Taming the beast: from image to print

This post is a glance behind the scenes at the work that I do to transform an image into a print. I always start with a drawing or a photo, and they always undergo several transformations until I have a design that I can print- here is what this transformation involves.

First I build this original image- the photo or the drawing- into a digital design that I like. This first transformation can be as simple as adding colours or shapes to a drawing, or cleaning a photograph, but it often means a lot more: applying filters, simplifying, fragmenting or repeating the image, recolouring it completely, layering other images in… there is a lot of experimentation at this stage, I start and I stop, I come back to it later, I tweak and I tinker.. Some images never make it past this stage, some undergo multiple mutations, and some just work in a flash.

Once I have an image I am happy with comes the puzzle: how to deconstruct it so that it works as a print? One option would be to use CMYK printing, which is a technique used in newspaper for example, and where all colours are separated into 4 layers of Cyan, Magenta, Yellow and Black (Key) which combine to give the impression of a whole gamut of colours.

I don’t think this would work for me- whenever I try to mock-up my artwork in CMYK, they look dull and muddy, and I am worried that I am too messy with registration for the images to come out clearly.

Instead I use a Simulated process for separating the colours out: I use a Photoshop tool to pick out each colour in turn and copy it over on a separate layer. I do this as many times as I need until I have a version that is close enough to the original design when looking at the separate layers together.

At this point I can have up to 20-30 colours/layers on my file, which would be a nightmare to print, so the next stage is combining colours where possible. I look for colours that are very similar or could work as a gradient of some sort and then merge these layers together. I do this until I have run out of possible colours to group- with simpler images this can be 2-3 layers, but on some of my recent works I have ended up with 7-9 layers. Of course, the more layers you have, the more details you can have in the print. This is useful to translate volume, luminosity or texture: I had a print recently where I used 3 red layers on top of each other to translate the feel of a velvety curtain on a print- I think I would have lost that certain shimmer and depth of velvet had I gone for 1 or 2 reds only.

Then all that remains is to halftone each layer to have positives ready for printing. Halftone uses dots or lines (or sometimes other shapes ) of different sizes to give the impression of a colour gradient: depending on the size and density of the pattern, the brain is tricked into seeing a specific colour- for example yellow dots on blue dots will read as green, and the comparative size and density of the yellow vs blue dots will impact whether we see it a yellowish green or a bluish one.

And finally I have an image ready to print! This approach doesn’t always work. I have some images that I have to abandon as I can’t get the layers to work- the final image it too dark, or too dull… More often that not though, it is quite satisfying to start with a full image and break it down into a number of printable layers. It feels like solving a puzzle or taming a beast!

Of course during the printing process, it goes back to being a beast, but for a moment, when my positives are ready to print, I feel like it is me who controls the process, and not the other way round!

From left to right: the original photo: the halftoned, separated design: the screenprint