Summer 1991

The desktop repro-house

Tools: Colour reproduction

Do-it-yourself reprographics are just around the corner, according to the suppliers. Brett Wickens urges caution

Asking a graphic designer if he recommends desktop repro is like asking a general practitioner if he recommends coffee-table surgery. Repro, or more specifically, colour reproduction, requires immense skill and talent. Presuming that designers want to know the technical intricacies and language of colour reproduction is specious and misguided.

The last few years have seen a renaissance of the physical design process. Typesetting and artworking, the traditional ‘analog’ processes of the designer, were among the last disciplines to be significantly altered by technological change, but have now been supplanted by a stream of modern digital communication whereby designs are as likely to be created and submitted on floppy disk as they are on paper.

The need for change was obvious. Designers, in a move to remain efficient, embraced the technology as the world around them opened up in terms of information transfer and access. Arguably, this has been a good thing, as it allowed designers to keep pace with their clients, who most certainly integrate technology at many levels within their organisations. It has also served to make more interesting the mundane facets of graphic design such as paste-up and artworking.

The down side to this process is twofold. First, practising designers fall victim to a variety of claims by hardware and software manufacturers, each extolling the virtues of their respective innovations. In some cases, these claims have been little short of incredible.

The second, and perhaps more serious problem is that young designers enter graphic design through an unrealistic world of screen-based information. Electronic pages are not printed pages – they have no tactile qualities. Special processes, such as embossing or foil stamping are foreign in this world. Colour becomes a weak red-green-blue approximation of ‘reality’, and unless you have expensive equipment, you are limited to viewing 256 colours at a time – versus the millions that the eye is able to detect.

As designers, we now know the joys of being able to compose typeset pages in-house through the use of the PostScript page description language. In a way, this has been a natural progression (much to the dismay of conventional phototypesetters). Designers always had a unique relationship with their typesetters; they understood the language in detail and were able to specify precisely what they required. The same empathic relationship cannot truly be said of designers and colour reproduction houses.

For most of us, the traditional repro process involves giving a repro-house a transparency, a crop guide and some base artwork. The repro expert scans the image, possibly on a Scitex colour repro computer, makes films and gives us back a colour Cromalin proof. Sometimes, if the job is really critical, he also makes plates and provides us with press proofs on the paper of the job. We compare the transparency and the proof and make, in general, non-technical comments regarding the quality and colour; it may then go back for re-scanning or proofing.

Few designers know exactly what happens during the process. From experience, we have learned that if colour proof doesn’t look right, then the best comment to make to the repro-house is ‘match-transparency’. Not ‘adjust the cyan dot gain minus fifteen percent’, or ‘perhaps your monitor gamma needs adjusting’. This is not to say that designers aren’t conversant with the terminology and processes of repro-houses, just that you are dealing with people who have spent many years learning their craft, as designers themselves would argue regarding their own profession.

So why is it that technology merchants are trying to convince us that desktop repro is just around the corner? Having saturated the market with electronic replacements for conventional artwork problems, technology suppliers are probably now viewing the repro part of the design process with a gleam in their eye. After all, repro-houses have been using computers for years – systems such as Scitex, Quantel and Crossfield. Heidelberg have just announced their completely digital PostScript printing press, eliminating the need for printing plates. As design companies have been modernising at a more or less parallel rate, it would seem sensible to integrate the various systems.

As any designer knows, working on screen is not easy. True, a complex piece of retouching can be achieved quickly and relatively painlessly on a Quantel or Scitex system. The problem is in the colour and quality differential. Screens are, by their nature, a source of transmitted light. Since the colour that we perceive is defined by reflective light, screen images too often look bright and unsubtle as a result.

The type of paper on which an image is printed also has an enormous effect on the quality of the reproduction. Repro software now has built-in transfer mechanisms for approximating the effect of printing on different stocks, but again, like the colours themselves, these are only approximations. Add enough of these approximations together and, at best, you have a compromised result, at worst, an unprintable one.

Colour proofing from these systems also leaves a lot to be desired. Thermal transfer and dye sublimation ‘hard copy’ used to be the most popular methods of providing immediate colour output from a system. Because of the comparatively low resolution of the imaging mechanism, the quality was invariably poor. Recently, however, we have seen the introduction of the Iris ink-jet proofer which actually sprays ink on to paper. The quality of colour reproduction on this system is very often superb – too superb in fact, because it prints a continuous tone image more akin to a photograph than to a half-tone screen. Moreover, it doesn’t handle type, especially small sizes very well.

The introduction of colour reproduction technology into companies will of course be dependent upon their requirements. There are very many successful installations of such equipment in the world, but they are primarily used for short lead-time publishing, where ultimate colour reproduction is less important than making a deadline. These installations, which have state-of-the-art input, manipulation and output terminals cost a great deal of money. Some of the larger US publishers spend nearly $4 million on their systems, in return for long-term savings.

At the other end of the scale, what can a small design company do to enter this area? By far the most sensible approach is a system based on a happy relationship between the designer and the repro-house. This may sound superficial, but look at the acrimony that resulted from designers forcing typesetters into becoming bromide bureaux. Many of them simply went out of business, taking with them their expertise. If we allow the same to happen with skilled repro technicians then we are in big trouble.

Quark Xpress software is generally the designer’s choice for page composition on the Macintosh. Scitex, realising that some kind of link between designers’ programs and their high-end systems would be desirable (and profitable), joined up with Quark to create a version of Xpress called Visionary. They also developed interpreters to allow the transfer of PostScript type and graphics on to their workstations for prepress correction and output.

The principle, which has now been adopted by other manufacturers, is simple. Images are scanned conventionally by the repro-house, which then provides low-resolution colour images on disk to use in your page make-up. You give the completed electronic document back to the repro-house when finished, and the mechanism flags-in the original high-resolution scan using your cropping and position data. The result is a set of usable separations with a minimum of planning by the repro-house. All image manipulation and colour balancing is left to the expertise of the repro-house. And you can still be given a variety of proofs – Cromalins, Iris, or press proofs, depending on your requirements.

For the foreseeable future, I cannot image a better link between design and reprographics. This workable relationship can be achieved with most existing Macintosh installations for relatively little expenditure. It leaves the expert parts of the operation in the hands of the experts, keeps the designer free to concentrate on designing, and avoids the need to install and train dedicated operators in-house, where they are forced to embrace the steep learning-curve associated with complex systems. The cost is palatable, and fidelity between idea and result can be maintained by the proofing method you choose.

Sadly, the rise in use of the term ‘desktop’ to describe user-friendly approaches to expert disciplines such as publishing, reprographics and design, has led to a decreasing awareness of what goes into being a skilled professional. Hardware and software manufacturers should re-evaluate how they address these professions to avoid belittling their skills and talents. Software does not make a mediocre designer an excellent one, nor does it make a designer a repro-expert or even a typesetter.

We are in danger of wrongly trying to blur the distinction between crafts, reducing the whole process to an oversimplified string of compromises. We should now be concentrating on fostering respect between disciplines which are vital for ensuring that what we create is realised satisfactorily and professionally.

Brett Wickens, designer, Pentagram, London

First published in Eye no. 4 vol. 1 1991

Eye is the world’s most beautiful and collectable graphic design journal, published for professional designers, students and anyone interested in critical, informed writing about graphic design and visual culture. It is available from all good design bookshops and online at the Eye shop, where you can buy subscriptions and single issues.