Boris Oicherman explores the highs & lows of colour management over the last 25 years and looks at some of the myths surrounding colour calibration.

Welcome to 20 years ago...

...Colour management is suppose to get us where we all where twenty years ago...

Tony Johnson, the former Technical Secretary of the ICC and one of those who defined colour management as we know it today, used to say this. Sounds like a paradox?

Let's remember colour management as it was at the beginning of the 80's. A printing house purchased a scanner for a six-digit figure. The scanner arrives in several boxes, accompanied by the service engineer who installs it and spends anything between a day and a week adjusting the scanner settings in order to produce a good set of separations for the offset press next door. A specially trained person called a scanner operator spends a considerable amount of time learning to play with all the knobs and buttons in order to control the scanner's output. Finally, he runs a test scan and produces a set of colour separations. After those separations are transferred to the plates, another professional staff member, the offset press operator, uses his skill, knowledge and some 'black magic' acquired through years of experience, to produce the final result of the whole operation: a high quality printed colour image. The scanner operator and the printer compare the scanned original picture and the print; they find their colours different, but nevertheless consider the result satisfying; they both knew what to expect, and they got what they expected. One scanner calibrated against one offset press and it worked as long as they were operated by staff who knew their jobs. They produced high quality prints... Judged also by today's standards.

The situation started to get more complicated in the 90's. Everything which could become digital started to become digital, in a strict accordance with Benny Landa's, the founder of Indigo Vision. As the personal computers and digital imaging technologies developed, the prices dropped, and scanners and printers became consumer, rather then industrial, products. Digital images were now originated by designers, photographers and artists, the very profession of scanner operator had started to vanish.

Developing global networks complicated the situation even more. The closed-loop colour model of one scanner, one printer clearly could not work any more, as the image files began to travel easily around the globe via the Internet. The originator of the image could not possibly know which printer would produce the scanned image. The press operator had no idea where the image is coming from. As the technical complexity of managing the image colour data increased, the quality expectations raised. Also every product which bore the label 'digital' was expected to perform at its best straight out of the box, while the definition of the 'best' was often left at the discretion of the user.

In this situation, a solution which would make this colour system manageable was impatiently anticipated in the market. Such a solution was expected to allow an unlimited number of scanners to prepare jobs for an unlimited number of printing presses. Whilst any number of other devices which capture or reproduce colour could also be added to the system at any time. In other words, this would be an 'open colour system' as opposed to the 'closed-loop colour' model.

The solution arrived in 1993 from a consortium established especially for this purpose; the International Colour Consortium (ICC). System architecture was defined, which included two major components, a 'Device Colour Profile' and a 'Profile Connection Space' (PCS). The colour profile contains information on how the scanner captures colours, and the printer profion how it reproduces colours. Also, and most importantly, instead of transforming the colours directly from the scanner to the printer, like in the closed-loop system, a 'mediator' is used, the PCS, an abstract non-existent device, the only purpose of which is to be able to communicate with all possible real devices. This is what became known as 'Profile-PCS-Profile' model, and it is what makes the system open: all the scanners talk to PCS, and the PCS talks to all the printers. Any number of scanners can be added to the system at any time, provided a colour profile for each is created, while no knowledge of the destination press is required. So, an open colour management system was established, and the problem could be considered to be solved. In theory anyway

However, the new technology met two major problems. On the one hand, naturally for any new technology, the implementation was far from being perfect. On the other, the customer's high expectation from the new digital technology. There was a clash, and the notion of, "colour management does not work," became a commonly heard phrase almost instantly, as the first colour-managed products hit the market.

What did this notion mean? It certainly meant that the products did not answer the expectations. The expectations, in their turn, were shaped largely by the combination of advertising slogans of the development companies and a lack of technological understanding of the market. They could be formulated as maintaining the 'correct colour' across the printing production chain. The 'correct colour' being a philosophical, rather than technical term, could not be delivered, simply because no one knew what it was. Thus, it was gradually replaced by 'consistent colour' which was at least technically definable, though not really achievable. This was due to colour gamut limitations inherent to different imaging devices. The scanner can capture colours which are not reproducible by the printer, who in turn can generate colours that are not reproducible by the monitor, meaning that the colour can be maintained consistent only within some fairly narrow limits.

Slowly but surely the term 'predictability' began to be used. If the scanner cannot scan all the colours in the original artwork and the printer cannot print all the scanned colours, what is left is to be able to 'predict' what is going to happen to the image and to take all necessary measures to minimise the damage. If the designer would be able to 'predict' how the colours will be printed while working with the page layout software, the design can be adjusted accordingly. If the photographer who uses image editing software can see how the photograph will change when printed in the magazine, he can make the necessary adjustments to his set-up. In other words, whoever prepares work for printing should 'know what to expect'.

But was this not exactly the function of the scanner operator in the eighties, to know what to expect from the system, and to use this knowledge in order to achieve the maximum quality within its limitations? It is this ability to predict the final colour which is the most important feature of colour management. A piece of software which will do the job of the professional from the past, yet not only for the multi-million printing house with one scanner and one printing press, but for every prepress operator, photographer, designer or secretary who receives an image from an unknown scanner and prepares it for print by an unknown printer.

Was this not what Tony Johnson meant? Welcome to 20 years ago!