fbpx

Ah Histograms, the photographers faithful friend. I will admit, I didn’t really feel they were that important but things change and now I would recommend eyeballing them now and again to every photographer.

So the video version of this is probably going to be easier to follow, so we’ll start there:

Let’s cover things off in brief:

What is a histogram?

A histogram is literally just a graph. A chart. A visual display of data. It’s like an intricate bar chart but the data is stacked based on it’s volume against a specific value. In our case, the data is pixels and they’re stacked based on their brightness from black to white.

What is a histogram used for in photography?

In photography, we usually use the histogram to read general exposure in an image.

The higher the dynamic range your camera body has, the wider your histograms range will be. If your camera has a high dynamic range, it means that you have more steps before black becomes white than someone with a lower dynamic range.

The difference between your black and white point is measured in stops. A stop is a measurement of exposure. You can find more about stops of light here. The amount of stops gives you your dynamic range. A Sony A7Riii has 15 stops whereas a Nikon D500 has 8.2 stops. The difference there is really quite enormous – you need to be more careful with a Nikon D500 than you do with a Sony when it comes to exposure in camera.

As the data is stacked up there in the graph, mountains and valleys will be formed – the peaks and troughs. These represent the majority and minority of pixels. If your peak is to the left, it means most of the data is in the shadows. The image could be underexposed, but it also may just be a dark scene. If your peak is to the right, it means that most of the data is in the highlights. The image could be overexposed, but it may also just be a super bright scene.

If the mountain is pressed up against the left side, you have black clipping. This means that the pressed part is completely black. Usually, this isn’t ideal.

If the mountain is pressed up against the right-hand side, you have highlight clipping. This means that the pressed part is completely white. Usually, this isn’t ideal.

If someone asks, “why do you shoot to the left”, this just means, “why do you underexpose your images”. The same is true in reverse.

If a tutor tells you to shoot to the right, it means that they want you to slightly overexpose the scene. It doesn’t mean blow your highlights and have the mountain pressed up against the right.

What should you actually care about?

As you can see from the above explanation, the lean of the mountain is only an issue in 2021 and beyond if there is something touching the sides. Any data pressing up against the side means there is data clipping there and usually, we do not want this. There are some exceptions, however, so times where clipping is desired include:

  • Black background portraits – you would want black clipping (data pressed to the left) here
  • White background portraits – you would want white clipping (data pressed to the right) here

In my personal opinion, keep an eye on your Histogram to assess two things:

  1. Is there any data pressed up against the sides?
  2. If you’re aiming for a specific look and feel, can you alter your exposure to shift more left or right?

Examples of images and their Histograms:

I’ll split this part into two sections using screengrabs from the EVF or Lightroom to illustrate what the histogram is saying at the time:

In-camera examples

This histogram has its mountain in the shadows portion of the histogram, to the left of the centre. You can tell that this image is underexposed (legitimately) by looking at the end of the trough. Where the data tails off on the right-hand side, it’s barely at the centre of the graph, right? This means there’s a huge expanse of the dynamic range that has zero data in it. This whole image could be exposed brighter by a couple of stops to get a better straight out of camera (SOOC) image.
This image is exposed much better in-camera. You can see the mountain is more of a bell curve with data in all of the histogram sections but none is pressed up against either side. This means we have a great SOOC file to work on, and no data is clipped.

In editing examples

When you get back into editing though, things get a little bit more wooly. The histogram points mentioned above are still true, data pressed against the sides is usually bad, and if there’s a huge gap between the last little tail of the data and the side, we need to shift the exposure one way or the other, but you can have a heavily leaning mountain depending on the “vibe” of the final image. I’ll explain:

So you see, the histogram is a useful tool to assess the literal situation in the scene, and also afterwards at editing. As long as nothing is pressing up against the sides, you’ll be fine.

Helpful? Please share and say so ❤️