Module f64 

Structs

EulerAngle

Represents a set of Euler angles.

Mat2

A 2x2 matrix. Data and operations are column-major.

Mat3

A 3x3 matrix. Data and operations are column-major.

Mat4

A 4x4 matrix. Data and operations are column-major.

Quaternion

A quaternion using Hamilton (not JPL) transformation conventions. The most common operations usedful for representing orientations and rotations are defined, including for operations with Vec3.

Quaternionx4

SoA. Performs operations on x quaternions

Quaternionx8

SoA. Performs operations on x quaternions

Vec2

A len-2 column vector.

Vec3

A len-3 column vector.

Vec4

A len-4 column vector

Vec3x4

SoA. Performs operations on x Vecs.

Vec3x8

SoA. Performs operations on x Vecs.

Vec4x4

SoA. Performs operations on x Vecs.

Vec4x8

SoA. Performs operations on x Vecs.

f64x4

Similar to core::simd.

f64x8

Similar to core::simd.

Constants

X_VEC

Y_VEC

Z_VEC

Functions

calc_dihedral_angle

Calculate the dihedral angle between 4 positions (3 bonds). Compared to calc_dihedral_angle_v2(), this function’s API is more clear if you have the bonds/connections, but not the positions. this might happen if you are doing certain vector operations, for example.

calc_dihedral_angle_v2

Calculate the dihedral angle between 4 positions. The positions must be in order by connection/bond, although both directions produce identical results. Compared to calc_dihedral_angle(), this function’s API is more clear if you have the set of positions directly.

det_from_cols

Calculate the determinate of a matrix defined by its columns. We use this for determining the full 0 - tau angle between bonds.

pack_quaternionx4

Used for creating a set of Quaternionx from one of Quatenrion. Padded as required. The result will have approximately x fewer elements than the input.

pack_quaternionx8

Used for creating a set of Quaternionx from one of Quatenrion. Padded as required. The result will have approximately x fewer elements than the input.

pack_slice

Convert a slice of any type to an array values, for use with SIMD. Padded as required. The result will have approximately 8x fewer elements than the input.

pack_slice_noncopy

pack_vec3x4

Used for creating a set of Vec3x from one of Vec3. Padded as required. The result will have approximately x fewer elements than the input.

pack_vec3x8

Used for creating a set of Vec3x from one of Vec3. Padded as required. The result will have approximately x fewer elements than the input.

pack_x4

Convert a slice of $f to an array of SIMD $f values, x-wide. Padded as required. The result will have approximately 8x fewer elements than the input.

pack_x8

Convert a slice of $f to an array of SIMD $f values, x-wide. Padded as required. The result will have approximately 8x fewer elements than the input.

unpack_quaternionx4

Convert a slice of SIMD Quaternionx values, x-wide to Quaternion. The result will have approximately x as many elements as the input. Its parameters include the number of original values, so it knows to only use valid lanes on the last chunk.

unpack_quaternionx8

Convert a slice of SIMD Quaternionx values, x-wide to Quaternion. The result will have approximately x as many elements as the input. Its parameters include the number of original values, so it knows to only use valid lanes on the last chunk.

unpack_slice

unpack_vec3x4

Convert a slice of SIMD Vec3x values, x-wide to Vec3. The result will have approximately x as many elements as the input. Its parameters include the number of original values, so it knows to only use valid lanes on the last chunk.

unpack_vec3x8

Convert a slice of SIMD Vec3x values, x-wide to Vec3. The result will have approximately x as many elements as the input. Its parameters include the number of original values, so it knows to only use valid lanes on the last chunk.

unpack_x4

Convert a slice of SIMD floating point values to plain floating point ones. The result will have approximately x as many elements as the input. Its parameters include the number of original values, so it knows to only use valid lanes on the last chunk.

unpack_x8

Convert a slice of SIMD floating point values to plain floating point ones. The result will have approximately x as many elements as the input. Its parameters include the number of original values, so it knows to only use valid lanes on the last chunk.