On Saturday, January 7, 2017 9:04:27 AM CET Joe wrote: > Is the screen truly off? If you look at it from unusual angles, can you > see any faint sign of the correct display?
I've further investigated what's going on, and yes, the screen is truly off-- i.e. the backlight is completely off. First an introduction. Screen brightness can be adjusted writing in "/sys/ class/backlight/intel_backlight/brightness" a value between 0 and max_brightness, in my case 1388. This value gets copied in "/sys/class/backlight/intel_backlight/ actual_brightness". Ultimately the brightness of the screen is decided by "actual_brightness", but you can't write directly to that file. When you switch off the backlight with "sleep 1 && xset dpms force off", the value in "brightness" doesn't change (for instance it remains equal to 1388), while only the value in /actual_brightness" is set to 0. Now, regarding my analysis, I've done it with the help of the usual ssh. When the screen is off, as I said this means that "actual_brightness" is 0, while "brightness" can be any value (0, 1388, 694, whatever), depending on the specific situation, but this is not relevant. Now, if I hibernate the system, and then resume it, what I find with ssh is that "brightness" is 1388, but "actual_brightness" is still 0, meaning the screen is indeed off. It's not possible to directly change the value in "actual_brightness", and in contrast to the usual behaviour, changing the value in "brightness" has no effect on "actual_brightness". As said in my previous posts, also "xset - display :0 dpms force on" doesn't change the value in "actual_brightness".
