There are no messages from the JVM or exceptions that are thrown when the canvas hangs. By the way, the application doesn't hang. Its only the canvas. The menu bar and textfields for the application still works after the canvas hangs. This explains why MouseRotate and MouseTranslate does not work either, after the canvas hangs.
I tried printing out 1+(dy*scale_factor). There is no pattern in the range of the expression when the canvas hangs. It will hang regardless of whether the expression evaluates to greater or smaller than 1 but have yet to reach negative numbers. Sorry, I am really bad at this. I am still new in J3D and have been stuck at this for weeks. Thanks for your patience. Jasmine Kwok ----- Original Message ----- From: "Mark Hood" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Wednesday, February 19, 2003 9:55 AM Subject: Re: [JAVA3D] Modified MouseZoom but canvas hangs > > Date: Wed, 19 Feb 2003 09:36:08 +1000 > > From: Jasmine Kwok <[EMAIL PROTECTED]> > > > > I checked for if (1+(dy*scale_factor)) to ensure that its greater than 0 > > before i go through the transformation but it doesn't help. > > If this is a view platform transform, is the scaling the same across X, Y, and > Z? Are you getting any BadTransformExceptions? (Catch BadTransformException > when doing a TransformGroup.setTransform() to be sure). Is there any message > from the JVM when you find your application hanging? > > When zooming at the extreme ends of the scale, it's not uncommon to get > non-congruent transforms due to numerical inaccuracy, especially if you're > accumulating incremental zoom factors into a single matrix. > > > I do not quite understand about the part on using Math.pow. Do you mean that > > I can try to replace if (1+(dy*scale_factor)) with Math.pow(2.0, exp)? And > > exp will be calculated according to the mouse movement? > > No, it was just an alternative way to compute a scale factor. I just suggested > it because it gets rid of the problem of scaling to 0 and produces a nice > geometric zoom instead of a linear zoom, without having to accumulate > incremental scale factors. But it's not at the root of your problem. > > -- Mark Hood > > =========================================================================== > 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". > =========================================================================== 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".
