I'm using Processing 2, and have multiple tabs. The exception is being thrown via an Interface, but the stacktrace isn't even showing the interface, only the original PDE filename.
I can of course println statements to try and figure out where all is going wrong, but that's a pretty tedious process. Is there a better way to get the accurate place the exception is occurring?
caused by: java.lang.ClassCastException: com.leapmotion.leap.FingerList cannot be cast to java.util.List at LeapTest$LeapDevice.getFingerVectors(LeapTest.java:134) at LeapTest$LeapDevice.getPos(LeapTest.java:119) at LeapTest.draw(LeapTest.java:41)
LeapTest is the original PDE file, LeapDevice is the interface where the exception is actually occurring.
I'm starting a project at work which is going to involve moving to and from areas in 3d space. In those spaces, multiple things will be happening and the user can leave and enter a space at will.
I'm deciding on what would be the best way to accomplish this and I've come up with these alternatives:
1. Center the camera in 3d space and draw the corresponding area objects. When the user indicates movement to another area, I transition the objects off the screen( left or right ) and bring in the new ones.
2. Draw *all* the areas and area objects as a linear progression starting from point 0 to point maximum and move the camera along the line of movement, left or right.
3. Draw all the areas and area objects in a circle with a stationary camera, and then rotate the circle a discrete amount clockwise or counter clockwise, depending on the users direction, to enter or leave a new area.
I'm trying to think of the pitfalls of each approach. Drawing the objects all at once at then having the camera move around to view them seems like the more simple approach, but I guess I'm worried about performance being an issue with all the area objects being updated, even if they aren't visible to the camera. Conversely, having to update object position and track their entry and exit from the left and right of the screen seems like a fairly complicated task.
Any insight or references to examples would be greatly appreciated.
The version of perspective without parameters sets the default perspective and the version with four parameters allows the programmer to set the area precisely. The default values are: perspective(PI/3.0, width/height, cameraZ/10.0, cameraZ*10.0) where cameraZ is ((height/2.0) / tan(PI*60.0/360.0));
But when I plug in the values for perspective and call it explicitly, my view looks different than if I called perspective() with no parameters.
Perhaps I am misreading the docs, but shouldn't perspective() and perspective(params) as stated above produce the same view? I'm using the OPENGL renderer.
I have been playing around with OpenCV and Daniel Schiffman's Kinect library this week with Processing, and everything has been working fine.
But after a update to ITunes 10.2.1, I can no longer access my capture devices through OpenCV or the processing.video.* packages, or any of the included video examples. At least I think ITunes is the problem, because I updated it yesterday and since, my video in Processing is broken. I've rolled back ITunes to 10.1.1 to no avail.
Does anyone have any tips for troubleshooting this? As I understand it, there should be no setup shenanigans with OSX to get video in Processing.
Using OSX 10.6.6 on a Macbook Pro and Processing 1.2.1.