PDF: The Cure for Dawdling Digital Files
Learn the basics of creating PDF files.
Managing and maintaining the digital workflow is one of the most challenging aspects of printing today. Whether we screen print or output to digital-color printing devices, the efficiency and accuracy of graphics production depends largely on how we manipulate and move digital files throughout the process.
As we make the transition from analog to digital production, we face a formidable puzzle of digital components and procedures, with many of the pieces awkwardly arranged in precarious, unpredictable sequences. In the past, I've written about the nature of digital file preparation and how to direct customers to produce digital files we can actually use. But sadly, as noble as our attempts may be, we are unsuccessful more times than not. The complexity of the digital workflow just presents too many opportunities for mistakes when it comes to preparing and moving electronic graphics files.
Among the issues we face are an array of font-related problems, including missing screen or printer fonts, substitution of incompatible fonts (or defaulting to the dreaded Courier face), unstable font metrics (font and character spacing), and character positioning issues. Variation in font construction among the major providers, as well as differences between font types (PostScript, Bit Map, TrueType, and Multiple Master) are related issues.
We can only watch in horror as the characters we saw on our monitors are rendered in unexpected ways during printing or final output. Some progress in stabilizing these font issues has been made in recent years, but unless we convert all type to curves or paths (which makes future text editing virtually impossible), many of the issues remain.
Compositing multiple elements and layers in complex image files also presents difficulties. We are at the mercy of graphic designers to make sure that all the images or elements they used in the original layout program to build the complete graphic are included (and are the correct versions) when the composite file is saved and transferred to us.
But frequently when we attempt to image a job, we discover that one or more of the images that should be included in the electronic file are either missing or need updating. Links may be lost when design elements are modified, and relying solely on visual inspection of the final graphic on a monitor is not enough to ensure that the image is correct. The more elements the design incorporates, the greater the problem.
Even if we're careful to include all graphic elements and build the design correctly, we may still experience unexplained shifting or changing of the elements when the graphic file is sent to an imagesetter or digital-printing device. Occasionally, this may stem from using a different version of the layout program to open and output the file than was used to create it. It also occurs frequently if we rasterize vector-based images from Illustrator or Freehand into Photoshop. Any placed EPS bitmap image may move unexplainably. In all these situations, our only recourse is to measure and compare the positions of placed elements on the print or film after it is output.
Finally, we must deal with the challenges of electronic delivery via e-mail, Internet, or direct modem. We never know (especially with e-mail attachments) whether we can open a file and whether the file is correct and complete. Nothing is worse than downloading a file for 45 minutes only to find an "unexpected end of file" error when the downloading process is 98% complete.
Additionally, we are challenged by the task of compressing or decompressing files. If lossy compression is used (where apparently redundant data is omitted by the compression algorithm), we can't be sure that it hasn't damaged the original file by deleting necessary data--the results will not be visible until film separations or the final color image are output. In many cases, a sequence of compressions may have been applied to the graphic, and our only hope of opening the file correctly is to decompress it using the identical sequence in reverse order. One slip, and we have alphabet soup!
The transmission situation is made even worse by the huge size of files today, especially large-format graphics, which can range from 450 MB to well over 1 GB in size. While fractal compression is becoming a viable means of lossless compression (where compression is based on commonality of color or other graphic attributes), we still face the hindrance of using courier and overnight delivery services to move the files on traditional magnetic or optical media. (For more about lossy and lossless compression, see my Prepress Wire column, "File Compression for Graphic Images," in the Sept. '98 issue of Screen Printing magazine, page 116.)
As if all of this were not enough, we battle the perpetual problem of platform incompatibility--Mac vs. PC vs. UNIX workstation vs. proprietary system. Even though noise about incompatibility has died down somewhat in the last year or so, it still remains an issue and can seriously interfere with digital throughput in prepress.
It is precisely for these reasons that the Adobe PDF workflow is beginning to take shape.
PDF to the rescue
PDF, which stands for portable document format, is a platform-independent graphic file format that accommodates good compression levels and allows us to move graphic files across platforms, networks, and the Internet without fear of compromising image integrity. With PDF, all the font issues mentioned previously vanish. The positions of imbedded images and other graphic elements are stable and maintained throughout the imaging process. In simple terms, what we see on the computer screen will reproduce exactly as it appears--with no surprises.
Creating PDF Graphics Files
1. Generate a graphic file as you would any electronic file for screen printing or your particular digital-color output device. Compose and preflight the graphic, making sure the file incorporates all elements of the image (including screen and outline fonts).
2. Set up Adobe Acrobat Distiller job options based on your compression preferences, output resolution, color space, etc. Do not convert native colors to RGB. File compatibility should be set to "binary" to minimize PDF file size.
3. Select the appropriate printer driver and PPD (the correct settings for your particular system are listed on Adobe's Website at www.adobe.com).
4. Determine whether you want the image in preseparated or composite format when converted to PDF.
5. Verify document size, fonts, spot colors (if any), trapping information, and OPI comments.
6. Print the graphic to a PostScript file, or, if your computer has sufficient resources, distill the file to PDF directly from the illustration program.
7. If the original file is saved as PostScript code, open the file in Distiller and convert to PDF based on the settings you established in steps 2-4.
8. View, proof, and print the PDF file.
PDF is rooted with the service bureau, where digital technicians have wrestled with digital files for many years. The conventional, non-PDF approach involves converting a file into PostScript code, which is sent to the RIP (Raster Image Processor). The RIP then processes and sends the layers of type, elements, and imbedded graphics (generally EPS, TIFF, JPEG, or nested EPS images) contained within the graphic to the output device. But the complex nature of layered graphic files often causes the RIP to choke and dump the files--usually an hour or more into the process, and most often when we were on deadline.
To help reduce this problem, Adobe introduced Distiller, a program designed to "distill" out redundant PostScript code from files. This streamlines files and reduces their size. The resulting code is called "clean code" because it slips cleanly through the RIP to the imagesetter or output device. Today, Distiller is the primary engine of the PDF workflow and one of four programs in the Adobe Acrobat family for managing PDF files.
PDF and graphics
While the vast majority of documents published in PDF take the form of manuals and on-line business forms (all government forms are PDF format, for example), the use of PDF as a high-end digital-prepress vehicle is relatively new. But as more and more service bureaus install and upgrade to PostScript 3, a key component of the PDF workflow, the capabilities of PDF technology will become much more common. Let's take a look at the specific benefits and concerns PDF introduces.
We'll start with fonts. PDF provides two different font options. The first is embedding fonts and the second is subsetting fonts. To take advantage of embedding, we must use either Type 1 PostScript or TrueType fonts. These fonts are known as outline fonts because they are represented by an outline that can be scaled to any required size and then filled.
For embedded fonts to work with PDF files, the screen and printer fonts used to create the graphic must both be available to Acrobat when the file is printed to PostScript. When the type is embedded, all font outlines, metrics, and sizes are also embedded in the document. This allows the document to display correctly on the computer screen and print without including the fonts with the file. And when a font is embedded, all of the outline information for all characters in the embedded font family are included in the document, regardless of whether they are used or not.
Subsetting fonts is a less desirable option, but it results in a smaller file size because only the characters used in the document are included in the subset. This approach can be somewhat limiting if we need to edit the distilled PDF file and the new character we need has not been included in the file. Additionally, Acrobat substitutes special serif and san serif faces if font information is missing, but preserves the desired character spacing. Font substitution also takes place if bitmapped fonts are used in the original file before it's distilled.
Transmission and storage of PDF files
The PDF workflow involves converting all continuous-tone images, vector graphics, bitmap graphics, and fonts to PostScript code, then using Distiller to compress and write the code into PDF format. The resulting PDF files are inherently smaller than conventional files. But how much smaller depends on the compression option we select when running Distiller.
One of the important distinctions with PDF is that the compression of file information is variable; we can apply different levels of compression to different types of data within the same document. This results in the smallest possible document size, with little or no loss of image quality. The menu of compression methods Acrobat provides includes JPEG, LZW, ZIP, and Run Length Encoding.
JPEG Acrobat's JPEG option offers five user-selectable levels of compression. Since JPEG is a lossy compression method, the greater the compression selected, the greater the loss in image quality. JPEG offers some of the highest compression ratios and, consequently, some of the smallest file sizes. It works best on continuous-tone data.
LZW This format is based on the Lempel-Ziv-Welch (LZW) data-compression algorithm, licensed from Unisys. It is lossless and works best when there are large areas of flat color. It is ideal for images created with vector-based drawing programs. It is not designed for use with continuous-tone images and can actually increase file size if applied to them.
ZIP Another lossless compression method, ZIP works in much the same way as the LZW method, but it can result in file sizes up to 20% smaller than comparable LZW files. It also does not work well on continuous-tone images.
Run Length Encoding This method is used for compressing monotone information. The most common example is compression of black-and-white text. Run length encoding is lossless and can result in very small file sizes.
Acrobat allows us to select compression information for each of the three types of PostScript images--color bitmap, grayscale bitmap, or monochrome bitmap. We can select downsampling resolution levels, as well as the specific compression formats for either automatic or manual compression. Be aware that different image types may require different compression levels to minimize file size. <B>Figure 1</B> shows typical settings for a PDF graphic file that will be used for screen printing
To create PDF files, we need the Adobe Acrobat 3.0 Suite of products, which includes PDF Writer, Distiller, Exchange, and Reader. The products are available from most computer software dealers.
Acrobat PDF Writer Acrobat provides two methods of making PDF files. The first utilizes the Acrobat PDF Writer, which can generate a PDF file from images created on just about any desktop illustration program. PDF Writer is not intended for high-resolution prepress work, but rather for publishing documents in PDF format for easy transport over the Internet, intranet, LAN, or between different platforms. It provides an excellent format for sending spreadsheets, word processing documents with simple graphics, and simple color proofs, especially if they are designed for viewing on a computer screen.
To use PDF Writer, we simply choose it as the printer when we're ready to output the graphic from our design application<B> (Figure 2)</B>. The graphic will be converted into a file with a .pdf extension. It then can be viewed and printed from any computer using Acrobat Reader.
Acrobat Distiller For high-resolution graphics, the best approach is to use Acrobat Distiller, which is designed specifically to convert PostScript code into the most compact and efficient file possible. We can use Distiller in two ways. Using the first method, we convert the file to PostScript code by selecting the "Print to File" option in the print dialog window (an option available in all graphic-arts programs). Next, we open the PostScript file in Acrobat Distiller. We then convert it to a PDF document, selecting the Distiller options that will satisfy the specific job parameters we've established.
The second method requires more RAM and additional hard drive space. The specific amounts will depend on the size of file we're converting. If these conditions are met, we can distill directly from the graphics program, creating a PDF file and bypassing the "Print to File" step in the process.
It's important to note a few rules for working with Distiller. When converting to PDF, make sure that all images are in CMYK format, not RGB. Secondly, be aware that the default settings in Distiller are intended for a low-resolution (computer screen or Web) application. You will need to change the settings for your intended use. Like any job you would send to a service bureau, or your own imagesetter, good preflighting procedures will result in fewer problems down the road. Make sure that you are only using Type 1 or TrueType fonts. Embed all fonts in the graphic file to avoid substitution problems. Also make sure that Distiller can locate the necessary outline and printer fonts when it converts the file to PDF.
Acrobat Exchange Acrobat Exchange is used to view and perform limited editing functions on PDF files after conversion. Specifically, it contains the Touch-up tool, which allows for minor text editing within the document. If extensive editing is necessary, you will need to acquire one or more of the third party plug-ins available for this purpose, such as Crackerjack.
Acrobat Reader (free) Once a file has been converted to PDF, it must be viewed with either Exchange or Acrobat Reader 3.0. Older versions of Reader will not work on files converted to PDF with Acrobat 3.0. PDF files can be viewed and printed from Reader, but not edited. Reader is available at no cost from Adobe's Web site (www.adobe.com) and will operate on any computer platform.
Benefits of PDF
Since PDF files are true PostScript, they are resolution independent. This means that we have the option to specify at what resolution the final document will be imaged. It may be a low-dpi device like a laser printer, or a high-resolution device like a laser imagesetter. The maximum resolution we can choose is 2400 dpi. The default settings in Distiller are 72 dpi and are intended for computer viewing, so they must be changed.
When preparing original files for Distiller, follow the usual and customary rules for halftone scanning resolution. Most commonly, this means using an image resolution that is twice the value of the desired output halftone line count. For instance, if the halftone needs a printed resolution of 100 lpi, the scan data should be 200 dpi. Monotone elements (primarily type) should be set for 1200 dpi or the highest resolution of your output device. For screen printers, this could be 300, 400, 600, 800 dpi, or higher. The device output resolution determines the settings we'll use in the Job Options dialog box under the Distiller Menu.
Preseparated or composite files
You have two file-type options in preparing the final PDF document. They are Preseparated and Composite. Preseparated files are larger and cannot be viewed by the end user. They are preseparated for output as color separations. One PDF page is generated for each color that is to be printed. So if our design includes CMYK plus one spot color, the preseparated file would contain five pages. This option makes for larger files, but it allows you to print spot color gradients, DCS files, duotones, tritones, quadtones, and colorized TIFF files.
Composite files are smaller and can be viewed by the user before they are sent to the RIP. They contain all the necessary information to separate and print the job. This includes UCR/GCR, trapping, OPI comments, halftone line count, angles, paper size, and solid spot color information. All of this information goes to the RIP where the file is rasterized and printed to the imagesetter or digital color printer.
Drawbacks to PDF
The limitations of working with PDF files include minimal editing and touch-up ability beyond those available in Acrobat Exchange. Complex PDF files are extremely difficult to edit without special plug-ins like Crackerjack. Additionally, because using PDF files for high-resolution printed graphics is a relatively new alternative, few service bureaus are experienced in handling the files.
Working with PDF files is not for the prepress novice. A good working knowledge of the prepress process and how color separations work is essential. As screen printers, we must also be intimately familiar with film preparation and our own printing requirements and limitations.
At this point, successful use of PDF files for output also requires some editing of PostScript code. And remember, PDF files can only be output directly to an imagesetter or digital color printer if you or your service bureau use a PostScript III compatible RIP. Presently, few service bureaus and even fewer screen shops have such RIPs. More details about the nuances of working with PDF files are available on a white paper entitled "Preparing Adobe PDF Files for High-Resolution Printing," which can be accessed at Adobe's Website (www.adobe.com/print/).
The future of PDF
PDF is a very exciting technology, and as its use in graphic arts increases, it's likely that we'll see many more useful plug-ins developed for the Acrobat Suite. It also means that we'll be able to customize how the program works in our own particular workflows. This may mean Hot Folders, where we drag and drop images for automatic conversion, or scripted routines where all the conversion work occurs automatically in the background.
Like any developing technology, PDF will gain strength through use. It is already a mature technology for document display and printing, and a growing favorite for Web use. Its utility for prepress is sure to follow suit.
About the author
Mark A. Coudray is president of Coudray Graphic Technologies, San Luis Obispo, CA. He has served as a director of the Screenprinting and Graphic Imaging Association International (SGIA) and as chairman of the Academy of Screenprinting Technology. Coudray has authored more than 70 papers and articles over the last 20 years, and he received the SGIA's Swormstedt Award in 1992 and 1994. He covers electronic prepress issues bimonthly in Screen Printing magazine. He can be reached via e-mail at firstname.lastname@example.org or email@example.com.