Allow me to present the Thorny Cube. This may seem like a repeat but I had only ever got around to presenting the Dethorned Cube in the New Puzzles section. Even that was never really a finished puzzle as it wasn't dyed and stickered.

Anyways Tanner aka Steryne had helped me via Skype to turn my POV-Ray code into something that could be printed. This was before I had SolidWorks. As a result I gave him the puzzle after I had got it turning as well as I could get it. He has now dyed and stickered it. These are his pictures.







Ok. Now for the the history of this puzzle. The spark that this puzzle grew from was a post by Hugo Mak that he made here back on July 9, 2010 . Its development started the same day when I started this thread here on TwistyPuzzles. The only 3D modeling software I knew at the time was POV-Ray and there are a ton of pretty POV-Ray pics in that thread showing every step of the design process. Tanner aka Steryne helped me turn this into something which could be printed. It took numerous Skype calls. Me sitting in front of my POV-Ray code trying to describe some rather complex geometry to Tanner who was using Alibre to make a model which could be output in a format suitable for 3D printing. Version 1 of the the Thorny Cube didn't work very well at all. It would just fall about in your hands. Still it was a great learning experience and Version 2 turned out much better. Its certainly a functional puzzle. I'm still not entirely happy with it. It locks up and pops frequently if you aren't very careful to keep everything in alignment but I think this is the best that can be done at this size and in this material. The puzzle has an edge length is 57mm. I think the only way to make things more stable is to make the small thorn pieces even bigger which means a bigger more expensive model. I may down the road make a Version 3 and I'm thinking of making it with an edge length of 100mm but I have many other puzzles I want to finish before I revisit this one again. So don't hold your breath. Plus as versions 1 and 2 were made in Alibre and I'm now coming up to speed with Solidworks, if there is a version 3 it will be totally made from scratch.

And now the big news...

I plan of offering Thorny Cube (version 2.0) on Shapeways starting later today. As this is my first puzzle and it does have turning issues I'm going with a very very low markup. The parts have too many faces to get into 1 model on Shapeways so I've broken it into two files. The first model will have all the parts needed to make the Dethorned Cube which actually turns very very well in my opinion but its so similiar to the Mixup Cube I'm not sure there will be much interest. The second model will contain all the thorn pieces. It includes 27 of each of the 3 types but only 24 of each type are needed to complete the puzzle. The extras are included as these are the parts that pop and as they relatively small its possible to lose them.

Part 1of2 will be $70.25 (so you can get a Dethorned Cube for well under $100 if you like... its a great puzzle)
Part 2of2 will be $36.52

So you can get a DIY kit for the Thorny Cube for $106.77. I believe Shapeways will charge you $6.50 for shipping on top of this.

Also Tanner has offered to help make finished puzzles for the more then reasonable markup of $33.23 so you can have a finished (assembled, dyed, and stickered) for $140. This doesn't include shipping to Tanner ($6.50) and the actual shipping charge to you.

More details to follow on my Shapeways shop soon.

Enjoy,
Carl

WOW, that is a great puzzle, it's nice to see it completed. I look forward to the video.

Great Puzzle!!!

Gus wrote:I look forward to the video.

The puzzle is no longer in my possession so Steryne is welcome to post a video if he'd like. I also plan on getting a set of stickers from Steryne and ordering a second puzzle for myself. I should be able to post a video then but that will be a while.

Till then you are welcome to check out the videos which show the turning quality in the Puzzle Building and Modding section. Link in first post. This was before the puzzle was dyed and stickered.

Carl

Dude, that's 2 puzzles completed! You're becoming one of us! One of us. One of us. One of.....

-Eitan

That looks absolutely epic! I can't even remember the first time you posted about you idea for it, and I really can't wait for the vid. Awesome job man!

wwwmwww wrote:The puzzle is no longer in my possession.

And that means I have to bother steryne with my usual questions.
He will hate me...

Lovely puzzle. I love how your projects are significantly influenced by those on the forums, and how 'in the know' we are on your projects.

More details added to the first post.

Carl

The Thorny Cube is now up on Shapeways.

http://www.shapeways.com/shops/wwwmwww

Still working on the look of my shop as I figure out how to do things. The description of the first part of the puzzle was about 5 times as long as I detailed what was needed for assembly, etc. But it appears there must be some character limit as Shapeways cut and dropped most of it. I'll get back to it later as I'm out of time for today.

Carl

I think I just sold my first puzzle on Shapeways today. If you are the buyer I'd love for it if you could send me a PM and let me know. I'm still working on my shop and I'm hoping to add some assembly instructions and maybe an assembly video as well. I'd just really like to know who my first customer is and also welcome any feedback they are willing to give.

Thanks,
Carl

My second Thorny Cube has been assembled and stickered. The edge length is 57mm and I'll get a weight for you soon Andreas. Here is the first picture I've removed the background from.



You Tube Video

Carl

that looks amazing! I want one

Oh my. I tend to dislike shape-shifters, but this is just fantastic.

@1:55 in the video is AMAZING

Awesome puzzle.

wwwmwww wrote:The edge length is 57mm and I'll get a weight for you soon Andreas.

Sounds promising.
A very good image!

Please excuse my lazyness Carl. I am sure this have been asked a few times. Is it possible to make one with out a corner bandaged to the core thus creating a pretty version of the puzzle. Or would that destroy the whole idea?

Sigurd wrote:Please excuse my lazyness Carl. I am sure this have been asked a few times. Is it possible to make one with out a corner bandaged to the core thus creating a pretty version of the puzzle. Or would that destroy the whole idea?

Yes and no. Let me explain. The whole idea here was to combine a circle cube and a mixup cube. In a circle cube you generally think of the circle centers as tied to the slice layer. You can't do that here as the slice layer is allowed to rotate by 45 degrees. If you look at a circle 2x2x2 you'll see the same pattern you do here. One of the corners has to be tied to the core. Keep in mind even Oskar's Mixup Cube has one corner attached to the core as the base of the puzzle is actually a Fused Cube. There are hidden 3x3x3 edges inside the puzzle that need to be kept in alignment and if a corner wasn't attached to the core these pieces could become out of alignment and you'd have no way to get them back in alignment again. So one corner must stay bandaged to the core. However you could cut the pegs off of the fixed face centers (see attached)
And this would free the circle centers to rotate on their own. However I would no longer consider this a circle mixup cube. Since the circle centers are free to rotate on their own I'd consider these baby-face like moves and not circle move. So you'd have a Mixup Baby-face Cube. There are many such puzzles which could fall into that catagory and Oskar has already made one of them. Look at his Jack's Cube. In his video he calls the turning on the circle centers a turn like a Circle Cube but on a Circle Cube the circles AREN'T free to rotate on their own. These are Baby-face turns and they make the puzzle much easier to solve.

Personally I'd prefer to see a version of Jack's Cube which had the circle centers bound to the core in the same fashion as my Thorny Cube, with pegs and grooves. I think that would make if much more interesting and far more challenging.

Carl

It would be interesting to see whether it's possible to make a Mixup Cube where all the edges and center pieces are directly above a core arm, using the shells mechanism to make the deep enough to move. If that is possible, then a 'true' Circle Mixup Cube could be made.

Luke wrote:It would be interesting to see whether it's possible to make a Mixup Cube where all the edges and center pieces are directly above a core arm, using the shells mechanism to make the deep enough to move. If that is possible, then a 'true' Circle Mixup Cube could be made.

I'm not sure how you are defining 'true' Circle Mixup Cube. Is it one of these?

(1) The circle centers turn with the slice layer. Since the slice layer can turn by 45 degrees this requires cutting up the already small pieces into many more even smaller pieces. Not sure at the moment how you'd control the alignment of the core but this may be doable. I think you'd have to attach the core to one of the face centers or edges, but what if that piece isn't in the slice layer or one of the circle centers that should turn? Actually not sure this is doable as I now see several problems with attaching the core to an edge or face center.

(2) The circle centers are fixed and don't turn with the face or slice layers. I don't think this is doable. You'd need a means to hold the core fixed but none of the exposed pieces can be attached to it. However this method doesn't require the circle centers to have 45 degree symmetry as the faces always rotate by multiples of 90 degrees.

(3) Something else... please explain.

(1) and (2) could both me made as apps where the core in held in place by the programing but I just don't see a way to make them as physical puzzles. That said I'd LOVE to be proven wrong.

Thanks,
Carl

My idea is based upon the first one. How I'd imagine it could be done is with an 18 armed core (following the geometry of a truncated rhombic dodecahedron). If we left it at that it wouldn't work, because the center pieces would be attached to the core. Therefore, I'd imagine a shells-like mechanism could create deep enough cuts that the pieces move independently from their axes (like how Megaminx centers are stationary, where as Pentultimate centers move). Hopefully that made sense.

I think option 2 might be possible if one were to use Carl's core design without having one corner physically attatched. All the corners would technically be secured via the internal 3x3 mechenism and the rest of the outerlayers pieces wouldn't jam as long as the were guided by the corners. The only thing with this design is whether the slice layer could be moved without the R and L layers.

The only thing with this design, is that baby moves would be possible as Carl stated, except that they would be physically attatched via the core to the circle face on the opposite sides.

Nonethless, I think this puzzle may be able to physically be built depending on the tolerances (ability to move the slice layer alone). I'll try to look more into this.

cubonkers112 wrote:I think option 2 might be possible if one were to use Carl's core design without having one corner physically attatched.

If one corner isn't attached to the core then the heart of this puzzle becomes a 3x3x3 and not a Fuzed Cube. So what keeps the inner face centers and 3x3x3 edges aligned with the corners?

cubonkers112 wrote:All the corners would technically be secured via the internal 3x3 mechenism and the rest of the outerlayers pieces wouldn't jam as long as the were guided by the corners.

Yes, the outer layer pieces are good if the corners are good, but how do you keep the corners good with respect to the other inner 3x3x3 pieces? Not saying it can't be done but I sure don't see an easy way to pull it off. I guess you could make the core out of a superconductor fixed in space in a large magnetic field with all the other pieces around it but short of that what is going to "hold" the core in the correct position?

Carl

The problem with the mechanism is trying to bandage the hidden 3x3x3 layers. If you want to, you could bandage a 3x3x3 edge to a Mixup Cube, however it would do nothing but mess up the internal puzzle, for two reasons. The first one is that you would need three edges to be bandaged for the three axes, and those would eventually cross and block moves, and secondly Mixup Cube moves couldn't be performed, because the internal 3x3x3 slice layer is rotated 45 degrees. Bandaging the 3x3x3 center pieces to the Mixup Cube center pieces also wouldn't work, because again problems would occur with the Mixup Cube turns. So, using traditional mechanisms, we have two options.

1) Bandaging 3x3x3 slice layers to Mixup Cube slice layers (as I've already stated, I highly doubt this is possible, unless you do something very weird to the mechanism.
2) Keep the 3x3x3 slice layer stationary whilst the Mixup Cube layers turn. This also seems next to impossible, because you'd need an almost frictionless 3x3x3 inside, and even if you had that, there wouldn't be anyway to control that layer.

I still believe that the answer to this is an 18 armed core and the shells mechanism to create deeper cuts.

Luke wrote:So, using traditional mechanisms, we have two options.

1) Bandaging 3x3x3 slice layers to Mixup Cube slice layers (as I've already stated, I highly doubt this is possible, unless you do something very weird to the mechanism.
2) Keep the 3x3x3 slice layer stationary whilst the Mixup Cube layers turn. This also seems next to impossible, because you'd need an almost frictionless 3x3x3 inside, and even if you had that, there wouldn't be anyway to control that layer.

Option 3 is to bandage the 3x3x3 slice layers to a corner as I did. I still view this as a 'ture' Circle Cube as that is what is done in the 2x2x2 Circle Cube.

Option 4. Make the core of the puzzle a spherical Geared Mixup Cube. That would keep the inner slice layer well behaved. I wonder if there is a part of the edges and face centers on a Geared Mixup that doesn't rotate. If so you'd want that part exposed on the surface of your sphere.

Luke wrote:I still believe that the answer to this is an 18 armed core and the shells mechanism to create deeper cuts.

Maybe... you want the circles to turn with the slice layer so keep in mind the circle centers must have 45 degree rotational symmetry. You also have 3 discontinous areas on the surface which turn with a slice turn; the slice layer and the two opposite circles. I'm not seeing how a shell mechanism will do that. But just because I can't see a way doesn't mean there isn't one. If you want to take a shot at it, I wish you luck.

Carl

[Edit] Added option 4 above. It just came to me.

Andreas Nortmann wrote:

wwwmwww wrote:The edge length is 57mm and I'll get a weight for you soon Andreas.

Sounds promising.
A very good image!

Puzzle weighs in at 116.3 grams. More pics to come as I clean them up.

Carl

Just sold my second Thorny Cube on Shapeways yesterday. Looks like someone took advantage of the 15% off yesterday. Thanks SO MUCH!!! Despite Tom not mentioning my shop in his post about the sale yesterday. (PLEASE NOTICE THE SMILEY... just playing with Tom)

No idea who it was and I'd love to know. Please PM me and I can help you get a set of stickers.

And also noticed I made the same amount off this one as I did the first despite the 15% off. That sounds very odd to me. Not sure I should point this out but what keeps someone from uploading a $1 model with a $99 markup? They could just wait till the next 15% off sale and arange to order many at the now reduced price of $85 and then have shapeways pay them $99 for each one sold. I sure hope the system is set up smarter then that. And NO!!! The Thorny Cube ISN'T a $1 model.... that wasn't why I asked.

Carl