I am also think this is an insteresting topic.
Before 3.0, Android graphics architecture is simple, especially for views animation. You know, to implement view animation, we engineer can use two way.
First way, just redraw every view till the animation's duration expires.
Second way, we can assign one special layer for every animated view, and perform animation in the surfaceflinger, this is just like iOS CoreAnimation done.
And after 3.0 version, android team use hardware acceleration method, because most engineer don't understand GPU's internal function and operation, they are happy, and thought Android animation will be good enough as iOS, but the fact is not.
I advice Android graphics team can re-construct the Graphics internal travasals logic. Talk to code, it may be:
frameworks/base/services/.../WindowManagerService.java
+SurfaceFlinger part.
+View/ViewGroup.java
So, if someone could clarify and point exactly the process how hardware accelerate in the view traversals, it will be much appreciated by most Android engineers, and I am also pleasure to work with this.
Yuandan
On Friday, May 11, 2012 8:40:59 AM UTC+8, Dianne Hackborn wrote:
I don't know what sessions there will be, but I think it unlikely for specifically this topic because Google IO sessions are focused on third party development not platform development.
Dear Android engineers,
I have a a kind request to conduct a session on Android Display
architecture.i seen many post related to android display system and
readding all together i am in a confused state.so why cant android
team think abt a session on display system to give a clear idea for us
and thus avoid foolish questions :)
Thanks and regards
Mansoor V.M
--
You received this message because you are subscribed to the Google Groups "android-platform" group.