Online since 2002. Over 3300 puzzles, 2600 worldwide members, and 270,000 messages.

TwistyPuzzles.com Forum

It is currently Tue Jul 22, 2014 12:36 pm

All times are UTC - 5 hours



Post new topic Reply to topic  [ 29 posts ] 
Author Message
 Post subject: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Sat Jul 26, 2008 11:03 pm 
Offline
User avatar

Joined: Mon Feb 04, 2008 8:57 pm
Video

Heres a video showing the movement of the V1 gigaminx and V1.5

After that you see a lengthy attempt at me playing with my Elite Tetrahedron prototype. I plan on redesigning it to be smaller and hopefully far more stable than it currently is.

If anyone has advice on what could be wrong with the puzzle and how to fix, please let me know. at the end of the video I show what the major issue in the design is. Keep in mind I have NOT sanded any of the sides of the pieces, only the faces, and edges so they would be slightly rounded.

_________________
Image


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Sat Jul 26, 2008 11:49 pm 
Offline

Joined: Sat Mar 22, 2003 9:11 am
Location: Marin, CA
I'm not sure of all the details of your design, but I think I can see the highlights of the issues.

The corners looseness I'm guessing is caused by the build-up being too 'holistic', with those corners being held down indirectly. In general there's the 'core' of a puzzle, and the core holds down some pieces, which hold down some more pieces, possibly etcetera, and each piece has some number of hops from the core to it. The greater the number of hops, the looser or just more poor working everything is. For a puzzle like this one, the interior core should build up to a pyraminx, and the outer corresponding pieces should connect directly to the interior core pieces.

The eastsheen 4x4x4 makes the same mistake with the outer corners, not having them connected directly to the interior corners, because it can't due to not using proper shells. That brings up another point about these designs in general - all interior pieces should be in 'shells', with an outer spherical face, interior spherical face, and all cuts around the edges being either cuts through planes which go through the center of the puzzle or cones whose apex is at the center of the puzzle. Cutting the very outermost visible parts along the visible cutting planes for aesthetic reasons is of course forgiveable, but as a general rule it's always better to not have cuts along planes which don't go through the center of the puzzle.

Your other big problem appears to be that the slice between the two layers of pieces jams constantly. I'm guessing that's caused by them basically being like one set of pieces but cut in half. The problem with such designs is that they only have one set of groove which both pieces sit in, and the grooves naturally shove the pieces inwards, into each other, and causes them to jam. The solution is to have something which amounts to two sets of interior grooves, with the pieces held separately. That way they don't catch on each other so much.

Hope that was helpful. I can explain more if anything is confusing.


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Sat Jul 26, 2008 11:53 pm 
Offline
User avatar

Joined: Tue Oct 09, 2001 6:14 am
Location: Orange County, CA, USA
Honestly it would be hard to say what the problem is without actually holding the puzzle, but I can hazard a guess or two.

The pieces on the point are floppy, as you demonstrated at the end of the video. That leads me to believe that either tolerances are not tight enough or that the pieces are not deeply embedded enough. In either case that would easily lead to pieces rotating slightly out of alignment and causing binding.

Is there any way to design this particular puzzle so that the pieces anchored to the core are the tips instead of the face centers? It would seem to me that this might increase stability somewhat.

_________________
http://chrono.org/puzzles.html
http://www.youtube.com/nechronius


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Sun Jul 27, 2008 12:03 am 
Offline
User avatar

Joined: Mon Feb 04, 2008 8:57 pm
The mechanism of the puzzle is Based on the V-cubes. The pieces all have been printed in SLS directly from a stl file that had zero tolerance adjustments after the raw cut from the paths. this means that the parts should fit together as tight as can be.

In theory even the corners should be tight but its not just the corner, its also those corner + pieces, all of them, the only perfect fitting piece in the puzzles are all of the center edge pieces.

_________________
Image


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Sun Jul 27, 2008 12:13 am 
Offline
User avatar

Joined: Tue Oct 09, 2001 6:14 am
Location: Orange County, CA, USA
If I'm reading things correctly, I believe that Bram said much of what I stated as well.

I'd see if there was any way to directly anchor the points to the center somehow. That would make the puzzle dramatically more stable, but of course would need an entire redesign.

_________________
http://chrono.org/puzzles.html
http://www.youtube.com/nechronius


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Sun Jul 27, 2008 11:45 am 
Offline

Joined: Sat Mar 22, 2003 9:11 am
Location: Marin, CA
Drewseph wrote:
The mechanism of the puzzle is Based on the V-cubes.


I don't know what this means. If you show pictures or renderings of the pieces individually, and photos or cross sections of sets of pieces put together, then I might be able to give a coherent and knowledgeable opinion, but short of that there's a lot of guesswork.


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Sun Jul 27, 2008 12:30 pm 
Offline
User avatar

Joined: Mon Feb 04, 2008 8:57 pm
Here are images of the pieces, the Ambient occlusion shows there are no gaps between the parts before printing them.

Image
Image
Image


Attachments:
DSC05120.JPG
DSC05120.JPG [ 100.2 KiB | Viewed 4716 times ]
Elite_Tetrahedron_AO_sm.jpg
Elite_Tetrahedron_AO_sm.jpg [ 98.29 KiB | Viewed 4711 times ]
Elite_Tetrahedron_sm.jpg
Elite_Tetrahedron_sm.jpg [ 90.56 KiB | Viewed 4715 times ]

_________________
Image
Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Sun Jul 27, 2008 2:31 pm 
Offline
User avatar

Joined: Sun Mar 11, 2007 3:11 am
Location: Oregon, USA
Gotta' agree with Jin, despite the fact it means redesigning every part (except possibly the spider). It should be possible to design one with the tips anchored to the spider instead of the face centers, since they turn on the same axis, but what a tedious redesign that would be!

But if there's a choice which part (tip or face-center) to anchor, I can see two major reasons to choose the tip: 1) the tip extends farther from the core, so any movement it has is magnified; 2) the tip has more pressures on it during use, so it benefits more from being anchored.


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Sun Jul 27, 2008 5:53 pm 
Offline

Joined: Sat Mar 22, 2003 9:11 am
Location: Marin, CA
Yeesh, okay, the pictures help a lot, and basically everybody's guesses are right. You've been a bit misled
by the Verdes cubes mechanisms - they're doing injection molding, you aren't, so you're going to have
to use more mechanism- than materials-based solutions to problems.

VeryWetPaint is right that it should be the corners should be the ones connected to the center, not the face centers.

Everybody is right that there should be an interior pyraminx. Really, you should think of this puzzle as being a
pyraminx with some sections for the 'extra' pieces carved out and the pieces added in there.

The most subtle point is this one: You have two layers of 'extra' pieces, one on the face center/edge side, and one
on the corner/edge side. The mechanism should be such that the pieces on one of the two sides will mostly stay in place
even if the pieces on the other side are removed (with the caveat that the pieces at the intersection with slices going a
different direction can move out, of course) .
That means a complete groove on the face center side, and a complete groove on the corner side. If you do that, then the
catching between the two layers should go away almost completely.

Hope that helps. Sorry to advocate a complete redesign.


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Sun Jul 27, 2008 7:26 pm 
Offline
User avatar

Joined: Mon Feb 04, 2008 8:57 pm
Bram, I don't understand anything you mean. basically I'll need a picture to understand your recommendation.

When I designed my MT to be a corners based mech everyone told me to do a centers based >_>

_________________
Image


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Sun Jul 27, 2008 8:46 pm 
Offline
User avatar

Joined: Thu Dec 21, 2006 5:32 pm
Location: Bay Area, CA
Drewseph wrote:
When I designed my MT to be a corners based mech everyone told me to do a centers based >_>

Shows you what we know ;)

As I recall, that advice was because everyone was preoccupied with the problem of the super long screws you would need for the corners based approach.
Good advice for that problem, apparently bad advice for puzzle stability.

So if you do a redesign, you will have to solve the issue of screw length. I suppose this could be solved by two part corners (basically a really long cap), but then the stability of the cap becomes a problem.

Lately I have been playing with aluminum tubing (cut to size as a way to get the effect of many small washers in a row). Aluminum might be too weak, but I'm guessing you could find steel tubing pretty thin (check out http://www.smallparts.com) and use that as a screw extender. Screw a set screw into the core and tapped tube, then fit the tube through the thin part of your piece up to the tip, and use a screw into the tapped other end of the tube. A lot of pieces, but in theory it should give you the strength you need.

Dave
P.S. I like the big picture for the detail, but it really makes reading the thread difficult as it scrolls about a full page to the right (12" laptop...). Could you replace it with a smaller picture and a link to the larger version?

_________________
Image
LitwinPuzzles.com has info on my puzzles.


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Sun Jul 27, 2008 10:06 pm 
Offline

Joined: Sat Mar 22, 2003 9:11 am
Location: Marin, CA
Drewseph wrote:
Bram, I don't understand anything you mean. basically I'll need a picture to understand your recommendation.

When I designed my MT to be a corners based mech everyone told me to do a centers based >_>


You asked for it. I'm not responsible for any internal bleeding to your ocular nerve which looking at this
picture may induce. All pieces of the same color are supposed to be identical shapes, my sketching skills
fell somewhat short of strongly indicating that.

This is a conical section view, much like some of your photos. Note the completely separate grooves for the
red and green pieces, so if you were to remove one of them the other wouldn't slide into the gap. Also
note that the purple parts are completely held down by the black parts which are held down by the core, so if you were to
strip out all the red and green pieces the puzzle would still move just fine.

Whether the corners or centers are the ones attached to the core is one of the least important details,
by the way, so long as they're held together near the core.


Attachments:
File comment: Isn't my artwork beautiful?
separate_grooves_mechanism.gif
separate_grooves_mechanism.gif [ 11.94 KiB | Viewed 3732 times ]
Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Sun Jul 27, 2008 10:40 pm 
Offline
User avatar

Joined: Thu Apr 05, 2007 8:05 pm
Location: Seed-nee
What, were you on a helicopter when you drew that?!?

Just kidding of course, I get what you're saying now too :)

I'm just going to throw my 2 cents in though; another item that would improve the motion would be to hollow out the large corner and edge pieces--think two part items. Removing these (relatively) huge masses would lower the second moment of inertia for these pieces considerably, which would improve the stability *during* turns by quite a bit.

I sense R2.0 coming, as opposed to some sanding... :(

_________________
я буду нуждаться пиво скоро.


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Sun Jul 27, 2008 11:15 pm 
Offline
User avatar

Joined: Mon Feb 04, 2008 8:57 pm
Thanks for that illustration, yeah its rough, but I get the idea. you are on the track with what my MT looked like, and my god the result of having so many grips = jamming galore.

The ET does not jam for the same reason as the MT though, it only jams becuase the smaller pieces and corners are loose. once I solve this it should work in its current mech. I have casted a second ET using the master molds directly from the SLS. my demo of the ET was casted from second generation pieces which I sanded round edges and polyurethane'd the surface to make it shiny then made a second mold using those.

anyways my second ET is still wiggly however it is much less loose than the first. (trade off it doesn't turn at all >_>, not sanded)

I'm convinced that the issue is in the outer face for the Grips on the Edge and corner pieces. if they are thicker in the direction of the outer faces, then it will push the scoop closer to the core thus tightening up the parts. I think all thats needed is:

1: more pillowed shape
2: An additive tolerance of the edge and corner grips
3: smaller design
4: hollow parts wherever I can get them to be

_________________
Image


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Sun Jul 27, 2008 11:18 pm 
Offline

Joined: Sat Mar 22, 2003 9:11 am
Location: Marin, CA
the.drizzle wrote:
What, were you on a helicopter when you drew that?!?

I have a severe visual processing problem, which usually is very subtle but makes my drawing ability
nearly nonexistent. (Yeah, ironic, isn't it?). Using a trackpad doesn't help either.


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Mon Jul 28, 2008 12:35 am 
Offline

Joined: Sat Mar 22, 2003 9:11 am
Location: Marin, CA
Drewseph wrote:
Thanks for that illustration, yeah its rough, but I get the idea.
you are on the track with what my MT looked like, and my god the result of having so many grips = jamming galore.

Well yeah, if you take unfinished pieces with no gap and non-rounded corners, pieces in grooves are going to jam :-P

My own approach is to give reasonable gap, round the corners, and use proper grooves. This is optimized for being able to take
unfinished pieces directly out of a low-quality 3d printer, put them together, and have them work immediately.


Top
 Profile  
 
 Post subject: Q
PostPosted: Mon Jul 28, 2008 12:46 am 
Offline
User avatar

Joined: Sun Mar 11, 2007 3:11 am
Location: Oregon, USA
Bram wrote:
the.drizzle wrote:
What, were you on a helicopter when you drew that?!?

I have a severe visual processing problem, which usually is very subtle but makes my drawing ability
nearly nonexistent. (Yeah, ironic, isn't it?). Using a trackpad doesn't help either.
I find that very unexpected because I was
certainly convinced you were mentally visualizing the parts. You put a lot of thought into the concept represented by that diagram,
so I put a lot of thought into visualizing the rest of it. It warrants thought.

While mentally visualizing all the parts I thought I'd identified an important issue (though not unmanageable). Then while I tried to draw a
diagram to illustrate the issue I thought I saw a built-in solution, possibly one you'd already thought of. It's worth sharing even if
Drewseph doesn't pursue it. Bram's ideas always lead to interesting discussion!

(Please forgive me Drewseph for vandalizing one of your excellent images.)
Attachment:
Tetrahedron fouling zone.jpg
Tetrahedron fouling zone.jpg [ 12.06 KiB | Viewed 5015 times ]

For the sake of discussion I lettered several of the parts and drew a dotted green circle to illustrate a possible location of one of the
concealed grooves, which would carry parts W, F, C, A in one of the slices illustrated.

The purpose of the diagram is to show that if parts A-A-A each attach to grooves in the center triangle they would have to be only partly
connected to the groove so they wouldn't overlap or block each other's movement at the orange splotches. Problem? On second thought,
no. Because part C rides in two adjacent grooves, it can extend into the orange "fouling zone" without blocking any other movements.
This might help parts A and C.

Then I noticed a lulu of an unresolved issue: if I assume the black pieces in your diagram refer to the edges B then this leaves the tips Q
unexplained. Each tip Q connects only to pieces that ride in the red groove (from Bram's diagram) which don't connect to the other slices
(due to the separate green & red grooves) nor to the rest of the puzzle. If the black pieces were supposed to be the tip pieces Q then
that leaves the edge pieces B unexplained, and the numbers don't add up either.

Aha, insight!

Earlier Bram agreed "there should be an interior pyraminx" but I propose we substitute a skewb. A skewb could provide a stable foundation
for the four tips (skewb's 4 fixed corners), the four centers (skewb's 4 floating corners), and the six edges (skewb's six faces). With that
one-word substitution Bram's entire explanation in this post
becomes a practical, workable solution. I never would have guessed. Bravo, Bram!


Top
 Profile  
 
 Post subject: Re: Q
PostPosted: Mon Jul 28, 2008 12:27 pm 
Offline

Joined: Sat Mar 22, 2003 9:11 am
Location: Marin, CA
VeryWetPaint wrote:
Earlier Bram agreed "there should be an interior pyraminx" but I propose we substitute a skewb. A skewb could provide a stable foundation
for the four tips (skewb's 4 fixed corners), the four centers (skewb's 4 floating corners), and the six edges (skewb's six faces).


Making the core a skewb makes a bunch of things easier, although it results in a bunch of unnecessary hidden pieces. If you make the core look like this then it's possible to avoid the extra pieces:

http://www.geocities.com/jaapsch/puzzles/sphere.htm?red=213&symred=1&angle=0,90,0

Come to think of it, the same approach works for building up a 2x2x2 to a 4x4x4, which is my new favorite way of making a 4x4x4.


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Mon Jul 28, 2008 4:20 pm 
Offline
User avatar

Joined: Mon Feb 04, 2008 8:57 pm
Bram, that is how the current core looks. my mech is a skewb mech, only use the centers instead of corners.

If I were to use the corners as support anchored to the core they would SNAP OFF from turning, they are already fragile as is and they are 2 times thicker than my MT

I'm not arguing with your advice. I am considering designing the next mech with your idea in mind, however I chose to anchor from the center for the very reason of the necks snapping on the corners.

_________________
Image


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Mon Jul 28, 2008 4:57 pm 
Offline
User avatar

Joined: Wed Aug 17, 2005 8:26 pm
Location: Saratoga Springs, NY
The Gigaminx that I designed and Aleh built had the same problem. If the master parts are exactly the right size, the shrinkage of the parts will result in the corner "block" being loose. If you were able to make the puzzle with all SLS parts, I am sure that you would not have the same problem. Aleh fixed this with a lot of fine shims to make it turn smoothly.

The other issue my Gigaminx may have had was the clearance in the interlocking shells. There are direct contact surfaces that need zero gap, but other surfaces that should have a small gap to avoid binding or pushing the parts out of alignment (see the attached picture). I did not have the clearance at the end of the tab, so that may have pushed the corner "block" out of alignment. I have since added the gap, but I may redesign it completely when I get my v-cubes and have a chance to study the mechanism in detail.


Attachments:
gx - xsect.jpg
gx - xsect.jpg [ 56.86 KiB | Viewed 3677 times ]

_________________
I will not Reason and Compare: my business is to Create. -William Blake

Videos on Youtube!

New Designs on Shapeways!
Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Mon Jul 28, 2008 5:13 pm 
Offline
User avatar

Joined: Mon Feb 04, 2008 8:57 pm
You know this may be why my Gigaminx 1.5 has such a perfect fit. I didn't sand anything on the master fdm parts in fact all I used was ABS colvent cement about 5 coats on all parts and then wiped it smooth with paper.

I think that may have enlarged the pieces to compensate for shrinkage from the molding casting process..... this no part was too loose or too tight. if I reprint the ET I'll choose FDM. I can't afford a complete frint of all the ET parts though. unless I find 2 people who are wiling to buy a gigaminx from me.

_________________
Image


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Mon Jul 28, 2008 7:30 pm 
Offline

Joined: Sat Mar 22, 2003 9:11 am
Location: Marin, CA
Puzzlemaster42 wrote:
The Gigaminx that I designed and Aleh built had the same problem. If the master parts are exactly the right size, the shrinkage of the parts will result in the corner "block" being loose. If you were able to make the puzzle with all SLS parts, I am sure that you would not have the same problem.


I'm not sure that it's shrinkage which causes the problem so much as it is jaggies. The problem is A LOT less of an issue for faces which are
close to perpendicular to the grain than for ones which are nearly parallel to the grain. It really pays to put some thought into direction
of grain when you're printing.


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Mon Jul 28, 2008 11:22 pm 
Offline
User avatar

Joined: Mon Feb 04, 2008 8:57 pm
Taylor, Good point!

_________________
Image


Top
 Profile  
 
 Post subject: Re: Q
PostPosted: Tue Jul 29, 2008 7:21 am 
Offline

Joined: Sat Mar 22, 2003 9:11 am
Location: Marin, CA
VeryWetPaint wrote:
Earlier Bram agreed "there should be an interior pyraminx" but I propose we substitute a skewb.


When you first said this, I thought you were just plain wrong, but after thinking about it some more, I realized that I'm the
one who was confused. The pyraminx core and skewb are easy to get mixed up, because they both contain the same number of pieces
in about the same places and the differences in how they move are subtle. In this case, the cuts are deep enough that the outer big pieces
do, in fact, correspond to a skewb, so the core should look like this:

http://www.geocities.com/jaapsch/puzzles/sphere.htm?red=150&symred=1&angle=359,111,0

The outer pieces can then be added in a manner very analogous to building up a 2x2x2 to a 4x4x4, or a 3x3x3 to a 7x7x7.

My visual problem is a little hard to describe, by the way. What I can do is much more 3d and 2d, and I don't 'see' things so much as I 'feel' them.


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Tue Jul 29, 2008 7:33 am 
Offline
User avatar

Joined: Mon Feb 04, 2008 8:57 pm
I really don't understand this sphere program you keep posting >_>

what is its purpose?

how does that represent the core at all? And my mech is already a skewb. I don't get how you can say something should be a diff mech only to come back and say keep it the same

_________________
Image


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Tue Jul 29, 2008 1:53 pm 
Offline
User avatar

Joined: Sun Mar 11, 2007 3:11 am
Location: Oregon, USA
Sorry Drewseph, our discussion has gotten a bit abstract. I think that's a healthy sign, since a stated goal of this thread is to
refine the mechanism, so it's on-topic to debate the theory behind the mechanism. I don't know if this abstract
discussion will lead to a better puzzle
but it's worth a try.

Jaap's "Sphere" applet is designed to explore rotational and spatial symmetries by mapping them onto the surface of a sphere.
It's not directly intended for puzzle design, but it's tremendously useful for exploring the schemes behind most twisty puzzles.
The relationship between cuts on a twisty puzzle can be represented by the relationship between the lines projected
onto the sphere, even though the twisty puzzle might not be built as a sphere.

Bram probably wasn't suggesting that the core would literally look like these sphere, but the sphere represents a map of
where the major parts fit relative to each other. The sphere itself isn't a puzzle until you do things to make it tangible, such as
cutting the slices and adding points of rotation. Here's Bram's sphere with some dots to help explain its relationship to a Skewb,
and hence its relationship with the core of your puzzle.
Attachment:
Sphere.JPG
Sphere.JPG [ 21.63 KiB | Viewed 3650 times ]

Assume the red lines are cuts, and that the parts can rotate around the dots on the triangular pieces. The triangular pieces
correspond to the corners of a Skewb, and hence half of them (magenta) correspond to the tips of the tetrahedron and the
other half (yellow) correspond to the face-centers. The six square regions correspond to the long edges of the tetrahedron.

Adding together Bram's previous comments, I believe he's using the sphere as a theoretical foundation for the enhanced
tetrahedron but there are several transformation steps between the theoretical sphere and the concrete tetrahedron. For
illustration let's continue the theoretical transformation: as before, let's assume we attach extensions onto the sphere to
transform it into a tetrahedron, as described in the previous paragraph. Now, following Bram's earlier observation, widen the gaps
between the extensions so we open deep canyons, but don't modify the spherical 'core.' This could make a stable
tetrahedron despite the wide gaps between the exterior pieces because the skewb-like core is intact and holds them in place.
Next cut grooves into the walls inside the canyons so the grooves follow the path of the puzzle's rotation. Finally insert two
layers of pieces into the canyons so one layer rides in the grooves in the inner wall and the other layer rides in the outer wall.

Good grief, no wonder you got lost. I think I know what Bram was trying to describe, but my description is considerably worse!

I think you understood Bram's concept earlier, so I'm just trying to help explain how the topological discussion ties into it.


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Tue Jul 29, 2008 2:50 pm 
Offline

Joined: Sat Mar 22, 2003 9:11 am
Location: Marin, CA
Drewseph, here's a question which may help clarify things. In your mechanism, if you were to remove all the
small pieces just leaving the face centers, corners, and big edges, would the remaining pieces hold together
and move nicely like a skewb or would the corners fall out? If my understanding of your pictures is correct, then
the corners would fall out very easily. What I'm advocating is that there be a literal skewb (yes VeryWetPaint, a
spherical shape for it is strongly preferred, so that higher layer pieces can slide over the same spot in
multiple directions).

It appears to be that the problem with your mechanism is that the corners are held down very indirectly, making
them very wobbly, so this sort of mechanism would fix that problem.

We also seem to have a disconnect with regards to the corners snapping issue. If the central core is a
self-contained skewb, then the difference between corners and faces is very subtle, the only real difference
being that one is slightly more wobbly than the other, as happens in an actual skewb. Your insistence that
there's a big difference as to which outer type connects to which inner type seems to indicate that we
have a miscommunication somewhere.

If the snapping problem really is simply one of the corner pieces being too long and thin (they sure do look
long and thin in your pictures) then the best solution is to simply make the corner pieces burlier. An
aesthetically ambiguous but effective method for doing that is to make the slices not exactly planar, so
that the corner pieces bulge out below the surface of the puzzle, and the pieces next to narrow. That
results in slightly non-straight visible slices, but doesn't interfere with the mechanism's functioning.

Or you could make the whole thing a cube instead of a tetrahedron. That would make the pieces a lot
less long and thin.


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Tue Jul 29, 2008 3:21 pm 
Offline
User avatar

Joined: Thu Apr 05, 2007 8:05 pm
Location: Seed-nee
Bram wrote:
Or you could make the whole thing a cube instead of a tetrahedron. That would make the pieces a lot less long and thin.


I was thinking the same thing! That would automatically add a pile of stability in one step.

As well, for "stronging" the central core of the puzzle, rather than printing / casting the parts, why not simply use a (production) skewb puzzle ball? I know they used to be available on ebay for something like $8 each, so if you got two of them you could build a very perfect, very strong sphere with the 150-level cuts already on it, and simply attach the next shell to that in one step. It would be big, no doubt, but it would remove a huge number of the obstacles too.

_________________
я буду нуждаться пиво скоро.


Top
 Profile  
 
 Post subject: Re: Video: Elite Tetrahedron + Gigaminx 1 vs 1.5
PostPosted: Tue Jul 29, 2008 3:38 pm 
Offline
User avatar

Joined: Mon Feb 04, 2008 8:57 pm
I wouldn't know how to transform an existing puzzle, nor do I have to tools to do so. so I'd rather print the parts and cast them than built them.

I see the arguments here. perhaps the tetrahedron is just plain unstable..... its a shame....

_________________
Image


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 29 posts ] 

All times are UTC - 5 hours


Who is online

Users browsing this forum: Google Adsense [Bot] and 11 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  

Forum powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group