Re: turning off (LCD+) backlight without suspending
> > ... Second time, my screen goes totally blank (I assume
> > it's driving the signal out the external connector.) So
> > this should be using less power since it's not driving the
> > LCD or the backlight, but driving external (even absent)
> > devices can be a little bad for battery power too.
> You miss the point, external monitors are expected to have their *own*
> wall-power cord. Therefore the laptop is not expected to give them
> enough power to run their display.
> Yes, a minimal amount of power will remain in use by the video card
> itself, to send its own signals in the new direction. But if your
> backlight power cost is significant *at all* then using the external
> mode will negate that cost; whatever gain that's worth, you will get.
I think Greg's trick takes me at least 75% of the way there,
but the ideal would be to tell the whole video subsystem
to shut down (apm --video-off). If someone can point me to
apm bios interface specs, maybe I can check if there is even
a low-level interface to do this.
Yes, Signal drivers are negligible compared to displays but
can be significant, esp. if they conform to an old standard.
For example, compare battery life when running with RS232C
active (such as doing syncs with a Palm device) vs. inactive.
We have to drive +12V and -12V, so a whole DC/DC power
subsystem powers up. (RS232C was defined is from even before
Video-out drives an analog signal (ancient spec, but not as
bad as RS232 I'm sure) for several feet, while internal
video can theoretically be very low power and digital-only.
But yes, compared to powering the display, this power usage
is minor, and perhaps not worth worrying about. Please
forgive me for arguing this manini point.
Thanks Greg and Heather for your help!