Benchmarks of ordito.selection against every library that computes the same thing. Each chart shows the time of every library relative to ordito on CUDA (1x, in green): a longer bar took longer. The table under each chart lists every input the function was measured on. See Benchmarks for how to read and reproduce these numbers.
Time relative to ordito (CUDA), geometric mean over the selection functions both implement; in parentheses, the number of functions. Log scale. Runs stopped at the time limit are left out.Values
Library
Time relative to ordito
Functions compared
ordito faster in
MeshLib
3x
5
2
Open3D
24x
1
1
PyVista
70x
2
2
MeshLab
74x
2
2
trimesh
99x
3
3
expand_vertex_mask
sphere_med (81 920 faces) · hops=1 · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
28.2 µs
1x
MeshLib
536.0 µs
19x
MeshLab
1.09 ms
39x
All inputs
Input
ordito (CUDA)
MeshLab
MeshLib
sphere_med hops=1
28.2 µs
1.09 ms
536.0 µs
sphere_med hops=8
57.9 µs
7.61 ms
4.94 ms
shrink_vertex_mask
sphere_med (81 920 faces) · hops=1 · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
27.7 µs
1x
MeshLib
939.3 µs
34x
MeshLab
1.25 ms
45x
All inputs
Input
ordito (CUDA)
MeshLab
MeshLib
sphere_med hops=1
27.7 µs
1.25 ms
939.3 µs
sphere_med hops=8
77.1 µs
9.91 ms
3.57 ms
submesh_from_face_indices
sphere_med (81 920 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
54.0 µs
1x
Open3D
1.27 ms
24x
PyVista
4.39 ms
81x
trimesh
6.36 ms
118x
All inputs
Input
ordito (CUDA)
Open3D
PyVista
trimesh
sphere_med
54.0 µs
1.27 ms
4.39 ms
6.36 ms
delete_region_keep_boundary
sphere_med (81 920 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
MeshLib
389.0 µs
0.91x
ordito (CUDA)
426.8 µs
1x
All inputs
Input
ordito (CUDA)
MeshLib
sphere_med
426.8 µs
389.0 µs
region_boundary_edges
sphere_med (81 920 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
MeshLib
34.4 µs
0.26x
ordito (CUDA)
133.2 µs
1x
PyVista
7.97 ms
60x
All inputs
Input
ordito (CUDA)
MeshLib
PyVista
sphere_med
133.2 µs
34.4 µs
7.97 ms
faces_left_of_contour
sphere_med (81 920 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
MeshLib
115.2 µs
0.61x
ordito (CUDA)
190.1 µs
1x
All inputs
Input
ordito (CUDA)
MeshLib
sphere_med
190.1 µs
115.2 µs
submesh_from_face_mask
sphere_med (81 920 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
80.9 µs
1x
trimesh
6.91 ms
85x
All inputs
Input
ordito (CUDA)
trimesh
sphere_med
80.9 µs
6.91 ms
submesh_from_vertex_indices
sphere_med (81 920 faces) · face_mode=all · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
89.1 µs
1x
trimesh
8.12 ms
91x
All inputs
Input
ordito (CUDA)
trimesh
sphere_med face_mode=all
89.1 µs
8.12 ms
sphere_med face_mode=any
79.5 µs
8.22 ms
Without a reference
No other library computes these, so they have no comparison. ordito (CUDA)'s median time on the largest input it was measured on: