![\"\"](\"https:\/\/thatkeith.com\/articles\/wp-content\/uploads\/2011\/08\/13-dither-thumb.gif\")
<\/a>When you look at designers today you see people who are faced with massive multi-skill requirements. Photographer? Check. (Everyone has a digital camera these days, even if they\u2019re not actually good at using it…) Photo retoucher? Check. Typesetter? Check. Repro specialist? Check. Copywriter? Check. (Although most designers still can\u2019t spell for toffee.) Production wizard? Check. Web designer? Check. Search engine optimisation expert? Check, and check again. The list goes on. This is both a blessing and a curse. Twenty-five years ago it wasn\u2019t like this – not as much, certainly. Before desktop publishing, designers still had to know about a wide range of stuff, but there were all sorts of specialists to call on and to do various parts of a project, from pasteup work to final repro artwork production. Things took longer to do, but you didn\u2019t have to know so much in detail beyond your immediate skillset.<\/p>\nThings changed fast, thanks largely to our digital friend the Mac, its chum Adobe PostScript, and a jostling crowd of software. Now we have to be able to turn our hands to almost anything if we want to keep the work coming in. Exciting? Absolutely. But stressful? Yes, just as much.<\/p>\n
Despite what some clients seem to think, design is a discipline that requires care, understanding and experience. Just remember to add \u2018flexibility\u2019 to the list and keep experimenting whenever you have the chance. It is more and more important to understand at least the fundamentals of the various disciplines of those around you, and preferably more than just that.<\/p>\n
The era of the specialist in the design industry is pretty much over; we have all had to become experts instead. Think of it like this: specialists do one thing well, which can make them vulnerable in this fast-moving industry. Experts do a number of things well but not exclusively, so they\u2019re more adaptable. Don\u2019t become a specialist, be an expert in your core discipline and be open to others too. The first is 20th century (even 19th C) thinking, the second is resolutely 21st century.<\/p>\n
But don\u2019t reach for the strong drink; you don’t have to be a jack of all trades, and this isn’t about pretending to be something you’re not. This is all about being more effective and more flexible, it is about having the ability to appreciate issues, processes and possibilities outside your core area of expertise.<\/p>\n
Take standard four-colour printing, for example. Sometimes looking at the print production process from a different point of view can throw up non-standard ways of getting to a creative end result. Normally, to recreate a full-colour image in just cyan, magenta, yellow and black inks the image will be separated into four greyscale channels, one for each ink, and then output as halftone dots – the lighter and darker parts of each separate channel will be reproduced using larger or smaller dots in a regular halftone grid. This is old hat, the way colour printing has been handled for over a century. More recently, stochastic colour separations, where same-size dots are scattered or grouped to reproduce different tones, has been used. Traditionally, this is a process that requires specialist prepress setups, and it is still pretty unusual. But since the first Macs shipped there have been ways to turn greyscale images into one-bit, black-and-white dithered simulations of tones. The process of creating a dithered black and white image from a greyscale one is simple in Photoshop, although this only gives you one flavour of diffusion dither. For this, the others options – 50% Threshold, Pattern Dither, Halftone Screen, and Custom Pattern – aren\u2019t any use to us.<\/p>\n
Believe it or not, there have been many different kinds of \u2018diffusion dither\u2019, each with its own slightly odd name. Floyd-Steinberg dithering is the true name for Photoshop\u2019s method, but Jarvis, Judice and Ninke (shortened but not improved to JaJuNi) is another one of many. the tech obsessives among you can refer to caca.zoy.org\/wiki\/libcaca\/study\/3<\/a> for a good summary of many different methods and how they all work, and the Mac history buffs should see bit.ly\/thunderscanhistory for an innovative early Mac use of this stuff.<\/p>\n Pick a strong image with bold colours and contrasts to start with, not anything too delicate. The resolution of your image is critical, as you\u2019re going to use the actual pixels as the printed dots in the final output. This means you\u2019ll need to think about the size it\u2019ll be in your layout then dial that in along with the pixels per inch value using Photoshop\u2019s Image Size dialog. For this, 200ppi (not 300ppi) is a good starting point – but you should definitely experiment when you can. (Do try resolutions of 300ppi and higher, but talk with your printer first as it gets hard to reproduce such fine dots cleanly and the paper choice becomes increasingly critical.)\u00a0If you’re a little confused by this then take a look at the article about resolution<\/a>.<\/p>\n Convert the image to CMYK, then go to the Channels panel and select the Cyan channel. Copy this, paste it into a new document, then go to Image > Mode and convert this from greyscale to \u2018bitmap\u2019, using the Diffusion Dither option. Copy and paste back into the Cyan channel and repeat with the Magenta, Yellow and Black channels.<\/p>\n When your image is zoomed out in Photoshop to 25% or smaller it\u2019ll look pretty much like the original continuous-tone image, but zoom in to 100% or more so you see every pixel cleanly and it\u2019ll have a very different feel. Don\u2019t get too worried though, the effect will be much smaller in print; it\u2019ll still look \u2018different\u2019 but it won\u2019t really be quite the explosion in a confetti factory that it might seem here.<\/p>\n At this stage you might decide that, for example, the black channel could do with a contrast boost to plug up the black shadows and hide the sparkling colour pixels. The trouble is, now that it is all dithered… you can\u2019t. This image is beyond regular manipulation, so you\u2019d need to undo a few steps first. Because of this, always keep your original un-dithered image safe.<\/p>\n Finally, save the image in TIFF or PSD format and import it to your layout. Check that the output resolution, or effective resolution as InDesign\u2019s Info panel calls it, is your desired value. If the image is scaled up or down on the page this will change, affecting the pixel-dithered screening effect dramatically.<\/p>\n Okay, this isn\u2019t rocket science. It isn\u2019t even particularly new. But what it certainly is is unusual. The printed results won\u2019t be perfect photographic recreations of the original image; there will be a slightly pointillist feel, particularly with slightly larger dots, and the colour can be a little \u2018sparkly\u2019. However, it does actually work, and it has a quality that sets the result slightly apart from run of the mill print output.<\/p>\n<\/a><\/p>\n
![\"\"](\"https:\/\/thatkeith.com\/articles\/wp-content\/uploads\/2011\/08\/13-dither-conversion-1-1.png\")
<\/a><\/td>\n![\"\"](\"https:\/\/thatkeith.com\/articles\/wp-content\/uploads\/2011\/08\/13-dither-conversion-2.png\")
<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n![\"\"](\"https:\/\/thatkeith.com\/articles\/wp-content\/uploads\/2011\/08\/13-dither-conversion-3-300x188.png\")
<\/td>\n![\"\"](\"https:\/\/thatkeith.com\/articles\/wp-content\/uploads\/2011\/08\/13-dither-conversion-4.png\")
<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n