my family uses that for some prayer ritual… its basically the two shell thing… it’s either both face the same way or one faces up while the other faces down. that’s it.
Three sided die doesn’t look much different than the wooden 2 sided dies pictured on the post. Sort of looks like a seed or pod. though there are other clever alternate designs.
I’d like to see a 1 sided die, just for lulz though and see what people come up with.
It’s a ping pong ball with a 1 written on it.
6 Likes
Möbius strip.
1 Like
I’m not sure what you’re arguing. If you’re happy with a biased distribution, then you don’t need two shells or two throws. Just throw a shell and call one side heads and one side tails. Done, no one’s stopping you. Or, as you say use two shells for four different options.
@chris0 and I were just discussing how to use two throws or two shells to make it perfectly fair, mathematically, even if the shells were biased. As it turns out, there is a way to do that, but no, it’s not nice and practical as it doesn’t give you the answer in one throw.
Oh dear, no, I’m not comfortable using it for anything other than a “disruptive” joke in a fun thread on bbs. I’d just use a four or six sided and assign evens and odds like a sane person.
1 Like
Yeah I get it I get it. I was being a dummy and you made a good catch on my bad math. I understand how to make it mathematically precise. All I am suggesting is that if it’s unfair-but-close-to-fair, then the Von Neumann method is fine. And I’m also suggesting that if it is grossly unfair then even though it may be mathematically precise, it will be “inaccurate” since the throws will take so long to achieve the varied outcomes needed to determine a true heads or true tails. That’s all. It’s a good thread.
Obviously a sphere is the best option, but i was curious what kind of weird, amusing things people could think up. Maybe some non-working die illustrations with optical illusions.
Indeed. If the shell is so unbiased that it lands curved-side-up 99% of the time, it will take 69 throws to have a better-than-even chance that there’ll be at least one pair of different throws…
Correct; it’d be faster and more definitive to use the revolver.
A sphere is a terrible D1. It keeps rolling and stuff, so you can never be quite sure it really “landed” on 1 or not.
3 Likes
Furthermore, it’s not a D1 at all, it’s a D∞. And it’s really really hard to tell which number between 0 and ∞ it landed on…
I think the best D1 would be a Gömböc:
Of course, it can take a while for it to settle down and come to rest, but when it finally does, you can take a good look at it, carefully note that the pointy side is “up,” and record that confidently in your book as a “1”.
1 Like
1 Like
A d8, d12 and percentile dice give me most of the possibilities I need
Discbot only works in the games category.
Ok, Serious entry. Basically two cones end-to-end. The cones are transparent, and the numbers are written on the bases of the cones. Could also be done with a d10-shape (this is basically a sanded d10) the inner face will always tilt upward on one side or the other.
ETA: Oops, never mind, this was the first and only solution proposed in the comments of the original Core77 article. I guess I should have RTFCOA (read the fucking comments on the original article)
3 Likes
Lovely sketch though.
And I guess someone took my Möbius strip D1 joke seriously?
2 Likes
