Turning of cacheAsBitmap
-> Scrolling got much slower (in the renderer there is nothing
changing/animating)
Am I right when I think all the fxg files are cached and using bitmap
images won't have any advantage (at least when cachingAsBitmap)?
Turning of usage of virtual layout
-> Can't see any effect
Removing Empty Groups
-> FEELS a little bit better
Actually I had no time to performance-test it (use intellij idea but
don't like the intellij profiler... and Adobe Scout ... and have to
switch to FlashBuilder)... but will post if I'll find out something that
helps...
Thanks for all the tips,
Christian
PS: If you are interested in the rederers current performance - I am
talking about that:
http://beta.goal-games.de/ -> (beta just working on Android)
... register, choose a team name, close all popus, swipe to the right,
click on the "zum Kader" - Button on the "Mannschaft"- Screen and start
scrolling)
Am 22.07.2013 06:01, schrieb jude:
Turning of cacheAsBitmap can help? Really?
CacheAsBitmap has some interesting performance implications. Sometimes it
is better to use it and sometimes, it is better to not use it.
I suggested this to test if something is changing the display list in a way
that cacheAsBitmap has to repeatedly recache the bitmap then it will be
much slower. I think there is a feature in Scout to see if this is
happening. You can also strategically use cacheAsBitmap when you are sure
the content will stay the same.
http://www.republicofcode.com/tutorials/flash/as3cacheasbitmap/
And an mxml item renderer can
improve performance?
MXML rarely outperforms good AS code.
Right. I should have mentioned that. I suggested this so you could see if
there is something in your code that is causing the performance loss. If
you create the same renderer in MXML you can use that as a baseline to see
how much performance you are gaining with your renderer.
When I mentioned FXG optimizations I thought that was an option in Flash
Builder but it's actually an option in Flash Catalyst. When you import an
FXG file it converts it into MXMLG converting an FXG Group into a heavier
Spark Group container. Choosing Optimize Vector Graphics just converts
MXMLG back into an FXG file. So in this case you're already using FXG.
