LISTSERV mailing list manager LISTSERV 15.5

Help for TYPO-L Archives

TYPO-L Archives

TYPO-L Archives


Next Message | Previous Message
Next in Topic | Previous in Topic
Next by Same Author | Previous by Same Author
Chronologically | Most Recent First
Proportional Font | Monospaced Font


Join or Leave TYPO-L
Reply | Post New Message
Search Archives


Re: Digital Typography


Joel Neely <[log in to unmask]>


TYPO-L Discussion of Type and Typographic Design <[log in to unmask]>


Fri, 25 Oct 1996 16:47:57 -0500





text/plain (55 lines)

Don Wilkes wrote:
> [snip]
> I'm afraid I still disagree that we need a new term. As you said:
> >All digital typographic forms are translated to discrete elements at some
> >scale. Even characters shaped on a 2500-dpi filmsetter are made up of
> >dots.
> It's simply a matter of scale. If the phrase "type design" and the word
> "typography" applies to making type that is normally output on an L300,
> then it applies equally to crafting type forms for lower resolution devices
> such as video monitors. In the latter, hinting is a more important issue,
> but that's about it. Glyphs are glyphs...
> \d

Whether we need a new term or not, I must disagree with the notion that
it's "simply" a matter of scale. To me, this is an example of a quantitative
difference so significant that it becomes a qualitative difference.

For example, at the practical level, grey-scale aliasing of letter forms
on a 2500-dpi film image is not only not worth the trouble, but may or
may not even be technically supported. However, on a roughly-80-dpi CRT
capable of displaying 256-level grey scale, if not thousands or millions
of colors, aliasing makes *considerable* difference in how the "same" letters
appear. The issue of "alias hinting", including the possibilities of
arbitrary combinations of type color and background color/pattern is *far*
from being settled AFAIK, but seems to me much more relevant (and possible!)
for those technologies with both user-perceptable grain and dynamic image

A computer controlling a display can certainly (at least in principle) adjust
which pixels to paint, and what color to paint them, by considering both the
previous target area (including surrounding pixels) and the specified result
of adding new content. This is / can be done in real-time as the image is
being generated.

OTOH, I doubt that many presses actually hold registration, to within a half of a
2500-DPI pixel. In addition, by the time the original 2500-DPI electronic image
goes through the film->plate->offset->paper (e.g.!) transformations before ever
coming to the viewer's eye, a considerable amount of analogue change has taken
place, independent of registration. Put these two together, however, and the
entire notion of what colors will appear in "adjacent" pixels (of a fixed grain)
begins to look highly dubious.

My respectful conclusion is that the scaling issues, and the considerable
differences in analogue effects and combinatorial possibilities, make the
pure-digital electronic rendering of images and type a different sack of
snakes than in the print media (although some of those snakes are clearly
of related species!)

Joel Neely [log in to unmask] 901-375-6586

Back to: Top of Message | Previous Page | Main TYPO-L Page



CataList Email List Search Powered by the LISTSERV Email List Manager