TeX specials & epsf with dvips & ps2pdf

In another thread, I asked about making screen dumps to produce
postscript files for inclusion into Plain TeX documents with epsf
macros. That works. I also use \special. If I run pdftex on the
TeX file, it rejects the \specials, so I use tex, dvips and ps2pdf.
This sort of works and looks ok in xpdf, but not when I print it out.
Fox example, a "pin" is a big black dot with a very thin line sticking
out of it (the line basically is an extension of a radius). It looks
ok in xpdf but when I print it out all I see is the big black dot, not
the stem. These pins are made with specials.

(1) Is there any way to get pdftex to accept specials?
(2) I have one idea to try to get the stems to appear. If it doesn't work,
I'll post the offending \special in a simple self-contained example.
Namely, I'll see whether making the linewidth larger in the special
helps.
--
Ignorantly,
Allan Adler <ara@zurich.csail.mit.edu>
* Disclaimer: I am a guest and *not* a member of the MIT CSAIL. My actions a
nd
* comments do not reflect in any way on MIT. Also, I am nowhere near Boston.

Allan Adler wrote:
> In another thread, I asked about making screen dumps to produce
> postscript files for inclusion into Plain TeX documents with epsf
> macros. That works. I also use \special. If I run pdftex on the
> TeX file, it rejects the \specials, so I use tex, dvips and ps2pdf.
> This sort of works and looks ok in xpdf, but not when I print it out.
> Fox example, a "pin" is a big black dot with a very thin line sticking
> out of it (the line basically is an extension of a radius). It looks
> ok in xpdf but when I print it out all I see is the big black dot, not
> the stem. These pins are made with specials.
>
> (1) Is there any way to get pdftex to accept specials?
> (2) I have one idea to try to get the stems to appear. If it doesn't work,
>     I'll post the offending \special in a simple self-contained example.
>     Namely, I'll see whether making the linewidth larger in the special
>     helps.

It sounds like someone is being "clever" and using 0 linewidth.

This is a NON scaleable feature, and defines a line
to be a single pixel wide. This (of course) defeats
the resolution independance of PostScript.

A single pixel is fine on  a 72 DPI monitor,
but "less good" on a 600 dpi laser printed,
and absolutely useless on a 3000 dpi
colour typesetter.

You may be able to prove this by rendering the postscript to
a raster (e.g. PNG) using ghostscript at various reslutions.

I'm less sure on a fix, but let's start with a diagnosis.

BugBear

bugbear <bugbear@trim_papermule.co.uk_trim> writes:

> It sounds like someone is being "clever" and using 0 linewidth.

Actually 0.1. I've changed it to 1 but it will take a while to see how
it looks on paper because the machine I'm using to do the work is not
connected to the printer.

> This is a NON scaleable feature, and defines a line
> to be a single pixel wide. This (of course) defeats
> the resolution independance of PostScript.
>
> A single pixel is fine on  a 72 DPI monitor,
> but "less good" on a 600 dpi laser printed,
> and absolutely useless on a 3000 dpi
> colour typesetter.
>
> You may be able to prove this by rendering the postscript to
> a raster (e.g. PNG) using ghostscript at various reslutions.
>
> I'm less sure on a fix, but let's start with a diagnosis.

Thanks for the explanations.
--
Ignorantly,
Allan Adler <ara@zurich.csail.mit.edu>
* Disclaimer: I am a guest and *not* a member of the MIT CSAIL. My actions a
nd
* comments do not reflect in any way on MIT. Also, I am nowhere near Boston.

Allan Adler <ara@nestle.csail.mit.edu> writes:

> bugbear <bugbear@trim_papermule.co.uk_trim> writes:
> 
>
> Actually 0.1. I've changed it to 1 but it will take a while to see how
> it looks on paper because the machine I'm using to do the work is not
> connected to the printer.

Changing it to 1 didn't help.

The appearance is different in xdvi and gv. The former fails to show the
pins whose tails are horizontal. gv shows everything clearly.

I sometimes do

1 -1 scale to reverse vertical but that doesn't seem to be what is
causing it.
--
Ignorantly,
Allan Adler <ara@zurich.csail.mit.edu>
* Disclaimer: I am a guest and *not* a member of the MIT CSAIL. My actions a
nd
* comments do not reflect in any way on MIT. Also, I am nowhere near Boston.