Rotatexyz does not work as expected when using with jit.matrix/jit.gl.multiple
Hello!
Please, help me understand what is wrong. While experimenting with creating multiple object instances using jit.gl.multiple, I noticed, that the behaviour of 'rotatexyz' differs when the the message is sent directly to jit.gl.gridshspe and when I'm modifying the jit.matrix, connected to jit.gl.multiple.
Here is a simplified example (screenshot attached):
Can you, please, explain what the problem is? Is this a bug, or am I doing something wrong?
Thank you!
Hey, Raja!
Thank you for trying to help me! To be honest I still don't understand why specifying the same numbers to matrix and gridshape behaves so differently.
When you send a 90º rotation to a single cube(or any multiple of 90 like 180, 270, 360, etc.), a single cube rotated 90º will still look like the same exact cube(because cubes have sides of equal length/dimensions)
In my example the cube is scaled to have non-equal sides. And I specifically created example with one and only cube, so there is no actual multiplication going and I'm not sure the coordinates of gridshape and matrix spaces are different...
Let me rephrase my initial question: Is it possible to rotate the scaled cube so it stand vertically by modifying jit.matrix? Here is the picture of what exactly I want:
Thanks again!
Exactly! Here is the patch that clearly illustrate the problem: when you are rotating the scaled object by modifying jit.matrix, the scaled object is heavily deformed while rotating on all axis, except the axis on which it was scaled:
Looks like a bug to me.
I might be wrong thinking, that this is necessarily a bug. John Jannon pointed me, that it is not a good idea to use local scaling, while object is rotating through a scaling matrix. If you control all aspects from jit.gl.multiple, the object rotates without any deformation:
