Forum changes: Editing of posts has been turned off until further notice.

Main Menu

Need some feedback on a dice resolution idea

Started by Teataine, October 16, 2007, 07:44:23 PM

Previous topic - Next topic


Hello everyone, first time poster.

Here's my "problem" - I've been toying with various systems recently, partialy for the fun of it, partialy with the intention of gradualy making my own system (isn't everyone?). I started with a classical percentile "roll under" mechanic, everyone is familliar with. You have an ability with a value of 0-99 (actualy, the system initialy restricted values to 20-80 more or less), you roll two d10, get a number, and if it's lower than your ability score, you succeed.

However, after some intense pondering on the ability-skill relationship and the problem with superhuman characters I started tweaking this. I went through several other systems, considered UA's flip-floping and so on. I thought, experimented...So the problem was to separate the chance of success and the degree of success. For example, if an average human's strenght score is 30, the strenght score of a giant or dragon or mecha should be very near 100 if not over. But that's awkward because it would mean while a weak person often fails to hit someone for example, a strong one hits him all the time, regardless of his clumsiness or skill. And blah blah. I'm sure most of you are familliar with these issues.

So anyway, after some time, this is what I've come up with, mostly as an experimet. The thing is, I can't decide if I really like it or if it's really clumsy. I'd like to hear your toughts on it.

1: Instead of two dice, everyone starts with a pool of four d10. When attempting a task, you roll your whole pool and assemble a number out of it, much like with percentile systems. For example if I roll a 4, 7, 2 and 0, and my ability is 36, I can make a 27 out of the four dice and pass the test.
2: Before rolling, you check your skills. If you have a related skill, you roll your whole pool. If you have additional ranks in the related skill, you get bonus dice to your pool, to a maximum of 7. If you don't have any skills that could be used, you only roll half your pool (two dice). If you have a skill that is somewhat usefull in the situation, but not directly related, you can add any ranks in that skill, and then halve your pool, rounded up.). For example if I don't have a Sword skill, and attempt to use a sword, I only get to roll two dice. If I have an Axe (2 ranks) skill, I can attempt to use that, and if the GM agrees, I add the two ranks to get a pool of six dice, then halve it...which gives me a pool of three. Still better than only two. If I have a Sword skill 3, I can roll a pool of 7 dice.
3: Abilities aren't limited to 100 anymore, but go up to virtual infinity. However, for humans they remain relatively low, about 20-60. By rolling more than two dice you can assemble numbers that are over 100, provided your ability allows it.

The only problem here is that after you pass the treshold of 50 on an ability and have four (or more!) dice, it becomes virtualy impossible to fail. There are no failures, only poor sucesses. Which is actualy one of the issues I've been trying to avoid in the first place. However the sucess is based both on someone's skill and his ability, albeit in a different manner, and I like that. In most games you would simply add your skill and ability together, and that causes some problems, especialy in point-buy systems.

That's the gist of it. I've also developed combat rules and so on, and in a way, this system seems to be very realistic, because, let's face it, when you are standing face to face with someone and try to hit him with a sword, it's pretty hard to miss. Staying alive in combat is much more dependent on active dodging and parrying, because your opponent isn't going to just slash and stab the empty air around you (which is what happens in most games).

Paul Czege


Have you seen the rules for Vincent Baker's game Otherkind? You roll a pool of dice, and then decide how to apportion them across multiple desirabilities, like achieving your goal, and not being harmed in the attempt. You could do something similar, where some of your dice add up to a success threshold, some to a defensive "not take harm" threshold, and the remainder add up to your degree of success.

My Life with Master knows codependence.
And if you're doing anything with your Acts of Evil ashcan license, of course I'm curious and would love to hear about your plans


I haven't read Otherkind yet, I do know however Vincent's other game, Dogs in the Vineyard and I learnt a lot from it. It was a big eye-opener for me, gaming-wise.

In combat I actualy have the rule that your pool gets divided between offense, defense and movement. So yeah, I did that.

Failure isn't that big a problem really, compared to Fate for example, where you only have variying degrees of success on The Ladder. You can fail only in a conflict with someone, when they are better than you.

I'd still like to hear any comments and critiques on my "system", however.

Oh and sory about any typos and grammatical errors, English isn't my first language.


I'm not sure I understand it clearly. In your example, you have rolled 4, 7, 2 and 0 and you choose two numbers to make percentile. So you can end up with 02, 74, 40 ect...
If I understand well, it's a little complicated but as you said, it could work but it's very difficult to miss an action if you have a high skill.
My dices are trained to roll high however so games like that will tend to destroy my hard training (in Québec, we throw them for one week in the fridge if they are poor :) Gamers mythology)
Link to my Dark Cataclysmic Fantasy RPG Kissanil:


Yes, you take two of the dice and make a "percentile", just like you would in a normal percentile system, except you get to choose from several dice and how to combine them. You can come up with either of those numbers (02, 20, 74, 47, 40, 42...whatever), but you would usualy pick the highest possible combination, because higher is still better (in combat for example), even if it has to be under your ability score.


Welcome to the Forge. I'm Christian :)

Your system has some aspects I like; die pools scale nicely, and you've got "positioning," i.e., a mechanical way of determining what matters most to you by using what we call Fortune in the Middle (declare action, roll (fortune), then distribute results).

The main question seems to be this: Is the handling time of your system proportionate to the additional effects you're getting out of it?

Imagine simply using a die pool system like White Wolf's or Burning Wheel's, except that you distribute your dice after you roll. Still got a bunch of D10s, but you just roll to determine the number of successes, then dish them out to things like offense and defense. That seems like it would be a bit faster and more intuitive than what you've got. But what doesn't it have? What does your way of combining dice into percentiles do that this other system would not?

It seems like the first answer to that is: you get more granularity. Someone rolling a 27 beats a 25, even if both rolled under their skills. Is that an important part of your design ideal? Does it make enough of a difference to justify the added time required to go through all kinds of combinations in your head when you roll 8D10 and figure out three percentile results?

And so on. What is it exactly that your system brings to the table?


Hi. :) You can call me Greg.

I don't think handling time increases too much compared to the Storyteller system. If you have an ability of 40, you look for a four and a zero, if there isn't any you look for a three and a highest possible second number, or a two and so on. I've tried it, and it isn't slower than dishing out a number of sucesses in my opinion. Sure, as I said, this was just an idea I was toying with and I will have to playtest it first to see if it's worth the trouble. Oh, and you would never have to roll more than seven dice. Most rolls are done only in combat and because you have to spread your dice between defense and offense you usualy don't get to roll more than five. In fact, three or four dice per roll seems to be the standard.

Granularity was definitely an issue, mostly on my player's wishes. They like a broad range of possible results. They'd much rather roll on a 1-100 scale than a 1-10 scale. So I was trying to apease them as well there. Since combat relies on oposing checks - attack and parry or dodge and so on, this finesse is important. Me and my players seem to feel it's more "justified" when someone parries your 25 attack with a 27 defense, than when he cancels out your 5 sucesses with 6.

The main reason why I switched to the "percentile pool" idea was because I wanted to get rid of the top cap of 100% probabiliy and make a broader range of variable sucesses instead . But I totaly agree with your arguments and the handling time will definitely be in the foreground once we move to playtesting, if it will clog up gameplay, I'll throw it out of the window. It doesn't really bring anything terribly fresh on inovative to the table, but the system alltogether is somewhat a combination of ideas. Combat works a bit like The Riddle of Steel, but it's much faster and streamlined since you don't have to roll on a bazillion tables each time you hit and so on. I'm trying to figure something out that will suit me as a GM so I can stick to it for most of my generic games. For other ocassions, other games and systems will do better.


I must have missed something. How do you determine the degree of success after selecting you pair of successful dice?

Btw, at 60 and 3 or 4 dice, you have around 90% chances of success, so you might want to reevaluate the skill limits.


Even with 2 dice and a skill level of 60 you have a 77% chance of making it. And yeah, with three dice you're looking at a 90.8% success rate. I won't even start calculating 4 dice :)

But that's not necessarily bad! Success is great. Skilled people should often succeed. The question rather becomes, do you still succeed when facing off with someone determined to stop you?

I also think you should roll more dice more often. The part I like best is where you prioritize different things and distribute your dice. So a simple skill roll is kind of meh, but when you have three different things and have to split your resources? That's great stuff (incidentally, that's how one of my first design suggestions here at the Forge worked, just with cards).

One question I have for you that might influence your design from this point on is this: what kind of play style do you want your system to promote? Are the decisions about distributing your resources going to be all tactical (how to best game the system to prevail against challengs)? Do they make a statement of the character's and player's priorities (risk my life and limbs for x)? Are they modeling something accurately?



Like your basic idea. Its funny, I read percentile in your post and had come up with the pool idea by the time I read it a few words later; great minds think alike. Anyway, for a first pass, I think I would recommend leaving it as a straight percentile system for normals, and then add bonus dice to super-normal characters and creatures. This way you don't have to worry about this infinite scale that normal characters suddenly become the ground floor of, and the percentile range can now be applied to other categories (so now you have the percentile range for humans, giants, dragons, giant planet consuming robotic creatures as old as the universe, etc.).

The pool can get out of hand fast, and make it tough for characters to fail (which is fine, if that is your angle). I would reserve it for characters that operate on a scale way beyond that of normal characters.

Just my opinion. Thoughts?


10-Cent Heroes; check out my blog:

Sync; my techno-horror 2-pager


Its seems that it was mentioned up there that there was a problem with failure becoming an impossibility on action that went unopposed. I think you might want to take a nod from Amber. In Amber, everything you do is opposed by another character's statistics (of which there are only four). So, if your trying to say, pick a lock, you'd be rolling against the character that set the lock, or if you prefer, you could just have the lock have a quality or some similar value and have both the player trying to pick the lock roll for his character and the game master would roll for the lock quality (or for the person who set the lock) and the better roll would win. Opposed rolls do add quite a bit to action resolution in terms of time and effort, but in situations where's there not enough tension to warrant it, than a high probability of success isn't a problem.
Nolan Callender


Bossy, I'm aware of that, that's why I broke from the classical percentile system in the first place. For example, in most percentile games you get an ability and a skill, add them up and get a percentile probability number...but once characters become really experienced, it gets clunky. Because if you have someone with a 60 ability and 40 skill they have 100 chance of success...and what then? Instead in my system, most people will succeed most of the time. That, of course provided nothing or no one opposes them and they are not performing an extraordinary or impossible task. Imagine you are standing in front of a person with a longsword in hand. Imagine he does nothing to stop you. How hard is it really to hit him? Even if you have never used a sword in your life before. The more experienced and able a person becomes, he gets higher average results, and less variable results instead. I'm totaly ok with that. I think it makes more sense that for example you only have 30% chance of lifting a rock, provided it isn't to heavy for you. If you are strong enough, you'll lift it most or all of the time.

Now, I know this causes some other problems instead, like Ken pointed out, but I'm in the process of trying to figure it all out.

masqueradeball: That's a good idea, I will try to fit it into the existing version of the mechanic. I mean, I already have oposing rolls, but I was wondering about situations when there isn't an actual person trying to stop you. Rolling against the lockmaker's skill when you pick the lock makes some sense, although it would get awful if you tried to follow it to the word (you would have to know the skills of everyone in the game world). Maybe the GM would get his own dice pool and roll for the lock, somewhat like in DitV conflicts.

Paul Czege

Hey Greg,

I think my concern (and why I referenced Otherkind) is that with your system there's always a best choice the player should make, and it's always glaringly obvious if the player takes a second to think through the options. There's always a best weapon to use, and there's always a best number you can make from the dice you rolled. And so the choices being made by the player aren't really meaningful choices at all. Game systems I enjoy put me in the position of making trade-offs between various desirable options. I'm not seeing any fun or strategy in your system. I'm just seeing no-brainer choices.

Sell me on it. Show me where the fun is.

My Life with Master knows codependence.
And if you're doing anything with your Acts of Evil ashcan license, of course I'm curious and would love to hear about your plans