h a l f b a k e r yOn the one hand, true. On the other hand, bollocks.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Please log in.
Before you can vote, you need to register.
Please log in or create an account.
|
*Pedants please see footnote.
Most dice give an evenly distributed random number over their range.
I propose dice of non-uniform shape which give values with different probabilities.
Example: A prism with extrusion surfaces of a series of widths.
It could have six sides of length 1,2,3,4,5
and 6 (times some constant).
The two end-caps would be rounded so a throw would never land on them.
When you roll this die it would be more likely to stop on a larger side. Some experimentation may be necessary to give a good range of probabilities - ideally they would be logarithmic I suppose.
This would mean that you had more chance of throwing a 1 than a 2, a 2 than a 3 and so on.
In addition, a good design would arrange the sides so that 'opposite' sides were approximately parallel.
Might be a useful building block for board game designers, role-players etc.
*The singular of dice is die. However I wanted to distinguish the title from death.
Liar's 8 Ball
http://www.halfbake...Liar_92s_208_20Ball
idea related to [suctionpad]'s suggested implementation [krelnik, Oct 05 2004, last modified Oct 17 2004]
[link]
|
| |
This might be possible with ordinary-shaped dice, by altering the centre of gravity and the order of the numbers on the faces. Or an alternative for board games would be a spinning wheel such as was used on "Game of Life". |
|
| |
Other possibilities would be exponential dice, polynomial dice, power law dice etc. Think of the fun you could have by using a power law die with the wrong constants... |
|
| |
I don't really understand what this would be needed for though. Presumably you're visualizing a more sophisticated situation than that which could be achieved by grouping numbers on a regular die i.e., if it's a 1, 2, 3 or 4 I'll go to bed. If it's a 5 I'll write a poem, and if it's a 6 I'll rape my cat.
But I don't see what this could be. |
|
| |
Nonsense - all sorts of games call for random numbers with uneven distribution, D&D being the archtype. I seem to recall someone posting an idea for dice with uneven distributions before, but if they did, it's long gone. Anyway, any dice idea will get my vote (my own collection includes spherical dice). |
|
| |
Hmm..spherical dice...maybe this should be in an idea of its own, but since I haven't actually posted an idea yet, why start now? |
|
| |
Fluid-filled spherical dice with a bubble-indicator: |
|
| |
Functioning much like a fluid level, the bubble would float to the top, to make it easier to determine exactly which number was actually rolled. (I've always hated those 80-plus sided dice for that reason) |
|
| |
Before you do something crazy like posting an idea, note that spherical dice are Baked (else I wouldn't have them in my collection). |
|
| |
Give your death many sides and put numbers several times. Example using a dodecahedron (20 sides, regular): |
|
| |
6: 1 time , 5: 2 times, 4: 3 times, 3: 4 times, 2: 5 times, 1: 6 times, (is that 21 surfaces?) |
|
| |
Its not logarithmic but at least not the standard all-the-same-probability die. |
|
| |
To make things really interesting, the die could have more than 3 dimensions. This could be easily done virtually. Well, maybe not easily, what do I know? |
|
| |
Yes spherical dice are baked, but have you ever seen one with a bubble to indicate top dead center? |
|
| |
Freefall: you post it, we'll toast it. |
|
| |
Dodecahedron: A polyhedron with 12 faces. This could screw up your odds, kbecker. |
|
| |
You can create a logarithmic distribution with a regular d6. |
|
| |
first roll:
1 - -1 and roll again
subsequent rolls:
1, 2 - -1 and roll again
3, 4, 5, 6 - stop |
|
| |
6 - 1 and roll again
subsequent rolls:
1, 2, 3, 4 - stop
5, 6 - 1 and roll again |
|
| |
As you move away from zero, every number is 1/3 as likely to come up as the one before it. |
|
| |