I always wondered . . .
- Thread starter parmasson
- Start date
Fritz_Brown
Super Moderator
The main reason might be the lack of a bell curve. With 2d# you get a nice sweet spot at average, and decreasing probabilities out to the extremes. With percentile dice, you get everything....
Rolemaster actually used its charts to enact the bell curve. You would notice that there would be a huge range of numbers in the middle with one average result. This would be followed by ranges going down to the pits, with very few individual numbers (for results). Going up, the groups would get smaller, until you usually had a few at the very top that were singles. Yeah, you had a bell curve, but you had to jerry-rig a table to get it. (BTW, I love RM. It is just very chart intensive - at least in the original editions, I don't know about this newfangled stuff they have out.)
You could actually tell, though, what your odds were by looking at the chart. If you had to make a 70 to break out of the mediocre range, then you knew it would a (70-mods)% chance of making it. Assuming standard dice, of course....
Rolemaster actually used its charts to enact the bell curve. You would notice that there would be a huge range of numbers in the middle with one average result. This would be followed by ranges going down to the pits, with very few individual numbers (for results). Going up, the groups would get smaller, until you usually had a few at the very top that were singles. Yeah, you had a bell curve, but you had to jerry-rig a table to get it. (BTW, I love RM. It is just very chart intensive - at least in the original editions, I don't know about this newfangled stuff they have out.)
You could actually tell, though, what your odds were by looking at the chart. If you had to make a 70 to break out of the mediocre range, then you knew it would a (70-mods)% chance of making it. Assuming standard dice, of course....
FASA Star Trek was nice for that. Made things very scientific, well, until you look at Mr. Spock's 105 intelligence score...
mike wightman
SOC-14 10K
2d anything doesn't give a bell curve... 
FlightCommanderSolitude
SOC-13
With 2dN, you're typically adding the results together, so you'll get a bell curve. With percentile dice you're just randomly generating a number between 1 and 100, so you have the same chance of getting 01 as 00 or any other number.
[That's right, isn't it? ]
[That's right, isn't it?
]Fritz_Brown
Super Moderator
Well, you get more of a pyramid than a bell curve, Sig.
3d# gets closer, though its no good reason to go for Handsful o' Dice.
mike wightman
SOC-14 10K
Sorry, I had to post that a bit quickly this morning
2dN gives a pyramid centred on the the average result, which isn't a true bell curve.
To get a bell curve, a minimum of 3dN are needed, although the more dice you roll then the more of a bell it becomes
2dN gives a pyramid centred on the the average result, which isn't a true bell curve.
To get a bell curve, a minimum of 3dN are needed, although the more dice you roll then the more of a bell it becomes
Sigg, you seem to have thought on this some. Does the increased dice lead to a normal (aka gaussian) distribution? I'd always wondered that as you increased the number of dice the 4th moment (kurtosis) of the distribution no longer bears the correct relation to the 2nd moment (variance) as you would find in the normal distribution. That is as you increase the number of dice the distribution becomes too "thin."Originally posted by Sigg Oddra:
2dN gives a pyramid centred on the the average result, which isn't a true bell curve.
To get a bell curve, a minimum of 3dN are needed, although the more dice you roll then the more of a bell it becomes
mike wightman
SOC-14 10K
3dN is almost a gaussian distribution, 4dN gives a much closer match to it, but as you increase the number of dice the "bell" flattens and starts to look like a cymbal rather than a bell.
Pacesetter did a very good percentile dice system. Their problem was they had only 2 interesting games(Chill & Space Ace?) & they published too many books too fast. I converted my D&D game to Pacesetter rules & found it quite good. My players really liked the system. It was easier than D&D's hodgepodge of rules, played fast & made combat very lethal.
Why do so few systems use percentile dice? At first blush it would seem to be an easy way to generate numbers with plenty of range to choose from. [/QB][/QUOTE]
Why do so few systems use percentile dice? At first blush it would seem to be an easy way to generate numbers with plenty of range to choose from. [/QB][/QUOTE]
I can say why I stopped trying to use percentile dice (after about a 3 year experiment): keeping track of the tens digit. Yes it sounds silly, but if you include beer intake in a game and as a GM need to roll lots of dice for groups of NPCs it gets a little dicey keeping it straight. Once I discovered I personally didn't need 1% resolution and could get a 1% chance on on 2D10 I went to 2D10. Converting from 2D6 was pretty easy in both absolute target numbers and modifiers.
PS: Thanks Sigg. It flattens with increased dice, interesting, I would have thought the opposite but of course never went to the trouble of normalizing the distributions.
PS: Thanks Sigg. It flattens with increased dice, interesting, I would have thought the opposite but of course never went to the trouble of normalizing the distributions.
fredofmars
SOC-12
John Scott Clegg and I worked out a conversion between Traveller skills and percentile dice (Runequest, etc.). It is at:
http://www.geocities.com/fredofmars/skill.htm
http://www.geocities.com/fredofmars/skill.htm
Fritz_Brown
Super Moderator
No, Kurega, but I have a d30. The numbers are hand-painted, too. (Well, hand-filled-in with paint, anyway....)
DaveShayne
SOC-12
I've got one. Aside from any issues regarding randomization I doubt I'd use it in a game situation. Too hard to read if you aren't sure of absolute honesty.
Similar threads
