Benchmarks of ordito.creation 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 creation 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
Open3D
0.71x
7
3
libigl
1.4x
2
1
MeshLab
15x
8
8
MeshLib
20x
6
5
trimesh
44x
14
14
PyTorch3D (CUDA)
111x
2
2
PyVista
270x
5
5
box
time relative to ordito, log scaleValues
Library
Median
Relative to ordito
Open3D
0.6 µs
0.02x
MeshLib
5.7 µs
0.21x
ordito (CUDA)
27.1 µs
1x
MeshLab
34.9 µs
1.3x
trimesh
92.9 µs
3.4x
All inputs
Input
ordito (CUDA)
MeshLab
MeshLib
Open3D
trimesh
—
27.1 µs
34.9 µs
5.7 µs
0.6 µs
92.9 µs
grid
count=32 · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
libigl
7.1 µs
0.23x
ordito (CUDA)
31.1 µs
1x
MeshLab
160.1 µs
5.1x
All inputs
Input
ordito (CUDA)
libigl
MeshLab
count=32
31.1 µs
7.1 µs
160.1 µs
count=512
36.7 µs
7.52 ms
43.51 ms
sphere_cap
subdivisions=3 · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
30.5 µs
1x
MeshLab
164.9 µs
5.4x
All inputs
Input
ordito (CUDA)
MeshLab
subdivisions=3
30.5 µs
164.9 µs
subdivisions=6
35.2 µs
12.03 ms
icosphere
subdivisions=3 · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
47.2 µs
1x
trimesh
433.4 µs
9.2x
MeshLab
2.10 ms
44x
PyTorch3D (CUDA)
5.11 ms
108x
All inputs
Input
ordito (CUDA)
MeshLab
PyTorch3D (CUDA)
trimesh
subdivisions=3
47.2 µs
2.10 ms
5.11 ms
433.4 µs
subdivisions=5
45.3 µs
26.74 ms
8.69 ms
2.53 ms
subdivisions=7
90.1 µs
693.00 ms
12.76 ms
64.11 ms
uv_sphere
sections=32 · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
Open3D
29.8 µs
0.50x
ordito (CUDA)
59.2 µs
1x
trimesh
756.0 µs
13x
MeshLib
770.1 µs
13x
All inputs
Input
ordito (CUDA)
MeshLib
Open3D
trimesh
sections=32
59.2 µs
770.1 µs
29.8 µs
756.0 µs
sections=512
74.9 µs
39.23 ms
7.98 ms
220.90 ms
sections=4096
3.45 ms
2.998 s
1.130 s
27.248 s
cylinder
sections=32 · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
Open3D
2.4 µs
0.07x
MeshLib
25.5 µs
0.75x
ordito (CUDA)
34.0 µs
1x
trimesh
202.4 µs
6x
All inputs
Input
ordito (CUDA)
MeshLib
Open3D
trimesh
sections=32
34.0 µs
25.5 µs
2.4 µs
202.4 µs
sections=512
27.9 µs
1.05 ms
19.5 µs
527.8 µs
sections=4096
29.9 µs
34.10 ms
151.3 µs
3.96 ms
cone
sections=32 · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
Open3D
1.7 µs
0.07x
MeshLib
13.8 µs
0.57x
ordito (CUDA)
24.1 µs
1x
MeshLab
41.2 µs
1.7x
trimesh
177.0 µs
7.3x
All inputs
Input
ordito (CUDA)
MeshLab
MeshLib
Open3D
trimesh
sections=32
24.1 µs
41.2 µs
13.8 µs
1.7 µs
177.0 µs
sections=512
30.9 µs
65.4 µs
846.0 µs
12.7 µs
402.8 µs
sections=4096
30.1 µs
290.1 µs
33.44 ms
76.4 µs
2.41 ms
annulus
sections=32 · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
36.5 µs
1x
MeshLab
39.7 µs
1.1x
trimesh
211.4 µs
5.8x
All inputs
Input
ordito (CUDA)
MeshLab
trimesh
sections=32
36.5 µs
39.7 µs
211.4 µs
sections=512
48.9 µs
76.6 µs
763.5 µs
sections=4096
44.1 µs
383.1 µs
5.63 ms
torus
sections=32 · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
Open3D
11.8 µs
0.27x
ordito (CUDA)
44.7 µs
1x
MeshLab
235.9 µs
5.3x
MeshLib
416.3 µs
9.3x
trimesh
461.7 µs
10x
PyTorch3D (CUDA)
996.5 µs
22x
All inputs
Input
ordito (CUDA)
MeshLab
MeshLib
Open3D
PyTorch3D (CUDA)
trimesh
sections=32
44.7 µs
235.9 µs
416.3 µs
11.8 µs
996.5 µs
461.7 µs
sections=512
40.4 µs
3.89 ms
4.42 ms
167.4 µs
13.60 ms
5.41 ms
sections=4096
52.5 µs
36.43 ms
18.18 ms
1.36 ms
47.95 ms
revolve
sections=32 · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
132.8 µs
1x
MeshLib
679.2 µs
5.1x
trimesh
702.7 µs
5.3x
All inputs
Input
ordito (CUDA)
MeshLib
trimesh
sections=32
132.8 µs
679.2 µs
702.7 µs
sections=512
106.8 µs
5.44 ms
11.14 ms
sections=4096
189.5 µs
14.92 ms
106.22 ms
extrude_polygon
ring_size=64 · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
63.2 µs
1x
Open3D
104.8 µs
1.7x
trimesh
302.9 µs
4.8x
PyVista
328.2 µs
5.2x
All inputs
Input
ordito (CUDA)
Open3D
PyVista
trimesh
ring_size=64
63.2 µs
104.8 µs
328.2 µs
302.9 µs
ring_size=1024
63.4 µs
1.54 ms
2.34 ms
1.55 ms
sweep_polygon
time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
90.9 µs
1x
trimesh
12.41 ms
136x
All inputs
Input
ordito (CUDA)
trimesh
—
90.9 µs
12.41 ms
truncated_prisms
face_count=1024 · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
57.1 µs
1x
trimesh
162.9 µs
2.9x
All inputs
Input
ordito (CUDA)
trimesh
face_count=1024
57.1 µs
162.9 µs
face_count=262144
145.6 µs
75.89 ms
parametric_surface
resolution=40 · surface=boy · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
39.6 µs
1x
PyVista
2.21 ms
56x
All inputs
Input
ordito (CUDA)
PyVista
resolution=40 surface=boy
39.6 µs
2.21 ms
resolution=40 surface=dini
55.1 µs
2.18 ms
resolution=160 surface=boy
54.5 µs
15.88 ms
resolution=160 surface=dini
54.5 µs
12.44 ms
resolution=640 surface=boy
60.8 µs
1.059 s
resolution=640 surface=dini
65.1 µs
842.81 ms
super_ellipsoid
resolution=40 · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
61.3 µs
1x
PyVista
1.82 ms
30x
All inputs
Input
ordito (CUDA)
PyVista
resolution=40
61.3 µs
1.82 ms
resolution=640
83.6 µs
954.23 ms
super_toroid
resolution=40 · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
57.6 µs
1x
PyVista
2.01 ms
35x
All inputs
Input
ordito (CUDA)
PyVista
resolution=40
57.6 µs
2.01 ms
resolution=640
55.6 µs
589.60 ms
random_hills
resolution=40 · time relative to ordito, log scaleValues
Library
Median
Relative to ordito
ordito (CUDA)
71.6 µs
1x
PyVista
2.16 ms
30x
All inputs
Input
ordito (CUDA)
PyVista
resolution=40
71.6 µs
2.16 ms
resolution=640
107.0 µs
930.79 ms
platonic_solids
builder=icosahedron · filter_name=create_icosahedron · n_faces=20 · time relative to ordito, log scaleValues