gxnoised

float  gxnoised(vector2 xy, vector2 &deriv)

float  gxnoised(float x, float y, float &dx, float &dy)

float  gxnoised(vector xyz, vector &deriv)

float  gxnoised(vector4 xyzt, vector4 &deriv)

Simplex noise is similar to Perlin noise, except with samples on a simplex mesh rather than a grid. This family of simplex noise functions uses a different lattice structure and a cheaper accumulation method compared to xnoise.

The various functions return noise value at a 4D (vector4 argument), 3D (vector argument), or 2D (a single vector2 argument or two float inputs) position. They also compute spatial derivatives of the noise function. The evaluated noise field corresponds to the same one used by gxnoise functions that return float.

Noise values will be in the 0-1 range. Nature of the noise field depends on the number of input dimensions. Higher-dimensional noise uses tighter noise elements, and the resultant noise field appears more structured and less smooth. Consider using the slower xnoise function in higher dimensions if this function gives you undesirable results.