D22468: Rewrite KScreen KCM as ConfigModule with outputs model and Kirigami

[Greg Varsanyi]

gvarsanyi added a comment.


  In D22468#496052 <https://phabricator.kde.org/D22468#496052>, @ngraham wrote:
  
  > I'm not sold on the proposal to use 1.25x increments, personally. For my personal use case, I like using 1.1x (13" 1080p screen) and I wouldn't want to lose that. Screen size and DPI combinations are so variable that I think it makes sense to support a higher level of precision. Apple can get away with their very low configurability because of tight control over the hardware that allows them to avoid selling machines where a given option wouldn't work.
  >
  > That said, I could get behind 0.25x increments if there's some "advanced" page or mods in the GUI that exposes 0.1 incrememebts, or even a spinbox or text box or something.
  >
  > Either is better than 0.2 increments though. As pointed out, that missed the all-important 1.5x.
  
  
  I think we'd all like to support everyone's use cases. My argument is that 0.25/0.75 steps are almost as important as 0.5. If devs/users are not concerned about too much granularity (e.g. artifacts, testing), this can be fixed easily by doing 0.05 increments - it's just a slider anyway.
  I also like the idea of adding a simple stepper with few predefined values, and an "advanced" configuration option for very precise settings.
  
  This makes me think: current KCM partially breaks if I set fractional scaling on Wayland to a non-integer value (the scale is not represented on the preview pane, and there is no selection in the scale dropdown). Probably a slider can better represent a value like 1.375x, even if it doesn't allow for picking it.

REPOSITORY
  R104 KScreen

REVISION DETAIL
  https://phabricator.kde.org/D22468

To: romangg, #plasma, #kwin
Cc: gvarsanyi, davidedmundson, ngraham, mvourlakos, knambiar, broulik, filipf, plasma-devel, LeGast00n, jraleigh, fbampaloukas, GB_2, ragreen, Pitel, ZrenBot, himcesjf, lesliezhai, ali-mohamed, jensreuterberg, abetts, sebas, apol, mart
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/plasma-devel/attachments/20190715/7b81d61e/attachment.html>