Benchmarks of ordito.validation 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 validation 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
20x
3
3
trimesh
100x
5
5
MeshLab
208x
5
5
libigl
275x
5
5
Open3D
589x
4
4
PyMeshFix
1000x
2
2
PyVista
6182x
1
1
is_watertight
sphere_med (81 920 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
528.1 µs
1x
trimesh
16.39 ms
31x
MeshLab
125.86 ms
238x
Open3D
834.92 ms
1581x
All inputs
Input
ordito (CUDA)
MeshLab
Open3D
trimesh
tangle_2
514.7 µs
61.31 ms
231.84 ms
7.94 ms
sphere_med
528.1 µs
125.86 ms
834.92 ms
16.39 ms
is_volume
sphere_med (81 920 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
127.0 µs
1x
trimesh
59.75 ms
471x
MeshLab
127.59 ms
1005x
All inputs
Input
ordito (CUDA)
MeshLab
trimesh
tangle_2
150.0 µs
61.93 ms
27.51 ms
sphere_med
127.0 µs
127.59 ms
59.75 ms
face_self_intersecting_mask
sphere_med (81 920 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
398.7 µs
1x
MeshLib
5.53 ms
14x
MeshLab
101.44 ms
254x
PyMeshFix
424.55 ms
1065x
Open3D
759.57 ms
1905x
All inputs
Input
ordito (CUDA)
MeshLab
MeshLib
Open3D
PyMeshFix
tangle_2
408.0 µs
50.21 ms
4.84 ms
194.23 ms
187.51 ms
sphere_med
398.7 µs
101.44 ms
5.53 ms
759.57 ms
424.55 ms
face_defective_mask
saddle (34 848 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
107.9 µs
1x
MeshLib
2.34 ms
22x
MeshLab
16.74 ms
155x
All inputs
Input
ordito (CUDA)
MeshLab
MeshLib
saddle
107.9 µs
16.74 ms
2.34 ms
saddle_graded
103.6 µs
16.57 ms
2.42 ms
is_vertex_manifold
sphere_med (81 920 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
138.0 µs
1x
MeshLab
22.63 ms
164x
Open3D
72.09 ms
522x
libigl
117.09 ms
848x
All inputs
Input
ordito (CUDA)
libigl
MeshLab
Open3D
sphere_med
138.0 µs
117.09 ms
22.63 ms
72.09 ms
fan_hub
157.9 µs
130.83 ms
16.14 ms
66.70 ms
vertex_manifold_mask
sphere_med (81 920 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
142.7 µs
1x
libigl
127.71 ms
895x
All inputs
Input
ordito (CUDA)
libigl
sphere_med
142.7 µs
127.71 ms
fan_hub
179.1 µs
137.27 ms
is_edge_manifold
sphere_med (81 920 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
84.4 µs
1x
PyVista
19.20 ms
227x
libigl
30.02 ms
356x
Open3D
31.63 ms
375x
All inputs
Input
ordito (CUDA)
libigl
Open3D
PyVista
sphere_med
84.4 µs
30.02 ms
31.63 ms
19.20 ms
fan_hub
92.0 µs
13.05 ms
25.05 ms
15.451 s
is_winding_consistent
dragon (871 414 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
207.6 µs
1x
trimesh
305.94 ms
1474x
All inputs
Input
ordito (CUDA)
trimesh
sphere_small
99.4 µs
bunny_decimated
95.4 µs
3.80 ms
bunny
110.0 µs
20.33 ms
dragon
207.6 µs
305.94 ms
happy_buddha
372.7 µs
383.15 ms
lucy
7.92 ms
10.565 s
face_orientation_bits
sphere_med (81 920 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
317.8 µs
1x
trimesh
19.61 ms
62x
libigl
42.59 ms
134x
All inputs
Input
ordito (CUDA)
libigl
trimesh
ribbon_long
187.8 µs
13.20 ms
9.64 ms
sphere_med
317.8 µs
42.59 ms
19.61 ms
face_flip_mask
sphere_med (81 920 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
148.7 µs
1x
trimesh
19.20 ms
129x
libigl
40.20 ms
270x
All inputs
Input
ordito (CUDA)
libigl
trimesh
ribbon_long
336.7 µs
13.54 ms
10.25 ms
sphere_med
148.7 µs
40.20 ms
19.20 ms
is_self_intersecting
sphere_med (81 920 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
214.2 µs
1x
MeshLib
5.61 ms
26x
PyMeshFix
474.55 ms
2216x
All inputs
Input
ordito (CUDA)
MeshLib
PyMeshFix
tangle_2
218.8 µs
5.57 ms
202.10 ms
sphere_med
214.2 µs
5.61 ms
474.55 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: