And how exactly is a shell defined? For example, does the original Professor Cube design, with extra layers slotted into a 3x3x3, count as a shells design? What about the similar Mixup Cube, with the middle layers slotted into the 2x2x2?

Rubik's Shells?

Rainbow Cube (from Compy Cube)?

-Doug

I think Bram was the first to suggest the shells design. It may have been in the context of Pentultimate design but somebody else probably remembers the details better than me.

I'm pretty sure I came up with the term 'shells', and popularized the concept on here. My main target for it was the pentultimate, but I did know the approach could be used for just about anything. I think the concept was independently invented by David Pitcher and Verdes, both of whom made actual puzzles with the approach long before anyone on here did. It also may be that Bob Hearn and Matt Shepit thought of it independently, I'm hazy on that. There are a bunch of old posts on here of me talking excitedly about designs with links to Jaap's applet illustrating different shells, although I'm hazy on whether anybody else understood what I was babbling about at the time.

So the first puzzle built using shells would be something by Verdes or Pitcher. For the first one built inspired by discussion on this site, it might depend on what you count as being a shells mechanism. I recall the Gigaminx and Toru/Rua as being major milestones it puzzle building technology.

If I recall correctly, Aleh was the first person to use shells to make a puzzle deeper cut when he hand built a Megaminx into the first pyraminx crystal, but i could be mistaken. The definition of shells is still not perfect, so this may not be what you are looking for.

The FTO uses shells, so David Pitcher's prototype would certainly predate that.

But I guess from the responses in this thread, the original 5x5 and the Mixup Cube are not shell puzzles?

Dave Pitcher's FTRD Uses a very shells-like mechanism.

To me the internal machinism looks manually designed in 3D rather than the typical 2D sketch revolved into 3D. That is, when you design a shells mechanism puzzle in 2D and then revolve the cuts, 100% of the volume is used. There are no gaps. Any gaps in the final product are small parts that have been left out. In the case of Dave's FTRD there seem to be lots of gaps between pieces where the interaction doesn't matter. The internal parts look less organic and more engineered.

One could argue that the Eastsheen 2x2 uses a sort of shell mechanism where the 3x3 edges and centers are hidden. Its patent dates to 1998.

The two 5x5x5 mechanisms before the Verdes mech were cylindrically based, hence their inability to scale to 7x7x7 and above.

Dave