Benchmarks of ordito.smoothing 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 smoothing 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
27x
9
9
PyVista
142x
2
2
Open3D
222x
4
4
trimesh
322x
3
3
MeshLab
347x
8
8
filter_mut_dif_laplacian
dragon (871 414 faces) · volume_constraint=False · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
971.2 µs
1x
Open3D
140.54 ms
145x
MeshLab
882.23 ms
908x
trimesh
1.527 s
1572x
All inputs
Input
ordito (CUDA)
MeshLab
Open3D
trimesh
sphere_small volume_constraint=False
252.7 µs
sphere_small volume_constraint=True
536.5 µs
bunny_decimated volume_constraint=False
285.4 µs
10.47 ms
2.88 ms
14.64 ms
bunny_decimated volume_constraint=True
621.0 µs
94.94 ms
bunny volume_constraint=False
269.0 µs
47.59 ms
11.58 ms
65.86 ms
bunny volume_constraint=True
575.0 µs
418.83 ms
dragon volume_constraint=False
971.2 µs
882.23 ms
140.54 ms
1.527 s
dragon volume_constraint=True
1.68 ms
8.496 s
happy_buddha volume_constraint=False
1.19 ms
1.104 s
397.56 ms
1.857 s
happy_buddha volume_constraint=True
1.98 ms
10.477 s
lucy volume_constraint=False
29.77 ms
36.927 s
12.056 s
43.836 s
lucy volume_constraint=True
51.03 ms
287.374 s
filter_laplacian_integration
saddle (34 848 faces) · implicit=False · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
84.4 µs
1x
MeshLib
4.78 ms
57x
PyVista
10.31 ms
122x
MeshLab
18.17 ms
215x
All inputs
Input
ordito (CUDA)
MeshLab
MeshLib
PyVista
saddle implicit=False
84.4 µs
18.17 ms
4.78 ms
10.31 ms
saddle_graded implicit=False
78.2 µs
18.59 ms
4.63 ms
9.97 ms
saddle implicit=True
1.46 ms
saddle_graded implicit=True
1.48 ms
filter_taubin
dragon (871 414 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
476.5 µs
1x
PyVista
153.29 ms
322x
trimesh
198.54 ms
417x
Open3D
306.02 ms
642x
MeshLab
867.52 ms
1821x
All inputs
Input
ordito (CUDA)
MeshLab
Open3D
PyVista
trimesh
sphere_small
129.7 µs
bunny_decimated
125.7 µs
10.29 ms
2.87 ms
3.05 ms
3.32 ms
bunny
107.4 µs
46.42 ms
11.62 ms
12.18 ms
14.19 ms
dragon
476.5 µs
867.52 ms
306.02 ms
153.29 ms
198.54 ms
happy_buddha
573.3 µs
1.093 s
392.43 ms
197.28 ms
256.77 ms
lucy
13.36 ms
36.409 s
11.933 s
4.869 s
8.334 s
filter_humphrey
dragon (871 414 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
1.07 ms
1x
trimesh
315.21 ms
296x
MeshLab
614.65 ms
577x
All inputs
Input
ordito (CUDA)
MeshLab
trimesh
sphere_small
152.9 µs
bunny_decimated
162.9 µs
7.82 ms
5.11 ms
bunny
229.7 µs
35.79 ms
21.43 ms
dragon
1.07 ms
614.65 ms
315.21 ms
happy_buddha
1.24 ms
790.17 ms
406.41 ms
lucy
28.55 ms
27.413 s
11.964 s
filter_scalar_laplacian
dragon (871 414 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
112.9 µs
1x
MeshLab
507.20 ms
4491x
All inputs
Input
ordito (CUDA)
MeshLab
sphere_small
80.6 µs
bunny_decimated
84.9 µs
6.50 ms
bunny
101.4 µs
29.24 ms
dragon
112.9 µs
507.20 ms
happy_buddha
151.9 µs
642.41 ms
lucy
5.31 ms
22.436 s
filter_normals
saddle (34 848 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
189.0 µs
1x
MeshLab
11.92 ms
63x
MeshLib
26.93 ms
142x
All inputs
Input
ordito (CUDA)
MeshLab
MeshLib
saddle
189.0 µs
11.92 ms
26.93 ms
saddle_graded
280.8 µs
11.72 ms
27.21 ms
filter_two_step
saddle (34 848 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
847.8 µs
1x
MeshLab
121.36 ms
143x
All inputs
Input
ordito (CUDA)
MeshLab
saddle
847.8 µs
121.36 ms
saddle_graded
982.1 µs
121.88 ms
filter_sharpen
dragon (871 414 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
325.5 µs
1x
Open3D
458.65 ms
1409x
MeshLab
891.58 ms
2739x
All inputs
Input
ordito (CUDA)
MeshLab
Open3D
sphere_small
61.0 µs
bunny_decimated
70.0 µs
11.91 ms
3.55 ms
bunny
84.7 µs
53.00 ms
15.84 ms
dragon
325.5 µs
891.58 ms
458.65 ms
happy_buddha
384.5 µs
1.146 s
574.37 ms
lucy
8.15 ms
39.053 s
16.905 s
smooth_region_fixed_rim
dragon (871 414 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
58.53 ms
1x
MeshLib
162.23 ms
2.8x
All inputs
Input
ordito (CUDA)
MeshLib
sphere_small
881.6 µs
bunny_decimated
3.36 ms
9.82 ms
bunny
6.24 ms
15.27 ms
dragon
58.53 ms
162.23 ms
happy_buddha
62.52 ms
245.41 ms
lucy
11.183 s
66.059 s
filter_spikes
dragon (871 414 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
1.07 ms
1x
MeshLib
55.89 ms
52x
All inputs
Input
ordito (CUDA)
MeshLib
sphere_small
96.7 µs
bunny_decimated
505.4 µs
8.09 ms
bunny
414.2 µs
9.19 ms
dragon
1.07 ms
55.89 ms
happy_buddha
1.04 ms
114.12 ms
lucy
3.30 ms
3.711 s
equalize_triangle_areas
dragon (871 414 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
944.5 µs
1x
MeshLib
65.00 ms
69x
All inputs
Input
ordito (CUDA)
MeshLib
sphere_small
300.4 µs
bunny_decimated
316.1 µs
8.43 ms
bunny
282.0 µs
10.39 ms
dragon
944.5 µs
65.00 ms
happy_buddha
1.15 ms
88.09 ms
lucy
26.75 ms
3.995 s
relax_keep_volume
dragon (871 414 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
1.71 ms
1x
MeshLib
68.72 ms
40x
All inputs
Input
ordito (CUDA)
MeshLib
sphere_small
408.5 µs
bunny_decimated
381.3 µs
8.56 ms
bunny
370.4 µs
10.43 ms
dragon
1.71 ms
68.72 ms
happy_buddha
1.94 ms
92.59 ms
lucy
53.76 ms
4.382 s
smooth_region_boundary
dragon (871 414 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
8.36 ms
1x
MeshLib
110.78 ms
13x
All inputs
Input
ordito (CUDA)
MeshLib
sphere_small
780.0 µs
bunny_decimated
1.21 ms
11.00 ms
bunny
1.17 ms
15.63 ms
dragon
8.36 ms
110.78 ms
happy_buddha
15.84 ms
123.38 ms
lucy
19.80 ms
4.752 s
filter_neighborhood_average
dragon (871 414 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
1.54 ms
1x
Open3D
681.76 ms
442x
All inputs
Input
ordito (CUDA)
Open3D
sphere_small
256.7 µs
bunny_decimated
232.9 µs
6.54 ms
bunny
315.1 µs
31.36 ms
dragon
1.54 ms
681.76 ms
happy_buddha
1.85 ms
879.92 ms
lucy
47.92 ms
20.387 s
smooth_region
dragon (871 414 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
183.88 ms
1x
MeshLib
814.42 ms
4.4x
All inputs
Input
ordito (CUDA)
MeshLib
sphere_small
2.01 ms
bunny_decimated
5.84 ms
14.57 ms
bunny
11.02 ms
52.32 ms
dragon
183.88 ms
814.42 ms
happy_buddha
198.57 ms
1.102 s
lucy
42.028 s
> 10 min
relax_approx
dragon (871 414 faces) · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
86.44 ms
1x
MeshLib
3.175 s
37x
All inputs
Input
ordito (CUDA)
MeshLib
sphere_small
244.9 µs
bunny_decimated
601.3 µs
9.40 ms
bunny
2.48 ms
30.58 ms
dragon
86.44 ms
3.175 s
happy_buddha
107.45 ms
5.123 s
lucy
2.611 s
> 10 min
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: