4 ms·
I'm assuming the loss of height comes from measuring distances directly between points instead of along the surface of the sphere. Much cheaper, but by design y
by frenchie14 8y ago
I'm assuming the loss of height comes from measuring distances directly between points instead of along the surface of the sphere. Much cheaper, but by design you'll cut through the sphere to connect each point
- slavik81 8y agoI think you're right. The circumcentre equation is just more complicated than the centroid calculation I was imagining. They're doing exactly what I had suggested in the third paragraph. There may also be a nice derivation of the triangle centre on the sphere coming from quaternion math that would avoid renormalization. I wonder...