Thanks to all the ideas, support, and advice from many of you, this project is complete. The results are summarized here:

http://home.ec.rr.com/specialpuzzles/RhomdoProject.htm

Now what? Perhaps the next attempts should include Trajber's Octahedron and Octaminx. What other puzzle ideas would be good? Does anybody have any ideas you would like to see prototyped?

A Trajber's Octahedron?

http://www.virtualpuzzlemuseum.com/8sidedpuzzles2of2.htm

Are you selling any of them? If you are how much for each of them? I would really love to add them to my collection of Rubik's/puzzles. Is it possible for me to Buy all 3? I really want them badly.

Although Trajber's octahedron would be nice
I think a new octahedron puzzle would be even nicer:

an octahedron similar to Trajber's but where the
rotation axes are not along the 6 corners but along
the 8 triangular center faces. Such a puzzle would
be the octahedron analog of Rubik's cube. The recent
rainbow cube is a sub-puzzle of this proposed
octahedron with only center triangles and edges -
but no corners.

Any ideas about such a puzzle ?

If you look at the following pictures:

http://crypto.cs.mcgill.ca/~crepeau/RUBIK/Meffert01/Slide__13.htm
http://crypto.cs.mcgill.ca/~crepeau/RUBIK/Meffert01/Slide__14.htm

you will realize that both Trajber's octahedron and
magic octahedron are puzzles relating to Rubik's cube
with only 6 axes of rotation.

What I am suggesting would have 8 axes of rotation,
as the rainbow cube.
If you look carefully at Trajber's octahedron you
will notice that the corner pieces are NOT shaped
in a way which is compatible with rotation around
center faces...

I really don't know how to even design the pieces
for the puzzle I suggest.

Now I understand your idea a little better. I agree with your assessment about the corner pieces. Take a look at David Byrden's java octahedron with twisty sides at:

http://byrden.com/puzzles/MainFrames.html

Is this concept a little closer to your idea?

The first octahedron puzzle with twisty sides shown on Byrden's site with only 3 triangles per face is Meffert's Skewb diamond which was released a couple years ago. Perhaps the second octahedron w/ twisty sides on Byrden's site which has 9 triangles per face is what you are talking about. Although it looks like it would be a great puzzle, there are no center cubies on each face.

I think that the puzzle being discussed would use the 8 axes of the rainbow cube as the centers for each face's rotation. I think that it would probably have a triangle or hexagon-shaped center piece and probably 4 diamond-shaped edge/corner pieces. I don't think each face would be divided equally into several of the same size shapes like the puzzles described on Byrden's site. I'm not sure yet though. I'll have to give it more thought before I could say for sure.

Doug.

I like these octahedron ideas, so much so that I've already started a puzzle that is exactly as you describe. But isn't this exactly Trajber's Octahedron?

How do you pronounce Trajber? I understand that Josef Trajber is Austrian. Is Trajber an Austrian name?

after thinking about it for a little while
I came up with the following picture about
what a face would look like on the puzzle
I describe:



I am now try to figure out an internal mechanism
for it...

If someone has a good 3D drawing software it would
be nice to see eight such faces connected into an
octahedron...

It sounds like you're describing an octahedron that was already made for some Time but far as I know isn't produced anymore.

http://www.virtualpuzzlemuseum.com/8sidedpuzzles2of2.htm

You may be able to find one for sale around.

That's so weird, I worked out the cuts for the dodec based on the 3x3x3 on friday night.

Thought i'd though of it first, but there you have it. I'll be making mine soon. It looks very nice, you've done well.

Meffert's Skewb diamond has 4 triangles per octahedron face, not "3" like I incorrectly stated in my above posting.

and yes, David Byrden's java octahedron with twisty sides is closer to that concept.

What a beautiful puzzle concept! The face reminds me of a Master Pyraminx, with fewer axes. This idea is a keeper.

On Joshua Bells Patent page you can see a patent for it from 1984 by Clarence W. Hewlett, Jr.

http://www.calormen.com/TwistyPuzzles/twisty.htm

I think I read somewhere (in Rubik's Cubic Compendium?)that Rubik also invented the mechanism for it.

Jaap