Perfect Exposure… one point of view

The title is, I suppose, provocative. Surely it goes against the ethos of creative photography to suggest that there is such a thing as a ‘correct’ exposure setting and result?

Absolutely right, yet there is, for every situation and image, an exposure that is perfect for your taste. Which will, naturally, likely be different from that of another photographer.

Camera manufacturers have put a huge amount of development work into ‘guaranteeing’ a good exposure to any user, the reason being that this seemingly basic function is not only critical but surprisingly resistant to being completely automated. It can be automated to a point, but the bottom line is that while the majority of situations can be calculated successfully by the camera, some cannot. And the ones that can’t are often the most interesting for photography, with the potential for producing striking images — namely high dynamic range and unusual.

Developing photographic skills means pushing some limits; I think we can all agree on that as a principle. You can push choice of subject, composition, emotional treatment, lighting, all kinds of things. This can happen without unusual and difficult lighting, which is always a challenge for calculating exposure. But when you do want to push the lighting situation, the camera’s automated system, however sophisticated, suffers from a major drawback – it cannot follow your taste.

In smart, predictive metering there are two approaches that a manufacturer can use, and they can be combined. One is to collect a database of all the ways in which people compose images and work out good exposure settings for each one. Then, as you shoot, the metering pattern can be compared with this database to find the nearest similar pattern, and apply that worked-out setting. Nikon, for example, claim to use tens of thousands of researched photographs for their database. The second approach is content recognition, an imaging science still in its infancy. The goal is to recognise subjects in the frame, such as faces and figures, on the reasonable assumption that what you focus on is likely to be your exposure priority. Smart metering gets smarter, but the problem is that its very sophistication makes it more difficult for any of us to know what decisions are being taken by the camera’s processor. And the camera manufacturers are understandably coy about giving away too much about their hard-won inventions. And personally, I don’t like handing over full control to an algorithm invented by a team of engineers.

There’s an argument that this sophistication and all the choices it has spawned in settings has done more harm than good. There are now more excuses than ever for dithering about, and attention gets taken away from the basics. After all, and this is worth remembering, it still all boils down to a single dosage of light on the sensor, and there are still only three things controlling this, as there have always been since the days of wet-plate cameras —  shutter speed, aperture and sensitivity.

The important and sometimes complex decisions to do with exposure are located a step back from where many people believe they are. First is knowing what you want from an image when you are in front of the scene — having a point of view about how what is in front of the camera should translate into a still image. Second is knowing which are the critical part (or parts) of the scene to hold at a particular brightness. And doing this fast and intuitively. Neither of these key decisions can be taken successfully by a camera – you’d need A.I. built into the machine for that.

So the following is what I see as the sequence, in time order. And it’s usually very fast, unless it’s a studio set. The on-the-spot actions often need to be close to instantaneous, especially with reportage. And in fact, most professionals are doing this so much of the time that they rarely think it through in such a formal, ordered way. It can, and probably should, be intuitive.

BEFORE

• Camera settings OK
• Metering mode familiar

ON THE SPOT

• Know what you want
• Scan for likely problems
• Recognise the key tones
• Risk of clipping
• Meter and shoot

AFTER

• Review

I’ll elaborate…

BEFORE

Camera settings OK – Means OK for the way you work, and in particular not leaving controls set to the last situation without checking (the ISO, for instance). Things to think about include having instant review on the screen after each shot, whether or not to show the highlight clipping warning, having the histogram quickly accessible if not actually displayed each time. Need to balance information against distraction.

Metering mode familiar – In a way, it almost doesn’t matter which metering mode you use, as long as you know very well how it behaves for any lighting situation. Familiarity is the key — if you know that the meter is likely to over- or under-rate a scene compared with what you want, you can compensate. The reason is that any adjustments you make will be departures from average/mid-tone.

ON THE SPOT

Know what you want – The million-dollar question, of course. Why are you taking the picture? What do you expect to get out of it? Can you picture the image as it should look?

Scan for likely problems – Are there any likely exposure issues? Think about what is in front of the camera before letting the metering system loose on it. For example, Is there a major hotspot likely to blow out? Does it matter if it does? Most problems are because the dynamic range of the scene is greater than the sensor can capture in one exposure. You might be able to solve them by thinking about viewpoint, focal length, composition.

Recognise the key tones – After knowing what you want (above), the second most important exposure decision. Often, but not always, what you’ve decided as the subject of the photograph will be the tone that needs to be set most critically. So, what in the scene is it, and how bright should it be? If a face, for example, is it Caucasian or East Asian (needs to be x times lighter than mid-tone)? Or black (needs to be x times darker than mid-tone)? The key tone may be only a part of the key subject, or in some circumstances may even be some other part of the scene, such as a background.

Risk of clipping – Very much a digital issue, and the subject of this month’s ‘Observation‘. However you feel about clipping, it does deserve consideration. Shooting to hold the highlights is simplistic as a rule for exposure (as I hope this month’s introduction to the exposure demonstrates), but clipping is usually better avoided or treated somehow. If there’s a risk, there are three kinds of solution:

CHANGE – Dealing with it on the spot by changing light or composition. Say a portrait is backlit and the background would be heavily clipped if the exposure were right for the face; you might want to add foreground shadow fill, or alter the backdrop. As another example, if there were a small bright hotspot doing nothing special for a shot, you might re-frame to crop it out.

COMPROMISE – Accept an exposure that isn’t perfect but leaves shadows or highlights just about acceptable. And then move on to…

TREATMENT – Relying on processing or post-production to solve problems is poor craftsmanship, but there are techniques that have their uses, such as highlight recovery in a raw converter, or HDR, which calls for multiple exposures.

Meter and Shoot- Depending on which metering mode you use, adjust to compensate for the key tone, and shoot.

AFTER

Review – Check on the camera screen, and adjust and re-shoot if necessary, and if there’s time. This is all about the kind of shooting you are doing and the situation you are in. If the action is fast, this can lose you shots. If a slow subject, like most landscapes, generally a good idea. Remember that what you see on the camera’s screen may not be completely accurate; it’s generated from a jpeg, not the raw file.

We’ll re-visit exposure in a few months’ time. It’s an issue that is at the same time alarmingly simple and infinitely complex. And Photoshop has nothing to do with it.

Some examples…

Knowing what you want #1…

High contrast, certainly out of the dynamic range of the camera, and an unusually small key subject, worse still in white, all combine to make a situation that you KNOW will be an exposure challenge. I had time to think about it before the Shinto priest stepped onto the bridge. The eye could see much more detail than this, so the important thing as always was to anticipate how it could and should look as a photograph. Important for me was SOME detail in the mid-tones on the bridge to give enough clues about the setting, and I felt I could live with some loss of information in the white robes.

Knowing what you want #2…

No time to waste at all for a shot like this, from a helicopter at low altitude (when things pass VERY fast below you). The tiny figure walking in the bottom right corner makes the shot by giving it scale and life – so much so that I cut the top off the spire above. So, the figure is a key subject, but tiny so not really anything by which to judge exposure. The tombs are more important for the exposure, and a quick glance says 2 stops more than average. However, the ground with its graves was if anything more important for me, because I wanted it rich and dark to give maximum contrast to the tiny figure. That meant 2 stops darker than average. A lot to think about in a second, but as I knew what I wanted, adjusting the exposure on the camera downwards was easy.

Deciding what the key tone is…

Quick-decision time again, and high contrast. There are two possible key tones, shown schematically below. One is the upper part of the woman’s blouse, a likely candidate for clipping is not careful. The alternative is the slightly less bright entire sunlit area. Going for the former would hold the texture of the fabric better, but lose more shadow detail. I went for the 2/3 stop brighter alternative.

Remember this...

Hide Sites

Tags: clipping, dynamic range, exposure, highlights, meter, shadows