American Printer's mission is to be the most reliable and authoritative source of information on integrating tomorrow's technology with today's management.


Mar 1, 2000 12:00 AM

         Subscribe in NewsGator Online   Subscribe in Bloglines

Can't sleep? Forget about counting sheep. Try counting the number of file format options available for saving files out of PhotoShop. There are more than 21-not including special formats supported by any import and export plug-ins you may have installed. Why do we need so many formats? Is one better than the other? Which format will maximize your production efficiencies? Before you can answer these questions, you need to understand the file format basics.

Raster and vector | These are the two main file types used in PostScript page production. Vector files describe objects using lines and shapes-e.g., files created in Adobe Illustrator or Macromedia Freehand. Vector files are either processed independently or imported into a page as an element. Because vector files are resolution independent, they can be output on many different devices at virtually any size. Vector file formats typically are limited to either the internal file format of the application or EPS (encapsulated PostScript).

Raster files are described as pixel (picture element) values. The amount of pixels and, in turn, the size of the file, are dependent on the resolution of the file. Typical raster files are scans and files created in paint programs. These files are resolution dependent, limiting their output value to a small range of sizes, based on the input or creation resolution values.

Many different raster file formats have emerged as application developers try to optimize the files that they create to handle tools and features available in their specific applications. Adobe, for example, has developed its own PhotoShop format in an effort to handle the features available in PhotoShop, such as channels used to store color plate or mask information. Scitex and other companies have developed special file types such as LW, which included RLE (run length encoding) to allow for good compression and edibility on these raster line work files. There are also special native OS (operating system) formats for many Unix, Macintosh and Windows products. While each of these file formats is different, they do have one thing in common-all describe pixel information.

Common formats EPS--Originally developed by Adobe Systems to support Adobe Illustrator and PostScript, this format supports both raster and vector information, making it ideal for images with clipping paths.

TIFF--A raster file that can support transparent backgrounds, colonization and some other features. Aldus originally developed it for use with PageMaker.

1Bit TIFF--This raster format is usually used in files where there is a need to describe only black or white pixels. It is commonly used to scan black-and-white line drawings and as a means of saving and exchanging scanned halftone-film files.

DCS--DCS was developed by Quark to extend support for pre-separated PostScript workflows. DCS 1.0 uses four high-resolution files, one for each color, since it is pre-separated, along with an EPS placement file. DCS 2.0 is a similar file format to DCS 1.0 except that the individual colors are not represented as individual color plates. Instead, they are represented as separate pages within a single file, making organization a little easier. Since DCS 2.0 is a pre-separated file format, it can also be used for scanned halftone screen files.

PDF--From its modest beginnings as a file format to facilitate the use of electronic paper in the corporate enterprise, PDF now has loftier goals-addressing most of the needs of digital paper and other publishing mediums. PDF is especially suited for publishing applications because it can contain both raster and vector information-including fonts. Other key features include device and page independence and cross-platform support with an available, free reader.

Device and page independence is becoming more important as publishing and output processes change. As production environments attain higher degrees of automation, there is a greater need to display or output the same file to multiple devices. While PostScript has been the format of record, it is very dependent on specific setups for each device. It doesn't support page independence, or the ability to view or extract a single page from the PostScript file.

When prepared correctly, PDF files are complete-all of the fonts and image files are included in a single file. This same file can be viewed on multiple OS platforms, including Macintosh, Windows and Unix. Thus it is an ideal soft proofing format for content-and if a color management program is in place-perhaps color, too.

PDF 1.3 facilitates a higher degree of file edibility, making it a potential choice for both a transfer and manufacturing format. Although it has been possible to edit parts of a PDF file for some time, the necessary tools weren't readily available. The recent release of Adobe Acrobat 4.0 together with PDF 1.3 offers greater editing possibilities. Limited editing of text as well as scanned objects (raster) and illustrations (vector) is possible through the use of PhotoShop and Illustrator.

PDF/X-1--This currently uses PDF 1.3 as its basic format, but adds some additional restrictions to ensure the resultant file maintains necessary process control for advertising and publishing needs. This "refined" PDF variation was created to address some of the concerns that had been addressed with the creation of the TIFF/IT-P1 format, but in the more portable PDF format. PDF/X-1 currently supports only process colors, but a second version X-2 is being discussed in committee to address extended functionality.

TIFF/IT--P1 addresses the digital ad distribution needs of magazine publishers, advertising agencies, prepress providers and printers-the professionals who belong to DDAP (Digital Distribution of Advertising for Publication). This officially designated standard is similar to Scitex CT/LW internal manufacturing format and includes three files for each page. In addition to the CT and LW, there is also a Page file, which functions to associate the CT and LW to each other. This format currently supports only process colors.

Compression--Lossey and lossless are two basic types of file compression. Both reduce raster file size-using a specific algorithm, lossey compression discards superfluous information. Lossless compression, by contrast, allows a lesser degree of compression, but without any file information loss.

While many different compression formats are available, only a few are readily supported in most applications. These include JPEG (lossey) and LZW and RLE (lossless).

Each file format has its own place in the workflow for which it was developed. This could be a specific task such as word processing, layout, image capture, retouching, etc. Choosing the ideal format is easier if you understand the value-added features of each.

Although specific output workflows play a role in determining which format is used in actual page production, other factors also must be considered. EPS files, for example, are better suited to silhouetted images than TIFF files, because the silhouette information is natively supported. While PhotoShop does allow you to retain that same information in TIFF files, it isn't part of the original TIFF file specification. Therefore, few layout applications support that feature. But if you need image "transparency" to allow one image to overlay another, TIFF would be a better format. Time and cost factors will ultimately determine which format is used.

Picture replacement--Automatic Picture Replacement (APR) methods were originally developed to allow designers to work with images in layout programs unencumbered by high-resolution files. Scitex APR and open prepress interface (OPI) are two common APR methods. OPI uses a high-resolution TIFF and low-resolution TIFF for the placement or proxy image. In addition, most OPI solutions also can use a DCS file. APR uses a high-resolution Scitex CT, TIFF or EPS file, along with a low-resolution EPS file. In all cases, a low-resolution "placement" file is used to build a page. During processing, the high-resolution image file is inserted in the document, replacing the placement file. With OPI, this replacement usually occurs on the print server, just prior to being sent to the RIP (although many recent RIP solutions include OPI handling in the RIP workflow). With APR, the replacement occurs at the RIP itself.

Application formats--Most software applications, such as Quark, InDesign, PageMaker and Microsoft Word, have their own internal file format, specifically designed to facilitate tasks that each of those applications can perform. But in some cases, specially designed translators allow you to import another application's native format. InDesign, for example, allows you to import the Quark Xpress native file format, and PhotoShop has one of the most extensive sets of translators. Generally speaking, however, applications' native file typically can be used only in that application.

Scitex, Barco, DS, Dalim, Rampage, PCC, Heidelberg (Linotype Hell) and others have developed their own internal formats to pave the way for the smooth integration of their output workflow solutions. Such formats were developed to allow each of these companies to create trapping, imposition, editing and other tools to facilitate the whole process. Others have chosen explicitly not to have their own formats for the express purpose of being 'more standard.' This would include Agfa, Creo (Prinergy) and Shira, to name a few.

Barco's GRO format is somewhat unique in that it maintains separate raster and vector objects in the GRO format. It then uses that format in the various workflow tools and modules that it has developed.

Heidelberg's Delta format was originally developed by Linotype-Hell for its Delta workflow solution. It is a rasterized format that retains its display list information. The display list is a descriptive map of the interpreted PostScript file that retains all of the original object (both raster and rasterized vector) for future processing. It is an integral part of all RIP processes.

Scitex LW/CT--The CT format is an uncompressed raster file that includes all of the original raster picture information assembled according to the final layout. The Linework file contains any information that was originally vector prior to rasterization and was rasterized at a higher resolution than the CT in order to maintain the necessary detail required for fonts, and other vector graphics. Since this raster file could be very large due to the resolutions used (typically in the 1800 dpi to 3500 dpi range), Scitex uses RLE compression to reduce the file size. In an effort to address a 248-color area limitation of the original LW format, Scitex developed NLW. This new format extends the available color areas to about 16,000, which allow for smooth LW blends and other value-added benefits.

Rampage and PCC both use a different but similar LW/CT format to that of Scitex. The POM format of DS is a rasterized object format that maintains object differentiation to support many of the features of its Taiga production systems. Shira uses the TIFF/IT format for its internal manufacturing format, although it isn't TIFF/IT-P1, a distinction that allows it to work with an extended color palette beyond process. Dalim uses its own internal manufacturing format that retains raster, vector and text information on separate layers for subsequent editing.

The advantages offered by each of these specialized file formats are evident when you look at the original applications and platforms. Nonetheless, the sheer number of available formats can make integrating work from different applications or computer platforms confusing and inefficient. Understanding how the file is used throughout the entire life of the process can help you determine the best format for your workflow.

Most file formats were developed to support a specific software application or process. What happens if you need to exchange the file with someone who lacks the specific application used to create the file? Better get that Snickers bar-you're not going anywhere for awhile. While some internal descriptions of file formats have been published to enable developers to build support in other applications, developing file format support can take a lot of time. As software applications mushroom and digital processes proliferate, a dizzying number of file formats are emerging. Supporting all of them in every application is impractical-hence the need for standards.

The road to standardization begins with industries negotiating among themselves to develop an acceptable format addressing global needs. Once the file format structure is agreed upon, an application to the international standards committees can be made to officially designate that format a standard. The internal structure is then published and available to virtually anyone who wants it.

Sounds like the ideal solution to file exchange problems. But the process of negotiating and applying can take significant time- three to five years, in many cases. A de facto standard is one way to avoid the wait. Under the de facto scenario, a company or groups of companies promote the active use of the format. PostScript, EPS and TIFF are all de facto standards-widely accepted and used, but not officially blessed.

De facto standards, however, can pose certain challenges. In the case of TIFF, there were many revisions to the format, which was originally created by Aldus Corp. in the early 1980s. Many of the other industries couldn't keep up the support of the changes in their individual applications. What could have been a way for Aldus to retain control of the industry merely resulted in production difficulties. As a result of this and other problems, industry groups are pushing for a select group of standard formats for the publishing process.