On Mon, 5/11/2026 8:02 AM, Robert Chippendale wrote:
I have just bought an old CRT monitor for my Pentium III system (which I am writing from). If I recall correctly, when I got it the contrast was set to 100, and the brightness was set to 64. Are these reasonable settings? Will they shorten the lifespan of the picture tube?
Also, I am getting an effect where objects on the screen have "shadows" or "tails" directly to the right. Does anyone know what, if anything can be
done about this?
Thanks in advance for any help you may be able to give
At a guess, brightness should be proportional to beam current,
and a higher beam current would wear the filament a bit faster.
You can adjust the contrast, for any level of brightness
you have selected. As the tube ages, you would likely be
lifting that brightness setting to compensate.
We were told that shadows off to the side of the main image,
were "reflections" related to the VGA RGB transmission lines
not matching something at the end of the cable. I personally
no longer believe this to be true, because LCD monitors run
with VGA analog feeds, don't seem to have those, even at 1920x1080.
The 400MHz bandwidth video card DAC is supposed to be sufficient up
to about 2048x2048 or so. It's really hard to tell what theory
holds in this case. I would be more willing to believe it has
something to do with the video amps and their response to a step input.
The part of VGA that is non-ideal, is the connector. A 13W3 uses
actual coaxial shaped ends on the connector, for the RGB connections.
Whereas the VGA just used separate pins for the R and the ground
for the R, and so on. Six pins for the three coaxial runs inside
the cable bundle. If the connectors were causing reflections,
then chaining two RGB cables together, should show more degradation
via after-image (the connector in the middle of the cable
run, being non-ideal). The H and V are not carried on coax.
I don't know if I can isolate the issue any closer than that,
via observations of permutations and combinations.
None of the engineers ever pontificated on the topic at work,
so there was no added color commentary there. I don't think
any of those engineers would know much about the video amps
in a CRT monitor (as to whether they're overdamped, underdamped,
critically damped, close to ideal or a bit non-linear).
And I don't recollect anyone testing "premium" cables versus
regular cables, to see if it made any difference. The VGA connector
is far from perfect as a concept, and all the VGA connectors have
the same shortcoming in the conversion from coax travel to pins.
You won't have an HDMI output on your Pentium III, but if you
have at least one more-modern computer, you could take an
HDMI output, use an HDMI to VGA adapter and drive the CRT that
way, as a test. Just to see if you can achieve an improvement.
Some CRTs had a "captive cable", which means you cannot shorten-up
the cable run at all. And if you wanted to make a gender changer
for a short cable test, you could take two DB-15 soldertail
and make up an adapter.
Sitting at my desk, is a male DB-15 soldered to a female DB-15.
There is one 1N4148 diode, a switch and some wiring. THAT
may be how I was signaling to the video card, what resolution
to send. Maybe that is how I was selecting between 1280x1024 and
1152x864 :-) That thing is full of dust, because EDID has been
around a long time now, making that kludge unnecessary. But that's
because the Trinitron I had, didn't have EDID.
Paul
--- PyGate Linux v1.5.14
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)