A search for an optimal resolution mechanic

[Username17]

I'm trying to understand the advantage in keeping Roll Under without it being anything other than an Appeal to Tradition. From a detached PoV, as I don't really play in d% resolution systems, I can only see the logic in Roll Over.

The advantage, and I mean that in the strict sense of there being literally only one advantage, is if you are checking binary success or failure against an unmodified target number. Like say, HERO system's skills. You have a skill like "Stealth 15 or less" and then you roll the dice and if you roll a 15 or less you "are stealthy", whatever the fuck that means.

The resolution of that is very fast. There is no addition or subtraction, just a simple comparison step. And if that is what you want to do with your skill system, it's good enough.

The problem is that unless you're making a pretty extreme "rules light" game, that isn't enough. You want to know how stealthy the character was. You want to know whether they were stealthier than another character. You want to know whether they were stealthy enough to get by someone who was "observant". And so on. And as soon as those questions come up, then Roll Under is 100% exactly the same as Roll Over except with a couple extra "multiply by negative 1" steps thrown in to make things more of a pain in the ass.

-Username17

[John Magnum]

Adding public modifiers to a d% roll under isn't too difficult, and is still fairly intuitive. Bonuses are positive numbers and you add them to your skill level, which is fairly intuitive. Penalties are negative numbers. It's the same operation, and it all basically makes sense.

But yeah, degrees of success and opposed rolls are annoying. Opposed rolls in particular are way, way, way nicer with Frank's additive system than the conventional roll-under system.

Is there an easy to way make 3d6-roll-under less insane? I guess since it's symmetric, you could still just do "roll, add your skill, compare to 18" and it would be the same odds. Or am I missing something?

[DSMatticus]

Is there an easy to way make 3d6-roll-under less insane? I guess since it's symmetric, you could still just do "roll, add your skill, compare to 18" and it would be the same odds. Or am I missing something?

You should compare to 21 or 20, depending on how you break ties.
3d6 on 7- succeeds on 3,4,5,6,7; 5 numbers.
3d6+7 vs 21 succeds on 18,17,16,15,14; 5 numbers.

I usually do this when I play any roll-under, because roll-under is fucking stupid. You just establish the system wide DC and modify that.

[tenuki]

How is a difference less of a number than a sum? Dude!

So you're saying that THAC0 is fine because the player can report that "they hit AC 3"?

-Username17

I believe I mentioned here and there that I consider D&D a shit game whose existence cannot be explained by anything but the moronic traditionalism of its fans. I don't care what is/was/will be in D&D. This thread is about a new resolution mechanic.

That aside, THAC0 is shit because it's unnecessarily obscure, much like your arguments. Hell, you're not even pretending to relate in any way to the roll-over/roll under discussion we're having here.

Unless you feel a desperate need to provide examples for shitty roll-over systems, that is.

[tussock]

The advantage, and I mean that in the strict sense of there being literally only one advantage, is if you are checking binary success or failure against an unmodified target number. Like say, HERO system's skills. You have a skill like "Stealth 15 or less" and then you roll the dice and if you roll a 15 or less you "are stealthy", whatever the fuck that means.

"Stealthy" would be a condition, it prevents people targeting you, and if you're all stealthy or not there then the enemy can't enter combat mode and is vulnerable to surprise. Which would probably just be left unstated.

The problem is that unless you're making a pretty extreme "rules light" game, that isn't enough. You want to know how stealthy the character was.

If you rolled a 13, you're stealthy "13". Higher being better allows more skilled characters to be "more stealthy", sometimes, just like an adding system.

You want to know whether they were stealthier than another character. You want to know whether they were stealthy enough to get by someone who was "observant". And so on. And as soon as those questions come up ...

He rolls too, and if he gets a 12 or less then I'm stealthier. No math once the die hits the table, just a comparison if there's two or more of them. If ten people roll it's really easy to find the best one, because the answer is sitting there on the dice.

If there's secret (GM-side) difficulty they have to beat that with their roll, like -4? They need to roll 5+ to beat it.

Stop subtracting shit, it's not necessary with roll under. Work out your modified skill, then roll the die, then call it out. Compare with challenges directly as desired. It's most functional with skills up to 20 (for d20), but there's a few workable solutions for higher ones, depending on how many times your system laps the RNG.

[DSMatticus]

THAC0 is shit because it's unnecessarily obscure

Face fucking palm? Let's do a basic attack roll in THAC0, 3d6 roll under, and 3d6 roll over:
THAC0: 1d20 vs skill - armor class +/- circumstantial modifiers (low skill is good, high roll is good)
Roll-under: 3d6 vs skill - armor class +/- circumstantial modifiers (high skill is good, low roll is good)
Roll-over: 3d6 + skill vs armor class +/- circumstantial modifiers (high skill is good, high roll is good)

Now which of those two systems actually has something in common? It's THAC0 and roll-under; the PC doesn't know enough to do the compare, and the DM needs two numbers to do the compare himself (and then he's doing all the math). To solve this problem, you proposed "smart roll-under." There is a smart version of THAC0, too. It's what Frank described. Here they both are:
Smart THAC0: 1d20 - skill vs armor class +/- circumstantial modifiers
Smart roll-under: -(3d6 - skill) vs armor class +/- circumstantial modifiers.

What do those two smart fixes do? They put PC information on ONE side of the compare and DM information on the other! But roll-over already fucking does that, so what the fucking fuck have you fucking accomplished? Answer: fuck-all. You are now debating the relative merits of:
1) 3d6 + skill vs armor class +/- circumstantial modifiers.
2) -(3d6 - skill) vs armor class +/- circumstantial modifiers.

They are mathematically equivalent. One is objectively less operations, favors addition over subtraction, and has intuitive directionality (bonus means add, penalty means subtract). And that's it. Roll-under is god awful terrible. It does everything roll-over does with more work and less intuitively. There is no better example of a pointless existence.

Edit P.S.: Roll-over also has infinite extensibility without additional complexity. Roll-under just fucking stops, and if you go past that point shit starts flipping signs or have you to start doing extra book-keeping. You are trading away convenience for an arbitrarily capped RNG.

[Lokathor]

So... that thing with the piles of d6s +X and you pick the highest, +1 for each 6 left over after picking the highest...

Would it work at all well with d20s? (or d12, or d10, etc) It seems like adding more sides to the type of dice rolled would keep numbers in the "average" range a lot, with occasional chances of rather low rolls and a very very small chance of getting the extra high rolls.

Using bigger dice lessens the impact of the +1 bonus and additionally makes that bonus less likely to happen. So using d20 I can guarantee that entire campaigns will happen without anyone seeing one of those +1 bonuses. At that point, why have it at all?

As for "average ranges", I am not even sure what you mean. What is an average range for a number?

"average range" here meaning "middle of the die", that is, a 1d20's average is 10.5, and so on.

[tenuki]

THAC0 is shit because it's unnecessarily obscure

What do those two smart fixes do? They put PC information on ONE side of the compare and DM information on the other! But roll-over already fucking does that, so what the fucking fuck have you fucking accomplished? Answer: fuck-all. You are now debating the relative merits of:
1) 3d6 + skill vs armor class +/- circumstantial modifiers.
2) -(3d6 - skill) vs armor class +/- circumstantial modifiers.

They are mathematically equivalent. One is objectively less operations, favors addition over subtraction, and has intuitive directionality (bonus means add, penalty means subtract). And that's it. Roll-under is god awful terrible. It does everything roll-over does with more work and less intuitively. There is no better example of a pointless existence.

Edit P.S.: Roll-over also has infinite extensibility without additional complexity. Roll-under just fucking stops, and if you go past that point shit starts flipping signs or have you to start doing extra book-keeping. You are trading away convenience for an arbitrarily capped RNG.

This whole THAC0 debate a red herring. Nobody actually wants to have this shit in his games, so let's not waste any more space, right? And we're all clear on the mathematical equivalence of additive roll-over and roll-under systems. I'll even agree that if you're using a purely additive system (i.e. one with no multipliers) roll-over is more convenient.

If you look at my original suggestion for a 2d10 system however, you'll find that does not easily convert to roll-over. You're not obliged to love my idea, but dismissing it purely on grounds of its being roll-under is intellectually lazy and crap.

The drawback of my mechanic is that you have to multiply a single-digit number by two or three sometimes to check for a good/critical success if you roll low enough.

I don't consider single-digit multiplication much of a math barrier though. It may not be quite as easy as d20+x, but it's a lot faster and less error-prone than adding two double-digit numbers to go past 100 in a d% system, for example. At any rate, you could always use a simple chart with three columns if you're worried.

The advantage of my mechanic is in how the probabilities for different success levels behave. With increasing ability, your chances for a basic/good/critical success rise in three separate logistic curves at different speeds.

Personally I think that's pretty flash and lends itself well to description: When two noobs fight each other, it's all about hit/parry with basic successes/failures. In a fight between two high-level swordsmasters, failed rolls basically do not happen and basic successes are below par. They blast each other with good and critical successes on offense and defense that are representative of their level of skill.

[Username17]

The advantage of my mechanic is in how the probabilities for different success levels behave. With increasing ability, your chances for a basic/good/critical success rise in three separate logistic curves at different speeds.

Oh wait, we're talking about your stupid fucking 2d10 roll under piece of shit where you divide your target number by various prime numbers to get degrees of success? Yeah, um... no. We are not discussing that in detail because that is one of the worst pieces of shit I have ever seen someone propose while sober.

At what point did you think it would be a good idea to require people to recalculate both good and bad 3e style save progressions to determine critical thresholds on every single action? That is insanity.

The rest of us are going to go back to talking about roll under systems that someone might actually want to use without having first been afflicted by stage four syphilis.

-Username17

[spaceLem]

Oh wait, we're talking about your stupid fucking 2d10 roll under piece of shit where you divide your target number by various prime numbers to get degrees of success? Yeah, um... no. We are not discussing that in detail because that is one of the worst pieces of shit I have ever seen someone propose while sober.

At what point did you think it would be a good idea to require people to recalculate both good and bad 3e style save progressions to determine critical thresholds on every single action? That is insanity.

The rest of us are going to go back to talking about roll under systems that someone might actually want to use without having first been afflicted by stage four syphilis.

It seems that the dominant opinion here is that rolling over is preferable to rolling under. I will happily accept that it has its advantages, although I remain unconvinced that it is objectively better (simply because without applying subjective weighting, that requires it to be better in every single aspect, which it is demonstrably not).

Either way, rolling under certainly isn't so bad that it's unplayable without the symptoms of an end stage STD, anecdotally at least plenty of people are capable of playing with a roll under mechanic without complaining, and there are still systems being published that use it.

However, if you hurl enough insults and profanity at everyone who states that opinion, maybe they'll go away, and you can be in total agreement with yourself? Or should I regard everything you say as hyperbole?

[RadiantPhoenix]

Either way, rolling under certainly isn't so bad that it's unplayable without the symptoms of an end stage STD

It's not roll-under that's so bad; it's repeatedly dividing target numbers that's a problem.

[spaceLem]

It's not roll-under that's so bad; it's repeatedly dividing target numbers that's a problem.

I have to admit, I'm not totally sold on that idea.

[wotmaniac]

However, if you hurl enough insults and profanity at everyone who states that opinion, maybe they'll go away, and you can be in total agreement with yourself? Or should I regard everything you say as hyperbole?

That's just Frank-speak. It is what it is.
He generally knows what he's talking about; and once you sift trough his ranting, you can usually get to the core of what he's talking about. Though, he does often need to be pressed for specifics -- which often don't fully show up until he decides to dedicate a whole thread to it.

Just grab a snickers and enjoy the ride.

[spaceLem]

That's just Frank-speak. It is what it is.
He generally knows what he's talking about; and once you sift trough his ranting, you can usually get to the core of what he's talking about. Though, he does often need to be pressed for specifics -- which often don't fully show up until he decides to dedicate a whole thread to it.

Just grab a snickers and enjoy the ride.

Heh. I have seen Frank be pretty insightful at times (ya hear that?), that's kind of what brought me here in the first place. Doesn't stop me from picturing his posts in my head with an angry Al Pacino voice though

[virgil]

I tend to imagine House, with more orange.

[tenuki]

Either way, rolling under certainly isn't so bad that it's unplayable without the symptoms of an end stage STD

It's not roll-under that's so bad; it's repeatedly dividing target numbers that's a problem.

There are two ways around that.

One is to multiply the roll result if it's low enough to qualify for a potential good or critical success instead of dividing the TN. You don't even have to do it for every roll. E.g., if my skill is 15 and I roll an 11, I don't bother.

The other would be to use a simple chart:

5/2/1
6/3/2
7/3/2
8/4/2
9/4/3
10/5/3
11/5/3
12/6/4
13/6/4
14/7/4
15/7/5
16/8/5
17/8/5
18/9/6
19/9/6
20/10/6

and so on.

You could print it in a corner of the character sheet like the step table in Earthdawn. I just didn't think this would be necessary.

If you use a chart, you could also smooth out some of the steps so you don't get raises in both the good and crit chance at multiples of 6. E.g., use 11/6/3 and 17/9/5).

[Username17]

If you use a chart, you could also smooth out some of the steps so you don't get raises in both the good and crit chance at multiples of 6. E.g., use 11/6/3 and 17/9/5).

That doesn't address the point where this is fucking stupid and way more complicated than it needs to be. You're already using a sloped RNG. Rolling an 11 on 2d10 is already twice as likely as rolling a 6 and ten times as likely as rolling a 2. Rolling multiple dice and adding them together allows you to make crits substantially less likely than normal successes with an entirely linear subsets. There's no point in division or charts, because a natural 2 or 20 on 2d10 are still one in a hundred chances.

You could just do roll over and set the TNs to 20, 25, and 30 for success, super success, and superduper success. It would be about a million times easier to explain and use and do basically everything you claim to want to do with the system.

-Username17

[tenuki]

You could just do roll over and set the TNs to 20, 25, and 30 for success, super success, and superduper success. It would be about a million times easier to explain and use and do basically everything you claim to want to do with the system.

-Username17

Not quite. If the target for a crit is 30, you can't get one at skill ratings below +9. The point where you are more likely to get a crit than anything happens at +20.

If you compare this to my suggestion suggested, you get a far wider usable range on the RNG, easily from 6 (1% crit, 2% good success, 12% basic success) to 30 (45% crits, 40% good success, 15% basic success).

[Lago PARANOIA]

I'm starting to enjoy dicepools more and more. They really make character sheets nicer to read and frankly rolling 8 6-sided dice is a lot better for most groups than rolling a 1d20+13.

However, there is a problem I have with dicepools. That is, the size can quickly get out of control. Personally, I think having to interact with more than 12 6-sided dice is just too much trouble from a time and usability standpoint.

So a couple of questions abound:

1.) In your opinion, what are the most dice that can be rolled at once before it just becomes too much of a pain in the ass to collect, roll, and count? My threshold is 12 dice, personally.

2.) Is there a way to reduce dicepool size by mixing in dice while still keeping the probability shift relatively stable? For example:

Say you have a mechanic where the maximum amount of 6-sided die you can roll at once is 6, with the TN normally being 3. These dice never explode. If you have a dicepool larger than that, you can trade in rolling 2 6-sided dice for one d12. The 'normal' target number for a d12 is 4-7. The 'extra' target number is 8-10, which provides two hits. If you get an 11 or 12, you get two hits and the dice explodes.

Yes, it's not merely a linear replacement. The intent is to have hitting rank 7 makes you significantly more awesome than the silly peasants in addition to tamping down on the dicepool size.

Is this fruitful, or should I just have automatic hits?

[wotmaniac]

2.) Is there a way to reduce dicepool size by mixing in dice while still keeping the probability shift relatively stable? For example:

Say you have a mechanic where the maximum amount of 6-sided die you can roll at once is 6, with the TN normally being 3. These dice never explode. If you have a dicepool larger than that, you can trade in rolling 2 6-sided dice for one d12. The 'normal' target number for a d12 is 4-7. The 'extra' target number is 8-10, which provides two hits. If you get an 11 or 12, you get two hits and the dice explodes.

Yes, it's not merely a linear replacement. The intent is to have hitting rank 7 makes you significantly more awesome than the silly peasants in addition to tamping down on the dicepool size.

Is this fruitful, or should I just have automatic hits?

That's an interesting take on it; but if you're using a specific die as the base standard for your dice pools, and then start throwing in other types of dice, I can see this as unduly complicating things. It's not hard, but easily confused because you are having to simultaneously be in both "d6 mode" and "d12 mode" ... a pile of d6s with a couple of d12s thrown in can be easily missed (especially if it's been a long night, or if you're in a hurry, or whatever).

I think auto-hits are the way to go .... just so long as you don't fall in to the insanity that WW did with Scion.


Oh, btw -- depending on my mood, my desired max is 10-15. I learned this the hard way after playing Aberrant.

[Lago PARANOIA]

Now here's a non-dicepool question, just for d20s.

Let's talk about damage for a bit. Rolling 1d20+6 (3E monk unarmed strike damage) is lame, but not experience-ruiningly lame. I just don't care for that amount of swinginess. However, rolling 1d6+24 is so lame that it's barely worth talking about.

The thing is, damage rolls on a linear/bell-curve range have several competing interests.

1.) You don't want to have the output be too impervious to the RNG. Having the damage output being almost completely dependent on the RNG is more acceptable, but you want to avoid it. 2d8+9 is a fine roll, 2d8+2 is irritating but acceptable, 2d8+20 is dumb.

2.) You don't want the dicepool size to get too large. That's a given, though.

3.) The obvious solution is to increase the dice size/amount alongside increasing the fixed operator... the problem of course being is that you either have Earthdawn's step-die chart (which is lame) or damage rolls become a hodgepodge of unpredictable RNGs that can't be generalized -- which is difficult to get a handle on.

4.) *D&D BONUS: A system that acknowledged your different weapon dice size would be kewl, too.

[Lago PARANOIA]

I think auto-hits are the way to go .... just so long as you don't fall in to the insanity that WW did with Scion.

My only problem with automatic hits is that for a long results curve they become pretty lame. Which isn't an issue for Shadowrun or Mouse Guard, but if you wanted to model the world of D&D or Champions you could have a situation where you rolled 9 d6s with 20 automatic hits. Which is lame.

[wotmaniac]

I think auto-hits are the way to go .... just so long as you don't fall in to the insanity that WW did with Scion.

My only problem with automatic hits is that for a long results curve they become pretty lame. Which isn't an issue for Shadowrun or Mouse Guard, but if you wanted to model the world of D&D or Champions you could have a situation where you rolled 9 d6s with 20 automatic hits. Which is lame.

yeah, your auto-hits should probably never exceed your dice pool -- if that's the case, it just might be time to go back to the drawing board.
In Scion, you could get your auto-hits literally in to the hundreds .... I know you're playing gods and all, but give me a fucking break.

[Username17]

You can make the effectively different sections of a dicepool bigger by making the dice hit more often. If you hit on a 5+, your difficulties can go up about 1 per three dice of awesome. If you hit on a 4+, difficulties can go up about 1 per two. If you hit on a 3+, then difficulties can go up about 2 every 3.

What that essentially means is that by lowering the TN, you can increase the threshold at which you need auto-hits to maintain sanity. Personally, I think that rolling more than 12 dice for a "normal" test is bullshit. But at the same time I see real value in people rolling 18 dice on specialist tests to show that they are super awesome. 24 dice is even OK to roll for "boss fights" and similar shenanigans.

So a major badass PC might need auto-hits if he is supposed to be attempting a task with a difficulty of more than 6 if his TN is 5+. But the cutoff for requiring autohits for the same badass PC would be more like 12 hits if his TN was 3+.

-Username17

[virgil]

Has there ever been a good resolution mechanic that used playing cards?