PDF 2.0 is the first real update since 2008, when Adobe handed PDF 1.7 to the ISO for future developments. 

PDF 2.0’s print-related new features can be summed up as improving automated workflows, especially across multiple printers and papers, and better colour handling, especially spot colours. How much it will be noticed is debatable, as people in the know suggest its main significance is to pave the way for the next-generation PDF/X-6, still in development for release in the middle of 2019. 

At the beginning of July, Adobe made a fairly low-key announcement of APPE 5, a new generation of its print file processing technology, Adobe PDF Print Engine. This is the technical underpinning licensed by many pre-press workflows and front-ends for platesetters and digital presses. Most people think of it as a RIP, although technically it’s classed as a renderer. 

While APPE 5 has useful technical tweaks that will work their way into pre-press engines over the next year, its more major significance is to implement important features of PDF 2.0, clearing one of the barrier to adoption. APPE’s main rival in the market, Global Graphics’ Harlequin, already supports PDF 2.0 features. However APPE still has the lion’s share of licensees and end-users in platesetters and digital presses. 

PDF 2.0’s long gestation is partly due to the curse of committees. It’s been under development by the ISO committee TC 171 SC 2 WG 8 since 2009 and was finalised as ISO 32000-2:2017 and published on 28 July last year. 

Adobe released the first PDF 1.0 in 1995, originally intending it as a universal digital document transfer format that could replace fax transmissions. It was followed rapidly by the much more press-friendly v.1.2, while v.1.4 was the breakthrough that saw PDF start to be adopted for print production in place of PostScript, EPS and native layout formats. 

In 2008 Adobe gave ISO the rights to PDF 1.7, to become ISO 32000-1. Since then, very little has changed in terms of new capabilities. PDF 2.0 could be regarded as PDF 1.8, but presumably the developers wished to emphasise that it’s no longer a proprietary Adobe format.

Since 2008, the only real PDF developments have been refinements of PDF/X, which define what can and can’t be included for particular ‘blind exchange’ applications. Although PDF/X is also controlled by ISO committees it gets a lot more oomph thanks to the very active Ghent Workgroup (GWG), a voluntary club of developers who create sub-sets of PDF/X for real-world needs in applications such as newspapers, magazines, flexo, digital, etc.

“GWG had significant input into the development of PDF 2.0,” says David Zwang, chair of GWG. “Our involvement was centred around two areas. The first has to do with colour space, especially for packaging. We tried to do it in PDF/X, but then found that the real problem is PDF, which just didn’t have that support. If something doesn’t exist in PDF it can’t exist in PDF/X either. For publication and commercial print the colour space support within PDF 1.x has been okay, because you’ve got CMYK, RGB, named colour spaces. But packaging involves things that go beyond just the named colour space, i.e. the root of colour, which is spectral.”

The other area GWG pressed for was the inclusion of process information, Zwang says. “PDF is normally thought about as content: ‘This is the content, this is what you’re going to print, these are the pieces and parts.’ But there’s always some process along with that that may or may not be printable. For instance, how do I indicate Braille? If I’m indicating a dyeline in a multi-vendor situation, how do I automate that process? What about multiple languages, where a dyeline is not called a dyeline? So we wanted to make sure that what people tend to call layers but are really called optional content in PDF, could support some of the nuances required to take this to the next step. But that’s not as significant an area as colour, which was the biggie!”

Since its publication last year PDF has gone through the classic chicken-and-egg stage. There’s been little point for design and layout application developers to add PDF 2.0 features, as there’s been nothing that could print them. On the other hand, the pre-press and print systems makers didn’t rush to support PDF 2.0 output because their end-users weren’t seeing any input files. 

APPE 5’s coming support for PDF 2.0 changes the equation – front-ends will get the new features as part of the normal APPE upgrade cycle and they’ll sit there quietly waiting for PDF 2.0s to reach them. Harlequin front-ends have already got this. 

Adobe steadfastly refuses to comment on future developments, but it’s safe to assume that now APPE can output PDF 2.0, it will soon start giving its millions of Creative Cloud users the ability to generate PDF 2.0 files as part of the automatic update cycle. A year ago Mark Lewiecki, Adobe senior product manager, blogged about PDF 2.0: “Adobe product teams are studying the workflow implications for users of Acrobat DC, InDesign CC, Illustrator CC, Photoshop CC, and the Adobe PDF Print Engine to see how new application features or tools could be developed to take advantage of new PDF capabilities.”

New capabilities

So what is PDF 2.0 and how much will it affect its creators and printers? In some ways it’s a clarification and a tightening of the definitions of existing features to enforce more consistency. This finally remove the arguments of whether CorelDraw or QuarkXPress PDFs are more problematic than ‘real’ Adobe output, and whether APPE and Harlequin do different things with, say, transparency (answer: yes, they have done, even though both conform to the standard). 

Here’s a quick run-down of the main print-relevant features of PDF 2.0, for which we’re grateful to Martin Bailey, chief technology officer at Global Graphics and Andrew Bailes-Collins, senior product manager at pre-flight developer Enfocus, for their explanations of what they actually mean in practise.

The transparency has been significantly rewritten for consistency. Previously it was possible for two RIPs to produce the same file differently, but both be correct to the PDF specification.

Output Intents can be set per page. These record how the file was designed and proofed, what printing process it was designed for and what ICC profile or colour standardisation was used. The page level lets you use different papers and print processes for say, covers and inners in the same book or magazine. 

A new Output Intent Dictionary supports viewing and proofing of PDFs with DeviceN and other spot colours. An important feature is to allow embedding of CxF spectral information for spot colours. CxF’s adoption has been held back because Adobe Illustrator and InDesign don’t support it, but given APPE 5’s CxF support it’s likely that they will do in future. 

Black Point Compensation can be specified at a page or object level rather than across the whole file. BPC adjusts the tone curve preserve shadow details given the maximum black achievable by the printer. 

Document Parts (Dpart) has some crossover with JDF. It puts job data within the PDF, describing how the job is to be printed and finished, so that for instance different pages or covers can be assigned to separate presses or paper input trays.

HTO (Halftone Origin) is mainly for consistency in step and repeat work – it makes sure that every item has exactly the same screening point of origin, so the rosettes are identical in each.

There are plenty of other new features, but they’re mainly to do with security, archiving, 3D, non-Roman fonts and coding. Others aid “accessibility”, which basically means Braille, reformatting for large type, and pronunciation hints for computer-to-speech reading. 

Next up: PDF/X-6

Bailes-Collins feels that APPE 5 in digital front-ends won’t initially make much difference, saying “The creation of PDF 2.0 based files won’t really start until Adobe supports PDF/X-6 in their applications, and I don’t expect that until late 2019.”

Adobe’s Lewiecki agrees, saying “PDF 2.0 is not a game-changer for commercial print workflows. Almost every new capability can be performed another way using existing tools and methods. But it will enable greater flexibility and automation in managing print workflows, eg via page-level output intents in multi-page jobs. It also has the potential to facilitate superior quality in a number of areas: eg by encouraging greater use of spectral data to specify spot colours. Note that PDF/X-6 compliance is essential to make use of some of the PDF 2.0 capabilities. But until the spec is published, probably sometime next year, don’t expect widespread support for the new features.”

So far PDF/X-6 is still in discussion with the aim of finalising it in October and publishing it (as ISO 15930-9) in June 2019. “PDF/X-6 is basically a PDF/X wrapper around PDF 2.0,” explains Zwang “That’s why it can’t come out yet.” 

Bailey says: “The bulk of PDF/X-6 is taking advantage of page-level output intents, and that kind of thing, as well as making it easier to make files that are highly accessible and still well formed for print. There have been some moves to add in more significant changes, but none of them have been widely enough accepted to make it into the standard.”

In conclusion, APPE 5 is an important milestone, but don’t expect PDF 2.0 to change the world, at least not before 2020.