Panda is great! But...

...there are some details that sorta spoil the fun.

One thing i dislike is for once the fact that the libraries are quite big. That's not really a problem, if your game is big, but an unoptimized "hello world" program took about 60 megabyte. Nodody would download a small game with such a size! That's an issue it owes to it's huge functionality...

Another thing i am sorta unhappy with is the matrix/vector functionality. I compared it to the implementation of the (supposedly discontinued) "Python Computer Graphics Kit" and i prefer the cgkit's version much more. For example you can multiply a vector (Vec3) directly with a transformation matrix (Mat4) via the normal multiply operator (*), which isn't really crucial, but a nice touch:

Panda3D:

myVector=Vec3(2.0,2.0,2.0)

myMatrix=Mat4()

myMatrix.scaleMat(Vec3(2.0,2.0,2.0))

Mat4(2, 0, 0, 0, 0, 2, 0, 0, 0, 0, 2, 0, 0, 0, 0, 1)

myMatrix.xformPoint(myVector)

VBase3(2, 2, 2)

myVector

Vec3(2, 2, 2)

(which isn't the correct result anyways... wtf?!)

CGKit:

myMatrix=mat4(1.0)

myMatrix.scale(vec3(2.0,2.0,2.0))

[2, 0, 0, 0]

[0, 2, 0, 0]

[0, 0, 2, 0]

[0, 0, 0, 1]

myVector=vec3(2.0)

myMatrix*myVector

(4, 4, 4)

Another one is the fact that when Panda3D is scaling a transformation matrix, it overwrites the matrix with a comletely new scale matrix:

Panda3D:

myMatrix=Mat4()

myMatrix

Mat4(1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1)

myMatrix.setTranslateMat(Vec3(2.0,3.0,4.0))

myMatrix

Mat4(1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 2, 3, 4, 1)

myMatrix.setScaleMat(Vec3(2.0,2.0,2.0))

myMatrix

Mat4(2, 0, 0, 0, 0, 2, 0, 0, 0, 0, 2, 0, 0, 0, 0, 1)

...does pretty much the same as...

myMatrix=Mat4()

myMatrix

Mat4(1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1)

myMatrix.translateMat(Vec3(2.0,3.0,4.0))

Mat4(1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 2, 3, 4, 1)

myMatrix.scaleMat(Vec3(2.0,2.0,2.0))

Mat4(2, 0, 0, 0, 0, 2, 0, 0, 0, 0, 2, 0, 0, 0, 0, 1)

...whereas CGKit is doing this:

myMatrix=mat4(1.0)

myMatrix.translate(vec3(2.0,3.0,4.0))

[1, 0, 0, 2]

[0, 1, 0, 3]

[0, 0, 1, 4]

[0, 0, 0, 1]

myMatrix.scale(vec3(2.0,2.0,2.0))

[2, 0, 0, 2]

[0, 2, 0, 3]

[0, 0, 2, 4]

[0, 0, 0, 1]

I have no clue why, but i am not too experienced with that stuff so far, so maybe i'm just thinking wrong. Nevertheless most of the time you don't need that kind of stuff anyways, Panda3D does a great job keeping these issues away from the user. I just had to use these vector calculations for my tree project.