From Sphericons to Countdown dice
HTML-код
- Опубликовано: 5 июл 2024
- How we made a better countdown/spindown die, inspired by sphericons.
Countdown d24: www.mathartfun.com/thedicelab...
Thanks to Alexandre Muñiz for the idea to make a countdown die based on a sphericon.
My article on Rolling Acrobatic Apparatus:
www.ams.org/journals/notices/...
You can 3D print your own sphericon and three-cone sphericon:
www.printables.com/model/1770...
www.printables.com/model/1772...
Inside of me there are two voices.
-One is a foppish prude who points out that for all this effort a simple multifaceted barrel shape would it be better
- The other a craze goblin who wants the pretty shiny!
It's such beautifully unique gemstone like shape. I love it.
A 24 (or even 20) sided barrel would not stay on one face very well. It would be very easy to jog the table and have it start rolling!
I think a 24-sided barrel would need a bigger radius to fit 24 readable numbers along a single edge. Plus it’d need to be wider so it doesn’t tip over. This solution is much more compact.
I guess you could use 2 10 digit spin down dice, then you can get up to 100
@@themixedmaster I had a mtg deck that regularly got me above 100 up during long games that I did that for
@@Elitekross play a different deck then lol
If you rolled it like a regular dice, what would the distribution graph look like? I'd love to see experimental data, although I'm sure no one wants to sit around rolling this thing 1000 times.
I have too much free time and plenty of spindowns... will report back if i manage to do it
Spindown dice tend to prefer the face-neighborhood of the number you started them on-this is true of normal dice too, but it doesn't matter so much since most dice are arranged such that the faces around each vertex have the same sums.
@@empressassassin9975 Have you done it yet?
It will depend on what you roll it on
I just got mine. I’m willing to do this
You make the Most intriguing videos ever.
I'm interested in the design and manufacturing of this shape. The kangaroo design tool looks pretty useful!
This is actually pretty cool! Nice work!
You need some gizmo that pushes the die up against a slope, so it rolls back the same tiny distance and counts down.
Meaning you have a single button to push for each decrement...
I love this video
Fascinating!
Feynman had an intreresting lecture about diffusion, brownian motion and random generators. If we generate a time series from this dice, which kind of distribution is obtained? I mean, it seems to have a regular roll pattern
Great idea
You make cool stuff.
Take my money now!
Would've been cool to have a different coloured light under the glass, such that when the current face is facing towards the camera it lights up. I suppose it's difficult to not also then light up the glass itself
I think the bigger issue is that it would light up the whole die, not just the face that is down on the glass.
I would like to try rolling as a die, thousands of times and chart the distribution of numbers it produces...
Then create a game to take advantage of that distribution
Make sure to get data on lots of different rolling surfaces - I bet it changes the distribution!
highly underrated
Very nice
Noice. extra-longated on the "oi"
this is so cool
DICE KINGS ARISE
Could you make a shape that rotates itself and the spherical joints its leaning on also rotate such that you could set the main shape onto a downwards shape corrisponding to a number and so the outer shapes their own number and like the dice here, the rotation would count the numbers through? Like a tetrahedron shape maybe?
I think I might need a picture to understand what you're suggesting...
this is even better for Magic the gathering- often players will use a few "refuge" lands at the start of a game and go up to 21life- which isn't on a D20! now if only someone would make a countdown D100 for those of us playing life gain decks
Guessing if you had 100 life, you'd probably do what Metroidvanias tend to do and have a token for every "additional" d20 of life remaining, yeh? And for the 21, either do binary "0/no dice is a life point" or just have an additional D1.
Not sure how close I am to what people do, perhaps there's better ways found by the people who actually do this stuff a lot?
Great idea. Wouldn't using Galapagos however make more sense to optimise the tipping distances?
Could be! I've never used it before.
Could we also get D42 Sphericons for stuff like commander? I probably won't buy it from the dice labs so this is for other producers for MTG products and such.
Fuzzy dice, the magical fruit....
At first I thought this was some kind of gomboc and would automatically roll to 1. You could use it as a literal countdown. How hard would it be to design something like that?
My understanding is that the Gomboc has to be very precisely made to work. And that gives you a single stable position, but many paths that go to it. My guess is that it would be very difficult to somehow arrange a single long path that ends at one point. Having distinct polygonal sides along the way seems that it would make it even more difficult to achieve.
@@henryseg Maybe if you cheat and weight it?
If you filled it with fluid, would you get a bubble indicating the current UP face?
Would a hexasphericon (made by revolving a hexagon about an axis through two opposing corners, slicing in half, rotating by 60°, and gluing) also work or does it share the same problem of faces not being vertical as the sphericon?
Right, again the paths on opposite sides of the shape are different, so sometimes you go over an edge.
@@henryseg can you also please clarify why the spherical has a single surface since it looks like it has the same surface before and after you twist the two cones..it still has edges in both configurations..hope to hear from you...
@@leif1075 it still has edges but there’s one continual face along the shape. You could drag your finger all across that face and never go over any edges.
@@themixedmaster I saw that buts that only one ONE SIDE..not the WHOLE shake as he says in the video..your finger will bit an edge halfway around the shape..see what I mean?
@@henryseg Hope you can respond when you can..thanks
Very nice, any chance at a d12? Tons of counters end at 10.
Interesting idea, although I think that if you did the same thing with only 12 sides, the shape you'd get would want to be just a regular rhombic dodecahedron. You'd still have the zigzag path, but it wouldn't be "enforced" by the shallow vs sharp angles.
@@henryseg An elongated rhombic dodecahedron might. I don't know how long you'd need to elongate it to get a 2:1 ratio of tipping along the path versus the unintended tipping direction(s), though.
Sounds like a great design solution for this problem, but I'm a little concerned about how difficult is to keep track of which number was on top if someone hits the table.
Have you considered interlacing several number series:
1,12,2,13,3,14,4,15,5...
Or
1,16,8,2,17,8,3,18,9,4,19,10,5...
So the user has to make 2 or 3 steps to move from N to N+1 but a little bump on the table won't make it difficult to deduce which number was showing up?
Changing the ordering to something non-standard would rather work against the aim of making an easy to use countdown die...
@@henryseg Yes, I get it, but I was proposing Error Detection/Error Correction versions (in the sense of Coding Theory). But yes, I guess that these variants will be way less intuitive...
Is there any way to make this die fair or close to fair? I would love it as a d20!!
I'd also like to know this :)
There are 6 identical "strokes" of numbers running "north" and "south" on the dice; "identical" in the sense that the face shapes repeat. With 24 faces total, each "stroke" has 4 faces. That means you could make 4 dice, each with the sequence starting at a different one of those 4 faces and the result would be fair if you drew a random dice from a bag and then rolled it.
Or since touching the dice with your fingers while drawing them opens it up to cheating, maybe use a dice cup with a lid that has a hole in only 1 of the 4 quadrants, letting only 1 of the dice out when you flip it. Or roll all 4 along with a color-coding d4 that determines which result to use.
I'm not clear on what "tipping distances" you are using. The red distance? I think a natural approach is to make the "energy barrier" the same for crossing over each edge. When it is rotating over the edge, it's center of mass is raising. Making the distance from each edge to the center of mass equal is equivalent to the equal potential energy condition.
I'm pretty sure that your energy barrier is equivalent to what we did. All faces are the same distance from the center of the die, and it turns out that all the dihedral angles along the path are also the same as each other. So the amount you have to raise the center of mass will be the same for all edges, as long as the red "tipping distances" are all the same.
Have you considered producing a d24 based off of the triakis octahedron?
The shape of the faces is not the best for drawing a number on...
How about a countdown Gomboc?
I heard the spindown D20 is patented by Wizards of the Coast. I'm not sure if it's true, and if it applies to other sizes of dice, but it might be worth checking out. Wizards is not exactly known for being nice about their patents.
I can find no other place on the internet where this is claimed, and plenty of different companies produce spindown dice. So likely you were thinking of something else. (e.g. the spindowns WotC provides often have a set symbol on them, and that symbol would be protected.)
@@starwed sounds fair enough. My local game store only had the spin-downs from WotC, and cheaper regular D20s from different brands. They said this was a patent issue, and I had never bothered to check 😅. Thanks for doing that for me, and correcting my claims 😀.
In terms of probability, is this sphericon die a fair die or is it only good for spin down purposes?
I would be very surprised if it is anywhere near fair.
Yo real talk, your website is EXTREMELY outdated and I can't find a way to order anything, I've clicked around 5 or 6 pages, looping end over end and I can't find any way to buy anything from you. Please please please HOW do I buy this die?
You can buy it from www.mathartfun.com/DiceLabDice.html
Are there any stores that sell this type of dice?
See mathartfun.com/DiceLabDice.html
Would it be possible to make something reasonably usable with at least 40 faces? Asking for the commander community within Magic the gathering
Could be... what would be the best number to go for do you think? Exactly 40? The more you get, the easier it will become to jog the table and have it skip a number (although our d120 isn't terrible for this, so there's a ways to go).
@@henryseg something in between 40 sides and 80 side as a spin down would likely handle the majority of games of commander/EDH where the starting life total is 40. Some times life total do get out of hand so upwards of 100+ as a spindown could also be potentially useful
I thought about this. After 3, the next odd number to use for rotational symmetry along the central axis is 5, and if there are 8 faces per zigzag like with the d24, that would get exactly 40. (I think the rotational symmetry has to be odd? But as I have demonstrated, I'm really bad at thinking in 3d, so I don't know.) Anyway, if you do that, the dihedral angles along the path might become too close to the dihedral angles that jump across the divides. Dividing up the triangles between the kites and rectangles gives 12 faces per zigzag, which gets you 60 total.
Request: D2
120 is the max for a fair dice.
Roll a d30 and a d6, subtract 1 from the d6 and multiply that by 30, then add the d30's result. (Or make a d6 with numbers 0, 30, 60, 90, 120, 150).
@@PystroThe D180 Is So Big
If I could subscribe harder I would