[1442] in Public-Access_Computer_Systems_Forum
Display resolution: summary and conclusions
daemon@ATHENA.MIT.EDU (Walt Crawford)
Fri Oct 9 14:30:15 1992
Date: Fri, 9 Oct 1992 13:21:13 CDT
Reply-To: Public-Access Computer Systems Forum <PACS-L%UHUPVM1.BITNET@mitvma.mit.edu>
From: Walt Crawford <BR.WCC%RLG.BITNET@mitvma.mit.edu>
To: Multiple recipients of list PACS-L <PACS-L%UHUPVM1.BITNET@mitvma.mit.edu>
----------------------------Original message----------------------------
Thanks to those who provided answers to my question about display
resolution. Since most answers actually went back to PACS-L, I
won't repeat them here.
To summarize what respondents said:
1. It's possible to increase *apparent* resolution using
anti-aliasing techniques (varying color/gray scale in
adjoining pixels). This does not increase *displayable*
resolution, however, and is mostly a software solution.
Incidentally, it's why U.S. television appears to have
much higher resolution than it actually has: the color
palette is nearly infinite, and the shadings increase
apparent resolution.
2. Monitors can accept and process signals with more
information than there are physical pixels on the screen.
Thus, it's not *fraudulent* (i.e., legally actionable)
to claim display capabilities greater than the physical
resolution of the picture tube. However, such displays
cannot be completely clear...the problems have been
described as "glare" or "blurriness" or whatever, but
come down to using the nearest available pixel...
Here, then, is what I think can usefully be said--and it's
very important if you're buying displays where you *need*
high resolution (or if you're designing software that will
*use* high resolution):
a. Many displays have a greater electrical resolution than
physical resolution. Attempting to use the full electrical
resolution will inherently result in less-than-clear displays,
since the pixels aren't really there.
b. If you can find out the actual horizontal display area (excluding
underscan) and the dot pitch, you can calculate the maximum
*physical* resolution. For any real-world purposes, this should
also be the maximum *useful* resolution. (The horizontal number
is always the larger of the two, and almost all modern displays
except page displays use a 4:3 hor:ver ratio.)
c. To do rough estimates, you can figure that most advertised
display sizes are overstated by an inch or two (because part
of the tube is masked off), and that the horizontal display
size is roughly 0.85 of the diagonal size.
d. Given all of that, some rough assumptions should include:
1. 1024x768 requires at least a 15" and probably a 16"
display at 0.31mm pitch, at least a 14" and probably
a 15" at 0.28mm pitch.
2. 1280x960 requires at least an 18" display at 0.31,
at least a 17" at 0.28, at least a 16" at 0.26mm.
3. No 14" monitor with dot pitch greater than 0.31
(which usually means 0.39 or 0.51) can clearly display
800x600; at 0.51, even 640x480 can't be displayed clearly.
e. HOWEVER: for easy reading, for most people, given typical
Windows/Macintosh text design, you really want text lines
to be roughly 1/4" high, which means an effective 72dpi.
That means that, for ease of use, appropriate figures might be:
640x480: at least 10.4" diagonal (in other words, the best
notebook LCD screens just barely make it)
800x600: at least 13" true diagonal--which most 14" displays
don't quite offer; a 15" display is probably advised
1024x768: at least 16.7" true diagonal (probably an 18" display)
1280x960: at least 21" *true* diagonal!
f. If someone claims 150 dpi screen resolution *in a color screen,*
check it out very carefully! That requires 0.17mm dot pitch,
which is far below any cost-effective manufacturing technique
that I'm currently aware of. (Monochrome screens are an entirely
different matter.)
My thanks to those who responded. Now, to get that 15" monitor
that *can* do 1024x768 correctly, but which really *should* run
at 800x600...
=Walt Crawford, br.wcc@rlg (BITNET), br.wcc@rlg.stanford.edu=
=Permission to forward or copy granted if signature retained=
=Usual disclaimers apply: my words, my opinions.=
To: PACS-L@UHUPVM1
