On 12/29/21 4:03 AM, George N. White III wrote:
> Zero-width lines in PostScript and PDF have always been problematic. In
> typography there is
> a "hairline", and zero-width was generally interpreted as the narrowest
> line the device could
> render, e.g., a hairline. Some devices did not do well -- there would
> be dropouts and rasterization
> artifacts on curves and sloping lines. I recall one scientific chart
> created by some specialized
> commercial package. I told the author it would not reproduce well in
> the journal. The author
> came back with the journal and was very pleased with the quality.
> Turned out the journal's
> Illustrator expert had printed the illustration, scanned it, and redrew
> it for publication.
>
> --
> George N. White III
That's interesting.
It seems that LibreOffice and other relevant applications and tools need
to distinguish between guide lines (used to help the user position and
align objects, but not show up in print) and lines that are meant to
show up in print (graph axes, tick marks, and background grid lines).
Until yesterday, when I saw Tibor's posts in the "Ask LibreOffice"
thread, I did not know about "zero width" objects. I do find "zero
width" somewhat confusing
Quoting the Postscript Reference Manual, 2nd Ed. (1995 13th Printing!) p 508:
setlinewidth:
A line width of zero is acceptable: It is interpreted as the thinnest line that can
be rendered at device resolution -- in other words, one device pixel wide. Some
devices cannot reproduce one-pixel lines, and on high-resolution devices, such
lines are nearly invisible. Since the results of rendering such "zero-width" lines
are device dependent, their use is not recommended.
My own experience: for many devices, the appearance of a specific pixel
depends on what is around it. On a crt, colors are generally generated by
some mix of RG and B pixels which may be rectangular. In print, colors are
often generated using half-tone methods. The appearance of a one-pixel
width line will vary with color choice.