This is a math library written in python for 2D/3D game development
which is also compatible with pypy. I made it while I was learning
more about the math used in graphics development and for personal use
in OpenGL related projects.
It’s still a work in progress.
It uses six to allow support between python2.x and python3.x.
To install the library just do
pip install gem
It will install the dependicies automatically.
The examples on how to use the library and more info are maintained on the github wiki:
- Transpose
- Scale
- NxN Matrix Multiplication
- NxN Matrix * N Dimensions Vector Multiplication
- 4x4 Perspective Projection Matrix
- lookAt 4x4 Matrix
- Translation (3x3, 4x4)
- Rotation (2x2, 3x3, 4x4)
- Shear (2x2, 3x3, 4x4)
- Project
- Unproject
- Orthographic Projection
- Perspective Projection
- lookAt 4x4 matrix
- Determinant 2x2, 3x3, 4x4
- Inverse 2x2, 3x3, 4x4
- Dot Product
- Cross Product (3D, No 7D as of now)
- 2D get angle of vector
- 2D -90 degree rotation
- 2D +90 degree rotation
- Refraction
- Reflection
- Negate
- Normalize
- Linear Interpolation
- Max Vector/Scalar
- Min Vector/Scalar
- Clamp
- Transform
- Barycentric
- isInSameDirection test
- isInOppositeDirection test
- 3D Vector swizzling, similar to GLSL
- 3D Vector idenitities
- Normalize
- Dot Product
- Rotation
- Conjugate
- Inverse
- Negate
- Rotate X, Y, Z
- Arbitary Axis Rotation
- From angle Rotation
- To Rotation Matrix (4x4)
- From Rotation Matrix (4x4)
- Cross Product
- Vector3D, Scalar Multiplication
- Logarithm
- Exponential
- Power
- Liner Interpolation (LERP)
- Spherical Interpolation (SLERP)
- Spherical Interpoliaton No Invert
- Quaternion Splines (SQUAD)
- Define using
- 3 Vectors
- Point and Normal
- Manual input
- Dot Product
- Normalize
- Best fit normal and D value
- Distance from plane to a point
- Point location
- Output
- Flip
- Rotate using Matrix
- Rotate using Quaternions
- Translate
- Output
- For spherical harmonics
- (l - m)PML(x) = x(2l - 1)PML-1(x
- Irradiance maps