Skip to content

tangent_space

Benchmarks of ordito.tangent_space 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.

Measured on GPU: NVIDIA GeForce RTX 5090 · CPU: Intel Xeon w7-2495X

Summary

0.1x1x10x100x1000xordito (CUDA): the reference, 1xordito (CUDA)1xlibigl: 86x ordito's time, geometric mean over 1 functions; slower than ordito in 1libigl86x (1)potpourri3d: 219x ordito's time, geometric mean over 1 functions; slower than ordito in 1potpourri3d219x (1)
Time relative to ordito (CUDA), geometric mean over the tangent_space 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
libigl 86x 1 1
potpourri3d 219x 1 1

vertex_tangent_frames

0.1x1x10x100x1000x10000xordito (CUDA): 283.0 µs, 1x ordito's timeordito (CUDA)1x · 283.0 µspotpourri3d: 367.96 ms, 1300x ordito's timepotpourri3d1300x · 367.96 ms
sphere_large (327 680 faces) · time relative to ordito, log scale
Values
Library Median Relative to ordito
ordito (CUDA) 283.0 µs 1x
potpourri3d 367.96 ms 1300x
All inputs
Input ordito (CUDA) potpourri3d
sphere_small 179.6 µs 4.56 ms
sphere_med 201.3 µs 63.91 ms
sphere_large 283.0 µs 367.96 ms

face_tangent_frames

0.1x1x10x100x1000xordito (CUDA): 34.3 µs, 1x ordito's timeordito (CUDA)1x · 34.3 µslibigl: 19.36 ms, 564x ordito's timelibigl564x · 19.36 ms
sphere_large (327 680 faces) · time relative to ordito, log scale
Values
Library Median Relative to ordito
ordito (CUDA) 34.3 µs 1x
libigl 19.36 ms 564x
All inputs
Input ordito (CUDA) libigl
sphere_small 33.9 µs 268.0 µs
sphere_med 30.9 µs 4.47 ms
sphere_large 34.3 µs 19.36 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:

Function Input ordito (CUDA)
halfedge_tangent_angles fan_hub 36.94 ms
halfedge_transport_angles sphere_large 55.9 µs