[00:08] <ochosi> i would start by initializing the scale to 1.0
[00:08] <ochosi> but obviously that's just part of the solution
[00:08] <ochosi> although... actually that may be fine in any case
[00:09] <ali1234> it will change the bug from "all monitors turn off when i plug or unplug one" to "all monitors scale resets to 1 when i plug or unplug one"
[00:09] <ochosi> the scale is in xfconf anyway and that's what it should try to load when going through the outputs
[00:09] <ali1234> which is slightly less annoying, but still really annoying
[00:10] <ochosi> did you set the scale by hand?
[00:10] <ochosi> i mean i guess you didn't
[00:10] <ochosi> cause otherwise xfsettingsd should have remembered it and applied it
[00:10] <ali1234> no, i never touched any scale setting
[00:10] <ochosi> all outputs that are connected are remembered in the "Default" profile
[00:11] <ochosi> so the scale is also retained there and that's what should be applied
[00:11] <ochosi> i'm just surprised you're the first one to run into this bug
[00:11] <ochosi> it sounds like i should have experienced it as well
[00:12] <ochosi> you could try to set the scale values in xfconf to 1.0 for one of your outputs
[00:12] <ochosi> just to see how that changes the behavior
[00:12] <ochosi> or to 2.0 if you want
[00:15] <ali1234> it wont change the behaviour
[00:15] <ali1234> because it calculates the screen size before the scale is loaded
[00:17] <ochosi> another issue i see with scale is that one time it's wrapped in #ifdef HAS_RANDR_ONE_POINT_THREE and the other time it's not
[00:18] <ali1234> yes
[00:18] <ali1234> there are a lot of problems with this code
[00:18] <ali1234> i'd have to read all of it to figure out hw to fix it properly
[00:20] <ali1234> https://bugzilla.xfce.org/show_bug.cgi?id=15971
[00:21] <ochosi> well, i'm afraid that would be the same for me too
[00:22] <ochosi> every time i have to fix something there i need to read the whole code
[00:23] <ali1234> well i'll ping perfinion with the bug, since he wrote this code
[00:24] <ali1234> i'll read the code tomorrow :)