Benchmarks of ordito.bounds 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 bounds 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
PyVista
0.11x
3
1
MeshLib
2.1x
1
1
Open3D
9.1x
8
8
trimesh
102x
2
2
libigl
104x
3
3
aabb
dragon (871 414 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
PyVista
1.3 µs
0.02x
ordito (CUDA)
56.3 µs
1x
MeshLib
276.6 µs
4.9x
Open3D
853.5 µs
15x
libigl
10.15 ms
180x
trimesh
14.09 ms
250x
All inputs
Input
ordito (CUDA)
libigl
MeshLib
Open3D
PyVista
trimesh
sphere_small
50.8 µs
bunny_decimated
59.3 µs
197.9 µs
24.9 µs
15.8 µs
1.5 µs
275.1 µs
bunny
60.4 µs
863.0 µs
30.9 µs
67.6 µs
1.3 µs
1.23 ms
dragon
56.3 µs
10.15 ms
276.6 µs
853.5 µs
1.3 µs
14.09 ms
happy_buddha
63.3 µs
12.12 ms
362.9 µs
1.03 ms
1.5 µs
17.12 ms
lucy
296.3 µs
557.04 ms
2.22 ms
76.14 ms
1.2 µs
445.21 ms
enclosing_diagonal
dragon (871 414 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
PyVista
0.5 µs
0.01x
ordito (CUDA)
71.2 µs
1x
libigl
4.33 ms
61x
All inputs
Input
ordito (CUDA)
libigl
PyVista
sphere_small
89.9 µs
bunny_decimated
74.4 µs
126.5 µs
0.6 µs
bunny
69.9 µs
371.2 µs
0.6 µs
dragon
71.2 µs
4.33 ms
0.5 µs
happy_buddha
78.8 µs
5.38 ms
0.9 µs
lucy
317.3 µs
353.07 ms
0.9 µs
points_in_aabb
sphere_large (327 680 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
60.8 µs
1x
Open3D
758.2 µs
12x
All inputs
Input
ordito (CUDA)
Open3D
sphere_small
43.5 µs
13.6 µs
sphere_med
52.1 µs
195.3 µs
sphere_large
60.8 µs
758.2 µs
crop_points
sphere_large (327 680 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
48.8 µs
1x
Open3D
589.3 µs
12x
All inputs
Input
ordito (CUDA)
Open3D
sphere_small
71.4 µs
10.3 µs
sphere_med
70.6 µs
154.9 µs
sphere_large
48.8 µs
589.3 µs
crop_mesh
sphere_large (327 680 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
67.6 µs
1x
Open3D
10.67 ms
158x
All inputs
Input
ordito (CUDA)
Open3D
sphere_small
84.8 µs
129.2 µs
sphere_med
73.2 µs
2.20 ms
sphere_large
67.6 µs
10.67 ms
oriented_bounding_box
dragon (871 414 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
1.58 ms
1x
PyVista
18.34 ms
12x
trimesh
159.09 ms
100x
Open3D
2.096 s
1322x
libigl
2.483 s
1567x
All inputs
Input
ordito (CUDA)
libigl
Open3D
PyVista
trimesh
sphere_small
907.9 µs
bunny_decimated
986.2 µs
8.67 ms
192.82 ms
4.47 ms
32.22 ms
bunny
1.13 ms
35.57 ms
938.30 ms
4.71 ms
69.56 ms
dragon
1.58 ms
2.483 s
2.096 s
18.34 ms
159.09 ms
happy_buddha
1.69 ms
3.138 s
11.295 s
21.53 ms
306.08 ms
lucy
7.98 ms
138.582 s
3.222 s
1.057 s
2.642 s
points_in_aabb_mask
sphere_large (327 680 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
24.3 µs
1x
Open3D
543.9 µs
22x
All inputs
Input
ordito (CUDA)
Open3D
sphere_small
27.6 µs
16.9 µs
sphere_med
24.3 µs
139.5 µs
sphere_large
24.3 µs
543.9 µs
points_in_obb
sphere_large (327 680 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
42.4 µs
1x
Open3D
763.3 µs
18x
All inputs
Input
ordito (CUDA)
Open3D
sphere_small
59.3 µs
16.0 µs
sphere_med
42.2 µs
188.8 µs
sphere_large
42.4 µs
763.3 µs
points_in_obb_mask
sphere_large (327 680 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
17.5 µs
1x
Open3D
750.3 µs
43x
All inputs
Input
ordito (CUDA)
Open3D
sphere_small
19.7 µs
16.8 µs
sphere_med
18.7 µs
276.4 µs
sphere_large
17.5 µs
750.3 µs
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: