Background
Now I’ve more or less done collecting images, my thoughts again turn to getting the best results from files in post production. This isn’t a significant focus in the OCA courses which, in my view, are more focused on conceptualism than formalism. I realise it is an area I revisit towards the conclusion of each course as my approach evolves and I see others working in post (mostly on the internet) and revisit books on the subject. George DeWofle’s Digital Photography Fine Print Workshop is useful to me, even if dated. He was a student of Ansel Adams and Minor White, and uses Photoshop to continue the craft of print making in the digital age. Midway through his book (p160), he observes:
The key to this process is perception, not a technical trick. If you can’t see the problem – brightness, contrast, color, softness, sharpness, or whatever it is – then no technique in the world, will make your print better.
He describes six aesthetic qualities of a digital print (there is no concern with contextualism in his book) that need to be worked on to develop the form: cropping, contrast, brightness, colour, defects, and sharpness. And discusses at length why the order of work is important and his approach to that work in Photoshop (and some other tools).
Recently, while fine tuning images (after making image-wide adjustments), I’ve felt my process a little mechanical – with adjustments and layer masks to target the adjustments, using a mouse. I’ve been feeling the pull to draw on the fine adjustments. Using the hand and pencil with years of learned control is very different to using a mouse and somehow more satisfying. DeWolfe has such an approach in his practice, which I’m going to try in my workflow as I finalise my project’s images.
Work flow
Here I describe my current work flow and how I plan to adjust/refine the approach. I do this mainly for my own record and to find clarity through having to write it down.
1) Input – negative and base copy (with broad adjustments)
I use Lightroom as file library. After importing a file, my current practice is to apply a tone preset (from the ‘film types’ available for Fujifilm cameras), make basic adjustments to exposure, add input sharpening, slight curves adjustment and image straightening if needed and sometimes cropping. So, the original file is adjusted non-destructively. This creatives complications if I want to rework the image later. In future, I’ll create a Photoshop copy and adjust using the camera RAW filter.
This will create two files that I’ll call negative (ie untouched apart from ‘film’ toning) and a Photoshop base copy (broad adjustments, white balance, and cropping if needed). DeWolfe emphasises the importance of first working on contrast and light (suggesting image is viewed in black and white to see this) before working on colour/colour balance – any changes in contrast and light also affect colour tones. I don’t always do this in order and find myself sometimes working circularly as a result. So, a discipline to introduce.
2) Optimising base copy – details and balance
DeWolfe uses a separate step in his workflow to optimise the base copy and retains this version separately from the version worked on for printing; more on that in the next step.
Optimising entails recovering any lost details in dng files and balancing the image (correcting broad areas that are either too bright or dark and adjusting contrast if necessary). To recover details DeWolfe uses external applications (eg nik sharpener) but technology has moved on since the book was published in 2006.
Lightroom now has an ‘enhance details’ option (takes around two minutes to process an image on my MBP) that can be used successfully on dngs from any camera. I also have Iridient X-transformer, which I bought when I was having problems with Fuji RAW conversion in previous versions of LR. I’ve not generally used either but could do so for selected and important images that contain significant details. I currently tend to balance later in my post processing process but now realise that this makes little sense as it disrupts any fine tuning of an image. From now on, I will enhance dng details where appropriate and always balance images in advance of detailed adjustments. I’ll check the Iridient vs built in LR options.
At this point one has a ‘base-copy’ image that has had broad exposure and contrast adjustments, has possibly been cropped, and has been optimised.
3) Proof – setup, overall contrast and colour balance
Contrary to common practice, DeWolfe recommends viewing the image in the correct proof set up at this point – otherwise it needs to be colour balanced again when printing. This is a bit of a revelation to me having spent time in the past wrestling with proof copies that required another lot of work before printing; it felt like going around in circles. It also occurs to me that if intending to output an image for screen (website, video, ebook) it perhaps makes sense to proof view in sRGB; I’ve noticed that some of my own images are not quite as satisfactory once converted on export and viewed in web browser. This would however make for an expansive workflow, given the steps that follow.
Levels adjustments are suggested to adjust overall contrast, brightness and colour balance. Any colour cast is corrected in a colour balance layer Local adjustments are made in the next step of the workflow. Once completed, an evaluation of a first print should be made before proceeding.
Working in this way would be a departure for me – proofing usually comes at that end and lead to no end of frustration.
4) Proof – local contrast and colour
This is the step where I’m dissatisfied with my current layers / mask approach. DeWolfe advocates using the ‘history brush tool’ to make marks (like an artist) and ‘move forward in a positive, courageous way’. It is the tool he uses for dodging, burning (through different blend modes); painting on local adjustments from snapshots of adjustments; and outlining to separate objects in detailed images; applying local hue/saturation adjustments. I’ll experiment with the techniques suggested but am wary of the time it could take when editing a number of images.
If subsequently printing on different paper profiles, the existing proof copy could be used as a base for a new proof copy and adjusted as necessary.
5) Final preparation for printing
This involves using a fresh copy of the image since it will be be resized and flattened for printing. I’ve never flatten my images prior to printing, but from what I can see online there are advantages in speed of printing and also output sharpening. It would also leave a final print file without layers – a print version that could be reprinted consistently.
DeWolfe first cleans up the image and then saves a flat version. He addresses any noise and sharpening. He doesn’t specifically address resizing images down (it was perhaps not a thing in 2006). However, there is a sound logic to applying sharpening after an image has been resized down, since its pixel dimension will have changed and the sharpening algorithms applied differently. His final steps are edge burning and final contrast tweaks using a gradient map (if he feels necessary).
6) Recap of files in workflow
- Neg – unprocessed in LR apart from application of camera ‘film’ tone.
- Base copy – cropped, broad contrast and colour adjustments, optimised detail, balanced image
- Proof copy – viewed using output colour profile, contrast & colour corrected, local adjustments, and cleanup
- Print – flattened, denoised, sharpened, final contrast adjustments, edge burning. Permanent record of a specific print.
Conclusion
A fellow student commented on a previous post about outputting to Instagram – observing that it sounded like a lot of work. They are partially correct but things perhaps always look longer when written than acted out. I don’t think it’s realistic for me to apply this kind of workflow to all images, just the ones destined for fine printing and large screen viewing. I’ll see how things turn out once I’ve worked through the process. I hope to arrive at a routine workflow that fits how I personally like to work with images before the end of the course!