Re: WAY OT: How to make PDF files open faster?

[Mr. Lance:]

I admit that I'm not very familiar with how PDF (or TIF) files work. Is there a trick that I can use to speed them up?

There are some techniques that can be used, depending on how the data is encoded in the PDF or TIFF files. Your options also depend on whether you have access to full-version 'Acrobat' (or comparable PDF authoring software), specialized image viewers, and batch PDF-to-TIFF conversion utilities.

PDF is basically a container for various types of data--raster images, vector images, text, etc. Various PDF readers handle the different types of data with differing levels of efficiency. For instance, full-version 'Acrobat' 7 is faster at opening raster data than 'FoxIt Reader', but 'FoxIt' renders vector images faster. However, as a generalization, a dedicated image browser like 'ACDSee' or 'Irfanview' will load a raster image faster than any PDF reader will load a PDF page containing the same raster image. Depending on the PDF reader and image browser combination you are using, a vector image may load faster (once rasterized) in an image browser than it will in a PDF reader.

Full-version 'Acrobat' has the ability to extract raster data directly from a PDF file, and (within limits) to rasterize other data in the PDF file which is not already a raster image. Generally you get the original resolution if you select "Choose resolution automatically," but 'Acrobat' will upsample/downsample to a resolution you specify. However, 'Acrobat' is not particularly efficient. Generally a dedicated PDF-to-TIF conversion utility like 'pdf2tif' (which will also do PDF to JPEG) will extract the images at their original resolution in much less time per page, unless the input raster data is exceptionally badly behaved. (Examples include images with different resolution horizontally and vertically, pages which are made up of "tiles" of raster images which can't be stitched back together into a single image, etc.) 'Acrobat' also won't output raster images that are longer than 20,000 pixels on a side--this means you have to downsample if the original image is more than 20,000 on a side (which is true of some scanned historical maps from, e.g., Kansas DOT).

In my experience, 'ACDSee' 7 is very, very fast in handling Group IV monochrome TIFF files (the vast majority of state DOTs and turnpike authorities who publish E-plans use either TIFF directly or in PDF documents), but it does have some overhead, notably an image database (to enhance image loading speed) which tends to divide into thousands of fragments on the HD and fool programs like 'Diskeeper' into thinking your HD is excessively fragmented. 'Irfanview' handles Group IV TIFF files acceptably but is noticeably slower. 'ACDSee' 7 will also rotate and crop Group IV TIFF files, but no longer has a multipage TIFF chopper (I use 'ACDSee' 4 for this purpose).

The loading time of a PDF file which has vector data will also vary considerably according to whether and how deeply the data is layered. Most state DOT online cartography that is not scanned is made available as PDF files which are plotted directly from 'MicroStation', which offers users the option of producing layered PDF files (generally with PDF layers corresponding to the data layers in the 'MicroStation' *.DGN file). My experience with 'Acrobat' 7 is that layered PDFs are often very, very slow to load, and the "Flatten layers" command in 'Acrobat' also takes a very long time to execute and is more likely than not to crash. The quick and dirty way of getting rid of layers is simply to print the PDF file to PDF using 'Acrobat Distiller'--the printed PDF will have just one layer. There will be a wedding-cake graphic (at least in 'Acrobat' 7) in the information bar for any PDF file with multiple layers.

PDF-to-raster conversion processes (whether extracting existing raster data or rasterizing vector data) can take a long time from start to finish and tie up the CPU to the point where your system becomes nonresponsive, so inevitably there is a tradeoff between the less-than-ideal alternatives of looking at the data directly in a PDF reader and converting it to some other format which is easier to work with.
.

Relevant Pages

  • Configuring IE to Use Adobe Reader 6.0 or Acrobat as Helper Applica
    ... To configure Adobe Reader 6.0.1 Open Reader, ... UNCHECK the box for "DISPLAY PDF IN BROWSER". ... >IE to use the Acrobat product as a helper applications. ... >Downloading a PDF File to Your Hard Disk ...
    (microsoft.public.windows.inetexplorer.ie6.browser)
  • Configuring IE to Use Adobe Reader 6.0 or Acrobat as Helper Applica
    ... To configure Adobe Reader 6.0.1 Open Reader, ... UNCHECK the box for "DISPLAY PDF IN BROWSER". ... >IE to use the Acrobat product as a helper applications. ... >Downloading a PDF File to Your Hard Disk ...
    (microsoft.public.windows.inetexplorer.ie6.browser)
  • Re: Just use acrobat, not Windows Journal Viewer for PDF files
    ... I double click a PDF file, or click on a web page PDF link and the Adobe ... Acrobat reader splash screen flashes onto the screen, ... > original link opens. ...
    (microsoft.public.windowsxp.basics)
  • Re: FindExecutable on Windows CE
    ... reset would likely remove it entirely (unless Acrobat was in ROM). ... Paul G. Tobey wrote: ... with .PDF files. ... way to do this is create one temporary pdf file and find the ...
    (microsoft.public.windowsce.app.development)
  • RE: Converting Word Hyperlinks into Acrobat
    ... came to the result that Acrobat has a problem. ... PDF, error messages popped up telling that bookmarks were missing). ... If you replace your hyperlinks by ordinary cross-references (as created via ... About the Adobe Acrobat settings in Word: ...
    (microsoft.public.word.docmanagement)
Loading