first of all, I am glad, that I am not the only one, having this problem.
J. Lee Dixon wrote:
This is very strange indeed. Somehow the order in which the spheres are calcutlated chages in dependancy to the y-value.Well, I played around with this for a while (very frustrating)... First I will say that yes I did see the problem. Certain spheres were blended with their background, whereas other spheres merely hid their background... very strange and unpredictable. At one point, I affect the problem by merely moving one sphere by 0.0000001 in the Y direction.
*g* I was trying to figure out, what exactly is different as well, after the message from Jan Koehnlein in comp.lang.java.3d.First, I commented out the "setDepthBufferFreezeTransparent()" call. Doing this, the scene rendered differently but still had the problem.
Ups - the same here!!! What is that?! Is that the point to start thinking (concerning a workaround?) Perhaps an Ordergroup with two subnodes (one for transparent one for opaque objects) or something like that could help... hmmmSecond, I commented out the call to create the cylinder in the middle. Poof! All the sphere transparencies worked. TRANSPARENCY IN JAVA3D IS OBVIOUSLY BROKEN. Adding one more thing to the scene should not make ANY difference.
Thanks, I take this as a compliment :-) I don't know if they did. Maybe they do know, as this mail is getting a few responses (I am so glad about that!)Has the Java3D team looked into this? This is a well documented example.
Yes, me as well. plus NT4sp6, ASUS V6600.BTW, I'm running J3D 1.2 and JDK 1.3.
Thanks for the time you spent playing around with this piece of code.
Best Regards,
Andreas Ebbert