libRR
/
Base
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Fixed projection matrix and made translation column-major

This commit is contained in:
Eric-Paul Ickhorn 2023-12-17 12:12:41 +01:00
parent b4235873b5
commit 6d7f2ba701
Signed by: epickh
GPG Key ID: F5EBBE013924D95F
1 changed files with 10 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
core/src-c/matrix.c
View File

@ -109,9 +109,9 @@ rr_vec4f_s rr_mat4f_multiply_vec4f(rr_mat4f_s matrix, rr_vec4f_s vector)
void rr_mat4f_translate(rr_mat4f_s *matrix, rr_vec3f_s xyz)
{
matrix->values[0][3] = xyz.x;
matrix->values[1][3] = xyz.y;
matrix->values[2][3] = xyz.z;
matrix->values[3][0] = xyz.x;
matrix->values[3][1] = xyz.y;
matrix->values[3][2] = xyz.z;
}
rr_mat4f_s rr_mat4f_rotate_x(float radians)
@ -166,31 +166,29 @@ rr_mat4f_s rr_mat4f_rotate_xyz(float x_rad, float y_rad, float z_rad)
rr_mat4f_s rr_mat4f_perspective(float fov, float near_plane, float far_plane, float aspect_ratio)
{
float plane_range = far_plane - near_plane;
float scale_x = (2 * near_plane) / (plane_range * aspect_ratio * 2);
float scale_y = near_plane / plane_range;
float scale_z = -(far_plane + near_plane) / (far_plane - near_plane);
float pos_z = tanf(fov / 2) * near_plane;
float a = tanf(fov / 2.0f) * near_plane;
float planes_sum = far_plane + near_plane;
float planes_range = far_plane - near_plane;
rr_mat4f_s matrix;
matrix.values[0][0] = scale_x;
matrix.values[0][0] = 1 / aspect_ratio;
matrix.values[0][1] = 0.0f;
matrix.values[0][2] = 0.0f;
matrix.values[0][3] = 0.0f;
matrix.values[1][0] = 0.0f;
matrix.values[1][1] = scale_y;
matrix.values[1][1] = aspect_ratio;
matrix.values[1][2] = 0.0f;
matrix.values[1][3] = 0.0f;
matrix.values[2][0] = 0.0f;
matrix.values[2][1] = 0.0f;
matrix.values[2][2] = scale_z;
matrix.values[2][2] = -(planes_sum / planes_range);
matrix.values[2][3] = -1.0f;
matrix.values[3][0] = 0.0f;
matrix.values[3][1] = 0.0f;
matrix.values[3][2] = pos_z;
matrix.values[3][2] = -(2 * planes_sum / planes_range);
matrix.values[3][3] = 0.0f;
return matrix;