A performance bug was filed against j3d1.2.1FCS, yesterday :
BugId 4427335 - Severe performance regression in 1.2.1 during attach/detach
Bug Summary : The time to perform an addChild()/detach() operation on a
BranchGroup get significantly lenghten after similar
operations is performed on the same BranchGroup.
This bug was exposed by one of our bug fixes in j3d1.2.1FCS, but we didn't
catch it in our pre-release testing. ;-( It will be fixed in our next release.
For now, we do have a workaround for whoever need it.
This bug will only surface if you've a BranchGroup, with a TransformGroup as
it's immediate parent, and detach and addChild is performed on that
BranchGroup. To workaround this bug, simply add a dummy Group node between
the TransformGroup and the BranchGroup.
TG TG
| ===> |
BG G <--- add a Group node between TG and BG, if you
| need to do detach() and addChild() at the BG.
BG
Thanks,
Chien Yang
Java 3D Team.
===========================================================================
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff JAVA3D-INTEREST". For general help, send email to
[EMAIL PROTECTED] and include in the body of the message "help".
