|A biannual newsletter from eLynx SwiftView Tools covering issues facing today's enterprises.|
Did you know that a page printed in color can contain 99 megabytes of information — or more? A letter size page printed at 600 DPI contains over 33 million pixels and each color pixel requires 3 bytes.
Compresion technologies reduce these sizes when saving to file, but most software must expand this compressed data back to its original size in memory for actual processing. Because of this heavy memory overhead, such software may run very slowly on a typical computer. With businesses today using color in their documents more than ever, it’s imperative to handle color images intelligently to reduce memory requirements. Otherwise, printing and viewing even simple documents with color could grind a computer to a halt.
SwiftView Tools has an innovative solution to the problem. Tiled Raster Technology represents color images as individual regions of black-and-white or color information. By analyzing the pixel data in a document, TRT can identify where color exists on a page and represent that information as accurately as possible, while keeping memory requirements low for those areas of the page that contain no color. Moreover, most pages are composed of a small amount of information (text or graphics) on a mostly-blank background. Look at all the blank space on the page you are reading now, for instance. TRT can identify these blank regions and optimize storage to represent them using almost no memory at all.
Many color documents use color on only a small portion of each page. For instance, a company might print custom letterhead in color, or add small color graphics to mostly black-and-white pages. Because so many color documents only use color on a small portion of each page, TRT intelligent storage can dramatically reduce memory requirements when handling large documents containing spot-color – up to a factor of 100 or more. And that means that SwiftView Tools won’t bog down your machine or get in the way of your other work.
Other tools try to solve the memory problem by compromising the representation of the document. With TRT, SwiftView Tools always represents each page precisely, down to every last pixel.
Fast & Accurate Display
But it’s not enough just to have an accurate digital representation of a document. This representation must be scaled and displayed on the screen quickly and with high fidelity. Often a design sacrifices either quality for speed, or speed for quality.
SwiftView Tools surpasses in both categories by using an optimized algorithm which takes advantage of TRT. TRT creates a quick, rough “first pass” version of the document which can be displayed at high speed. Then it uses the high fidelity information of the document image to fill in the rough edges, resulting in a crisp, accurate display of the document. A 600 DPI image may be scaled by a factor of eight when displayed in a typical window. This means that 8x8 = 64 pixels of input data are represented by a single pixel on the display! What color is that pixel? TRT uses area weighted averaging to calculate the pixel value, resulting in the best mathematical representation of the pixel. The high speed first pass quickly presents an image to the user’s eye and this allows time for the high fidelity image calculations.
TRT uses an adaptive algorithm to decide when to use the rough first pass plus an additional refinement pass, or to simply use the high fidelity image directly. The result is a smooth, enjoyable experience as users zoom and pan around the document.
The TIFF specification has been, for many years, the mainstay of imaging and document archival systems. TIFF is simple, widely used, supports color, has good compression methods, and is output by all scanners. But now there is a new contender for the archive niche: PDF/A.
PDF/A is a defined subset of Adobe’s popular PDF specification. PDF/A was originated as a joint effort between NPES — The Association for Suppliers of Printing, Publishing and Converting Technologies, and AIMM — the Association for Information and Image Management, International. The intended function being: the storage of PDF files in long term document archiving applications.
Why PDF as an archive format? Given the huge number of PDF viewers currently in existence, and the ready availability of PDF software tools from vendors other than Adobe, PDF has emerged as a format likely to be useful far into the future.
Why define a subset of PDF? PDF is a complex specification — and continues to grow more so. Some PDF features, such as audio and video, are deemed unimportant for archiving business documents. Another key requirement is that all files be 100% self contained. A PDF document, typically, has many elements that rely on external information; perhaps the most common example is the font files resident in the OS. PDF/A requires that all fonts used as text in the PDF/A document, be stored in the document. This can significantly increase the size of files that are only a few pages, but insures that the document can be accurately reproduced, even years in the future when the original OS (and font) may no longer be in use.
The PDF/A specification defines two levels of compliance: PDF/A-1a and PDF/A-1b. A PDF/A-1b compliant document insures that the original visual form of the document can be readily recreated. A PDF/A-1a compliant document maintains both the visual appearance of the document (PDF/A-1b), and also the structure of the document.
SwiftConvert and SwiftView Pro output both TIFF and PDF/A-1b compliant files. SwiftView uses advanced rendering technology to achieve high accuracy, down to individual pixels. The rendered image is then saved as an exact PDF/A copy. In the past, rendered images produced the most accurate output, but could result in substantially larger files. However, PDF/A allows the use of the new JBIG2 compression technique. JBIG2 compression results in far smaller file size giving SwiftConvert the best of both worlds, high accuracy and small file size. Moreover, whenever the original PCL file contains text in a usable format, SwiftConvert can (optionally) store that raw text in the PDF/A file, making it suitable for both search and cut-and-paste operations.
Does this mean that TIFF archives should be migrated to PDF/A? Not necessarily, TIFF is a far simpler format than PDF/A. The number of legacy applications and documents using TIFF are huge, insuring that TIFF will be around for a long time. TIFF is also deeply imbedded in the workflow of many companies, changing it will not be a trivial task.
Tiff is not a "dead" format, in fact, it is designed for easy extension, and various standards bodies continue to modernize it, keeping it competitive. For example, to see the TIFF/IT enhancements for prepress digital data exchange click here, they have also recently defined JBIG2 compression in TIFF. Microsoft Imaging has lately added new tags that support the addition of raw text in their TIFF files. SwiftConvert, following Microsoft’s lead, can also store the raw text in a TIFF file. This means that TIFF, the reigning champ, will continue to give the new kid a contest for king of the archive hill.
To read more about TIFF, PDF, and JBIG2, and to compare them as well as other formats supported by SwiftView Tools:
Please click here to send feedback.
Is there someone else in your organization that would benefit from our e-newsletter? Click here to let us know.
Copyright© 2007 eLynx, Ltd.