Concerning the RNG and Modifiers

[Shazbot79]

This is a subject carried over from the Game Design forum. Frank Trollman suggested I post here for general feedback.

So, the question I have is about modifiers pushing action resolution off of the random number generator.

As I understand it, in some games (d20) the modifiers get so big that success becomes almost guaranteed and the RNG becomes completely arbitrary.

What I'm wondering is, at about what point does this start to happen and what practical effect this has on actual gameplay.

Is it just a question of smaller numbers being easier to balance or does this have some dire effect on a game that I don't see?

[Thymos]

It's a question of what ratio of your result is determined by the bonus and the random number generator.

Some systems scale better than others. DnD for example scales horribly. When you have no bonus your entire result is the d20. When you have a +10 then about half the result is the d20. With a +20 most of the result is the bonus. Once you hit +40 (which is not hard at all) the d20 stops mattering as much. At +100 you might as well not really bother with the d20. When these numbers vary as much as they do it's almost impossible to create DCs that work.

Some other systems scale really well. Shadowruns hit system actually causes the bonuses to scale completely, as your bonus gets larger your random number range also increases.

The dire effect of the d20 is that not only is it hard to set DC's, but at the higher level the d20 actually matters less than it does at the early levels.

[Grek]

Being off the RNG is bad because it means there is no longer variablity in a part of the game where there is supposed to be variable outcomes. If the game, for whatever reason, requires that the players fail rolls occasionally, and they never fail, then the game is no longer running like it should.

Going off the RNG is usually the result of adding a bunch of bonuses in the system without considering the possibility of a player getting more bonuses than expected. It happens alot in DnD, for example, because the rules are constantly being added to and being written by different people, few of whom have a terribly good grasp on mathmatics or balance.

[RobbyPants]

It's a question of what ratio of your result is determined by the bonus and the random number generator.

Some systems scale better than others. DnD for example scales horribly. When you have no bonus your entire result is the d20. When you have a +10 then about half the result is the d20. With a +20 most of the result is the bonus. Once you hit +40 (which is not hard at all) the d20 stops mattering as much. At +100 you might as well not really bother with the d20. When these numbers vary as much as they do it's almost impossible to create DCs that work.

Some other systems scale really well. Shadowruns hit system actually causes the bonuses to scale completely, as your bonus gets larger your random number range also increases.

The dire effect of the d20 is that not only is it hard to set DC's, but at the higher level the d20 actually matters less than it does at the early levels.

While this is all technically true, I don't think it's really the problem. The only reason you care about the final result of the roll (d20 + modifiers) is to see if you hit some target number. Assuming your modifiers scale as does the TN, then the RNG is still a pretty big part of the equation to see if you succeed or not.

The problem is if either the modifiers or TN scale too quickly, or not quickly enough. Once the difference between the two becomes greater than the range of the RNG, then you have guaranteed success or failure. That's the problem.

Edit:
So, what I'm saying is even if you're rolling d20 + 100, if your TN is 110 or 111, then everything's fine. That d20 matters plenty.

[Kaelik]

The problem is that with setting a TN, if the bonus makes up too large a portion of the RNG, you can't balance possible failure between someon really good, and someone kinda good.

The Spot checks of typical party consist of:

Druid: +40
Rogue: +25
Wizard: +15
Fighter: -1.

There is no monster that the Fighter could see that the Rogue could not see, ever. If the fighter even has a chance of seeing, then the Rogue and Druid don't have to roll at all.

Likewise, anything that the Druid even rolls on is going to be impossible for most of the party to see.

That's the "problem."

This can be design intent, or it can be a flaw, depending on your goals.

[TheWorid]

Would it not be agreeable to make chance of success based on proportion rather than difference? So the functional distinction between two opposed characters with 2 and 6 attributes would be 1:3 rather than +4? Forgive me if my terminology is off.

Are there any games that already do that (or do that well)?

[Username17]

Basically no matter what your RNG actually is, you have a basic setup that is as follows:

• You have a base chance of success. This is generally going to be somewhere between 50% and 2/3.
• When you have a "favorable" situation (being better than your opponent, attacking from concealment, whatever) you will receive a bonus to your chance of success that is supposed to feel noticeable.
• If you have more than one such favorable situation, you can either add the modifiers together or ignore one of the modifiers.

So if you use a flat RNG, you can add about 3 "noticeable" bonuses. After that, you've already hit the end of the RNG, so the fourth and subsequent noticeable bonuses won't be added because you'll already be at 100%. This is a universal fact whether your flat RNG is a d6 (4+ with a +1 bonus is 3+, then 2+, then 1+), or a d20 (10+ with a +3 bonus is 7+, then 4+, then 1+). Or d% or a d12 or whatever. Because that is how an uncurved die roll works.

If you use a curved die roll, the first modifier is bigger than the third. So depending on how curved your die roll is, you can squeeze in 4 or 5 - or even more bonuses that are individually noticeable without hitting the end of the RNG.

Which one you go with depends largely upon what you want your game to do when characters attack weaker and damaged enemies from concealment while benefiting from a temporary buff spell.

-Username17

[Ganbare Gincun]

Carried over from the Game Design Forum:

Resource management basically is your game. It's the thing that people are actually making choices against during the game. Without resource management, characters just play themselves.

You can do charges, or cooldown, or ragebar, or drain, or WoF, or defense commitment or some combination, or something else. But there has got to be a reason that your presence is actually required. Otherwise people will wander off and play smash brothers.

So Frank - what are the various pro and cons of these various resource management systems? Which ones do you prefer, and which ones do you run screaming from?

[Crissa]

There are also the situation like things that affect all members of the party - it's possible to make situations where one person cannot succeed and another cannot fail.

In D&D with its death traps... That's dumb.

-Crissa

[Username17]

Resource Management.

• Charges The idea is that you have a limited number of super moves that you can do. This encourages nova attacks and hit-and-run tactics, and it increases the value of get out of jail free cards like super dodge and substitute because your enemies are doing the same thing.
• Rage Bar The idea is that you have to do stuff in combat to unlock your super maneuvers. Done right, it encourages people to grind in longer combats and play defensively until they can let loose with their super moves.
• Cool Down The idea is that you have to wait after using any maneuver before using it again. Encourages "rhythms" where characters use moves in the same order over and over again.
• Warm Up Like Rage Bar, except that it's a simple amount of time in warming up a power before it is ready to use. This encourages ambushes.
• WoF The characters have access to random abilities each round, which makes them use different abilities in different rounds but does not otherwise create patterns.
• Drain The characters suffer a penalty on rounds after they use their super moves, which encourages players to attempt to time their ability usage to occur as close to the end of battle as possible.
• Unlimited The characters can use whatever ability they want, whenever they want. This tends to create ability spam.

Now that's basically all there is. And all of them work and all of them have their uses. The thing I really can't stand is when someone thinks that they've invented a new system by increasing the granularity to the point where it is hard to keep track of. So instead of giving someone 3 charges and making a super fireball use up one charge, they might hand out 39 charges and make a super fireball use up 12. That's the same damn thing, but it involves multiple digit math problems to get there, which is retarded.

This is why "mana" systems blow ass in table top.

-Username17

[Crissa]

Of course, systems like Winds of Fate in a video game turn into whack a mole where you're playing against yourself as much as you're playing against an opponent. Random chance could totally bone you and you just press restart again.

So given there's differences to how you're playing - table top, pbm, mmo, etc - it'll give different disadvantages to each system.

-Crissa

[Shazbot79]

I don't think I've seen an example of Rage Bar resource management mechanics in a tabletop rpg.

It's been a while since I've really played any video games, but as i recall a lot of boss fights used to get tougher the closer the player would get to victory. I mean, you'd get half their life bar down and suddenly they go apeshit and start pummeling you to oblivion...

[RandomCasualty2]

Some systems scale better than others. DnD for example scales horribly. When you have no bonus your entire result is the d20. When you have a +10 then about half the result is the d20. With a +20 most of the result is the bonus. Once you hit +40 (which is not hard at all) the d20 stops mattering as much. At +100 you might as well not really bother with the d20. When these numbers vary as much as they do it's almost impossible to create DCs that work.

No. Not at all. You're thinking of damage rolls and other accumulation based rolls.

In a straight pass/fail roll, such as an attack roll or a saving throw, The d20 never stops mattering unless your bonus beats out the DC by a huge amount. Having a +20 against a DC 30 check is the same as having a +10 versus a DC 20 check.

The idea of the RNG in such a game is that both the DC and the bonus will constantly improve and match the enemy.

[A Man In Black]

I don't think I've seen an example of Rage Bar resource management mechanics in a tabletop rpg.

The build-up stat doesn't actually have to be on the character doing the build-up. Lots of games have "This ability (can only be used | is much more effective if used) when the enemy is weakened X amount."

[Username17]

Heck, 4e D&D has a number of abilities on that are on charges, but kick in when the character (or monster) is bloodied. As a charged power, you'd expect to use it on round 1, but because it requires a loss of hit points in combat first, people end up using it later in the fight. Most versions of D&D have some vampiric stuff in there, where characters can build up something or other by attacking enemies. All that Master Vampire crap, where you attack people and build up a power pool and spend the pool to do super stuff.

-Username17

[hogarth]

As I understand it, in some games (d20) the modifiers get so big that success becomes almost guaranteed and the RNG becomes completely arbitrary.

What I'm wondering is, at about what point does this start to happen and what practical effect this has on actual gameplay.

Personally, I find it's a problem when, in the sort of task everyone ought to be have some chance for success (e.g. diplomacy or making a saving throw in D&D), an average task for a specialized PC is an impossible task for a non-specialized PC, or an average task for a non-specialized PC is an impossible task for a specialized PC. So in the d20 system, that corresponds to a spread of +10 between the modifiers.

Note that it's just a subset of tasks I'm talking about, though. There's nothing wrong with impossible tasks in many cases -- only when you would reasonably expect the whole party to be able to contribute.

[Orca]

There's nothing wrong with impossible tasks in many cases -- only when you would reasonably expect the whole party to be able to contribute.

Or in cases where you might think that anyone should have a chance of success, however small - noticing the assassin sneaking up, for example.