PDA

View Full Version : Original System Scaling dice system



RFLS
2017-06-13, 01:00 PM
I'm writing a custom system for my setting and am trying to settle on a dice system. I don't have anything set in stone yet, so I'm looking for input on current ideas and suggestions for new ones. The dice system will be used in a skill only system (no base ability scores). Ideally, it would have a tight range of outcomes. Here's what I've been considering.

Scaling Dice:

Essentially, you get a bigger average number the better you are at a skill. Multiple dice are used to ensure a bell curve on each roll. The system would be 1d4 > 1d6 > 2d4 > 2d6 > 3d4 > 2d8 > 4d4 > 2d10 > 5d4 > 2d12 > 6d4.

Pros: Scaling is tightly controlled, and you're not keeping track of numerical modifiers.
Cons: It covers a very wide range of outcomes, and it requires a lot of dice.

Mordar
2017-06-13, 02:21 PM
I'm writing a custom system for my setting and am trying to settle on a dice system. I don't have anything set in stone yet, so I'm looking for input on current ideas and suggestions for new ones. The dice system will be used in a skill only system (no base ability scores). Ideally, it would have a tight range of outcomes. Here's what I've been considering.

Scaling Dice:

Essentially, you get a bigger average number the better you are at a skill. Multiple dice are used to ensure a bell curve on each roll. The system would be 1d4 > 1d6 > 2d4 > 2d6 > 3d4 > 2d8 > 4d4 > 2d10 > 5d4 > 2d12 > 6d4.

Pros: Scaling is tightly controlled, and you're not keeping track of numerical modifiers.
Cons: It covers a very wide range of outcomes, and it requires a lot of dice.

Have a look at the EarthDawn Step system - it is based on the idea of a given Step (rank) rolling a combination of dice that generates an average that (generally) corresponds to that rank. For instance:


Step 4 is 1d6
Step 5 is 1d8
Step 8 is 2d6
Step 9 is 1d8 + 1d6

You get the idea.

The added wrinkle was that ED used exploding dice (roll the max value on a die and you get to reroll and add the result), but that isn't a necessary consideration.

NOTE: First edition used d4 and d20 in the steps, but the latter edition(s) eschewed them as too variable. Review the 3rd edition steps for a smoother implementation.

- M

Knaight
2017-06-14, 03:52 AM
Not having base ability scores doesn't really mean much (basically any die system can work without them, although there's a few roll and keep versions that really benefit from them).

The current system is a bit wonky. Some quick numbers:
Dice: 1d4 > 1d6 > 2d4 > 2d6 > 3d4 > 2d8 > 4d4 > 2d10 > 5d4 > 2d12 > 6d4.
Min: 1, 1, 2, 2, 3, 2, 4, 2, 5, 2, 6
Max: 4, 6, 8, 12, 12, 16, 16, 20, 20, 24, 24
Avg: 2.5, 3.5, 5, 7, 7.5, 9, 10, 11, 12.5, 13, 15

The minimum is particularly odd, given that it goes down a couple of times and fluctuates wildly. The maximum is actually a pretty cool behavior, with the average at least increasing consistently but in a pretty weird way. There's also heavy oscillation in how tight the distribution is. There are also some general trends, so I'm going to assume a few design conditions - if any are wrong, correct me.

A higher skill does better on average.
A higher skill has an equal or higher maximum.
A higher skill makes you more likely to succeed at any difficulty.
The change in average with one skill level is fairly small and level.
Higher skill produces a higher range of outcomes, while making average outcomes more likely.

Yours currently meets the first two, but not the last three - and if you look at Anydice graphs of these against each other (using at least) you see a lot of curve intersections that are pretty noticeable. 2d12 being much better than 6d4 at difficulties 18-24, and 2d10 being much better than 5d4 at difficulties 16-20 really stand out. 2d10 is also worse than 4d4 at difficulties 2-9.

So, options. I've been playing with a reverse dice pool recently that's working out pretty well (skills are a number, difficulties are a collection of a dice that produce either a 0 or a 1, roll under your skill), and if you're intentionally going for something esoteric there are a few options. Otherwise, the things that seem to fit what you're looking for, at least in most respects:


Conventional Dice Pool - Skills are a number of dice that you roll, which are either 0s or 1s (usually d6 or d10, with about a 1/2 to 1/3 chance of being a 1).
Large Die Dice Pool - The d6 system stands out here, but basically you have a pool of larger dice with wider individual ranges, you roll them, you add them. Generally these use either d6 or d10.
Growing Polyhedral System - 1d2, 1d4, 1d6, 1d8, 1d10, 1d10 + 1d2, 1d10 +1d4, 1d10 + 1d6, 1d10 +1d8, 2d10, 2d10 + 1d2, ....
Set Matching Semi-Additive Pool - You generally have a number of d6, where you take the best result. In addition, sets count as 1 higher for every match in the set (so 5,5, is 6, and 5,5,6 is 7).
One Roll Engine - You roll a bunch of d10. Anything not in a set is discarded, and any roll with no sets fails. For rolls that succeed higher sets represent doing better (7,7 is higher than 3,3) and wider sets represent doing things faster (4,4,4,4 is wider than 10,10). I wouldn't use it (mostly because it's the distinctive work of one designer), but I would absolutely look closely at it.


There's also some more esoteric systems out there you might like. If you're fond of keeping ranges tight, want skill to matter, and want to avoid numerical modifiers, here's my proposal:

You have exactly N skill levels (say 6).
You then roll exactly NdX always (say 6d6), and sort the dice from highest to lowest.
Your skill is the place of the dice you pick, and that is your result. This does make skills better if lower, but there's ways to flip that, and just using the terms "first, second, third..." suggests it.


The system doesn't handle modifiers easily, but there are a few ways to handle it, and it's a problem shared with your proposed system; I assume it's not particularly concerning. Opposed rolls can also take a bit, as you might have to compare down the line a bit, particularly at high levels.