10 May
2001
10 May
'01
11:35 a.m.
Chuck McManis skrev:
>How do you define resolution at the monitor level?
I've always been able to
>get entirely different resolutions out of monitors, and why would they
>disagree? If it's an analogue signal, wouldn't the monitor just sweep along
>and project whatever is input?
Color monitors have a color mask, and that color mask
has a 'dot pitch'
which defines where you can display pixels. If you attempt to display more
pixels on a line than there are holes in the mask, then you will get
banding artifacts.
Perhaps I'm being a brute, but according to my definition, that is all right.
As long as the monitor doesn't give up the ghost or loses sync, I don't mind
how bad the image looks. Not in a clinical sense, anyway.
Next there is frequency response. The amplifiers that
connect to the color
guns have something called a 'slew rate' which is the rate at which they
can change their output color. If you put to many pixels side by side then
you will start seeing color degradation due to the fact that the amplifier
can't get to the new color fast enough. If you display several columns of
vertical white lines on a black background you will see (in cheaper
monitors) that the leading edge of the while line is not white, its gray.
And as you increase the number of lines the white lines will get grayer and
eventually you will have just a gray screen.
Also acceptable.
You "can" drive monitors all over the map,
eventually you will destroy them
if you allow the horizontal output driver to over heat. However, getting an
acceptable signal out of one is more constrained.
But a pixel rate is meaningless without an accompanying sync rate. When they
say 800?560, they must intend at a particular frequency, right?
--
En ligne avec Thor 2.6a.
If I don't document something, it's usually either for a good reason,
or a bad reason. In this case it's a good reason. :-)
--Larry Wall (perl) in <1992Jan17.005405.16806(a)netlabs.com>