Hi Kevin, Thank you for the confirmation!
To make the test case simpler, so I did it that way. What I originally did is to put the picking function in a behavior, so I guess there should no synchronization problem for that. ---white On Tue, 12 Nov 2002 15:11:27 -0800, Kelvin Chung <[EMAIL PROTECTED]> wrote: >Hi White, > > Yes, it is a Java3D bug, regression from v1.2.1 > >4777750 - PickShape BoundingPolytope fail to compute intersection correctly > >is filed for this and will be fixed in next release. > >However, the application need some change to invoke >picking function AFTER java3d render the >first frame. It is because current architecture of >picking is perform asynchronously with other Java3D >threads.As a result the data structure responsible for >picking may not update when if it is done before first frame >render. > >Good test case ! > >Thanks. > >- Kelvin >----------------- >Java 3D Team >Sun Microsystems Inc. > > > > >White Morph wrote: >> Hi All, >> >> Here is the problem: I have a PickTool object with a PickBound as its PickShape. >> The pick mode is PickTool.GEOMETRY. The scene graph consists of 1 polygon. >> If there is any point of the polygon located inside the PickShape, then >> the PickTool object should intersect with the polygons, so it should return >> some PickResults, right? but if fails sometimes. >> >> I guess this is also not because of numerical error of calculation, because >> the bound object can correctly detect if the point is inside it while the >> PickTool fails to do that. >> >> Thanks---white >> =========================================================================== 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".
