[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-bugs-dist
Subject:    [Bug 235821] using kwin shadows with oxygen creates artifact
From:       Thomas Lübking  <thomas.luebking () web ! de>
Date:       2010-04-30 22:53:18
Message-ID: 20100430225318.BDD443EA53 () immanuel ! kde ! org
[Download RAW message or body]

https://bugs.kde.org/show_bug.cgi?id=235821





--- Comment #12 from Thomas Lübking <thomas luebking web de>  2010-05-01 00:53:14 \
--- (In reply to comment #11)
> With my patch, if you completely disable any painting in
> kcommondecoration::paintevent (with a return statement at the first line), you
> have a shadow painted and buttons on top of it. Looks like total alpha channel
> to me, and not broken. Is it ? Or what do I miss ? 

yes - "what casts the shadow" ;-)
(try the B II decoration to see what i meant)

> their own shadows, which oxygen did. Then I re-added the possibility to use
> kwin shadows on user request, which triggered a bug in kwin. 
i'd rather say the most important reason is that generic shadows cannot handle
aribtrary decoration (alpha)masks in a least efficient way.
(the shadow plugin is slow mainly due to some quite cumbersome code :)

> My view is: either we live with this performance drop for oxygen + fixed kwin
> shadow (which is _not_ the original design), or we drop the possibility to have \
> kwin shadows + oxygen (and leave the bug open).

...or have the decorations export their alphachannel to the effects for
multiplication =)
I'll work out a fix for the shadows this WE and then the overhead is quite
minor (compared to forcing naive bottom up painting as it is with this patch)

> However, shadows will still be made slower for decorations that use alpha channel, \
> right ?
yes, but that's not the same level.

> I can also revert my change and leave the original bug open. But modifying
> oxygen with ugly masks for this is not an option as far as I'm concerned ... 
I second Fredrik here: revert 1121187, that's no way a solution.

-- 
Configure bugmail: https://bugs.kde.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.


[prev in list] [next in list] [prev in thread] [next in thread]