Which gaming system has the best mechanics and why?

[billd91]

The problem I've found is that the most popular bell-curve systems--GURPS, HERO--are "roll under."

Put bluntly, I dislike roll-under systems, and I hate with a flaming, furious passion GURPS' version of it. I hate the nebulousness of degree of success, I hate the idea that as a player you NEVER have any real idea of what the "target" level of success is. Is a "standard" success good enough? Do I need to roll 4 under the the obstacle? 5 under?

For that reason alone I've pretty much discounted ever playing HERO as well (sorry [MENTION=19675]Dannyalcatraz[/MENTION]). I just can't stand the basic 3d6-roll under resolution.

It wouldn't be all that much work to make them roll over. If most successes in Champions are 11 and under, you'd have to switch to 10 and over and get the algebra right on the other elements. Instead of trying to roll under 11+OCV-DCV, it would be roll over 10+DCV-OCV. Or you could have the target to hit be 10+DCV and roll 3d6 adding your OCV. It becomes a lot more like the D&D system of rolling, adding mods, against a target success number with a little judicious algebra.

I'm a little surprised at the vehemence of your reaction to roll-under games. It's all just rolling methods. I don't know why rolling under would be worse than rolling over. It certainly makes calculation of your chance of success very easy when rolling under.

 

[innerdude]

I'm a little surprised at the vehemence of your reaction to roll-under games. It's all just rolling methods. I don't know why rolling under would be worse than rolling over. It certainly makes calculation of your chance of success very easy when rolling under.

To a point you're right, in that the math isn't all that hard to reverse around. I think it's mostly due to the general mental conception/paradigm of action resolution ---- that the difficulty of resolving a problem has no bearing on the outcome, only the skill level of the person performing the task. It's just......backwards to me.

So one person has a 12 in metalworking, and another has a 15. For any given metalworking problem, if they both roll an 11, they've both succeeded, yet if they both roll a 13, one of them has succeeded and the other one hasn't........regardless of how "difficult" the task at hand actually was.

As a GURPS player, it makes it hard to "eyeball" a situation and know if it's something your character is capable of, should be considering doing in the fiction, etc. It's just a constant metagame irritant about the entire system, trying to rationalize why a given roll is good enough, or not good enough to succeed, when a "roll over a target number" system makes it exceedingly clear. Oh sure, a GM could say, "Well, technically you 'failed' your roll, but what really happened is you succeeded, but it just took 4 times longer than you would have hoped."

But why set up a system like that in the first place, where "degree of success" is entirely GM fiat? Savage Worlds is much, much better in this regard, where the idea of degree of success ("getting a raise") is hard baked into the mechanics.

Of course the obvious answer is, "Well that's your GM's job to convey difficulty of a task through the fiction and the narrative." And to a point, setting "difficulty" ratings is arbitrary no matter whether it's roll over or roll under. There's just something wrong in my head about the way GURPS does it by default.

 

[Jhaelen]

Much of the time the licensed game or narrow-niche game will do it's thing much better than the broader-genre, mutli-genre or universal system. But not every time. Some games, however much they may specialize, just aren't that good.

That's true of course. E.g. I don't consider ICE's MERP a system that gives you an enjoyable play experience in Tolkien's world. The One Ring RPG, however, manages that perfectly.

So, I suppose there might be some exotic narrow-niche that isn't sufficiently covered by dedicated RPG systems, so a universal system ends up being the best choice for it. But I'd wager that's a very rare thing.
In any case, a dedicated system will have a better chance of being the best system for the niche it's been designed for than a universal system.

 

[JDulle]

I've enjoyed playing multiple d20 rules sets, though lately Pathfinder has been the game played most with our group. I'm not sure d20 is the best rule's mechanics though. I prefer more versatile systems -- Atomic Highways or d00Lite. Both of those systems seem to allow for more cinematic use to me.

 

[Argyle King]

To a point you're right, in that the math isn't all that hard to reverse around. I think it's mostly due to the general mental conception/paradigm of action resolution ---- that the difficulty of resolving a problem has no bearing on the outcome, only the skill level of the person performing the task. It's just......backwards to me.

So one person has a 12 in metalworking, and another has a 15. For any given metalworking problem, if they both roll an 11, they've both succeeded, yet if they both roll a 13, one of them has succeeded and the other one hasn't........regardless of how "difficult" the task at hand actually was.

As a GURPS player, it makes it hard to "eyeball" a situation and know if it's something your character is capable of, should be considering doing in the fiction, etc. It's just a constant metagame irritant about the entire system, trying to rationalize why a given roll is good enough, or not good enough to succeed, when a "roll over a target number" system makes it exceedingly clear. Oh sure, a GM could say, "Well, technically you 'failed' your roll, but what really happened is you succeeded, but it just took 4 times longer than you would have hoped."

But why set up a system like that in the first place, where "degree of success" is entirely GM fiat? Savage Worlds is much, much better in this regard, where the idea of degree of success ("getting a raise") is hard baked into the mechanics.

Of course the obvious answer is, "Well that's your GM's job to convey difficulty of a task through the fiction and the narrative." And to a point, setting "difficulty" ratings is arbitrary no matter whether it's roll over or roll under. There's just something wrong in my head about the way GURPS does it by default.

weird...

I literally have the exact opposite opinion; I prefer GURPS because I feel like I'm more often making decisions from my character's point of view and what makes sense given the in game situation as opposed to making decisions from an abstract player's point of view and what makes sense given how the dice work.

Degree of success is pretty easy to understand.

 

[dracomilan]

Alternity is a pretty solid roll-under.

 

[Umbran]

Degree of success is pretty easy to understand.

Degree of success is pretty easy for you to understand.

Some mechanics just don't fall out intuitively in some people's brains. There are a few reasons why this construction will tend to be more difficult than other constructions for some folks: The more steps and relative calculations you have to do, the more likely it is to not become intuitive to some players. And, it has been documented that while individuals vary, the human mind is better at addition and multiplication than subtraction and addition - so roll-under is generally just that tad more difficult, requires that extra processing step, and is thus that much more likely to never become intuitive to some.

 

[Argyle King]

Degree of success is pretty easy for you to understand.

Some mechanics just don't fall out intuitively in some people's brains. There are a few reasons why this construction will tend to be more difficult than other constructions for some folks: The more steps and relative calculations you have to do, the more likely it is to not become intuitive to some players. And, it has been documented that while individuals vary, the human mind is better at addition and multiplication than subtraction and addition - so roll-under is generally just that tad more difficult, requires that extra processing step, and is thus that much more likely to never become intuitive to some.

In my mind, subtracting isn't even involved. It's comparing two numbers and determining which one is higher. That's actually less intense than adding the variety of d20 system modifiers.

I have 12 DX; I rolled 10. Is 10 less than 12?

Margin of success asks for an additional step. How much difference is between the two numbers?



Also, I wasn't condemning the poster that I quoted. I simply found it interesting that my view was literally the exact opposite. Despite having more rules, I find GURPS vastly more intuitive than the majority of D&D editions I've played.

 

[innerdude]

Well, in a certain sense, you COULD simply conflate "margin of success" as being equivalent to the "target number" in GURPS. In other words, some challenges are a "Simply roll under your skill, regardless of how much you succeed"; some might be "Roll at least 2 under your skill"; some might be "Roll at least 5 under your skill." That makes sense. This also directly scales with skills, since the higher a skill rating, the more likely you are to succeed by the required margin of success.

This is essentially the same as "lowering" the player's skill in the background without telling them. "Okay, normally they have a 16, but for this check they have to roll below 13." But even then there's a problem, which is that the more skilled a PC you are, the more punishing the penalties are. A PC is penalized MORE heavily by coming from outside the second standard deviation of the bell curve back to the middle. A character with a 16 skill is penalized exceptionally more by having their skill rating lowered to 13 than someone with a 13 skill is penalized by having their skill lowered to 10. And that's just screwy.

Because now to do skill penalties right, the GM has to adjust the penalty differently depending on where the character's rating is within the bell curve. It becomes a massive headache trying to even visualize the "relative" difficulty of any given task, instead of simply saying, "A task has difficulty X, roll your dice and see if you meet or exceed it."

The problem is too, by default GURPS doesn't seem to tell GMs how to consistently apply this, or even if they should. I do recall GURPS explicitly telling GMs not to make players roll all the time, and to be judicious as to whether a roll is called for at all, but I don't recall anything directly relating to degree of success. I've only read the GURPS 4e rulebooks all the way through a couple of times, so maybe I'm wrong about that, but I don't recall any explicit mention of degree of success except for critical success/failure.

The place this most bugs the living daylights out of me for this is active defenses. It feels completely wrong that a defender doesn't have to account for an attacker's margin of success, they simply have to make a standard success against their own fighting skill. So a guy with a 16 skill can roll a 6 for an attack (for the uninitiated, rolling lower is better in GURPS) and a guy with a 12 can roll a 12 for his defense, and the clearly far less skilled defender still successfully parries. With two opponents in the upper limit of skill range (15+), winning a fight basically comes down to sheer luck. "Attack." "Parry." "Attack." "Parry." For round, after round, after round.

As a side note, I also hate that when you act in a round is purely a construct of your speed rating. Fastest guy ALWAYS goes first, the end. There's no element of chance, no potential for a narrative of, "You get distracted for a split second and the opponent gets in the first attack this round." To me this also detracts from the teamwork aspect, because you can never really plan moves based on your combat order unless you're the guy who goes first. I would absolutely use the Savage Worlds initiative mechanic in GURPS, giving a bonus to people whose speed reached a certain rating (i.e., with a speed rating of 6.75 you automatically re-draw any cards 4 or lower, etc.).

If you don't use degree of success at all, GURPS to me is basically incoherent, because at that point your skill level means nothing other than your skill level. It provides zero meaningful information about the nature or difficulty of a challenge, meaning the player cannot make rational decisions about what to do. It's basically reverse dissociation---the character would know perfectly well in the fiction how difficult a challenge appears to be, and how likely he or she is to succeed at a task, but the player can't visualize that information based on what's on their character sheet. Without accounting for degree of success, you're basically playing a game of "Schrodinger's challenge rating." The actual difficulty of the challenge isn't known until AFTER you resolve the dice throw.

And heaven help you if a player ever has to ask, "How does that actually work?" in GURPS. Because the answer will be scattered across four different supplements, and will require a half-dozen rolls each from the player and GM to resolve. (This is somewhat better in GURPS 4e, since the core rules contain all the previous 3e compendia. But in 3e, it's pretty much EXACTLY like this.)

 

[The Crimson Binome]

Well, in a certain sense, you COULD simply conflate "margin of success" as being equivalent to the "target number" in GURPS. In other words, some challenges are a "Simply roll under your skill, regardless of how much you succeed"; some might be "Roll at least 2 under your skill"; some might be "Roll at least 5 under your skill." That makes sense. This also directly scales with skills, since the higher a skill rating, the more likely you are to succeed by the required margin of success.

It's not always an amount under your skill, though. Sometimes, for an easier task, you might be able to get away with "fail by a margin no greater than 2"; and it's been a few years since I've read the books, but I seem to recall that the easiest checks could allow you to succeed if you fail the check by a margin of up to 7. Effectively, you could gain a temporary bonus of +7 to your skill rating when you are performing an extremely easy task under optimal conditions, and this is the excuse for why a professional pilot might have a relevant skill rating of 12 or so - because that would shoot up to a 19 during routine operations.

Really, a roll-under system is great when your chance of success isn't going to be modified very often. If you're almost-always making a straight check, then it's trivial to compare the number you roll against the relevant skill rating and determine success or failure. If you're going to frequently adjust for difficulty, then an add-up system works better; for example, you could modify GURPS to run with 3d6 and add your modifiers in order to hit a Target Number of 18.

Comparison of numbers is easier than addition, and addition is easier than subtraction.

The place this most bugs the living daylights out of me for this is active defenses. It feels completely wrong that a defender doesn't have to account for an attacker's margin of success, they simply have make a standard success against their own fighting skill. So a guy with a 16 skill can roll an 6, and a guy with an 12 can roll a 12, and the clearly far less skilled defender still successfully parries. With two opponents in the upper limit of skill range (15+), winning a fight basically comes down to sheer luck. "Attack." "Parry." "Attack." "Parry." For round, after round, after round.

From a game design standpoint, there are good reasons why they do this. Now, the particular implementation fails for the very reason you suggest - it's far too easy to create a character with a high parry value - but that doesn't mean the idea lacks merit.

Basically, it comes down to putting a clamp on min-maxxers, and keeping defense relevant. If you're familiar with D&D 3.x or Pathfinder, then you know how attack bonuses eventually render AC to be meaningless - the Amulet of Natural Armor +2 becomes a joke item when it changes your AC from 22 to 24 and the enemy is swinging with +30 to hit. And that's a big chunk of the game which is invalidated. AC is really supposed to matter.

For the functional range of weapon skills in GURPS (roughly 9 to 18), the linked parry chance will always be significantly lower (a skill rating from 9-18 generates a parry value from 7-12). If you think that it's too common for a success on the attack roll to be negated by a simple success on the parry roll, then instituting a margin of success would end up with the inverse problem - success on the attack roll would very rarely be stopped by any sort of defense roll. It's difficult enough to roll under a 9 on 3d6, but you'll hardly ever roll under a 9 by more than the margin by which someone rolls under a 12 on 3d6.

Instituting a margin of success system would also greatly incentivize the sort of ridiculous skill ratings that tend to derail gameplay. With a simple success system, there's little incentive for raising your effective skill rating (after called shots) above 14, because each further point gives you a decreasing benefit - you won't succeed that much more often with a 15 than you would with a 14. With a margin of success, it's an arms race to boost your score as high as possible, because each point of skill buys you extra insurance against the defender having a good defense or rolling well, and you negate that chance on a 1-for-1 basis.

With a simple opposed check, your chance of hitting is equal to the chance that you succeed on your roll multiplied by the chance that they fail on their roll. As you increase your chance closer to 100%, the overall chance of success asymptotes out to whatever their chance of defending is. You still benefit from raising your accuracy, but there are diminishing returns, and the defender is never put into a position where they have zero chance of defending. Really, there just needs to be a cap on defenses to prevent ridiculous stalemate fights. (I would suggest something like 50% as the maximum chance of a successful defense roll.)