I believe that everything was off by a factor of 2. What would be awesome is
if regardless of screen resolution or DPI, I could always assume that the
scaling magic was related to a width of 768, which I could create as a
constant and problem solved. That is what I did for the renderer and it
worked like a charm, but I don't know if that will work for android or all
iPads??

I have a bigger issue now. The TypeErrors are gone, and all of the measuring
seems to have been massively reduced, but the performance is not much
better, only slightly. When tethered to Scout, the iPad button click still
takes about 5000ms to slide the list onto the screen. Untethered, I would
estimate it to be about 3000ms. So now the question comes down to two parts:

1) Would an actionscript itemrenderer make a noticeable difference? I am a
little daunted to try it because I never have, and you already know that I
have 30 labels and radiobutton and I know there will be a lot of overrides.
But if you believe it will make a significant improvement, I could give it a
try.

2) When the list moves up and completely covers the content behind it, how
much does the content that it covers affect performance? I always assumed
that it all stayed in the display list and was simply covered up
temporarily. But on the main screen, I have 20 dropdownbuttons, buttons, and
textinputs in heavily nested groups as well as 3 lists with itemrenderers,
only one of which is visible at a time. If that all has to be removed and
redrawn when the list transitions in and out, then maybe I need to trim the
fat out of that code as well???

Any thoughts are appreciated. The desktop version works great, but right
now, the mobile version just does not feel like a professional app :(



--
View this message in context: 
http://apache-flex-users.2333346.n4.nabble.com/Scout-What-does-this-mean-tp14126p14156.html
Sent from the Apache Flex Users mailing list archive at Nabble.com.

Reply via email to