Processing (some people call it editing) is probably the last surviving craft of photography, which makes it all the more surprising that it’s not widely done in a craftsmanlike fashion.
Well, there are two extreme statements in one, so I’d better justify myself.
First the idea of processing as a craft (and one of the last remaining at that). Digital has done away with the many skill sets that were often needed for shooting film, for one extremely basic reason — photographers had to predict what the image would look like. There was no preview, or immediate after-view. And large cameras were made with craft and needed craft skills to operate well. There were tilts, shifts, filters to tape behind the lens, all kinds of arcane stuff that continued into developing the film (sometimes for longer, sometimes for less to control the contrast). It went on, but this isn’t an article about photographic nostalgia. It’s a very good thing that the switch to digital, advanced electronic automation and mechanical streamlining have simply made operating a camera easier and more efficient, and that’s a good thing. Just as you don’t need to be a mechanic to drive a car, you don’t need to love engineering to operate a camera.
But why is processing a craft? For two reasons.
The first is that a craft is an artisanal skill, which with teaching, practice and experience, can be improved to a level of excellence. In fact it needs effort and time to become excellent. It’s not a ready-made skill set.
The second is to set it apart from being creative. Craft is not creative. It’s not where you invent things and let your imagination soar. It’s where you perfect things, hone them. And that’s exactly the job of processing. The word creative is used too often just to promote things that aren’t, like a clever digital filter. If you want to use post-production (or post-processing if you prefer that expression) creatively, then you’re in the different world of photo-illustration. That’s fine but it’s not photography, which is about capture.
However, processing has become MORE important than ever it was, because now we do it for ourselves instead of giving film to a lab. Moreover, the processing tools get better and better. Or rather, more and more, which may not necessarily be the same thing! The Raw file you shoot can contain much more shadow and highlight information that can ever be displayed on a screen or in a print, and modern raw processing ‘engines’ like Adobe’s ACR can pull remarkable details and qualities out of a file. Lightroom, Photoshop, DxO Optics, Capture One all have similar tools, and yes, there’s a lot to learn and a lot you can do. This brings me to my second claim, that there’s a lot of poor processing out there. The range of what you can do with the software puts many people off, which results in inadequate processing. Many people simply don’t know how to make the most of what is in the Raw file, and some of them would rather not find out because it’s nerdy and technical and not what they signed up for when they bought a camera. At the other end of the scale, however, are people for whom a software box full of tricks is an invitation to play. This gets encouraged by the class of processing tutorial that goes “Hey, did you know you can do this amazing effect in Photoshop or whatever?’ The result, more dramatically bad than failure to process, is over-processing, typically through over-saturation and the double-edged invention of tonemapping. Here and next month, I want to show you how to avoid either and how to make processing work sensibly for your images.
Don’t sweat the software
This is important. Processing software is an industry, and like the other industry we all buy from, camera manufacturing, is competitive and full of claims. It’s also hard to understand in depth. You’ll see a lot of strong opinions on the internet about which software is best, but most is not fully informed. When someone writes that. for example, Capture One or whatever is better than Lightroom or whatever, and that the colour rendering is “simply better”, what they almost always mean is that they haven’t gone beyond the defaults.
If you open a Raw file in any software, before you touch the sliders it’s presented to you with a certain look. It has to be, because an original Raw file is almost unviewable — dense, flat and greenish — and needs some basic things done to it by the Raw converter, including de-mosaicing (applying the blocky patchwork of colours from the colour filter array in front of the sensor into a smooth colour map), applying white balance and a strong gamma curve. On top of this, there’s how much sharpening to apply, colour rendering, and so on. These are the defaults, and they’r meant to be the starting point, the base settings that you can then adjust if you like. Don’t confuse them with the possibilities — with what you can actually do. Some users get very partisan about this subject; for a measured expert view, see this article by Martin Evening [https://lightroomkillertips.com/brilliant-article-martin-evening-lightroom-vs-capture-one-pro/]. Basically, all the main software contenders do their job very well. My examples are using ACR (Adobe Camera Raw), which is what Photoshop presents you with when you open a Raw file. Lightroom has more or less the same engine. Capture One, DxO Optics and the rest do the same things with different defaults, different interface, and sliders/tools that have relatively small differences in how they’re applied.
Calibrate your monitor, or…
…or at least do an eyeball calibration, like the Display Calibrator Assistant on a Mac…
Monitors differ, and what you see on yours won’t necessarily look that way on someone else’s — or when printed. Time spent getting a reliable work space is well spent! This is my workspace when I’m processing. I use an Eizo CG, and always have, because they’re the most accurate monitors, self-calibrating, and they’re 10-bit, meaning a billion colours. At the same time, I mirror the display on a MacBook Pro for comparison, to see how the image will look to most viewers: over-contrasty and that silly shiny finish to the screen. I use Photoshop and ACR because I like software that does one job at a time. Logitech mouse and a small Wacom tablet…
Industry standard optimal
Are there industry standards in processing? Best practices? Yes, there certainly are, although you won’t easily find them in one place. Where you will get good advice, if you ever get deeply involved in book publishing, is from any pre-press company, because they handle image files from all kinds of sources and have to make sure they’re fit for repro. Even when many are very much not. Pre-press people, and digital print shops also, are at the front end of image processing, and they want consistency. Pre-press involves quite a bit more than a well-processed image, but it really helps if that’s the starting point. Here below, we’re examining proofs for one of my books at XYDigital in London. Note the controlled lighting of a viewing booth with colour-corrected lamps…
So, regardless of the special needs of an image, which is what we’ll come onto next, it’s useful to know what the objective industry standards are. Follow the basic procedure below, and you will at least guarantee a competent image. Think of it as a starting point. There are no issues of taste and personal judgment here, just a few objective standards, the most important of which are:-
- Overall average exposure
- Neutral colour balance
- Detail visible in highlights and shadows
- Black and white points set
This is basic optimisation, and please note that this is NOT the same meaning as optimising compression, which you’ll also see used. Here’s a walkthrough with a sample image…
Optimisation essentially means preparing a TIFF from a Raw file to expected industry standards, with a just-full dynamic range and qualities that a typical viewer would expect, and which appear normal. The qualities include overall brightness, contrast, colour saturation and sharpness.
It has an increasing importance as extremes of processing become more possible with the latest software. Over-processing and failure to process adequately are the two widespread ills that it corrects, and optimisation involves a lot of avoidance—that is, not using adjustment sliders that superficially offer interesting effects.
Optimisation ignores creative and stylistic choices, and so is not necessarily a guaranteed way of processing. Rather, it’s a starting point. One example of this is if you had shot a misty or foggy scene. Setting the black point would not normally be appropriate, because while it would ‘smarten up’ the image, it would lose the delicate misty feeling in the one and the subtlety of colours and tones in the other.
Step 1: assessment
Once you’ve opened the Raw image, before making any adjustments, stop and consider two things:
- what had you expected or wanted when you shot the image, and how close does it look to this now? 2. Looking at it objectively, as if a viewer coming fresh to the image without preconceptions, are its image qualities all as expected?
Step 2: lens profile
The software will detect the camera-lens combination, so checking this removes vignetting and geometric distortions, which are common with all zoom lenses. Adobe advise that their colour correction, offered as a separate checkbox, is more effective at eliminating colour fringing than the lens manufacturers’ own profiles, so check this always.
Step 3: default sharpness
There are just two possible occasions for sharpening: 1. to compensate just a little for the slight softening that digital capture creates, and 2. a more aggressive one just before display (printing or screen display). Perform only #1 during Raw processing; the ACR defaults shown here are recommended—25%, radius 1 pixel, detail 25%, no masking. Some other software, like Capture One, have more aggressive defaults.
Step 4: global adjustment
In preparation, make sure that the upper left and right boxes on the histogram are checked, so that they each have a white outline. Clipped shadows will appear blue, clipped highlights red.
Also beware that the following sliders are effective at what they set out to do, but carry the penalty of non-traditional appearance (ie tend to give an illustrative look rather than a traditional photographic look): Highlights, Shadows, Clarity. These three use local tone-mapping algorithms.
a) Colour balance
By default, aim for making obvious neutrals, such as grey metal, concrete and so on being neutral (this means the RGB values should be more or less equal to each other). If the light has an obvious cast due to the time of day, such as warm low sunlight or bluish shadows on a sunny day, consider whether you want to final colour to have a slight cast; this is your taste entirely. Colour temperature is from blush to orange-ish, while the lower slider works the other colour axis, from green to red.
Make an overall judgment. You can always go back to this later
This is my preferred technique, but there are others (see Curves below). Reduce Highlights and raise Shadows by no more than 30% (more and you risk a non- photographic look). The effect of these two is to flatten the image, so raise Contrast accordingly (perhaps around 20%). In order, one by one, this is what happens…
d) Whites and Blacks
Also known as setting the black and white points. Often forgotten by many people, but important if you want to make full use of the tonal range. In effect, you’re stretching the range to fit the scale. Raise or lower the Whites slider until just below when the red warnings appear for clipped highlights. Then lower (you hardly ever need to raise) the Blacks slider until just above when the blue warnings appear.
An alternative c) above is to use Curves. This is very traditional and gives you control over highlights, shadows and contrast all by changing the shape of a single curve, as shown, but needs experience. Most pre-press people in my experience use this tool. It works in a different way from Highlights and Shadows — not necessarily better or worse, just different.
Here I want to introduce you to the idea of processing with a well-defined purpose. If you shoot Raw, digital processing is as essential as printing a film negative was in a wet darkroom. It converts the image data recorded by your camera into the best possible viewable image. In the professional way, you should not think of processing either as a boring duty or as a chance to play around. To be effective, processing needs to follow your shooting and the ideas you had at the time of shooting. It should never be a creative activity. If that sounds dogmatic and controversial, consider what your primary creative medium is. If you see yourself as a photographer, then creativity is inevitably concentrated on the moment of shooting, the exposure. It’s perfectly legitimate, of course, to broaden your ‘job description’, and if you see yourself as a creator of images by any means, then you are really a photo-illustrator. As a photo-illustrator, you can be as fanciful and imaginative as you like, and make full use of all the possibilities in Photoshop and other image-changing software, and yes, be creative during the processing. But if you are first and foremost a photographer, the camera is where the creative action lies. It isn’t compatible to try and be creative in two places, shooting and processing.
For photographers, processing is a craft. A very important one, and one that can take years to perfect. But nevertheless a craft, subordinate to the actual photography.
The main purpose of most processing is to do full justice to the Raw image file. This contains much more tonal information than can ever be reproduced in a print or on a normal screen, which its why the Raw processing stage is so important, particularly for ‘difficult’ images, such as those with a very high dynamic range. For most images, the default is simply to optimise it as above, which means preparing a TIFF from the Raw file in a way that simply looks the way most viewers would expect it to look.
A step up from that is applying more skill and more techniques to really make the most of the image, in ways that may seem small to the casual viewer but which from a professional point of view enhance or reduce different areas and/or subjects within the frame. The image below has had considerable local adjustment work done, yet it still appears ‘normal’…
This is what it looked like before processing…
The changes are important but not hugely dramatic. Here below is a composite screenshot of the local adjustment — 4 radial filters, one grad filter and one brush. Quite a lot of work, but as in any craft, it’s in the detail!
In the next article, we’ll continue with 1. Bringing processing into your workflow by making it an extension of your shooting, 2. Deciding on the look that you want, and 3. Using local adjustment tools for real craftsmanship instead of trying to do it the cheap and easy way with tone mapping sliders.