D3s [Archive] - Wizards Community - Pages Merged by Solauren

D3s [Archive] - Wizards Community

Post/Author/DateTime Post
Orcmonk89
12-28-05, 06:58 AM What is this? I know it's a die, but cannot locate an image.
Nurgan_the_drunked
12-28-05, 07:07 AM It's shocking, they never sell d3s in dice bags or offer them in games stores anymore. We always got them in my day. Just because they're against the rules of natural reality to have it! Pathetic excuse, I tells ya ;)

Seriously, there is no d3 die (or if there is, it's another die with different numbers on). To get a result for a d3, roll a d6, and halve it, rounding up so:
1-2: 1
3-4: 2
5-6: 3

Hope that helps
Milburn
12-28-05, 07:40 AM OR, another way that is sometimes used is to roll a d6 like this-

1/4=1
2/5=2
3/6=3

I have heard of a d6 with numbers from 1-3 on, but i dunno about that. There is not a shape that makes a fair d3, without having different shaped faces, trust me.
Joe Khol
12-28-05, 07:50 AM The two ways I've see proper D3s produced is a D6 numbered 1 to 3 twice and a prism with rounded ends. The latter 'rolls' even worse than a D4 though and never really looked right to me.
Planar Wanderer
12-28-05, 08:04 AM Theres also 1d4 re-roll on a 4, that works as well.
Orcmonk89
12-28-05, 08:14 AM Cheers, I thought it was physically impossible, but I saw 1d3 written and thought 'I don't have this, and I can't see how it'd work', so I asked.
Thanks again.
Chris
CzarGarrett
12-28-05, 09:10 AM Should they ever form an actual d3, the Universe will implode on itself.
babysamurai
12-28-05, 09:20 AM Should they ever form an actual d3, the Universe will implode on itself.

Yeah, I think Stephen Hawkings has a theory on the d3.
kayelbe
12-28-05, 09:27 AM Theres also 1d4 re-roll on a 4, that works as well.

That actually messes with the "random chance" aspect of dice rolling. Someone explained it to me once, but I forget the details. You don't have a 1 in 3 chance of rolling a 1, 2, or 3.

The 1-2=1, 3-4=2, 5-6=3 method is the most "random".
Seeker95
12-28-05, 09:29 AM They have 'em.
They're just not "true polygons".
d3 (http://www.advancinghordes.com/images/gamescience/gms_43t3k.gif)
And another (http://shop.rpg.net/images/uploaded/d3.jpg)
Majorafire77
12-28-05, 09:32 AM That actually messes with the "random chance" aspect of dice rolling. Someone explained it to me once, but I forget the details. You don't have a 1 in 3 chance of rolling a 1, 2, or 3.

The 1-2=1, 3-4=2, 5-6=3 method is the most "random".
That makes no sense. Why would you be more likely to get a 1 on a die roll than a 2 or 3 (or vice versa)? There's no reason why the other method would be more random; the only advantage is that it might take less time (rolling 1 6-sided die once as opposed to 1 4-sided one multiple times).
Nurgan_the_drunked
12-28-05, 09:34 AM That actually messes with the "random chance" aspect of dice rolling. Someone explained it to me once, but I forget the details. You don't have a 1 in 3 chance of rolling a 1, 2, or 3.

The 1-2=1, 3-4=2, 5-6=3 method is the most "random".
It's still 1 in 3, if you re-roll, no more or less "random" than halving a d6. The chance gets skewed if you say a 4 is a set thing (if you get a 4 treat it as a 2, for example) but the chance approaches 1/3, as you roll the dice more. As there are 3 equally likely events it can happen as. You just run the risk that you'll be rolling 4's 'til the cows come home.

D6's are better, 'cos you just roll them once, and they do actually roll properly. But d4s work
babysamurai
12-28-05, 09:35 AM They have 'em.
They're just not "true polygons".
d3 (http://www.advancinghordes.com/images/gamescience/gms_43t3k.gif)
And another (http://shop.rpg.net/images/uploaded/d3.jpg)

Interesting...they kind of look like distorted jellybeans.
MrBrown
12-28-05, 10:04 AM That actually messes with the "random chance" aspect of dice rolling. Someone explained it to me once, but I forget the details. You don't have a 1 in 3 chance of rolling a 1, 2, or 3.

I'd like to see the proof of that. :D

Really, if the rolls are independent from each other, then a d4, 4s rerolled, will do it as well. And, if you roll your dice properly, they should be.

Computer random number generators are a different thing.
Tyler Do'Urden
12-28-05, 10:11 AM That actually messes with the "random chance" aspect of dice rolling.Not really. You're disregarding one quarter of the results, so the three quarters that remain each constitute one third of the acceptable possibilties.

You've got a 25% chance of rolling a 1, a 25% chance of rolling a 2, a 25% chance of rolling a 3, and a 25% chance of trying again. Since you never accept a result of 4, the other three possibilities fill in the gap.
Milburn
12-28-05, 11:31 AM He is probably right, the Monty Haul doors theory doesn't look like it works until someone explains it to you. Thrice. With diagrams.

I know, i tried to get it across to a high aptitude maths GCSE class, i was the only one who got it and the teacher didn't want to try.
Elendur
12-28-05, 11:55 AM The Monty Hall Doors problem involves some knowlege of the outcome. That's certainly not the case with d4, reroll the 4's.

I have a d5, it's freaky. 2 sides are larger than the other 3, but it's distribution is even.
Milburn
12-28-05, 11:58 AM Yeah, i know you can't compare the two, i am just saying how probability does not always work the way you epxpect it.

I don't think i would want to roll a dice that had uneven sides, at all. I know I wouldn't allow one in my games.
Bob the Great
12-28-05, 12:07 PM Assuming we have a completely even D4, and we re-roll 4s, how is there not an even chance to get a 1, 2 or 3. Really, i'd like you to explain it. The way I see it, you've got a 1/4 chance of any given number. Thus one in four rolls is a one, one in four is a two, one in four is a three, and one in four is disregarded completely. That would seem to give us an even chance. Please explain.
Planar Wanderer
12-28-05, 02:04 PM As other people have mentioned there is a problem called the monty Hall problem (http://en.wikipedia.org/wiki/Monty_hall_problem) which is similar (although I'm fairly sure it doesn't apply here) that is indeed a pretty wierd piece of probability which has the unfortunate downside of being absolutely correct despite this ;) . Like I say I don't think it applies here but its similar enough that it would be an easy mistake to make
Koga:The ninja trick
12-28-05, 02:20 PM That Monty Hall crap looks like the insane drivvle of someone in an asylum.

#1: The game-show host DOESN'T pick a door, what? Did the guy who write this think there's some massive goat conspiracy?

If you are given a chance to win a car from one of three doors, and the gameshow host allows to switch doors before opening the one you originaly picked, it does not increase your chances at all. The gameshow host is simply giving you the option of remaking your descsion, therfore it's still a 1/3 chance.

The same applies to the argument about a d4 re-roll 4s versus a d6 halved. It's like comparing Jelly brands. You may like the look and taste of one over the other, but in the end, it's still jelly.

The Koga swears, he's going to one day throw the words "logic" and "scientificaly proven" into a long sentence pattern and get people to eat mud.. :rolleyes:
TheLastOfTheFallen
12-28-05, 02:40 PM The Monty Hall problem has two things that need to be cleared up.

1) It actually does work. Sure it goes against your instincts, but it does work.

2) It has absolutely no bearing here. The Monty Hall problem has to do with knowing the outcome, and you cannot know the outcome of the d4 roll. Furthermore, the Monty Hall problem hinges on you being able to pick your result - to say "I want door number 3, not door 2." You cannot say that with a d4 roll - you cannot influence the result of a d4 without changing the dice weight or whatnot.

Edit: here is something that supports this statement

From Wikipidia

"The problem would be different if there were no initial choice"

A d4 roll is entirely random. There is no initial choice. The probability of getting 1, 2, or 3 on a d4 is not changed if 4's are re-rolled. If someone has a proof that the odds are different, I'd love to see it.
Solaris
12-28-05, 03:10 PM The probability of getting 1, 2, or 3 on a d4 is not changed if 4's are re-rolled. If someone has a proof that the odds are different, I'd love to see it.

I'll save everyone the trouble; here's a proof that the odds are identical.

We want to find the probability of obtaining any given result by rolling a d4 and re-rolling 4s.

Obviously we can never obtain a final roll of 4. For any other result, 1, 2, or 3, the probability of getting it on the first roll is 1/4. To this, we must add the probability of getting it in two rolls: 1/4 for rolling a 4, multiplied by 1/4 for then rolling the result, is 1/4^2. Likewise, we must add the probability of getting it in three rolls, 1/4^3; four rolls, 1/4^4; five rolls, 1/4^5; and so on, forever (we could roll arbitrarily many 4s before finally rolling the particular result).

The total probability of rolling it is therefore given by the infinite sum 1/4 + 1/4^2 + 1/4^3 + .... This is an instance of a "geometric series", which has the general form a + ar + ar^2 + .... In this case, a = 1/4 and r = 1/4. When the absolute value of r is less than 1, as it is here, a geometric series has a finite sum which is equal to a / (1 - r). Substituting 1/4 for a and r gives a / (1 - r) = (1/4) / (1 - 1/4) = (1/4) / (3/4) = 1/3.

Each of the outcomes 1, 2, and 3 therefore occurs with probability 1/3, as expected.
crazy_cat
12-28-05, 03:11 PM What is this? I know it's a die, but cannot locate an image.

*Orcmonk89*

Blimey!! If I'd known my question was going to kick start such a debate on probability and statistcs I would have asked over at www.hardmathsproblemsandlogicforstatisti ciansorpeoplewithtoomuchtimetospare.org. Anyways thanks :)

*goes away to play D&D using a D6 (or D4) to roll for a D3*

*/Orcmonk89*

Note: The link is made up - it doesn't work (unless there is such a website, in which case it might work but its not this posters fault, honest :) )

:D
rbrt_Spade
12-28-05, 03:52 PM That d4 idea is neat. But to the op, what make you think of making this thread, I thought everyone would figure out the d3 and d2 from math and gaming groups :D . I use d6 and div 2.
rhrobinson1
12-28-05, 04:21 PM Calling Marilyn Vos Savant!
Marylin_Vos_Savant
12-28-05, 04:45 PM You rang?
Milburn
12-28-05, 05:11 PM Whooooopppsss....... I really did not want to compare Monty Hall with the D4 thing, I was just saying that Maths does not always work how you expect it!
Dig
12-28-05, 05:29 PM if you wanna see a picture of a d3 click here

http://www.gamestation.net/products.asp?dept=1048
Archangel_James
12-28-05, 05:49 PM I have the d's 3, 5, and 7, and although d5 and d7 are influenced by your hands--they'd work best in a dice cup--the d3 is a perfect random number generator (it's not technically a die).
Dave Stebbins
12-28-05, 06:47 PM I have a couple six-sided d3s and love them. As others have noted, they are just a regular six-sided die numbered from one to three twice. They roll well, and I get a charge out of it when I get the chance to use them.

I also like to use eight-sided d4s because they roll in a much more pleasing manner than the standard tetrahedron d4s. I still need to get one more so I can do high-level magic missiles with only eight-sided d4s. Yeah, just need one more die.

I'm not hooked or nothin', I can quit whenever I want to. :D

-Dave
Seeker95
12-28-05, 11:03 PM Dig, it is occasionally a good idea to read a thread before posting to it. Your link is a duplicate.
Orcmonk89
12-29-05, 05:39 AM That d4 idea is neat. But to the op, what make you think of making this thread, I thought everyone would figure out the d3 and d2 from math and gaming groups :D . I use d6 and div 2.

Well, I wanted to clarify if there was not already a D3 out there, and as the PHB listed it, I assumed there must be. Our gaming group has not met yet, so I couldn't ask them.
Aeolus91
12-29-05, 10:04 AM if you wanna see a picture of a d3 click here

http://www.gamestation.net/products.asp?dept=1048

My brain just imploded.
Secrets Untold
12-29-05, 10:20 AM We all know they exist, but how on earth do you roll the stupid dice? It's like a d2, you could make one, but a polygonal d3 or d2 is phisically impossible. Every thing needs a base, and 3 dimensions. 4 sides.
Seeker95
12-29-05, 11:36 AM It's like a d2, you could make one, but a polygonal d3 or d2 is phisically impossible.Who said it has to by polygonal?
It just has to have an equal chance of landing on any given number.
For a two-sider, use a coin, poker chip, washer or even compact disk.
The Sinister Rogue
12-29-05, 11:41 AM Or a pizza.
Rolof
12-29-05, 11:47 AM If someone already said this, my bad. But I think I do see
where probability could be effected by the d4 and reroll 4's
approach. Because any given die, over the long run, results
in an average roll equal to half it's largest value +.5, a d4
will average 2.5-I think what the person who first brought
it up was trying to say was that probability will even out in
the long run, compensating rolls of 4 with rolls of 1. (4+1=
5, divide by 2 = 2.5, the "perfect average" d4 roll)

So, if you rolled a d4 to get d3 results, and got 1-3 on the
first try, there would be no effect on probability. If you had
to try over because of a 4 though, I think there would be a
slight bend to the curve, if you counted your results over a
lengthy period of time.

As for d2's, you can use the d6 as already mentioned, but I
have also sometimes used a flipped coin. Heads=2, tails=1.
kayelbe
12-29-05, 11:52 AM If someone already said this, my bad. But I think I do see
where probability could be effected by the d4 and reroll 4's
approach. Because any given die, over the long run, results
in an average roll equal to half it's largest value +.5, a d4
will average 2.5-I think what the person who first brought
it up was trying to say was that probability will even out in
the long run, compensating rolls of 4 with rolls of 1. (4+1=
5, divide by 2 = 2.5, the "perfect average" d4 roll)

So, if you rolled a d4 to get d3 results, and got 1-3 on the
first try, there would be no effect on probability. If you had
to try over because of a 4 though, I think there would be a
slight bend to the curve, if you counted your results over a
lengthy period of time.

As for d2's, you can use the d6 as already mentioned, but I
have also sometimes used a flipped coin. Heads=2, tails=1.

Thanks for your reply. That is the "theory" I was referring to, but couldn't quite remember.

I originally said it wasn't perfectly random. I didn't say the "re-roll 4's" wouldn't suffice.
Seeker95
12-29-05, 02:00 PM That is the "theory" I was referring toThat "theory" is the law of averages, and has nothing to do with individual rolls.
Solaris
12-29-05, 03:42 PM And its alleged effect on this situation was completely refuted by my earlier post.

To put it more simply, though, the results are not affected because the outcome of a roll is independent of the outcomes of previous rolls. If you toss a fair coin 99 times and get heads every time, then what is the probability of getting heads the next time? Still 50%. It's the same with the die: it doesn't care if you rolled a 4 last time and are rolling again; the odds of getting 1, 2, 3, and 4 are still each 25% on the next roll.

The simple math in my previous post showed that this results in a final probability of 1/3 for outcomes of 1, 2, and 3, no matter how many times you have to re-roll the d4. In short, there is no skew. If someone wants to claim there is, then they'd better get busy trying to find problems with the mathematical argument.
Waterhammer
12-29-05, 09:12 PM I can make my d6 roll as a d3. I can make my d10 roll as a d5,and I can make my d6 roll as a d2.
Seeker95
12-29-05, 10:02 PM Yeah? Well I can make my d10 roll as a d100, d50, d25, d20, d10, d5, d4, and d2. :P
Ruhl-Than Sage
12-29-05, 11:52 PM That Monty Hall crap looks like the insane drivvle of someone in an asylum.

#1: The game-show host DOESN'T pick a door, what? Did the guy who write this think there's some massive goat conspiracy?

If you are given a chance to win a car from one of three doors, and the gameshow host allows to switch doors before opening the one you originaly picked, it does not increase your chances at all. The gameshow host is simply giving you the option of remaking your descsion, therfore it's still a 1/3 chance.

The same applies to the argument about a d4 re-roll 4s versus a d6 halved. It's like comparing Jelly brands. You may like the look and taste of one over the other, but in the end, it's still jelly.

The Koga swears, he's going to one day throw the words "logic" and "scientificaly proven" into a long sentence pattern and get people to eat mud..

Here is a quick and simple explaination for why the Monty Hall Problem works the way it does.

There are three doors, only one of them has the prize behind it. So when you make you intial selection you have a 1/3 chance of making the right choice.

Conversely (and more importantly) you have a 2/3 chance of making the wrong choice.

If you've made the right choice (1/3 chance) than you not changing doors would mean you win.

If you've made the wrong choice initially (a 2/3 chance), Monty Hall by revealing one of the wrong choices eliminates the only other wrong choice, meaning if you change doors you win.

So because you initially had a 2/3 chance of making the wrong choice, by changing your answer you have a 2/3 chance of winning.
Ekranoplan
12-30-05, 12:49 AM I can make my D6, rolling as a D3, roll as a D20.
Gilean
12-30-05, 07:37 AM And its alleged effect on this situation was completely refuted by my earlier post.People just ignore it. No asking explanations, no nothing. So simple, yet so hard to understnad for some reason...
Milburn
12-30-05, 07:55 AM Well, flipping a coin is DEFINATELY not random. Well, i dunno about your american coins, but all our english ones are biased towards landing tails up.

A poker chip should be even though.
Orcmonk89
12-30-05, 09:20 AM Indeed they are Milburn.
Seeker95
12-30-05, 09:32 AM Well, flipping a coin is DEFINATELY not random.Notice that the reference was to a fair coin.
Flipping a coin is not random.
Flipping a fair coin is.

Post/Author/DateTime	Post
Orcmonk89 12-28-05, 06:58 AM	What is this? I know it's a die, but cannot locate an image.
Nurgan_the_drunked 12-28-05, 07:07 AM	It's shocking, they never sell d3s in dice bags or offer them in games stores anymore. We always got them in my day. Just because they're against the rules of natural reality to have it! Pathetic excuse, I tells ya ;) Seriously, there is no d3 die (or if there is, it's another die with different numbers on). To get a result for a d3, roll a d6, and halve it, rounding up so: 1-2: 1 3-4: 2 5-6: 3 Hope that helps
Milburn 12-28-05, 07:40 AM	OR, another way that is sometimes used is to roll a d6 like this- 1/4=1 2/5=2 3/6=3 I have heard of a d6 with numbers from 1-3 on, but i dunno about that. There is not a shape that makes a fair d3, without having different shaped faces, trust me.
Joe Khol 12-28-05, 07:50 AM	The two ways I've see proper D3s produced is a D6 numbered 1 to 3 twice and a prism with rounded ends. The latter 'rolls' even worse than a D4 though and never really looked right to me.
Planar Wanderer 12-28-05, 08:04 AM	Theres also 1d4 re-roll on a 4, that works as well.
Orcmonk89 12-28-05, 08:14 AM	Cheers, I thought it was physically impossible, but I saw 1d3 written and thought 'I don't have this, and I can't see how it'd work', so I asked. Thanks again. Chris
CzarGarrett 12-28-05, 09:10 AM	Should they ever form an actual d3, the Universe will implode on itself.
babysamurai 12-28-05, 09:20 AM	Should they ever form an actual d3, the Universe will implode on itself. Yeah, I think Stephen Hawkings has a theory on the d3.
kayelbe 12-28-05, 09:27 AM	Theres also 1d4 re-roll on a 4, that works as well. That actually messes with the "random chance" aspect of dice rolling. Someone explained it to me once, but I forget the details. You don't have a 1 in 3 chance of rolling a 1, 2, or 3. The 1-2=1, 3-4=2, 5-6=3 method is the most "random".
Seeker95 12-28-05, 09:29 AM	They have 'em. They're just not "true polygons". d3 (http://www.advancinghordes.com/images/gamescience/gms_43t3k.gif) And another (http://shop.rpg.net/images/uploaded/d3.jpg)
Majorafire77 12-28-05, 09:32 AM	That actually messes with the "random chance" aspect of dice rolling. Someone explained it to me once, but I forget the details. You don't have a 1 in 3 chance of rolling a 1, 2, or 3. The 1-2=1, 3-4=2, 5-6=3 method is the most "random". That makes no sense. Why would you be more likely to get a 1 on a die roll than a 2 or 3 (or vice versa)? There's no reason why the other method would be more random; the only advantage is that it might take less time (rolling 1 6-sided die once as opposed to 1 4-sided one multiple times).
Nurgan_the_drunked 12-28-05, 09:34 AM	That actually messes with the "random chance" aspect of dice rolling. Someone explained it to me once, but I forget the details. You don't have a 1 in 3 chance of rolling a 1, 2, or 3. The 1-2=1, 3-4=2, 5-6=3 method is the most "random". It's still 1 in 3, if you re-roll, no more or less "random" than halving a d6. The chance gets skewed if you say a 4 is a set thing (if you get a 4 treat it as a 2, for example) but the chance approaches 1/3, as you roll the dice more. As there are 3 equally likely events it can happen as. You just run the risk that you'll be rolling 4's 'til the cows come home. D6's are better, 'cos you just roll them once, and they do actually roll properly. But d4s work
babysamurai 12-28-05, 09:35 AM	They have 'em. They're just not "true polygons". d3 (http://www.advancinghordes.com/images/gamescience/gms_43t3k.gif) And another (http://shop.rpg.net/images/uploaded/d3.jpg) Interesting...they kind of look like distorted jellybeans.
MrBrown 12-28-05, 10:04 AM	That actually messes with the "random chance" aspect of dice rolling. Someone explained it to me once, but I forget the details. You don't have a 1 in 3 chance of rolling a 1, 2, or 3. I'd like to see the proof of that. :D Really, if the rolls are independent from each other, then a d4, 4s rerolled, will do it as well. And, if you roll your dice properly, they should be. Computer random number generators are a different thing.
Tyler Do'Urden 12-28-05, 10:11 AM	That actually messes with the "random chance" aspect of dice rolling.Not really. You're disregarding one quarter of the results, so the three quarters that remain each constitute one third of the acceptable possibilties. You've got a 25% chance of rolling a 1, a 25% chance of rolling a 2, a 25% chance of rolling a 3, and a 25% chance of trying again. Since you never accept a result of 4, the other three possibilities fill in the gap.
Milburn 12-28-05, 11:31 AM	He is probably right, the Monty Haul doors theory doesn't look like it works until someone explains it to you. Thrice. With diagrams. I know, i tried to get it across to a high aptitude maths GCSE class, i was the only one who got it and the teacher didn't want to try.
Elendur 12-28-05, 11:55 AM	The Monty Hall Doors problem involves some knowlege of the outcome. That's certainly not the case with d4, reroll the 4's. I have a d5, it's freaky. 2 sides are larger than the other 3, but it's distribution is even.
Milburn 12-28-05, 11:58 AM	Yeah, i know you can't compare the two, i am just saying how probability does not always work the way you epxpect it. I don't think i would want to roll a dice that had uneven sides, at all. I know I wouldn't allow one in my games.
Bob the Great 12-28-05, 12:07 PM	Assuming we have a completely even D4, and we re-roll 4s, how is there not an even chance to get a 1, 2 or 3. Really, i'd like you to explain it. The way I see it, you've got a 1/4 chance of any given number. Thus one in four rolls is a one, one in four is a two, one in four is a three, and one in four is disregarded completely. That would seem to give us an even chance. Please explain.
Planar Wanderer 12-28-05, 02:04 PM	As other people have mentioned there is a problem called the monty Hall problem (http://en.wikipedia.org/wiki/Monty_hall_problem) which is similar (although I'm fairly sure it doesn't apply here) that is indeed a pretty wierd piece of probability which has the unfortunate downside of being absolutely correct despite this ;) . Like I say I don't think it applies here but its similar enough that it would be an easy mistake to make
Koga:The ninja trick 12-28-05, 02:20 PM	That Monty Hall crap looks like the insane drivvle of someone in an asylum. #1: The game-show host DOESN'T pick a door, what? Did the guy who write this think there's some massive goat conspiracy? If you are given a chance to win a car from one of three doors, and the gameshow host allows to switch doors before opening the one you originaly picked, it does not increase your chances at all. The gameshow host is simply giving you the option of remaking your descsion, therfore it's still a 1/3 chance. The same applies to the argument about a d4 re-roll 4s versus a d6 halved. It's like comparing Jelly brands. You may like the look and taste of one over the other, but in the end, it's still jelly. The Koga swears, he's going to one day throw the words "logic" and "scientificaly proven" into a long sentence pattern and get people to eat mud.. :rolleyes:
TheLastOfTheFallen 12-28-05, 02:40 PM	The Monty Hall problem has two things that need to be cleared up. 1) It actually does work. Sure it goes against your instincts, but it does work. 2) It has absolutely no bearing here. The Monty Hall problem has to do with knowing the outcome, and you cannot know the outcome of the d4 roll. Furthermore, the Monty Hall problem hinges on you being able to pick your result - to say "I want door number 3, not door 2." You cannot say that with a d4 roll - you cannot influence the result of a d4 without changing the dice weight or whatnot. Edit: here is something that supports this statement From Wikipidia "The problem would be different if there were no initial choice" A d4 roll is entirely random. There is no initial choice. The probability of getting 1, 2, or 3 on a d4 is not changed if 4's are re-rolled. If someone has a proof that the odds are different, I'd love to see it.
Solaris 12-28-05, 03:10 PM	The probability of getting 1, 2, or 3 on a d4 is not changed if 4's are re-rolled. If someone has a proof that the odds are different, I'd love to see it. I'll save everyone the trouble; here's a proof that the odds are identical. We want to find the probability of obtaining any given result by rolling a d4 and re-rolling 4s. Obviously we can never obtain a final roll of 4. For any other result, 1, 2, or 3, the probability of getting it on the first roll is 1/4. To this, we must add the probability of getting it in two rolls: 1/4 for rolling a 4, multiplied by 1/4 for then rolling the result, is 1/4^2. Likewise, we must add the probability of getting it in three rolls, 1/4^3; four rolls, 1/4^4; five rolls, 1/4^5; and so on, forever (we could roll arbitrarily many 4s before finally rolling the particular result). The total probability of rolling it is therefore given by the infinite sum 1/4 + 1/4^2 + 1/4^3 + .... This is an instance of a "geometric series", which has the general form a + ar + ar^2 + .... In this case, a = 1/4 and r = 1/4. When the absolute value of r is less than 1, as it is here, a geometric series has a finite sum which is equal to a / (1 - r). Substituting 1/4 for a and r gives a / (1 - r) = (1/4) / (1 - 1/4) = (1/4) / (3/4) = 1/3. Each of the outcomes 1, 2, and 3 therefore occurs with probability 1/3, as expected.
crazy_cat 12-28-05, 03:11 PM	What is this? I know it's a die, but cannot locate an image. Orcmonk89 Blimey!! If I'd known my question was going to kick start such a debate on probability and statistcs I would have asked over at www.hardmathsproblemsandlogicforstatisti ciansorpeoplewithtoomuchtimetospare.org. Anyways thanks :) goes away to play D&D using a D6 (or D4) to roll for a D3 /Orcmonk89 Note: The link is made up - it doesn't work (unless there is such a website, in which case it might work but its not this posters fault, honest :) ) :D
rbrt_Spade 12-28-05, 03:52 PM	That d4 idea is neat. But to the op, what make you think of making this thread, I thought everyone would figure out the d3 and d2 from math and gaming groups :D . I use d6 and div 2.
rhrobinson1 12-28-05, 04:21 PM	Calling Marilyn Vos Savant!
Marylin_Vos_Savant 12-28-05, 04:45 PM	You rang?
Milburn 12-28-05, 05:11 PM	Whooooopppsss....... I really did not want to compare Monty Hall with the D4 thing, I was just saying that Maths does not always work how you expect it!
Dig 12-28-05, 05:29 PM	if you wanna see a picture of a d3 click here http://www.gamestation.net/products.asp?dept=1048
Archangel_James 12-28-05, 05:49 PM	I have the d's 3, 5, and 7, and although d5 and d7 are influenced by your hands--they'd work best in a dice cup--the d3 is a perfect random number generator (it's not technically a die).
Dave Stebbins 12-28-05, 06:47 PM	I have a couple six-sided d3s and love them. As others have noted, they are just a regular six-sided die numbered from one to three twice. They roll well, and I get a charge out of it when I get the chance to use them. I also like to use eight-sided d4s because they roll in a much more pleasing manner than the standard tetrahedron d4s. I still need to get one more so I can do high-level magic missiles with only eight-sided d4s. Yeah, just need one more die. I'm not hooked or nothin', I can quit whenever I want to. :D -Dave
Seeker95 12-28-05, 11:03 PM	Dig, it is occasionally a good idea to read a thread before posting to it. Your link is a duplicate.
Orcmonk89 12-29-05, 05:39 AM	That d4 idea is neat. But to the op, what make you think of making this thread, I thought everyone would figure out the d3 and d2 from math and gaming groups :D . I use d6 and div 2. Well, I wanted to clarify if there was not already a D3 out there, and as the PHB listed it, I assumed there must be. Our gaming group has not met yet, so I couldn't ask them.
Aeolus91 12-29-05, 10:04 AM	if you wanna see a picture of a d3 click here http://www.gamestation.net/products.asp?dept=1048 My brain just imploded.
Secrets Untold 12-29-05, 10:20 AM	We all know they exist, but how on earth do you roll the stupid dice? It's like a d2, you could make one, but a polygonal d3 or d2 is phisically impossible. Every thing needs a base, and 3 dimensions. 4 sides.
Seeker95 12-29-05, 11:36 AM	It's like a d2, you could make one, but a polygonal d3 or d2 is phisically impossible.Who said it has to by polygonal? It just has to have an equal chance of landing on any given number. For a two-sider, use a coin, poker chip, washer or even compact disk.
The Sinister Rogue 12-29-05, 11:41 AM	Or a pizza.
Rolof 12-29-05, 11:47 AM	If someone already said this, my bad. But I think I do see where probability could be effected by the d4 and reroll 4's approach. Because any given die, over the long run, results in an average roll equal to half it's largest value +.5, a d4 will average 2.5-I think what the person who first brought it up was trying to say was that probability will even out in the long run, compensating rolls of 4 with rolls of 1. (4+1= 5, divide by 2 = 2.5, the "perfect average" d4 roll) So, if you rolled a d4 to get d3 results, and got 1-3 on the first try, there would be no effect on probability. If you had to try over because of a 4 though, I think there would be a slight bend to the curve, if you counted your results over a lengthy period of time. As for d2's, you can use the d6 as already mentioned, but I have also sometimes used a flipped coin. Heads=2, tails=1.
kayelbe 12-29-05, 11:52 AM	If someone already said this, my bad. But I think I do see where probability could be effected by the d4 and reroll 4's approach. Because any given die, over the long run, results in an average roll equal to half it's largest value +.5, a d4 will average 2.5-I think what the person who first brought it up was trying to say was that probability will even out in the long run, compensating rolls of 4 with rolls of 1. (4+1= 5, divide by 2 = 2.5, the "perfect average" d4 roll) So, if you rolled a d4 to get d3 results, and got 1-3 on the first try, there would be no effect on probability. If you had to try over because of a 4 though, I think there would be a slight bend to the curve, if you counted your results over a lengthy period of time. As for d2's, you can use the d6 as already mentioned, but I have also sometimes used a flipped coin. Heads=2, tails=1. Thanks for your reply. That is the "theory" I was referring to, but couldn't quite remember. I originally said it wasn't perfectly random. I didn't say the "re-roll 4's" wouldn't suffice.
Seeker95 12-29-05, 02:00 PM	That is the "theory" I was referring toThat "theory" is the law of averages, and has nothing to do with individual rolls.
Solaris 12-29-05, 03:42 PM	And its alleged effect on this situation was completely refuted by my earlier post. To put it more simply, though, the results are not affected because the outcome of a roll is independent of the outcomes of previous rolls. If you toss a fair coin 99 times and get heads every time, then what is the probability of getting heads the next time? Still 50%. It's the same with the die: it doesn't care if you rolled a 4 last time and are rolling again; the odds of getting 1, 2, 3, and 4 are still each 25% on the next roll. The simple math in my previous post showed that this results in a final probability of 1/3 for outcomes of 1, 2, and 3, no matter how many times you have to re-roll the d4. In short, there is no skew. If someone wants to claim there is, then they'd better get busy trying to find problems with the mathematical argument.
Waterhammer 12-29-05, 09:12 PM	I can make my d6 roll as a d3. I can make my d10 roll as a d5,and I can make my d6 roll as a d2.
Seeker95 12-29-05, 10:02 PM	Yeah? Well I can make my d10 roll as a d100, d50, d25, d20, d10, d5, d4, and d2. :P
Ruhl-Than Sage 12-29-05, 11:52 PM	That Monty Hall crap looks like the insane drivvle of someone in an asylum. #1: The game-show host DOESN'T pick a door, what? Did the guy who write this think there's some massive goat conspiracy? If you are given a chance to win a car from one of three doors, and the gameshow host allows to switch doors before opening the one you originaly picked, it does not increase your chances at all. The gameshow host is simply giving you the option of remaking your descsion, therfore it's still a 1/3 chance. The same applies to the argument about a d4 re-roll 4s versus a d6 halved. It's like comparing Jelly brands. You may like the look and taste of one over the other, but in the end, it's still jelly. The Koga swears, he's going to one day throw the words "logic" and "scientificaly proven" into a long sentence pattern and get people to eat mud.. Here is a quick and simple explaination for why the Monty Hall Problem works the way it does. There are three doors, only one of them has the prize behind it. So when you make you intial selection you have a 1/3 chance of making the right choice. Conversely (and more importantly) you have a 2/3 chance of making the wrong choice. If you've made the right choice (1/3 chance) than you not changing doors would mean you win. If you've made the wrong choice initially (a 2/3 chance), Monty Hall by revealing one of the wrong choices eliminates the only other wrong choice, meaning if you change doors you win. So because you initially had a 2/3 chance of making the wrong choice, by changing your answer you have a 2/3 chance of winning.
Ekranoplan 12-30-05, 12:49 AM	I can make my D6, rolling as a D3, roll as a D20.
Gilean 12-30-05, 07:37 AM	And its alleged effect on this situation was completely refuted by my earlier post.People just ignore it. No asking explanations, no nothing. So simple, yet so hard to understnad for some reason...
Milburn 12-30-05, 07:55 AM	Well, flipping a coin is DEFINATELY not random. Well, i dunno about your american coins, but all our english ones are biased towards landing tails up. A poker chip should be even though.
Orcmonk89 12-30-05, 09:20 AM	Indeed they are Milburn.
Seeker95 12-30-05, 09:32 AM	Well, flipping a coin is DEFINATELY not random.Notice that the reference was to a fair coin. Flipping a coin is not random. Flipping a fair coin is.