Polish all pieces to a glass-like finish and over-lube it? If that cannot reduce fiction to workable levels, nothing can.

^ For that, you'd need 2 bottles of lube and about a week of sanding

Volunteer. jk

he hasnt actually built it yet

May I ask the Author: does this design mechanism patented?

I understand, that is for 3x3x3, 5x5x5, 9x9x9, 17x17x17, 33x33x33, etc forms only. But there is a good alternaive for big and biggest cubes!
Maybe, some mind working of it - can bring result for apply to any number of slices...

So... is this project still going?

I doubt it. I still may be built eventually, but I haven't heard anything.

By the way, it's a good idea to avoid bumping topics. "Bumping" is when you post in an old topic, making it come to the top of the forum, making people think there is new information. Don't worry about it this time, since you're new, but try not to do it next time.

-pi (Eitan)

Hi Jerry,

Yes, it has been built by Claus earlier this year. It must have been a lot of work. Unfortunately, some pieces are way too loose and fall out. Also,some colored caps on the edge pieces disappeared.

Watch the YouTube video.
Check out the photos below.

Oskar

So, did it work? Suspense building!!!

I too would like to know if it worked. I mean, why was this not announced even if it was a failure...

Please, upload a video to youtube or something. Even if it failed, or turned horribly, we still need to be shown proof that it was built. Not that I would want to buy/own that massive behemoth, but the puzzle community needs to e updated. This is a big milestone for the puzzling community.

Wow, I only just heard about this. Would be amazing if it actually worked. Come to think of it, would be rather amazing even if it didn't work! Either way it's a step in the right direction towards actually making a fully functional 17x17x17

To my understanding, the puzzle is not yet finished but it looks like it's not going to work.

Thanks for bumping this topic!! If you hadn't done that we wouldn't have know there was a living prototype of the beast! Thumbs up frop me to make an video, and claus, don't push too hard while assembling it, or you'll turn crazy

Greg

P.S: Bumping is very rarely good.

Hmm, thats not exactly great news, but I'm sure it would still be very worth it to finish assembling it so that there is at least proof of it's existence.

I agree. Better to have a horrible quality physical 17x17x17, or even a non-working one, than to not have it at all.

Sorry if I offended you guys; I just had to see if this actually worked.

In my honest opinion it's not a problem at all. I'm sure there are a lot of others at the forum who wanted to know what ever happened to this project, and thanks to your bump, we now know

Agreed with Katten. This is one of those threads that I was looking forward to seeing the progress. If you didn't bump this, I'd have forgetten about it.

PICTURES. NOW.

No offense taken, but please, people, calm down.

I'm very excited to see it, even if that turns horribly

As excited you all may be, I would understand that a person wouldnt be too eager to show off a $3500 paperweight

NGW - agreed! I cant see any other reason to explain why we havent seen it already then

I can think of many reasons-one just happens to be the designer doesn't want to show it. Simple really and we have to respect his wishes. The second reason that comes to mind is an argument that ensued when another larger order puzzle was shown. We don't want that again.

Please stop asking guys-if he wants to show it he will. Until then, trust that it has been done. Take Oskar's word for it.

So the design is actually 4cm each way? That's amazing and problematic in so many ways! On the one hand, it'll fit comfortably in your hands, which is a plus, and on the other hand, the minute size of the layers would make turning an inner layer a pain. Still, it's a brilliant design. When I first read this article, i thought it would be about 2' each way, but i was wrong. Shows you how valuable small things can be.

It'd be hilarious if someone tried speedsolving a prototype and an eighth of the puzzle popped.

It's 10cm^3, not 4cm.

Sorry, I was thinking inches, not centimeters.

Still, only 10cm?! The 11x11 is like twice as big each way!

You want to see pictures? Here they are:






If you work on the black (brown) face the pieces at the opposite face will fall apart.

If I made the screws stronger it is to much tension in the cube.

General, the friction between the pieces are very high.

its fantastic...even if ya cant use it...

Oh my GOD...

This is the most EPIC puzzle EVAR It lives!!!

Thank you so much for showing us this beast! It is just amazing, really, although it can't be scrambled or solved.

When you say the pieces on the lower face fall apart f you hold it with the black face on top, does that mean you have to reassemble everytime???

I love the puzzle although it's too small to be functionnal (well).

Greg

Dear Claus,

First of all, thank you for taking the risk, the time and the investment to build this puzzle. It must have been an enormous task to identify, sort, dye and assemble all pieces. We knew from the beginning, that the result would not be speed solving quality. Unfortunately, the puzzle hardly turns at all. I am sorry about that.

Please share your evaluation (and perhaps a video) of this puzzle in its present state, and in its building experience. How did you manage to build it at all? What things are wrong with this designs? What aspects did work out well?

I wonder if and how we could make the "Over The Top" 17x17x17 work. New materials? Bigger size? Other design parameters (filleting radius, shell thicknesses, shape, tolerances)? Different design style (pagoda, binary, linear, ...)? Or should we better abandon the idea that we can ever go beyond Leslie Le's incredible 12x12x12?

Again, thank you!

Oskar

I would have to believe something with so many small parts would need a level of smoothness not possible from SLS. Injection with polished molds and precise mold lines would be ideal. I don't think any 3D printing method can get smooth enough and also be cost effective (Polyjet might have promise but is expensive).
I also don't think silicone rubber molding can be precise enough given registration errors (I commonly have between 0.0 and 0.5mm shifts in m mold halves even with pins).

Dave

Holy ****! That's incredible, even if it doesn't work to a decent level.

I would really hand sand and lube every part just to see it work, offer is up now

I bet it could work it just needs work

Thanks for showing this. I know it is disappointing when something isn't a huge success but it may just be the 2nd part of a long list of events leading to a fully working puzzle.

That is a lot of pieces. I am sorry for getting in peoples' faces for posting on this seemingly dead topic.



Fixed.

I have not failed. I've just found 700 ways that won't work.
-Thomas A. Edison

even if it's just one way, at this rate you'll have a good puzzle $2,450,000 later.

but really thank you for showing this, it shows that sometimes risks must be take.

I know it's been said a couple of times, but thank you so much for posting those pictures even though the project didn't go as well as planed. It's still amazing and without a doubt a fantastic achievement!

From reading earlier posts, I should point at that bad turning quality due to so much friction was foretold. So technically, it did go according to plans. I mean, you CAN twist it, right? Even if it's extremely unstable, it can probably be turned very carefully, which would then make it a working 17x17x17.

it is suitable for casting? We can start a group for casting all parts, them we will have smooth pieces

Now here's an idea: the mechanism you're working with is original, right? Then why can't it be streamlined down to smaller sizes? Specifically, I believe the largest fully functional cubes on the planet now are the 12x12s (and I don't believe there are many of them), so why not redesign your blueprints to a 13x13? Since the # of pieces goes up dramatically whenever you add a cube in each direction, a reduction of 4 cubies would likely reduce the # of pieces WAY down (Not to mention it'd be cheaper). Plus, you'd be working with a simpler design; and once you've mastered it, you can go on towards rebuilding (and possibly redesigning) the 17x17. You'd also be in possession of the first 13x13 in the world.

Just an idea.

@jerry533482 This mechanism only works for 5x5x5, 9x9x9, 17x17x17, 33x33x33, etc. If you'd notice in the mechanism pics, you'd see that it requres the number of layers to be fixed to a certain pattern.

Yes but a 17x17x17 could be bandaged into a 13x13x13.
Plus it will be a few years before I make my 21x21x21 so I like jerry's idea.

Why not perfect the 9x9? It's not the same mech as the v cube, right? Perfect a smaller design, then move on to more layers.

That's not a bad idea! Just make sure whoever does it has mechanism pictures.

So would this design be cubic? It'd probably be a bit easier to keep the puzzle even if it were a square shape; and the more even it is, the more likely it is to be fully functional.

OH MY GOD

First of all I say a BIG Thank You to Oskar. I am proud to be part of this huge project.

I would like to be part of every project Oskar will starting in future.

I am still believing in this design. I know it will work. But the pieces are to small. Please see the picture.


There exist more than 800 pieces with nearly the same size of this both (the other 750 are a littte bit bigger). And this is one of the mayor problems.

If I try to turn one layer the distortion between some pieces are to big and the pieces will fall part.

I think it is not possible to cast this!!

What will happen, if I scale the stl's with factor 1.8 or 2.0? The grip will be better?

What is with the tolerances? They will be double, too. Is this a problem?

Sometimes the biggest parts of the puzzle hooked with the core. I think it will be a workaround to design the core as a sphere.

I will ask Oskar to give me the permission to send the files to a 3dprinter. Karl-Heinz (incredible) promised to sponser a lot of the printing costs. He don't know what this will mean!!

Can this be successful: FDM printed parts in double size? In this case there will be lot of sandings necessary.

Those two little part are cute

For the scale factor, well yes, you can scale it to 2, this will make a working puzzle, i bet, but you have to ask oskar to scale it and then remake tolerances, because 0.2 mm on each face is just too much!

I hope this project will literally see the day and work just fine. I know it can!

Greg