DCS, OPI and friends
DCS
DCS stands for Desktop Color Separations. It is a variant of the EPS format invented by Quark to make working with colour separations easier.
The original DCS format (version 1) had five files. The master file contained a colour, but low resolution, version of an EPS graphic. This is used for proofing. Then there are four other files containing the cyan, magenta, yellow, and black separations. Each of these is a black and white graphic (since separations do not include colour information). As Quark (or whatever) prints colour separations it includes the appropriate separation file from each included DCS graphic. The master file includes comments giving the names of each of the separation (plate) files. Because of file naming problems, DCS version 1 is not usually portable between Macintosh and DOS.
DCS version 2 was expanded to support extra plates (spot colour) and to allow a single-file version of DCS. The single file is just the separate files joined together with a header - it has to be split up before any part can be printed.
OPI
OPI stands for Open Pre-Press Interface. It was invented by Aldus. It is not a file format but a way of working with smaller files to speed up DTP work. It relies on faking EPS files. A page might use 10 megabytes or more of EPS files, and this would just be too slow if it wasn't for OPI replacing this with a few kilobytes.
With OPI, graphics are placed into the system, and handled by an OPI input processor. It hides the original ('hires') graphic somewhere, then makes a much smaller file ('lores') for the user to use in place of the original. This is best called a 'stub' file, not lores. It keeps the preview from the original EPS.
A DTP program doesn't look at the PostScript in an EPS file, except for the header comments. It just uses the preview, then when printing, dumps a complete copy of the EPS file into the print file. OPI relies on this by removing all of the contents of an EPS file except for the header. It then adds a special comment which it can recognise, giving the name of a place where the original EPS file has been kept.
As a file is printed it has to pass through an OPI process. It looks for the comments it inserted, and when it sees them, it puts back the original files that it has been keeping.
There is a standard for OPI, but each implementation I have seen has some major variations from the standard. Or standards - there are two official standards. You can download a copy of the original standard (58k) or the new standard (270k) (better, but widely ignored) from Adobe's ftp site in Acrobat PDF format.
Low resolution
The term low resolution or lores is used to refer to three different things, to avoid confusion.
- One is the preview part of an EPS file, usually a bitmap at screen resolution.
- The second is the master file in a DCS graphic, which contains PostScript suitable for proofing, not typesetting.
- And the third is the stub file generated by OPI processing.
Printing without an OPI
When you print a page that has used OPI (stub) graphics, it's important to make sure that the OPI gets a chance to put back in the original graphics. Most systems will set this up automatically, but in some cases you might be able to print directly to a printer. In this case the stub files are not expanded. What happens?
With some OPI products, the stub file has no image in it, so it will print as a blank picture. With other OPI products, the stub file contains a rough-and-ready version of the original picture, enough to give an idea of what it looks like when printed.
See also
EPS in ten easy stages
Making an EPS file, with a shameless plug
for PSAlter and its facility to convert
PostScript to EPS.
Go back to our PostScript introduction
Go to Quite At Home.
