Hey forum,

Here is My latest puzzle, the Compound Crystal. It is a vertex-turning trapezo-rhombic dodecahedron.

The puzzle has two different kind of cuts: turns around the three fol and turns around the four fold verticies. during the four fold turns, four separate groups, each containing four parts, are passed around the axis of rotation. If only four fold turns are applied, the composition of each part group will never change. The introduction of the three fold turns allows parts to be passed between the aforementioned groups.


Video


solved:



four fold turn:



Three fold turn:



Scrambled with a few moves:




I am a bit busy right now, but I will update this thread with more information later on.

Great work, I absolutely love this puzzle. A phenomenal concept; it's probably my favourite from you! Well done.

By the way, do these parts ever get scrambled (the parts in the red circle)? My mind can't visualise how they become seperate.

Can the top and bottom three fold corners rotate?

Great puzzle!

That thing is CRAZY Eric!

PuzzleMaster6262 wrote:Can the top and bottom three fold corners rotate?

Great puzzle!

I'll answer this one, having seen the cad file and argued about it with Eric

No. They don't turn, and it kind of unnerves me, since we typically don't make puzzles with layers bandaged that don't have to be. But since it's Eric's design, it's Eric's prerogative. It'll certainly make it much more difficult to solve.

Eitan

We have here 6 rotations with n*120° and 6 rotations with n*90°
It stays in shape and therefore does not jumble.
Must be fudged...
and a great achievement.

Amazing. That is a very complex puzzle.

Thomas

Wow this puzzle looks awesome! I love the shape.

Andreas Nortmann wrote:Must be fudged...
and a great achievement.

Agreed... its a great achievement but it doesn't appear to be fudged to me. Are you sure that it is?

Carl

wwwmwww wrote:

Andreas Nortmann wrote:Must be fudged...
and a great achievement.

Agreed... its a great achievement but it doesn't appear to be fudged to me. Are you sure that it is?

It does not change shape => non-jumbling.
Possible moves are independent from permutation => non-bandaged.
It has a no multiple origins.
It has no "traditional" platonic or prismatic (?) axis system.

It has stored cuts but that is no contradiction,

Proove me wrong and I have to make up another category.

Not sure if you are agreeing with me or trying to prove it is fudged. Looks like you are asking me to prove we even need a "fudged" classification.

Andreas Nortmann wrote:It does not change shape => non-jumbling.

Shape changing is NOT equal to jumbling. The Mixup Cube changes shape yet it doesn't jumble. And a spherical helicopter cube wouldn't change shape yet is does still jumble.

Andreas Nortmann wrote:Possible moves are independent from permutation => non-bandaged.

Agreed... though I'd be inclined to say doctrinaire instead of non-bandaged.

Andreas Nortmann wrote:It has a no multiple origins.

Agreed.

Andreas Nortmann wrote:It has no "traditional" platonic or prismatic (?) axis system.

Agreed... I think. I don't see how this means it must be fudged though.

Andreas Nortmann wrote:It has stored cuts but that is no contradiction,

Agreed.

Andreas Nortmann wrote:Proove me wrong and I have to make up another category.

Fudged to me means the exact geometry doesn't allow a turn but the turn has been enabled via the use of voids, cuts which have some width, etc. A prime example would be Oskar's Illegal Cube. Notice how the pieces aren't flush against each other at the surface. This puzzle is still doctrinaire as after any permutation it looks just as it did before (if the stickers were removed).

Carl

A spherical helicopter cube does change shape when you look at the positions of the cuts, so Andeas' implication is right (at least, in the way he meant it).

Yet again, having seen the CAD file, I can attest to the fact that this puzzle is NOT fudged. Actually, the last time I spoke with Eric, we were both still at a bit of a loss for how to systematically fudge a puzzle.

-Eitan

Such a strange symmetry, yet no jumbling or shapeshifting. This puzzle has pieces that are not similar to any other puzzle I've seen before on a puzzle.
This is taking doctrinaire twisty puzzles to the limit

Very well done! You've created a most unusual and beautiful puzzle.

The first question that comes to mind is whether there are any other puzzles that can be made with this geometry. For example, could it be made face-turning on the rhombic sides, while allowing pieces to exchange halves via three-sided vertex turns? Or could the puzzle be made simpler (at the risk of being made too simple) by unitizing all of the smaller pieces into single edge pieces, and allowing them to turn in an "illegal" manner? The geometry seems pregnant with possibilities.

Sjoerd wrote: This is taking doctrinaire twisty puzzles to the limit

This was the intention of this puzzle! I don't know how shallow doctrinaire cuts could get much more complicated!


Just to clear a few things up:

Andreas Nortmann wrote:Must be fudged...

No fudging here, just normal cuts

PuzzleMaster6262 wrote:Can the top and bottom three fold corners rotate?

no, the top and bottom corners do not rotate.

Luke wrote:By the way, do these parts ever get scrambled (the parts in the red circle)? My mind can't visualize how they become separate.

Eric Puzzle.jpg

I find it interesting that this question has been asked. When showing the puzzle to Bram Cohen he also thought that they could not be separate. The fact is that every stored cut on this puzzle can be executed. Any pair of those parts that cannot be split by their adjacent three-fold turn can be moved to the three fold turn that is adjacent to it (using four fold turns) and that parts can be separated.

Hey...... this is kinda cool!

Strictly speaking, Andreas is right This puzzle is jumbling! Edit: Oh wait, I just re-read Andreas's comments. He never said it was jumbling - he thought it must be fudged. Sorry Andreas I tried!

Ok, now let me qualify that statement - this is a jumbling geometry. However these cuts are all quite shallow and the geometry is locally what I guess you could call "perfect". So, with cuts this shallow, the pieces affected by the jumbling moves have not yet come into existence. So I'm not really sure how we should classify this - it's a pretty typical jumbling puzzle from deepcut up to a certain cut depth - but before that none of the jumbling-affected pieces exist yet. I think...

I know I haven't contributed alot to the forum recently, and that's because I have been super busy working on something rather significant that hopefully (if I'm succesful) everyone will learn about in several months, but I still check in from time to time. So I don't really have time to investigate this new phenomenon . But someone explore it keeping what I said above in mind. It's a gut feeling I got after looking at this puzzle for about 30 seconds, but I'm already about 90% sure I'm correct. Someone do some awesome in-depth analysis!

Peace,
Matt Galla

Allagem wrote:Ok, now let me qualify that statement - this is a jumbling geometry. However these cuts are all quite shallow and the geometry is locally what I guess you could call "perfect".

I think I understand now, it is like the 'problem' with the icosahedron- the geometry will not jumble until certain layers overlap, forming certain groups of pieces that can be interchanged in new ways.

Monopoly wrote:

Allagem wrote:Ok, now let me qualify that statement - this is a jumbling geometry. However these cuts are all quite shallow and the geometry is locally what I guess you could call "perfect".

I think I understand now, it is like the 'problem' with the icosahedron- the geometry will not jumble until certain layers overlap, forming certain groups of pieces that can be interchanged in new ways.

Exactly! Although I should point out that the icosahedron jumbles nearly immediately. Virtually every icosahedron-cored puzzle will have jumbling pieces (an example of a non-jumbling icosahedral-cored puzzle is gelatinbrain 1.2.1). It would seem that this geometry has a more substantial jumble-less zone.

Peace,
Matt Galla

PS Hope I'm right about this. Remember I haven't done a full analysis to check - it's just a gut feeling

Sorry to derail, but I saw an argument here which begged the question: can there be a non-spherical puzzle whose geometry permits jumbling, but not shapeshifting?

NType3 wrote:can there be a non-spherical puzzle whose geometry permits jumbling, but not shapeshifting?

One can argue that Fairly Twisted does not shapeshift. Once you bring all centers back to their original state, the puzzle is brought back to its original shape, albeit possibly scrambled.

Oskar

NType3 wrote:can there be a non-spherical puzzle whose geometry permits jumbling, but not shapeshifting?

Yes, the Constellation Six without circular centres. I also have a concept which I have have momentarily discussed with Tom which would be similar to the Constellation Six, but in a completely different form.

gingervergo wrote:

Andreas Nortmann wrote:Must be fudged...

No fudging here, just normal cuts.

In this case:
It is ... difficult.

Sorry for the slight bump but I think it is worth it, since this question has not been answered yet.

Luke wrote:By the way, do these parts ever get scrambled (the parts in the red circle)? My mind can't visualise how they become seperate.

Eric Puzzle.jpg

I would say: NO. These two pieces are implicitly glued into one.
Eric: was that intentional for aesthetic reasons?

Seeing as what Eric said earlier, is this puzzle like a Boublezed puzzle in any way? It seems too strange to be plain and boring and doctrinaire.

Hopefully this video will answer any questions!

Sorry for the low production value, I am busy and I sort of winged this.

Awesome! Finally understood this puzzle! Seeing the bridge piece separate from the base piece felt like a great mystery solved! Bravo! Bravo!

I can't believe I missed this puzzle!
It's yet another one of those my mind can't grasp and therefor it's added to my wanted-list together with edge turning rhomb and more madness and those other new concepts I also dont understand at all, but dream of getting to know in the futur

Will it be for sale?