Filters

In the very simplest of terms, a filter is a piece of glass, resin or plastic which is attached in front of your camera lens and which will change the light entering your camera to some degree. There are literally hundreds of different types of filter all of which are designed to apply a different effect. I could not hope to cover all the different types of filter here, so I’m just going to cover the three I use, and why I use them. Incidentally, when I’m taking pictures of animals I very rarely use any type of filter so this information only applies to my snapshots and landscapes.

Most DSLR lenses have a thread on them which is specifically designed to screw on a filter. Bridge Cameras often have filter threads too but compact cameras often don't.

Filter threads on a camera lens

There are basically two types of filter:

1. Those that attach directly to the thread on the lens:

2. Those which are placed in a filter holder which is, in turn, attached to the thread on the lens:

Because camera lenses are all different sizes, their thread diameters are all different too, so if you are buying a filter which screws directly to the filter threads of one lens, it may not fit the others meaning you will have to buy other filters for your other lenses. This gets expensive.

Filter holders get round this problem: you buy a filter holder and one set of filters. This clips on to an adaptor ring which is then screwed to the filter threads. You simply need to buy adaptor rings to fit each of your lenses which is much less expensive than buying separate filters.

Circular Polarising Filter (often abbreviated to CPL). These filters work like magic and apply an effect that cannot be replicated in post processing. They reduce reflections by shutting out reflected light and also help to cut through atmospheric haze and make existing colours more pronounced, especially blues. Their effect is quite profound and the animation below shows the dramatic effect this filter can have as you rotate it. Because of this I think I will always use a CPL in certain circumstances. How they work is far too complicated for this article but if you’ve got a good head for physics then there are plenty of articles on the internet that explain it, all of which were beyond my simple brain.

Graduated Neutral Density Filter (often abbreviated to GND). These filters work very simply, they are darker at the top and get increasingly more transparent until they are completely clear just passed halfway.

These filters allow you to compose pictures where, for example the sky is very bright. If the sky is bright then you will either have to expose your camera for that, which means that whilst the sky looks good the ground is in complete shadow or, if you expose for the shadows the ground looks nice whilst the sky will just become a bright white blob. By putting a GND on, this has the effect of darkening the sky which means you have a better chance of exposing both the ground and the sky correctly. These filters are commonly bought as either “Hard Edged” where the dark portion of the filter becomes transparent very quickly or “soft edged” where there is a gradual transition from the dark part of the filter to the transparent. You would use a soft edged filter when there is no distinct horizon, for example in a mountainous environment. You would use a hard edged where you can see the horizon such as in a seascape or when using a telephoto lens where a hard edge, when zoomed right in, becomes a soft edge.

There are also variants of this type of filter, for example a Reverse ND Graduated filter which is darker at the horizon line and gets lighter (more transparent) towards the top and these are designed for sunset and sunrise pictures where the horizon is brighter than either the sky or the ground.

Now you can get away without using one of these filters and a common way of doing this is to take a few pictures of your scene at different exposures so that some have the ground properly exposed, some have the sky properly exposed and you just “blend” them in your post processing software to give you a an image where everything is exposed. Taking a series of pictures at different exposures is called ‘Bracketing’. It’s a process that takes time but because you many want want to get it correct “in camera” rather than relying on software it’s something you may want to experiment with.

Neutral Density Filter (often abbreviated to ND). These filters are simply dark or very dark pieces of glass/resin/plastic. The point of these is to completely darken the whole scene and is generally used when you want to have a long exposure. You know those pictures you see of waterfalls with silky smooth water or clouds smudged across the sky? Well, they’ve probably been taken with a ND filter which has meant that in order to get that picture the camera needs to be mounted on a tripod and the shutter speed will have been several seconds long. It is possible to replicate the effect of an ND filter in software and again the trick to doing this is to take lots and lots of images and blend them in post-processing and so long as the static elements in all these pictures (e.g. mountains) can be lined up then the moving elements (e.g. water) when blended will appear to take on the same effect as if you had made a long exposure. Good post-processing software makes this quite straightforward.

Something else to consider: as we have discussed, both ND and GND filters work by reducing the amount of light coming into the camera. You can buy different strengths of these filters, to meet the needs of different lighting situations. The strength of the filter will always be advertised when you buy it and may even be printed on the filter itself (in small type in a corner of the filter which won't interfere with your photographs). However, the numbers that are used to tell you what strength these filters are is not very intuitive. The numbers will either represent Optical Density, or they will represent Filter Factor. Now the maths behind these numbers is more complicated than I want to deal with so staying with my tradition of keeping things simple the following table lists the Stop Reduction and the Transparency, Optical Density and Filter Factor equivalents:

Stop Reduction	Percentage Transparency	Optical Density	Filter Factor
0	100%	0	0
1	50%	0.3	2
2	25%	0.6	4
3	12.5%	0.9	8
4	6.25%	1.2	16
5	3.12%	1.5	32
6	1.56%	1.8	64
7	0.78%	2.1	128
8	0.39%	2.4	256
9	0.19%	2.7	512
10	0.10%	3	1024

And so on. So, each time you half the amount of light, you increase the Optical Density by 0.3 and double the Filter Factor. Incidentally, halving the light coming into the camera is called "a one stop reduction". So using a 0.3 filter is a "one stop reduction"; using a 0.6 filter is a "two stop reduction" and using a 0.9 filter is a "three stop reduction". So whenever you hear phrases like "increase by two stops" or "decrease by two stops" it is simply shorthand for the amount of light you are reducing or letting through. Some brands use Filter Factors with the letters "ND" (for Neutral Density) prefixing, e.g. "ND64" (a 6 stop filter) or "ND1000" (a 10 stop filter with the filter factor rounded down from 1024).

Check out my other articles

WTF is HDR?

What is HDR exactly and how can you use it to your advantage?

Let’s start with the definition.  HDR stands for High Dynamic Range, and this requires a bit more explanation. Dynamic Range is the term used to describe the range of luminosity within an image. Imagine you are standing on a beach watching sunrise: the sun is the most luminous object in the scene. The light from the sun will be lighting up the atmosphere and any clouds and reflecting off the water, so these will be the next most luminous aspects of the scene, then finally the sun will be shining on the beach and rocks, some of which will be casting deep shadows, so these will be the least luminous parts of the scene. All these ranges of luminosity, from the dark shadows cast by the rocks to the bright sun itself describe the dynamic range. And in this particular example, the dynamic range is extreme large going from extremely bright to extremely dark. If I were to take a single photograph of this scene I could do one of two things:

1)      Expose for the sky and sun and leave the beach, rocks and shadows in darkness or
2)      Expose for the beach, rocks and shadows and completely blow out the sky.

But, unless you are looking for a high contrast picture, neither of these scenarios is ideal and both will result in a photograph which fails to accurately capture the scene in front of you. This is because the sensor in your camera isn’t capable of capturing such a large range of luminosity – it can either expose for the really bright stuff or it can expose for the really dark stuff, but it can’t do both at the same time. And whilst the dynamic range of sensors is improving all the time, we are a long way from being at the stage where such a large variation in luminosity can be captured in a single image.

There are, however, a couple of things you could do to help.

The first is to use a Graduated Neutral Density (GND) filter. As discussed in this article, half of the GND filter is completely transparent and half is partially opaque, this serves the purpose of darkening the brightest part of your image whilst leaving the darker areas untouched and thus reducing the overall dynamic range of the scene. GND filters come in different “strengths”, and the “stronger” the GND filter you use, the darker it will make the brightest areas. In a situation like the one described above, this may work in a lot of instances but when we place a filter in front of the lens and point it directly at the sun we run the risk of introducing reflections into the image (light reflects off the lens, then off the back of the filter and then back into the lens). In addition, it is unlikely that even the strongest GND filter by itself would reduce the dynamic range of this particular scene sufficiently and you may have to stack more than one which in turn increases the risk of introducing colour casts into the image (Only really, really expensive GND filters are truly colour neutral) so what’s the alternative?

Well, if you’re prepared to put in the effort in post processing software, you can create an HDR image and this is the purpose of this article.

The principle of an HDR photograph is extremely simple but there are a couple of caveats. You will have to use software to create, and bring the best out of the image. To make your life simpler you should be using a sturdy tripod when taking the initial pictures, it will become apparent why in a moment. And finally you should either be in Shutter Priority mode or in full Manual mode with your camera for this to work well.

Okay, so we’ve established that taking one picture of our scene is not going to cut it, but what if we were to take lots of pictures, each exposed differently (by varying the shutter speed) so that in each one a single element of the total was well exposed? So for example, one or two pictures are exposed for the sun, two or three for the brightness of the clouds and the reflection in the sea, and a few more to properly expose the sand, rocks and shadows. On their own these would be useless but by using software we can blend these images to create one image which contains the properly exposed elements of all of them. By using a tripod to ensure the camera doesn’t move during the different exposures all the different images will align very easily and the end result will be a sharp and properly exposed image of the whole which can then be further adjusted in post processing to give you the final, desired result. And that, dear reader, is what an HDR image should be.

A few years ago, HDR photography got a bad name for itself when the internet was awash with overly processed pictures which, frankly, looked terrible; all sharp bold lines and pale haloes around edges, but thankfully that particular fad has passed and HDR is a legitimate way of dealing with situations where one picture alone just won’t cut the mustard.

Check out my other articles