PDF and PDF/X Standards Demystified

PDF and PDF/X Standards Demystified

Initially, the concept of PDF/X was intended to facilitate the complete exchange of reliable digital data, particularly in a publications advertising workflow. Complete exchange means the file can be relied upon to be correct without further technical discussion or involvement.

PDF/X is not yet another file format—a PDF/X file is a PDF file. More specifically, however, it is a restricted subset of the PDF format. What’s a “restricted subset?” Well, PDF files can contain lots of elements that have nothing to do with printing, like sounds, movies, and scripts. They can also contain a lot of elements that could result in incorrect output, like RGB images, missing bleeds, or fonts that are not embedded. PDF/X limits or restricts the inclusion of data that can compromise the integrity of the printed output from a PDF file. Think of it as a PDF file for print that contains what’s needed to print correctly and without the excess baggage of non-print-specific elements. In addition, the PDF/X specification defines how a prepress production tool will render the PDF/X file. For example, the fonts embedded within the PDF/X file must be used when it is output rather than any similar fonts which might be resident at the output device.

Content in a PDF/X file is limited according to four “conditions”:

  • Content that is required: a conforming PDF/X file must contain this key or object.
  • Content that is prohibited: a conforming PDF/X file cannot contain this key or object.
  • Content that is restricted: where and how certain keys or objects are used is limited by pre-established rules.
  • Content that is recommended: it’s suggested that the conforming PDF/X file contain this key or object, but it does not have to.

 

Here are a few examples of PDF/X limitations. In order to be a PDF/X-1a-conforming file, a PDF file cannot contain a movie, as it has no place in print manufacturing, so movies are prohibited. Annotations, like notes, are allowed, because they do have value in print production workflows for the purpose of document approval. However, most forms of annotations will print, so if they have been placed inside of the live area of a page, they could result in faulty output. Annotations are allowed in the PDF/X specification, but only outside of the bleed area of the document; in other words, they are an example of restricted content. Specific general information (key-value pairs) like the title, creator, and producer of the file is important to print production workflows, so they are required in PDF/X documents.

Who “created” PDF/X? The Digital Distribution of Advertising for Publications (DDAP) and Newspaper Association of America (NAA), seeing the emerging Portable Document Format as a viable means to digitally deliver advertising to publications, first requested a prepress-specific standard for PDF.

The Committee for Graphic Arts Technologies Standards (CGATS) originally developed the PDF/X standard (DDAP and NAA are voting members of CGATS). The first PDF/X standard was ratified by CGATS and ANSI in 1999. Adobe is an active member of the standards bodies responsible for the development the PDF/X standard, but PDF/X is not, as some people erroneously believe, another Adobe product. It is a completely non-proprietary accredited standard for the exchange of digital data.

As with any other application or file format, PDF/X is ever evolving. The first format, PDF/X-1, officially called ANSI PDF/X-1:1999, was based on PDF 1.2. By the time it was ratified, Adobe had released Acrobat 4, so PDF 1.3 was available in the marketplace. In a sense, it was outdated before it had a chance to be of much practical use. The International Standards Organization (ISO) expanded the ANSI standard in a project originally called ISO 15930. The committee decided that there was a need for several types of PDF/X files based on different production workflow environments. ISO ratified a version of PDF/X in 2001, officially called ISO 15930-1:2001. This replaced the old ANSI PDF/X-1:1999 and was based on PDF 1.3. PDF/X-1a:2001 was intended for the blind exchange of digital data primarily for ad distribution. Originally, there were two conformance levels: PDF/X-1:2001 and PDF/X-1a:2001. The difference between them was small. PDF/X-1a:2001 prohibits the use of encryption as well as OPI objects; otherwise, the two were identical. Since then, PDF/X-1 use had entirely ceased and only PDF/X-1a is actually implemented.

In order for a PDF file to conform to the PDF/X-1a specification, it must meet certain criteria. All fonts must be embedded and legally embeddable. Images must be encoded as CMYK, DeviceGray, DeviceN, or spot color spaces. No RGB or Lab color is allowed. The trapping key must be set to “on” (true) or “off” (false). The file must contain correct media box and trim box or art box definitions. The trim box or art box cannot exceed the media box array. The media box is established by the paper size defined when a file is printed to an output device from a layout application. Think of it as “virtual paper.” The media box must be of sufficient size to encompass the trim size of a document (and should be large enough to accommodate bleed and outside objects like crop marks as well).

PDF/X-3, originally called ISO 15930-3, was originally ratified by ISO in 2002 and is very similar to PDF/X-1 except that it allows device-independent, or “three-color,” color spaces, like RGB or Lab. While PDF/X-1a is geared entirely to CMYK-based print production workflows, PDF/X-3 can be used for workflows in which alternate color spaces are important, such as for photographic output or fully color-managed environments where ICC profiles are applied to images or other color elements based on a particular output intent.

PDF/X-2 is based upon the PDF/X-3 specification but allows for the “partial exchange” of data. Partial exchange means that some of the data required for the successful output of the file can reside at the output provider’s site, not necessarily within the PDF/X file itself, including such things as fonts or image files. This is very different from the “blind exchange” of PDF/X-1a in that it requires some sort of communication between the sender and receiver of the file. It is intended for OPI-like workflows, in which the print provider may keep high-resolution copies of images, while the content-creators work with low-resolution proxies of those images. When the PDF/X2 files arrive at the print shop, the high-resolution images can be swapped for the low-resolution images.

The entire PDF/X family was updated in 2003 to be based upon the PDF 1.4 specification. Since then, several new members have been added to the family and the purpose of the specification had been expanded from the original goal, of helping with the exchange of prepress files, to helping preserve and aid distribution of other types of content. NPES (The Association for Suppliers of Printing, Publishing and Converting Technologies), has long been the secretariat of the ISO TC 130 group of standards, based on that organization’s deep interest and experience in the graphic production world. Visit www.npes.org and navigate to the Standards Workroom if you’re interested in looking at the actual minutes of the meetings where industry experts hammer out these standards. The actual published ISO standard is not freely distributed but must be purchased. It is available directly from the ISO Central Secretariat (www.iso.org) or, in the U.S., from ANSI (www.ansi.org).

Here’s an overview of the current lineup of PDF/X standards:

PDF/X-1a:2003 (ISO 15930-4:2003) 

PDF/X-1a: 2003 is the current conformance level of the PDF/X-1a standard and was updated to include newer technologies:

  • Transparency—while transparency is part of the PDF 1.4 specification, partial transparency in PDF/X-1a is prohibited. If transparency is used to create content, it must be flattened before converting the file to PDF/X-1a:2003.
  • JBIG2 compression—JBIG2 is a standard for bi-level image compression and very effective for things like copydot scans. It is prohibited in PDF/X-1a:2003 apparently because of difficulty obtaining licenses to the intellectual property from the JBIG2 standardization group.
  • Encryption—remains prohibited. PDF/X-1 (without the “a”) has been eliminated.

 

PDF/X-2:2003 (ISO 15930-5:2003)

PDF/X-2:2003 is the new conformance level of the internationally accredited PDF/X-2 standard. It is considered a superset of PDF/X-3:2002 because it allows for the use of device-independent color spaces like Lab and RGB based on ICC profiles, but the exchange of that data is not blind. This version of the PDF/X family was only first ratified in 2002 and has been little used in production. It’s expected to be used on OPI workflow scenarios.

PDF/X-3:2003 (ISO 15930-6:2003)

PDF/X-3:2003 is the current conformance level of the PDF/X-3 standard. PDF/X-3 allows for the use of device-independent color spaces such as Lab and those based on ICC profiles as well as CMYK. Like PDF/X-1a: 2003, it has been updated to include these new technologies:

  • Transparency—while transparency is part of the PDF 1.4 specification, partial transparency in PDF/X-1a is prohibited. If transparency is used to create content, it must be flattened before converting the file to PDF/X-3:2003.
  • JBIG2 compression—JBIG2 is a standard for bi-level image compression and very effective for things like copydot scans. It is prohibited in PDF/X-3:2003 apparently because of difficulty obtaining licenses to the intellectual property from the JBIG2 standardization group.
  • Encryption—remains prohibited.

 

PDF/X-4 (ISO/DIS 15930-7) 

When you export a PDF file from an Adobe Creative Suite 3 application, take note that there is a new PDF/X option available for PDF/X-4: 2007. To date, PDF/X-4 is under development and is a draft only. Based upon the PDF/X-3 standard, the biggest difference between PDF/X-4 and X-3 is that it allows for live transparency in the file. PDF/X-4 lets users take advantage of RIPs that can directly digest PDF files, such as those based on the Adobe PDF Print Engine, which don’t require transparency flattening, as PostScript RIPs do.

PDF/X-5 (ISO/DIS 15930-8) 

Derived from the PDF/X-4 specification, PDF/X-5 allows for externally referenced images and profiles.

Tags:
By WSPA

Leave a Reply

Your email address will not be published.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

*