging side: You can see that the skin on the forearms is lighter than the face (that’s due to her being close to the coke machine with her face). Thus, here, we cannot get arms and face exposed perfectly at the same time (without changing the setup). Always expose for the face first (of course there are exceptions to any rule). I added about 1 stop here.

That’s basically the whole process – no secrets. Sure it takes some time to get the composition right etc. but the overall technical aspect is very simple.

Observe the characteristic of light. If you get the exposure right, there is a minimal amount of postprocessing to be done, giving an overall more natural look. Related to the exposure metering discussed above, you also want to look out for the overall characteristic of light on the face. A lot here is subjective and most comes with experience. Let me just make you aware of a few things to keep in mind.

1. Make sure that the face is evenly light. Avoid hot spots in the face. For example in the shot above we are very close to hot-spotting on the left cheek, right under the eye – granted it was a hot day and close to the end of the shooting where skin moisture accumulates that reflects more. A bit of hot-spotting is ok, because it adds 3D structure to your image, but you need to keep it under control, otherwise you get way too much contrast to handle or you even blow out the highlights. You can see in the shot below that the face is significantly more evenly lit – still having enough structure.
2. Try to get a bit of light into the eyes to light up the iris. There are several ways of doing it. Sometimes it suffices if your model slightly lifts up her or his head so that the eyes are not shielded from the light by the bones of the eye sockets. This is harder if your model has deep eye sockets. You can also help here a bit using a reflector (or even a flash – but that would not be natural light…). If you are only a bit short on light, you can also slightly increase the exposure on the eyes in post-processing. Attention: this will not work well, when you are shooting film: there is often not enough detail in the shadows… A good rule is: get it right from the start and avoid post processing (in the sense of fixing) whenever possible.
3. Make-up can support the charateristic of light, by emphasizing facial structure. A bit of foundation and supporting make-up on the cheek bones can make all the difference. In principle you can do some of the things in post-processing but it is a lot of work in order to get it right and it still does somehow look unnatural.

The shot below is a good example of the above. You can see that the skin is evenly lit and cheek bones are emphasized for structure. The iris is slightly light up to add character.

In a nutshell, that’s the whole process. Now let’s talk about some tweaks that I ignored above. They are not so mission-critical, however will still influence the main result.

Tweaks.

Focal length. Focal length is always on the tricky side. For everything protrait-ish I tend to shoot 90mm-150mm (equivalents). I say equivalents because I do regularly shoot crop size sensors – I am strong believer in the m43 systems especially at base iso: good enough. If you are shooting an APS-C sensor this means you have to multiply your lens’ focal length by about 1.5-1.6 (depending on make) and by 2 for m43.

Longer lenses (those with a larger focal length) compress perspective. This is a good thing for portraits as it makes faces look more pleasing, allow more easily to blur the background, and you do not need to stick the camera into your model’s face. On the other hand, I also love to shoot shorter focal length, 28mm in particularly. They work great for environmental portraits and larger prints – but not so much for close(r)-ups.

Aperture control. If you want to get a nice background blur you will need to shoot a large aperture (i.e., the lens opening is larger. Bigger is not always better though. When increasing the aperture (smaller aperture number) you get a shallower area that is in focus and when your model moves, her or she can easily move out of the area of focus. In the view of the above, as rules of thumb, for around 50mm – 90mm equivalent I would probably shoot around f/2.8 give or take. For longer focal length say 135mm or 150mm I would shoot around f/4. This typically gives plenty speed (keeping your shutter speed high), enough background blur for pleasing results, and at the same time enough depth of field allowing minor model movements without leading to problems.

White balance. Today’s cameras tend to do a good job at guessing the white balance. However I regularly find myself required to fine-tune white balance to communicate a mood. That’s why we want to shoot RAW. White balance is not something that is contained in the data – it is added to it by the camera. If you shoot jpeg it is hard coded into the jpeg file and changing it later can (and typically will) lead to a degradation in quality. in RAW you can change the white balance without any loss in quality. You can also recover some of the highlights and shadow detail from RAW that you cannot from jpegs – that’s a different story for some other time though. So to get a good white balance you need something neutral in the picture. Typically something grey will do, if the grey is somewhat neutral-ish (that’s exactly why people use grey cards). Put the grey thing where you want to shoot your model and take a shot of the grey thing. Then normally work with the model. Later in your favorite raw-converter (I prefer Lightroom) read of the white balance from the first grey shot and copy it to all the other images from that particular spot. This should give you an accurate white balance. You can also deliberately warm up or cool down the shots to communicate a certain mood – experiment with it. For example the very first shot is on the coolish side, the second on the warm side, and the last is somewhat netural. If you do not have a grey object, you can also often read of the white balance from the white area in eyes, provided that they are not super close to channel clipping – you will see what I mean once you try and run into it…

Shooting film. Shooting film changes the game a bit. For example, when shooting digital you expose for the highlights, i.e., you try to prevent highlight clipping vs. when you shoot analog you will expose for the shadows, because the process is a negative-process (provided you shoot C41 or black and white and not slides). Also, you will need to be extra considerate because you do not have any retakes and you cannot check the LCD. Try shooting film along with digital to get some feeling for the metering and some extra confidence. Also, for most films, you might want to deliberately overexpose them one to two stops, i.e., rate them lower to get a smoother rendition. For example I shot Portra 160 at ISO 75 and Fuji 400H at ISO 100.

I hope you enjoyed that post. Go out and shoot and share your results – TSJ.