Tag Archives: technique

The Composition Question

I recently posted on my Facebook page, asking people to let me know if there’s anything about photography that they still struggle with. A common complaint (well, the only one, but from two posters) was that people often don’t understand why their photographs don’t reflect the way they remember a scene.

I suspect that a simple grounding in composition techniques may help here. The human eye, in conjunction with the brain, is able to filter out extraneous detail, especially in a three-dimensional image constructed from the two images from each eye combined, whereas a photograph is not processed in the same way, so you need to be aware of how you can draw people into an image, keep their interest, and avoid distracting them with extraneous detail.

I’m listing here some techniques that I use in my photography to try to keep pictures interesting, and retain the impact of a scene as viewed through the naked eye.

Rule of Thirds

One of the most basic rules of composition is called the ‘Rule of Thirds’. This is a quick and easy way to make a photograph more pleasing, though it has variations and even detractors, which I’ll come to in a minute.

The basic principle is this: if you are photographing a subject that does not fill the frame, try placing the subject at any of the four points of intersection between imaginary lines which divide the picture into thirds both horizontally and vertically. This makes the image more pleasing to the eye. Here’s an example:

The Rule of Thirds in action
The Rule of Thirds in action

Opinion is divided over why the Rule of Thirds works, and there are even some who believe it is only an approximation of a different rule where objects should be placed on any of the four 45° diagonals drawn from each of the four corners of the image. This is known as the diagonal method.

The rule of thirds also applies to horizon lines. Try to place the horizon in your frame along one of the dividing lines that cut the frame into thirds, rather than exactly in the centre. This places clear emphasis on either the sky or the ground, avoiding the conflict created by bisecting the frame, when it’s not clear which is more important.

Leading lines

The photograph above also demonstrates the use of leading lines. Using a foreground detail, it is possible to lead the viewer’s eye into the photograph to a point of interest. This is often done using a straight or curved diagonal line starting at or near a corner of the image. The line of surf on the beach starts near the bottom left hand corner of the image and moves diagonally until it meets the rock stack, reinforcing its status as the subject of the image.

The footpath leads directly to the tree, drawing the eye into the frame

By being aware of foreground elements, you can help to ensure the image does not inadvertently draw the eye away from the subject of interest, which causes conflict and makes an image less pleasing to look at.

Leading lines can also be implied. If you are photographing people or animals, the direction they are looking in will also lead the eye. Try to ensure they are looking into the frame, not outside it, which will naturally draw the viewer into the picture.

The puffins are looking left-to-right, so placing them at the left of the frame draws the viewer in. Note the puffins and the glacier in the background are also following the rule of thirds.
The puffins are looking left-to-right, so placing them at the left of the frame draws the viewer in.

Keep it simple

A common mistake in photography is to try to include too much of a scene in a single frame. This can disorientate the viewer and lead to a lack of engagement with the photograph. If you are photographing an object, try to avoid distractions by careful framing; ensure foreground objects don’t compete for attention, using leading lines to link them to the main subject or eliminating them altogether. Use an appropriate aperture and/or focal length to blur a distracting background (wider apertures and longer focal lengths will decrease depth of field).

Highgate cemetery is full of tombstones, trees and statues. Cropping in on one of the statues, and blurring out the background, keeps the composition simple.
Highgate cemetery is full of tombstones, trees and statues. Cropping in on one of the statues, and blurring out the background, keeps the composition simple.

Obviously there are instances where featuring a lot of information is the required effect. Where this is the case, think about using leading lines again to draw the eye from one group to another, and try to create a sense of unity by either showing the entire group of subjects in the scene, or cropping so that the group fills the scene to all four edges.

Find an unusual vantage point

A photograph is about more than just its subject. The location of the camera is implicit in every scene, and careful choice of vantage point can turn a good photo into a great one. Try to avoid shooting from eye level – this is how most people see objects every day, and adds nothing new to an interesting subject.

Squat on your haunches, or even place the camera on the ground, to increase foreground interest and emphasise the height of the subject. When doing this, look out for converging verticals – if you have to tilt the camera too far backwards, objects (particularly buildings) will look like they are leaning. This isn’t necessarily a bad thing, just be aware of it and whether it adds to or detracts from the effect you are trying to achieve.

A low vantage point shows more of the interesting rocks in the foreground
A low vantage point shows more of the interesting rocks in the foreground

Many photogenic landmarks have typical views which everyone who visits feel they have to capture, like ticking off a list. To make your photos stand out, find a vantage point that shows your subject in a new light – try walking around it, viewing it from all angles, or see if there are buildings or staircases nearby that allow you to view it from higher up than ground level. Look for interesting juxtapositions with other objects in the area, and see if you can use two or more of the intersecting points from the Rule of Thirds.

A wind turbine, a common sight these days, becomes almost unrecognisably abstract when viewed from directly below.
A wind turbine, a common sight these days, becomes almost unrecognisably abstract when viewed from directly below.

Use context

Sometimes a scene can be rather abstract, so including familiar elements will provide context, particularly in terms of scale. An unfamiliar scene can confuse the viewer, so placing, for example, a human figure somewhere in the photograph serves two purposes: it orients the viewer so they know which way is ‘up’ (where it’s ambiguous), and also provides a sense of how large the subject is.

The photographer in the foroground emphasises just how high the waterfall is. Cropping out the top of the waterfall also serves to emphasis its scale.
The photographer in the foreground emphasises just how high the waterfall is. Cropping out the top of the waterfall also serves to emphasis its scale.

Balance opposing elements

If you are able to find elements of a scene that complement each other, try to make them balance, rather than compete. One of my favourite photos from Iceland is this one of a fjord in the Westfjords region, where most of the techniques I’ve described above have been used.

Ísafjorður fjord in Iceland
Ísafjörður fjord in Iceland

Note first of all that the horizon is not in the centre of the image, but more towards the top, approximately one-quarter of the way down the frame. A clear leading line follows the edge of the stones on the bank of the fjord, which has been enabled by a low vantage point. The composition is simple, with sharply defined dark and light areas (stones, water, mountain, sky) and the leading line of the bank continues into the outline of the mountain, creating a balance which makes the elements merge together into a whole.

Natural framing

Use of foreground to frame a subject can create a pleasing effect. A printed photograph always looks better in a frame, and you can create your own within the photograph itself using overhanging tree branches, windows or other holes, or walls. Always be aware of your surroundings so you can find something to frame your subject if it’s there.

The reeds in the foreground frame the photo and add depth
The reeds in the foreground frame the photo and add depth

Colour palette

One final tip I’ll pass on, though it’s not always feasible, is in a way an extension of the ‘keep it simple’ rule. Where possible, try to narrow the colour palette of your photograph as much as possible. Look for the predominant colour in your chosen subject, and try to match the foreground and background using similar colours (or neutral tones such as browns, blacks, greys and/or whites); or frame the image so that only one colour dominates.

Borough Market in London is a riot of colour, however ripe tomatoes are always red.
Borough Market in London is a riot of colour, however ripe tomatoes are always red.

If your subject itself is neutral in colour, try to keep the rest of the photograph based around one or two predominant colours.

This is more a stylistic tip than a rule, as obviously there are occasions where lots of strong, bright colours work well, but if you keep it in mind you may find you can add the extra element to a photograph that makes it stand out in your collection.

And finally…

Once you’ve got into the habit of applying these techniques to your photography, start to consciously make an effort to break them. Photography is, after all, an art form, and you are free to express yourself however you see fit. So be daring, and do something crazy now and then. You never know where your next great image may come from.

The natural symmetry of the church, broken only by the bell tower, makes it an ideal candidate for placing right in the centre of the image.
The natural symmetry of the church, broken only by the bell tower, makes it an ideal candidate for placing right in the centre of the image.

Tilt-shifting at windmills (and other tall buildings)

What does “tilt-shift” mean to you?

Get that thought out of your head right now. Yes, tilt-shift lenses CAN be used to do that, but there’s so much more to them. To start with, the above effect was created in Photoshop, but when you tilt a lens in the opposite direction you can actually achieve a WIDER depth of field by aligning the focal plane with, or close to, the ground, without having to stop the lens down to f/11 or smaller. But that’s not why I use these lenses.

Take a look at this photograph of the railway arches near where I live:

It was taken with a 17mm lens from ground level, meaning that to include the top of the viaduct in the photo I had to tilt the camera back. With a rectilinear (i.e. not fisheye) wide angle lens, this has the result that all vertical lines start to converge, distorting the shapes in an unpleasant, almost vertiginous manner.

Luckily, the 17mm lens in question has a shift feature (the tilt wasn’t used for the next shot). So I shifted the lens upwards and aligned the back of the camera to be parallel with the walls of the viaduct:

Now the vertical lines on the structure are vertical in the photograph, and yet I still have the top of the viaduct in shot. This is because the 17mm TS-E lens has a much larger image circle than a standard 17mm lens. When you shift the front element upwards, the image circle also moves so that the portion that hits the sensor is towards the top of the scene. Using the same lens but without shifting would have resulted in a photo with only the bottom portion of the arches in shot, and far more pavement than would be reasonable.

This shot of Watford Fire Station was taken with the lens shifted up partially, meaning the building is centred vertically in the frame, rather than towards the top:

_MG_0279An extreme example (when you stand a bit closer to a building than is strictly necessary) shows how the perspective towards the edges of the larger image circle can be a bit overblown:

However with a slightly less wide lens (24mm in this case) and a building that’s further away, shifting the lens can result in very pleasing ground-level shots of tall buildings:

Apologies for the lack of windmills in this post. I just couldn’t resist the pun.



Graduated filters

It’s a little late in coming, but I did promise a lesson in how to use graduated filters. It follows on nicely from what I spoke about in my post about the Zone System, so please familiarise yourself with that one before reading this if you haven’t already; this one will make more sense that way.

I mentioned that the range of brightness that a camera can ‘see’ is much narrower than that of the human eye. In most cases, you can isolate an interesting part of an image and bring out the detail where it’s relevant, however there are many situations where this simply doesn’t work. For example, if you are taking a landscape photograph that includes a lot of sky, especially late in the day, the foreground will generally be three or four stops darker than the sky. By metering for the foreground, the sky will be blown out, but if you stop down to increase detail in the sky, the foreground will usually become too dark.

Of course, with RAW files and sophisticated post-processing tools such as Adobe’s Photoshop Lightroom, it can be easy to fix such photographs afterwards, but this won’t always work, as often the sky is so overexposed that the highlights end up edged with grey as colour detail is lost. HDR is another option, but is time consuming, and the effect is not to everyone’s taste (myself included). It is far preferable, I believe, to achieve the effect you want in-camera. So how do you do this?

Take the following picture of the church at Prestbakki in northern Iceland as an example:

Original image
Original image

There is some great detail in the foreground, with the gravel and the weeds among the flagstones clearly discernible, but the sky is almost pure white. I could stop down by 2 stops to bring the clouds into detail, but then we lose the interesting textures on the ground:

Underexposed by 2 stops
Underexposed by 2 stops

However, by applying a 2 stop neutral density graduated filter (usually referred to as an ND grad) with the dark half placed so that it covers the sky, I can keep strong detail in both halves of the picture:

2-stop ND grad filter applied to sky
2-stop ND grad filter applied to sky

This is a particularly strong setting, mainly for illustration. The best use of ND grads is when you can’t tell it’s been used. Compare the next two photos of the river Lagan in Belfast:

Without filter
Without filter
With 2-stop ND grad
With 2-stop ND grad

The first was taken as is, and for the second I used an ND grad to drop the sky down 2 stops so that it was the same brightness as the river. The contrast between reflective surfaces (such as water) and sky will always be less than that between, say, grass and sky, so it’s possible to make the whole picture appear more uniform. This is another common use for ND grads where you won’t necessarily get the same loss of detail caused by excessive differences in brightness.

Finally, an example of where an ND grad really comes into its own:


Thirlmere, Lake District
2-stop ND grad used

This was taken at Thirlmere reservoir in the Lake District. The sky was turning some beautiful colours, but the bracken was a lovely shade of orange/brown and the drystone wall provided a nice way to draw the eye into the picture, so I placed my usual 2-stop ND grad on so that I was able to capture the lot.

Note that there are many types of ND grad filters. They vary from 1- to 4- stops, and in the degree of gradation, with a hard edge for clearly defined horizons, and softer edges for when foreground detail (e.g. trees, tall buildings) intrude on the sky and would be spoiled by excessive darkening. It’s a good idea to have more than one to hand, but you can vary the softening effect by using the aperture. A narrower aperture (higher ‘f’ number) will harden the boundary between dark and light halves, so open up as wide as you can if you want a softer edge without changing filters. You can also buy screw-on ND grads which rotate independently of the lens, but you can’t move the horizon up or down with these, so it’s better to get the rectangular acetate filters in the Cokin or Lee series.

I hope this lesson has also been helpful. Please use the comments below to let me know if I’m striking the right tone. I’d also like to see any photos you take using ND grads; you can link to flickr if you have any you’d like to share.

The Zone System

What I’m going to try now is the first in a series of lessons, where I’ll show you some aspect of photography of which the average point-and-shoot snapper may not be aware. Hopefully if you put these lessons into practice you’ll be able to achieve the sort of photographs you’ve always wanted to. Let me know if I sound too patronising; that’s not my intention at all.

The zone system was developed by the American landscape photographer Ansel Adams, and the principle is simple. A camera can ‘see’ a more limited range of brightness than the human eye, so it needs to be adjusted to take into account lighting conditions. Most cameras try to render a scene so that the overall lightness is what’s known as 18% grey, or exactly half way between pure black and pure white. If we number these brightnesses, 0 for pure black, 10 for pure white and 5 for 18% grey, then by filling in the numbers in between we have the 11 zones of the zone system. Each zone is 1 stop brighter than the previous one.

The zones mapped to an exposure meter
The zones mapped to an exposure meter

Most modern SLR cameras, digital or film, allow you to over- or underexpose by up to 2 stops in auto or semi-automatic mode, which can be limiting, so the first thing to do when using the zone system is to switch to full manual mode. The next thing to do is to set your camera to spot metering mode; refer to your camera’s manual if you’re not sure how to do this. You are now equipped to try the zone system.

Each zone has a description of what it represents:

0 – Pure black
1 – Near black, with slight tonality but no texture
2 – Textured black; the darkest part of the image in which slight detail is recorded
3 – Average dark materials and low values showing adequate texture
4 – Average dark foliage, dark stone, or landscape shadows
5 – Middle gray: clear north sky; dark skin, average weathered wood, grass
6 – Average Caucasian skin; light stone; shadows on snow in sunlit landscapes
7 – Very light skin; shadows in snow with acute side lighting
8 – Lightest tone with texture: textured snow
9 – Slight tone without texture; glaring snow
10 – Pure white: light sources and specular reflections

What you need to do to apply it is to find a key part of your photograph that you can assign a zone to, then set the exposure accordingly. In the illustration above, I have mapped the camera’s exposure meter onto the middle 5 zones, but that does not mean you are limited to just these zones – by using full manual mode you can go beyond these zones but you need to count!

Now let’s take an example. Suppose you are in a forest on a sunny day, where the grass on the floor is partly lit by sun and partly in shade. The most effective way of exposing such a scene is to meter on the sunlit grass, and the rest will take care of itself. Point your camera’s spot metering zone at the grass so that the grass fills the metering zone, and adjust the exposure so that the camera says ‘0’ – i.e. perfectly exposed (or zone 5). You can then recompose, and because the camera is on manual, the exposure will not be affected. The result will be something like this:

Spot metered

A camera set to full auto would have seen the silhouetted trees in the background and overexposed by a stop or two, to look something like this:


Note how the vivid green in the foreground in the first picture has been bleached out, and the stump is not as isolated as a feature as it was.

Generally you will be able to find a part of your picture that fits within zones 3-7 so you can see your camera showing you how far over- or underexposed you have made it. However if you want to venture further into the more extreme zones, here’s where you need to count. To expose for zone 2, set the exposure to underexpose by 2 stops, then decrease the aperture or the shutter speed by 3 more clicks of the wheel (if your camera meters in 1/3 stop increments, as in the illustration above) or 2 more clicks (if 1/2 stop). For zone 1, use 6 or 4 clicks, and 9 or 6 for zone 0. To achieve a silhouette, for example, you should ensure the object you want to appear in silhouette is at zone 2 or below, but bear in mind that you may lose detail in the background if it is not bright enough, so take another reading from the background to ensure that you keep it in zone 5 or higher. If the contrast between the background and the object isn’t great enough, you probably won’t be able to achieve a perfect silhouette.

I hope this has all made sense; by all means look on the internet for more about the zone system and how a camera meters, but if you do get any good results using this system I’d love to hear from you.