[Date Prev][Date Next] [Thread Prev][Thread Next]
[Date Index] [Thread Index] [New search]

Re: Halftones in PDF files



Yesterday, there were a few postings with regards to how one could (or should)
set screen values in PDF files via FrameMaker and/or drivers and/or Acrobat.

A few thoughts:

(1) In general, in current publishing workflows, and especially PDF-based
workflows, the industry has moved away from the setting of screening values
in the image data itself, whether that be in Photoshop format files, PostScript
image data, or PDF image data. These values include frequencies, dot pattern
type, and angles. With very rare exceptions, these values are now generally
set globally for a publication at RIP time by the RIP operator using console
operations of the RIP itself. In fact, many RIPs are set to totally ignore
and override any and all screening values in the incoming PostScript in favor
of the RIP's own values. This is also true for many if not most desktop and
departmental PostScript printing devices. You almost need to go through fiery
(pun-intended) hoops to force a certain screening via PostScript. Such devices
typically use OEM-proprietary halftoning algorithms.

(2) Early Windows PostScript drivers did provide a supposed ability to set
screening parameters globally for a printer device for the duration of the
print job. With the low-resolution devices of the day, those settings had the
effect of allowing the user to either get a "finer" screen at the expense
of gray shades or more gray shades with a "coarser" screen. The settings also
effected how polygon fills were rendered as well as images. We specifically 
removed that capability for a few reasons:

        (a) Higher resolution devices (240 dpi and 300 dpi replaced by
        600 dpi and 1200 dpi devices for the most part) mitigated the need
        to make these tradeoffs

        (b) Per (1) above, such values were ignored for increasing numbers
        of devices that used proprietary screening techniques to overcome
        dot size and gray shade count limitations of conventional screening.

        (c) The values only applied to applications that did NOT generate
        their own PostScript. In other words, they were ignored by Photoshop,
        Illustrator, Quark XPress, PageMaker, Freehand, InDesign, Acrobat,
        and other such publishing tools.

        (d) The values were overriden by any EPS that set their own values.

There is no intention whatsoever to restore such capability to the drivers.

(3) If you really want to create PDF files that have embedded screening information,
you must do it via other means that are not associated with the PostScript
drivers. You can put screening information into vector artwork and images via
EPS output via Illustrator and Photoshop, but understand that you will need to
set the correct options to do this. The Distiller's job options must be set to
"preserve halftone information." This is NOT normally set. This of course does
not guarantee that such screening information will be honored by whatever
device receives PostScript output by the print function of Acrobat or
Acrobat Reader. As an alternative to forcing screening values via EPS
imported into FrameMaker (or similar application), you could probably write
some code for the "prologue.ps" file to force certain screening for every
image. This would also be used in conjunction with the "preserve halftone
information" joboption.

(4) Unless you were trying to achieve a particular artistic effect, I
would personally never recommend setting any screening information prior
to RIP operation. As I recall, the original thread was trying to use
screening as a form of copy protection. Per above, given that many printers
ignore the embedded screening settings, this will work globally. Personally,
I am underwhelmed by someone sending/licensing/selling me crippled content
that has the underlying assumption of lack of any trust.

        - Dov
        


At 3/25/2002 02:55 PM, Michael Cudmore wrote:
>Thomas Michanek wrote:
>> If I have understood you correctly, you want to specify the line screen
>> frequency (and possibly angle) of the halftone rasters generated in
>> PostScript code for graphics or text in color or grayscale. Is that correct?
>
>Yes.
>
>> FM doesn't have an option to specify this, so it uses the screen settings
>> specified in the printer or printer driver. If you have a PostScript
>> printer driver that has a PostScript setting for halftone screens, you
>> could possibly use that driver when creating the PS file. 
>
>The postscript driver I use, AdobePS 4.5.2 for Windows 9x set up with
>the Distiller 5 PPD, does not have a direct option for halftone screens,
>just a resolution setting. For the physical Postscript printers I have
>used with various versions of the AdobePS for Windows driver, this
>resolution setting DOES seem to determine the halftone screen used, but
>this doesn't seem to be the case for the Distiller printer. 
>
>Which means two possibilities: 
>1. the driver for Distiller acts differently than a physical printer,
>and does not add a suitable setscreen line to the PS; or,
>2. the driver for Distiller acts the same as for a physical printer, and
>adds a setscreen line to the PS, but the Distiller program ignores this
>setscreen when making the PDF. Although this should not be happening,
>since my job options include the setting to preserve halftone screens,
>and halftones in EPS artwork ARE being preserved.
>
>However I have not taken the obvious step of inspecting the PS generated
>by the driver for Distiller and comparing it with PS generated from the
>same FM document by the driver for my Xerox N2125 PS. (I guess my recent
>use of the Acrobat 5 PDF port, which pipes the PS straight into
>Distiller rather than creating an intermediate PS file, and which mostly
>works pretty seamlessly, means I am out of the habit of managing,
>moving, inspecting and deleting PS files!) I will see what comes out of
>a few initial tests. 
>
>> One option could be to postprocess the PostScript file before Distilling,
>> and insert a "setscreen" command at an appropriate place in the code.
>> The problem is finding a place that will work for all PS files...
>> 
>> I have a script for inserting such a command in the PostScript code
>> generated by FM 4.x and 5.x on UNIX, that will correctly set the halftone
>> screen for the whole document (at least for printing; I haven't tried
>> Distilling). I don't know if this would work for the PostScript created
>> by FM on Windows, but it basically does the following:
>> Immediately after the "%%EndSetup" line, insert the line
>> "<frequency> <angle> currentscreen exch pop exch pop setscreen"
>> (insert your desired frequency and angle values, in lpi and degrees)
>
>Thank you; I will also try this method. 
>
>Cheers,
>
>Michael
>
>--
>Michael Cudmore


** To unsubscribe, send a message to majordomo@omsys.com **
** with "unsubscribe framers" (no quotes) in the body.   **