The Forge Archives

A little inspired by Pyron's posting on "Consistency" perhaps, I thought up a way to eliminate modifiers in my combat system.

A totally unmodified roll would be D12 to beat target number. Very simple. But if you then add concepts like adding modifiers here and there you get a very muddled thing. Especially with both positive and negative modifiers.

Originally I thought of a system to handle disadvantaged situation by diminishing the dice by 2 for every level of disadvantage, like this:

No handicap: D12
1 point of handicap: D10
2 points of handicap: D8
3 points of handicap: D6
and so on.

However, the reverse (advantage) needed to be dealt with by reducing the target number or adding a +2 to the die roll. In the end I got those modifiers anyway.

But reading especially the recent posting by Pyron, I got inspired to do it a little differently, like this:

No handicap: 1D12
1 point of handicap: 2D12, keep the lowest
2 points of handicap: 3D12, keep the lowest
3 points of handicap: 4D12, keep the lowest

This works with advantages too:
No advantage: 1D12
1 point of advantage: 2D12 keep the highest
2 points of advantage: 3D12 keep the highest.

This doesn't quite behave the same way modifiers does, but that's not necessarily a bad thing.

Also it feels pretty natural to use it:
"Oh you aim with the bow for a round? Here, you get another D12"

or

"You only want to make a quick slash as you run past him? That's not gonna be very easy. Here roll another D12 and pick the lowest"

Somehow it feels like it's really making sense.

Now I just need to figure out how to create a nice skill system in the same vein without using the skill+stat approach.

The reason modifiers are so cumbersome to deal with is because they are always changing depending on the situation. So adding or subtracting from the dice roll, changing the individual die, the number of dice rolls, or the target number will still bring up the same response when a roll is called for: "What am I rolling?" This question is layered. In it the player can mean what polyhedron to pick up and roll, how many of said die, does he want high or low, what stat is this roll on. what is the target number for the roll, and so on.

Now some may find what you're working with here a little more intuitive than modifiers, but more intuitive does not mean it is intuitive.  I believe you'll still hear "what am I rolling?" fairly often because who wants to keep track of all of those variables?

Quote from: Pale Fire
This doesn't quite behave the same way modifiers does, but that's not necessarily a bad thing.

Well. IMO, it works exactly the same way modifiers do. Adding an advantage increases your chacne of success, adding a disadvantage decreases your chance of success. If you're interested in this mechanic, check out Rules Lawyer for Free, an RPG.Net collumn by Sergio Mascarenhaus. As far as I know, he's the first person on the net to formalize the mechanic you're using. It's what prompted me to write up the game that's up at my site as a sort of prototype. It's sort of a designer's Exploration of Setting. "That's an awesome mechanic!" I thought. "I wonder what would happen if I designed a system around it?"

Now, I'm not sure how concerned you are with the math of this mechanic. However, doing it the way you are causes a big big jump between 1 die and 2 dice of either sort. A better way (assuming that a smoother curve is better) is to roll 2d6 and add them. The next step is 3d6, add the two highest (or lowest), then 4d6, and so on. This is nice because it gives you a smooth curve, and you have decent range (2 - 12) without needing large or unusual dice. (How many people do you know with four or five d12s?)

Quote
Now I just need to figure out how to create a nice skill system in the same vein without using the skill+stat approach.

Well, skill + stat actually works pretty well with this mechanic. You just construct a dice pool form your ratings, and the GM modifies it to taste.

What I would like to see with this mechanic, though, is some way for one point of difficulty to be more or less equivalent to one more die. That is, your rating would serve equally well as a *target* number for a roll, as it does for *making* rolls. This way the GM doesn't have to roll dice for NPCs. If the PCs are getting attacked, the players roll their defense dice against the PCs' attack skill. If the PC's are attacking, they roll their attack dice against the PCs' defense skill.

This is tricky (probably impossible) with a curving mechanic like we've been talking about, but since we're on the subject of cool dice mechanics and modifiers I thought I'd mention it. :)[/quote]

Quote from: Pale Fire
This doesn't quite behave the same way modifiers does, but that's not necessarily a bad thing.

Well. IMO, it works exactly the same way modifiers do. Adding an advantage increases your chacne of success, adding a disadvantage decreases your chance of success. If you're interested in this mechanic, check out Rules Lawyer for Free, an RPG.Net collumn by Sergio Mascarenhaus. As far as I know, he's the first person on the net to formalize the mechanic you're using. It's what prompted me to write up the game that's up at my site as a sort of prototype. It's sort of a designer's Exploration of Setting. "That's an awesome mechanic!" I thought. "I wonder what would happen if I designed a system around it?"

Now, I'm not sure how concerned you are with the math of this mechanic. However, doing it the way you are causes a big big jump between 1 die and 2 dice of either sort. A better way (assuming that a smoother curve is better) is to roll 2d6 and add them. The next step is 3d6, add the two highest (or lowest), then 4d6, and so on. This is nice because it gives you a smooth curve, and you have decent range (2 - 12) without needing large or unusual dice. (How many people do you know with four or five d12s?)

Quote
Now I just need to figure out how to create a nice skill system in the same vein without using the skill+stat approach.

Well, skill + stat actually works pretty well with this mechanic. You just construct a dice pool form your ratings, and the GM modifies it to taste.

What I would like to see with this mechanic, though, is some way for one point of difficulty to be more or less equivalent to one more die. That is, your rating would serve equally well as a *target* number for a roll, as it does for *making* rolls. This way the GM doesn't have to roll dice for NPCs. If the PCs are getting attacked, the players roll their defense dice against the PCs' attack skill. If the PC's are attacking, they roll their attack dice against the PCs' defense skill.

This is tricky (probably impossible) with a curving mechanic like we've been talking about, but since we're on the subject of cool dice mechanics and modifiers I thought I'd mention it. :)

Hello,

If I'm not mistaken, the first game to use this system was The Window. The second edition of Immortal uses it as well (and as far as I can tell is a lightly modified version of The Window).

Anyone know of any others? (And no, simply using lots of different-sided dice at once doesn't count; that goes all the way back to RuneQuest.)

Best,
Ron

Ron,

I have been toying with a polyhedron dice system like this, but the first published game I saw that used it was DragonStrike, that bad, bad, bad TSR version of HeroQuest (board game) with and even worse video (that, bad as it was, was still better than the movie). It kind of put me off the idea of using the polyhedrons as a means to show levels of effectiveness or difficulty. But the really big reason is that there's dice with even numbers up to 12, then a big gap until 20 and then an even bigger gap until 30, an d30's aren't all that common. I suppose in d14's, d16's, & d18's became more common (or even available???) then I'd consider this idea again.

Is anyone reading the whole first post? Pale Fire's final mechanic (1 point of advantage - roll 2 dice, keep the highest, etc) is nearly the same mechanic as penalty dice/bonus dice from Over the Edge.

I'd look here also. [url http=http://www.indie-rpgs.com/forum/viewtopic.php?t=2509]Symetry [/url]
This is Walt's system. It's not exacly what your getting at, but pretty similar.

As was mentioned, the first die has big effect, although it's highyl dependnet on your target number also. With a 2 negative modifiers and a 11/12 chance of succes, you still have a 77% chance. (Or reversing that, with 2 positive modivers and a 1/12 change, you have a 23% chance of success)
With a high chance of success, modifiers don't change it all that quickly, but with a low chance, there is a large effect.  
It's an ok system, as long as the number of modifiers present at a time was pretty low. Also you don't have any granularity on modifiers. With additive modifiers you can have a +/- 1 or a +/- 2, +/- 3 ect. You could have double modifiers with this, but  their pretty severe.

[/url]

Quote from: Ron Edwards
Anyone know of any others? (And no, simply using lots of different-sided dice at once doesn't count; that goes all the way back to RuneQuest.)

Well, Ron, this isn't what Pale Fire was talking about, but if you're referring to using die sizes to rate abilities, then Sovereign Stone is probably the best commercial example that I'm familiar with. Fable also does this also.

If you're talking about the actual mechanic that Pale Fire presented, it doesn't really have much in common with the Window at all. :) However, I've been told that Don't Look Back uses the d6 version of the mechanic I proposed. Can't verify that, sice I don't have that game.

Jack, I don't quite see what you're getting at. Do away with modifiers and everything related to them? Might not that turn out to be a problem?

Paganini, I'm gonna check out Sergio's columns (downloading it for later reading) Thanks for that hint.

As for the maths, I was already to only use +/-2 modifiers for the D12 rolls. It turns out that adding a die doesn't differ by using modifiers all that much. It does differ, but the question is if they differ in a bad way or a good way.

Compared to the modifier version the extra die differs less than +/-9% in most cases (or equal to a +/- 1 modifier on a D12), except for in very special cases.

This means that the extra dice corresponds to a modifier between 1 and 3 on a D12. This seems to be ok.

As for the smoothness. I'm not overly concerned about it, because I only want to track significant changes. So the lack of precision is more of a feature. It simplifies the task for the GM to make estimates. Besides, who is able to pinpoint a 10% change here and there anyway? Especially considering that the intended application is mostly resisted rolls like combat.

As for the many D12s, I use dice pool damage resolution using D12s so people are gonna need them anyway :)

And the concern that modifiers building up might not work: You're not supposed to stack negative/positive modifiers. Two +1 advantages does not equal a +2 advantage. Basically in my modifier system 1+1=1.5 or something like that. Not that you're supposed to count them. There's gonna be a guide rather than fixed modifier rules. I'm calling it handicap and stuff but that's just for easy comparison to the diminishing die version. The numbers aren't gonna appear in the game.

Quote from: Pale Fire
As for the smoothness. I'm not overly concerned about it, because I only want to track significant changes. So the lack of precision is more of a feature. It simplifies the task for the GM to make estimates. Besides, who is able to pinpoint a 10% change here and there anyway? Especially considering that the intended application is mostly resisted rolls like combat.

I'm not sure you understand what I was getting at though. "Smoothness" was one of the reasons you didn't want to go with a straight polyhedral rating system, IIRC. I'm not talking about the size of the jumps from step to step, I'm talking about the *differences* between those sizes. Think of it this way:

Using a d12 as a base die, then adding additional d12s to it is similar in practice to this:

You start with a d30. When you modify, you bump to a d20. When you modify again you bump to a d12. Then you bump to d10. It works out the same way: Biiig gap between 1d12 and 1d12 pick highest (or lowest). Smaller gap between 2d12 and 3d12.

If you do it this way, there's basically a stairstep between 1d and 2d. If you do it with d6, add *two* higest (or lowest)  it's much more even.

It's the nature of a bell curve system that the differences get smaller as you go along (that is, step 0 and step 1 are further apart than step 1 and step 2). However, the stair-step thing is going to be visible in play. It puts a big breakpoint right on the 2d level. The returns from the effort of going from 1d to 2d are much much bigger than the returns from going from 2d to 3d. So, if going from 2d to 3d is as hard as going from 1d to 2d, no one will ever do it. :)

Quote from: Pale FireJack, I don't quite see what you're getting at. Do away with modifiers and everything related to them? Might not that turn out to be a problem?

That depends. For some people stocking Coke instead of Pepsi is a problem, but assuming you have a group who are like minded on the subject, it shouldn't be. It's just a matter of acceptance and a lack of desire to do any form of "modifier juggling"

One game I know of that does this is Sandman from Pacesetter. I'm not sure if this is the Pacesetter House System which would've been found in the original CHill, but it worked like this:

It was percentage dice. When attempting an action, you tried to roll 40 or less. This is called Standard Chance. If the GM judged the action to be especially difficult, then you rolled at a Reduced Chance, which is -20% So a roll of 20 or less is required.

Skills are on a star system you can have a skillof one to three stars. Stars raise your Standard Chance:

*     60%
**   70%
*** 95%

Not exactly a linear progression, but it works. Likewise, the Reduced Chance is just the above -20%

This is not for all tastes, but you don't have to spend time doing all the math to see what the target number is or any of that. It's a simple judgement call on the part of the GM but, unlike some games where the judgement call can be anything on from 1-100 or X number of "Successes" from a dice pool, it is an either-or situation. That is, either the action is either Standard or Reduced Chance with no spectrum to deal with. (also automatic or impossible as many games state and ignore thus)

Some people don't like this sort of thing, calling it "coarse granularity" or something, but let's think about this. WHat is really gained by all the granularity? When you roll the dice, it either comes up success or failure and that's that. Basically, it just slows down the handling time a little, and may cause problems later if the player is the sort to figure out that the GM's calculations were off by one crucial point.

But, like I said, not for everybody but it is workable and has been published, as it turns out. But Pacesetter also went out of business in the 80's, so what do I know?

Actually Jack, what I indended was exactly like that.

"You're lying down, well that counts as being impaired"

"One hand grabbed? Hmm... probably disadvantaged"

"Both hands grabbed but you're a martial artists using your legs as well? Only a disadvantage for that too"

Basically on the negative side you'd have something like:

Disadvantaged (2 dice pick lowest)
Impaired (3 dice pick lowest)
xxxxx (4 dice pick lowest)

and correspondingly for advantages:

Advantage (2 dice pick highest)
Superior situation (3 dice pick highest)
(I don't think more than 3 dice would ever be applicable)

The "coarse granularity argument" exactly mirrors my thinking. Who can see changes of 10% in real life anyway? No better add more randomness to cover for those things and use simple modifiers. Detail does not necessarily make things more realistic.

Paganini:

No my objections to the handicap system with diminishing dice was that I needed modifiers anyway. And lots of different dice. And it wasn't totally clear. Since I'm basing my game on D12 the smoothness isn't a problem. Just count down from D12: D12-D10-D8-D6-D4-D2. The problem was that how would I deal with other characters interacting with static defense values of disadvantaged characters? Modifiers seemed to be the inevitable (maybe this is what you're getting at, I can't "count up" from D12?) alternative which would make the diminishing die thing meaningless.

"Coarse" granularity (large increments of success probability) versus "fine" granularity (small increments of success probability) are not advantages or disadvantages in and of themselves. The important thing is matching up the degree of granularity with the amount of action that's being resolved in the roll. In general, coarse granularity works better when resolving a larger amount of action.

When it takes the cumulative effect of many rolls to accomplish something, finer granularity becomes more useful. That's why in some systems you DO care about a mere 10% or even 5% difference. For example, imagine a contest in which the first character to roll 10 total successess wins. If both contestants have an equal (say, 50%) chance to score a success on each round, then their overall chances of winning the contest are also equal. But if one of them has a 55% chance instead, that character will win the contest more than 58% of the time, while the character with the "small" disadvantage will win less than 35% of the time. (The other 7% will be ties.) If the smallest increment available was an increment of 20%, so that the contestant with the advantage gets a 60% chance per round while the other gets only a 40% chance, the stronger character has a nearly 10 to 1 advantage over the weaker (88% wins vs. 9% wins).

To take the example to extremes, consider a battle between two equally matched armies, with a small slope to the battlefield. You could say that the downhill advantage gives one side a larger chance to win the battle -- say, 60% instead of 50% -- or you could say that fighting downhill gives each blow struck a very small additional chance, something like .01%, of being effective. Over the course of the battle, the miniscule advantage applied over tens of thousands of individual attacks adds up to the 10% advantage for winning the battle. If you were resolving the battle with one roll, a relatively coarse granularity gives you what you need to represent the effect of the slope. But if you were resolving the same battle blow by blow, you'd need a much finer granularity to represent the effect of the same slope.

Anyhow, here are the probabilities for Pale Fire's d12 system


Chance of rolling over target number

Target Number  1   2   3   4   5   6   7   8   9  10  11

1d12         .92 .83 .75 .67 .58 .50 .42 .33 .25 .16 .08
2d12 highest .99 .97 .94 .89 .83 .75 .66 .56 .44 .31 .16
3d12 highest   *   * .98 .96 .93 .88 .80 .70 .58 .42 .23

1d12         .92 .83 .75 .67 .58 .50 .42 .33 .25 .16 .08
2d12 lowest  .84 .69 .56 .44 .34 .25 .17 .11 .06 .02 .01
3d12 lowest  .77 .58 .42 .30 .20 .13 .07 .04 .02  **  **

*  > .995
** < .005


There's a lot about this system that I like. At a TN of 6, it's equivalent to the "high granularity" variant of Symmetry in which only the "level dice" are used. (The distinction between "take the lowest" for adverse modifiers versus "take the highest" for favorable modifiers takes the place of Symmetry's "marked side means failure" for adverse modifiers and "marked side means success" for favorable modifiers.) And there's no reason, at middle of the road difficulties, you couldn't add on more dice in either direction. (At low TNs, there's room for more dice for disadvantages, and at high TNs, there's room for more dice for advantages.)

One feature to note is that at high TNs, negative modifiers are devastating to one's chance of success, while at low TNs, positive modifiers quickly reduce the chance of failure to practically nil. This could be seen as desirable, or not. Do you want success at difficult actions to be reasonably likely only if there are no disadvantage factors? Do you want the chance of failure at an easy (TN 1) action to be reduced more than eightfold by a single advantage factor?

The flip side is that at low TNs, negative modifiers quickly boost the chance of failure while at high TNs, positive modifiers quickly boost the chance of success. This could work fine in play as long as you're aware of this behavior when assigning TNs and advantages/impariments. Beware of deprotagonizing situations where a character has a very high skill, which should be adequate to succeed despite some circumstantial disadvantages, but even after a low TN is set the disadvantages end up producing a lot of whiffs. This occurs because at most parts of the table, the effects of advantages/disadvantages overwhelm the effects of an equal number of points of change to the TN.

If I recall correctly, on an earlier thread Pale Fire spoke favorably of systems that tend to "drive" toward guaranteed success or failure when favorable or unfavorable factors accumulate. This mechanism does just that.

- Walt

Quote from: Pale FireAs for the smoothness. I'm not overly concerned about it, because I only want to track significant changes. So the lack of precision is more of a feature. It simplifies the task for the GM to make estimates. Besides, who is able to pinpoint a 10% change here and there anyway?

The average human is able to detect changes of around 2%.  Our minds(and senses) Look at things within 2% of one another and judge them "equal."  This is why I prefer percentile resolution systems - given a base of 50%, a one-point change is a 2% difference.