weird VGA video problem on a laptop: the screen doesn't come on

Greg Stark (gsstark@mit.edu)
16 Jun 1997 20:11:48 -0400


I have a weird problem with the screen on my laptop that may be a hardware
fault, but even so may be caused by Linux. When I turn the computer on the
screen usually doesn't turn on, i don't even see the video BIOS copyright
messages. I know this sounds like a hardware or BIOS problem, but bear with
me, i really think it's somehow being triggered by Linux somehow.

The problem also occurs if the console blanking screen saver comes on (if i
have apm configured to turn the monitor off). With linux 2.0.25 running zgv (an
random svgalib program) or X repeatedly eventually caused the screen to come
on, but under 2.0.30 this trick doesn't seem to work any longer. The problem
never occurs when doing a soft reboot if the screen was working before the
reboot, and the problem rarely occurs if i wake up the screen blanking screen
saver immediately after sleeping, but usually occurs if i wake it up after
sitting for a long time.

So, question 1 is, do other people with the same hardware see this? This is a
Ascentia J/10 laptop, it has a C&T 65548 video card, and i've updated the BIOS
to the most recent on the AST site ("PlugnPlay BIOS Rel 3.11", "Phoenix
NoteBIOS 4.0", and "Keyboard BIOS Rev 1.08").

People have suggested it may be that the video card is set to drive the
external monitor which i don't have, but i can't fix the problem by merely
hitting the function key combination to switch that. Nor does the problem seem
to correlate in any way to when i hit that combination (to restore text mode
after running X).

My pet theory is that Linux is doing something funky when it turns the screen
off that somehow leaves the card in a strange state where neither the external
nor the internal screen is being driven and the bios function to switch them
doesn't work. Or maybe some kind of PnP card initializatin or lack thereof is
putting the card into a strange state? Or possibly the video card is getting
hot and that's causing some kind of problem with the initialization sequence?

I believe this is somehow related to Linux for several reasons. One, is that
the problem seems to go away when i boot to Windows . Also i find it
suspicious that the problem got much worse with 2.0.30 over 2.0.25. I wouldn't
be surprised if the fault ultimately lay with a video bios bogosity, but it's
the kind of thing that should be possible to work around.

Question 2 would be, anyone have any idea what's causing this problem or how
to fix it?

greg