Is that possible to set the exact radius for a icosphere somehow? Or if not, how scale/size is related to its radius?

I tried several tutorials but when I measure it in my game, it does not match. The game engine I programmed is exact, so I think the problem is that the radius is not correct in the model.

  • 2
    $\begingroup$ The size property when adding an icosphere corresponds to the radius. What about it leads you to believe it is not correct? Have in mind that at their default subdivision they are not totally geometrically even. $\endgroup$ – Duarte Farrajota Ramos Feb 21 '18 at 18:16

You can't do this out of the gate. It is by default not proportional in size on the X & Y axis. This shape is way more about equal triangles than it is about radius of a sphere.

Look at the Width and Height properties when the object is at full scale.

Rotate about Z axis 17°

Look at the Width and Height properties when the object is at full scale again.

You might be thinking...

What If I rotate about Z axis 8.5°

now check your Z value and your line of thinking should get you the answer.

Once all your bounding box dimensions equal out, scale it up until it reaches 2.00, and the apply the scale.

Once you have done this, scale will represent the radius of your icosphere.

  • $\begingroup$ Also, you may want to look up the terms: inscribed & circumscribed, as this may affect the actual radius value you want to reference. $\endgroup$ – Rick Riggs Feb 21 '18 at 18:36

As you most likely know, curved surfaces are approximated by several connected flat surfaces. Because of this the radii will differ depending on how you measure them. You need to be consistent. For an icosphere, you could possibly measure the raduis in several ways.

  1. Distance from origin to vertex. This will be the greastest distance, i.e. the icosphere will be completely inscribed in a perfect sphere that has the same radius and shares the same origin. This is what you set by defining the Size parameter when adding an icosphere.

  2. Distance from origin to centre of face. This will be smallest distance, i.e. a perfect sphere with this radius that shares the icosphere's radius will be completely inscribed in the icosphere.

  3. Any other way that suits your needs, as long as it's consistent, e.g. midpoint of edges.

With the default resolution (2 subdivisions) you can't achieve consistent, perpendicular radii in all three dimensions, regardless of how you measure them. If you need three perpendicular radii, you need to use at least 3 subdivisions. This will result in a vertex where the limit surface intersects each of the three global axes.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.