Try using Flex 4 to see if it makes a difference.
Thanks for the suggestion.
I've been keeping to 3.5 to reduce the number of moving parts...
But now I've tried with Flex 4.0.0 SDK (and using FB4-beta2, by the way)
The same results. (although instead of a 'grey' box, I now get a 'white' box)
(and these results hold on both Chrome and IE7)
For bonus points: when I try to use Flex-4 SDK from cmd-line ant,
[having redefined FLEX_HOME=c:/INSTALL/Adobe/AdobeFlashBuilderPlug-inBeta2/sdks/4.0.0 (the correct path)]
c:\Data\Programs\Hiaim-FlexBuilder3\videocomp\build.xml:33: The class not found in jar file: mxmlc.jar
By the way, to aid in testing this,
is there a way to detect which version of globalplayer.swc is loaded?
Ok file a bug with a simple test case.
Loaded by the compiler? Playerglobal is built into the FP. You can test
for various FP10 APIs to see what player version the FP is actually running
your SWF in.
Ok, i've filed the bug: https://bugs.adobe.com/jira/browse/FP-4168
I assume the compiler records the version to be used in the .swf,
and the FP binds/uses to the appropriately restricted 'api' for backward compatibility?
[like: refusing to admit to NetStream.nearID if the app was compiled for FP-9]
Still, it would be nice to discover what globalplayer.swc is being used at runtime.
Because sometimes the tools find old versions, or web servers or browser caches bring up older versions,
and it would be nice if i could display the version actually being used to save confusion.