Photo here.
I seem to remember that there was something special about the way it rolled down an incline, and that its name started with an “o”. Sorry, that’s all I got.
It’s very hard to see from your photo, but it looks vaguely like a Reuleaux Triangle - Like the rotor of a Wankel engine.
Thank you, beowulff, but that’s not it. I’m sorry the picture is so bad.
I tried to draw it in Google SketchUp, but failed miserably. Anyone know a good online library of 3D shapes that I could browse through?
Is it one of these?
It looks like a cross between a teardrop and a clam shell…would that be an accurate description of what you saw?
That’s very interesting; I hadn’t heard of that before. I don’t think that’s it, but when reading up about the gömböc I found a reference to the oloid, which looks very close to the OP’s photo, and starts with an “o”.
Looked like a fun challenge. Once Dr. Strangelove nailed the shape (interesting, that), I gave it a shot… only took me about two mins. Check it out!
Whoops, looks like I needed to pinch the other end as well. It’s cool though. I want to replace the tires on my Jetta with those.
Not quite, but that’s interesting nevertheless, thank you!
Yes, those were my first associations, too.
That’s it! Great!
See it rolling here.
Wow, great video and soundtrack! I had actually tried a similar approach (draw a circle, pull it into a cylinder and then rotate one end), but I couldn’t figure out SketchUp’s user interface.
I should add, though, that as you can see in the Wikipedia article, the far end is not only round, but also tapered, if that makes any sense. So rather than a teardrop and a clam shell, it’s more like two clam shells set back-to-back at a 90 degree angle.
Its a chinese fortune cookie!
I just realized that this thread is related to this other thread in that the surface of an oloid can be cut into four pieces that are each a chunk of a circular oblique cone.
I don’t know if that info helps anyone draw an oloid, but it helped me create one with Mathematica.
It’s weird that these two threads were on the front page of GQ at the same time.
That’s good for visualizing, but the surface isn’t really four cones. See The development of the Oloid (PDF).
In Figure 2, the line AB represents the straight line the oloid will have touching a plane as it rolls on it. A and B are both parametrized by t, and are both varying simultaneously. You can also see that the line from (0,1/2,1) (which would be the apex of your approximating cone) to U will intersect the disk Pi[sub]2[/sub].
I think if you instead took the convex hull of the semicircle of Pi[sub]1[/sub] with y < -1/2 and the semicircle of Pi[sub]2[/sub] with y > 1/2, you *would *get a shape made of four cones.
ZenBeam, you are correct. I drew the convex hull of two semicircles. My mistake.