My laptop is set to blank its display after 10 minutes, and turn it
off after 15.

(**) Option "BlankTime" "10"
(**) Option "StandbyTime" "0"
(**) Option "SuspendTime" "15"
(**) Option "OffTime" "0"

Alas, when I have an external USB mouse plugged in, it only ever goes
blank.  If I xset force dpms off, then it will turn the backlights off
on the internal monitor, and an external LCD, but then come back on at
a random time later, from seconds to 10 minutes later.  The display
remains blank though, but with the backlights turned on.

Eventually, I tracked this down to random 1 or 2 pixel spontaneous
moves of the mouse, and demonstrate that in the following example:
       while read -n 1 -e -s -t 10 c ; do  #read one character at a time with a 10 second timeout
           echo -n `date` -- ; ord "$c"
       done
       echo  #print a blank line every 10 seconds
    done
  ) < /dev/input/mice


Sat Feb 16 12:00:58 EST 2008 -- 24
Sat Feb 16 12:01:00 EST 2008 -- 255
Sat Feb 16 12:01:00 EST 2008 -- 0


<snip a few minutes of blank lines>


Sat Feb 16 12:06:39 EST 2008 -- 8
Sat Feb 16 12:06:39 EST 2008 -- 0
Sat Feb 16 12:06:40 EST 2008 -- 1
Sat Feb 16 12:06:40 EST 2008 -- 8
Sat Feb 16 12:06:40 EST 2008 -- 2
Sat Feb 16 12:06:40 EST 2008 -- 0


I find it odd that 1 or 2 pixel moves (presumably caused by light
variations in the optical mouse's detector) is enough to light the
backlight, but a bigger movement is required to unblank the display.
I suggest this means the X server needs to have a configurable
threshold of number of number of pixels of mouse movement in a certain
polling period, needed to retrigger the DPMS timer.

Until then, I need to unplug my external USB mouse every time I leave
my laptop so that I can turn off the displays.  This is particularly
annoying, since closing the laptop's screen turns the backlight off
through the dpms method using ACPI detection of the lid switch.  Since
this gets erroneously turned back on, my laptop's backlight is still
on even when its display is folded close.

Still having the problem today?

Brice

Yes.  On an
01:00.0 VGA compatible controller: nVidia Corporation G84 [GeForce 8600M GT] (rev a1)
running unstable (which can't currently run X at all due to a segfault in
nvidia_glx.so, but this was the case the last time it ran)
and

00:02.0 VGA compatible controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c)
00:02.1 Display controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c)

running xserver-xorg 1:7.4+4 and xserver-xorg-video-intel 2:2.9.1-1 (not
brave enough to upgrade it, given how my last upgrade went with the nvidia
machine)

And every other machine I have access to these days regardless of
distribution.  I suspect all optical mice suffer from 1 pixel jitters, and
being a hardware problem, it has to be worked around in software.

I have same problem, but the strange thing is that if the monitor is put to
sleep by the screensaver then id does not wake up spontaneously!
Best,
     Roman.