D9848: Updated the blur method to use the more efficient dual kawase blur algorithm.

[Alex Nemeth]

anemeth updated this revision to Diff 25350.
anemeth added a comment.
Restricted Application edited projects, added KWin; removed Plasma.


  Reverted the change of putting the blur strength configuration into a vector, because it resulted in some complications.
  Doing it with the huge switch case might be unpleasent for the eyes but it does not require more unnecessary variables and
  (in my opinion) is cleaner and more maintainable this way.
  
  I strongly disagree with the intent to remove these magic numbers and let the user only change the offset or the iteration number.
  There aren't enough strength steps between the steps if we make one of these values constant.
  If we do this we remove a big part of customization and KDE is all about customization.
  Those 15 set of magic number numbers is the only price we have to pay for the customization.
  
  Said that, if the core members of Plasma or KWin decide/vote to make the offset constant and let the user only change
  between the 4 or 5 iteration number to make the code a bit leaner at the cost of customization then I will not be against the decision.

REPOSITORY
  R108 KWin

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D9848?vs=25305&id=25350

BRANCH
  master

REVISION DETAIL
  https://phabricator.kde.org/D9848

AFFECTED FILES
  .arcconfig
  effects/blur/blur.cpp
  effects/blur/blur.h
  effects/blur/blur.kcfg
  effects/blur/blur_config.ui
  effects/blur/blurshader.cpp
  effects/blur/blurshader.h

To: anemeth, #plasma, #kwin
Cc: broulik, romangg, zzag, anthonyfieroni, mart, davidedmundson, fredrik, ngraham, plasma-devel, kwin, #kwin, iodelay, bwowk, ZrenBot, progwolff, lesliezhai, ali-mohamed, hardening, jensreuterberg, abetts, sebas, apol
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/plasma-devel/attachments/20180114/a8c7276e/attachment.html>