PDA

View Full Version : The Maximal Skill Problem



Lord Raziere
2013-10-19, 03:35 AM
I think I've noticed something in skill-based rpgs. Specifically, they might claim to be more flexible about the character creation process, but all the ones I have tried so far (Exalted, Shadowrun) seem to default to a class-like arrangement. you often end up with a bunch of hyper-specialized characters anyways despite their talk of having any skill you want and being open-ended.

The result is that skill-based rpgs seem to be in many ways, just unbuilt classes, with far more potential to choose bad choices that screw over the character, one problem I've noticed in particular is the maximal skill problem, in that you often have to get your skill up to maximum competency just for it to be effective at all. there is no room for generalists, sure you can make one but these are generally considered useless.

In many ways, this is oddly more limiting than classes. with classes at least you know your limitations and what you are playing, the limitations and capabilities are built right in and the class is more a base to build upon from there, a solid foundation and character concept. with skill-based ones, you don't know what you are getting. without the right combination of skills, it might not work. you can at least count upon the fact that a class is clear whether its bad or good right from the get-go, but with skills its not clear because its all spread out and is like a bunch of blocks: you don't know if the structure you put together will work until you are finished. which means you tend to stick to structures you know. the structures people know of course, are classes.

Sure its more customizable, but that just means there is more room to make bad choices and holes to fall into. I say this as someone who likes skill-based systems in theory, but find myself disappointed in how they work practically.

However, there are some systems that are better- Eclipse Phases package system makes sure the character isn't hyper-specialized, even if its technically just picking a few sub-classes and combining them together. FATE with all its narrative flexibility allows you a lot of leeway in how you go about using the skills you have, but I count FATE less as a skill rpg and more as a narrative one. the true power is the narrative, not the skills used to uphold it.

I don't know, I guess this is just a realization I've had and just wanted to share it. I guess what I'd want, going forward from this is to somehow find a way to make a skill-based system live up to its ideal that you can take any skill you want and have a viable character concept and not just fall back into classes-but-without-the-classes. Don't know if this problem is in other skill-based rpgs so I don't know, my experience might be biased.

nedz
2013-10-19, 06:20 AM
This is just the principle of specialisation.

A group of specialist characters, who between them cover all the bases, will be more effective than a group of generalists.

This is as true in real life as it is in most game systems. The type of system is irrelevant.

Wulfram
2013-10-19, 06:44 AM
I think specialisation is a good thing in a party based game.

Arcane_Snowman
2013-10-19, 06:44 AM
It's an interesting theory, although I can't say I agree with it.

Hyper-specialization is a matter of people generally being interested in being as good as possible at what they do, you see this in class systems as well, but most of the numbers are already given to them so there's less modification possible. With a class-less system the balance ends up a lot more dependent on the GM than with a class system, and as such it can more easily be skewed because it's a lot easier for the GM to make hyper-specialization necessary to succeed.

It's true that it's easier to be led astray by a point buy system than a class system as there are more options, but I also think that a lot of the time there's a severe disconnect between the benchmarks of the game, compared to what is possible to acquire for a starting character. More often than not, people end up making what could be equated to TO characters, the Face is a Diplomancer, the Warrior an Ubercharger etc.

With the two first examples you've given, I find that Exalted is a horrible mess system-wise and the attempts at rectification haven't really impressed me.

Shadowrun is a pretty big offender regarding my point, it's fairly easy to get caught up trying to give your character a ridiculously high bonus to your task of choice, because it doesn't give a great feel of what is good enough. And this can be really messed with if the GM isn't paying attention to benchmarks.

I cannot speak of Eclipse Phase, but I think discounting FATE seems a bit arbitrary, the skills have a pretty huge impact on the character's capabilities, even if the Aspects allow for narrative-based bonuses and you get rewards for out of the box application of them.

For examples of some pretty good class-less systems, I'd recommend Ars Magica and Legend of the Five Rings, the latter being somewhat more of a middle-ground between the two things, as you do have a "class" but are free to purchase abilities, skills etc. with next to no restrictions.

Rhynn
2013-10-19, 06:58 AM
It's a matter of degrees. Shadowrun 4E happens to give you less resources to work with, relative to the number of options, than any other edition; in Shadowrun 5E, the highest or second-highest skill priority let you max out your primary skill and then splurge on secondaries (e.g. you can have 10/15 skills at 5, with 6 the maximum to start with). Obviously, if you spend your high priorities on something else (metahumanity, magic), you'll have less for skills. On the other hand, 5E removed some of the limits 4E imposed (limiting you to 4 in skill groups, etc.).

In GURPS, in a 100-150 point campaign, you're pretty much bound to have enough points to get decent levels in a wide variety of skills, while still specializing in one thing (probably 1-3 skills, 1 attribute, maybe a few supporting advantages) over others.

In Aces & Eights, you can't even specialize in combat (except by taking some advantages, which doesn't count for nearly as much as gunfighting experience during play will), and buying skills gets more and more expensive; what you're good at is largely determined by your ability score rolls (although you can buy up the ones you like a little).

RuneQuest since Mongoose took over (so editions 4th through 6th) has limits on how much you can spend on one skill at character creation, so you are forced to generalize to some degree. Buying multiple combat skills (at least beyond two) is pretty much a waste.

Fading Suns does something similar to Eclipse Phase: you create your character by choosing packages for the stages of your life, which can be focused towards a particular thing (combat, social, etc.) but will always get you some "incidental" skills, giving you a broader skill base.


However, in the end, "specialists are better at their specialty than generalists" is a logical statement (as in "X = X"). It's the definition of the words. You're not really going to get away from it.

Grinner
2013-10-19, 07:00 AM
Yes, this is a common problem in game design. The question really is "How do you know what to set Difficulty Classes to?" or "How does the game deal with optimization?". When you set DCs so that a generalist could make it, a ridiculously optimized character, exploiting every rule possible, will breeze past them. If you then raise them to counter this, that will leave the generalist out of commission.

This solution is to avoid such mechanics altogether, but they're so deeply ingrained into tabletop gaming that people have difficulty adapting to anything else.

Interestingly, I've seen two games that mechanically reward balanced characters: Sufficiently Advanced (http://www.rpgnow.com/product_info.php?products_id=51719) and A Wanderer's Romance (http://www.stargazergames.eu/games/a-wanderers-romance/).

Rhynn
2013-10-19, 07:07 AM
This solution is to avoid such mechanics altogether, but they're so deeply ingrained into tabletop gaming that people have difficulty adapting to anything else.

Alternatively, you can limit specialization. A RQ6 character can't really specialize in any one skill beyond +50% from the starting value (or whatever it ends up at).

Shadowrun 5E uses Inherent Limits to limit your successes, and generally has cut down on dice pool bonuses (with more Limit increases), making boosting your dice pools both harder and less useful. The Inherent Limits also reward balanced attributes to a degree: they're figured by averaging attributes (although in each case, one attribute weighs double).

One way to reward generalization in play is a lot of individual challenges. In cyberpunk games, like Cyberpunk 2020 or Shadowrun, I always try to spread my characters' skills out so they have a lot of basic competency (combat, movement, social, driving, first aid, computers), because when I run such games, I regularly throw the PCs out of their comfort zones, split the party, etc. You can't just go "oh, we have a guy who does first aid" or "we have a guy who knows how to drive," and not just because odds are decent that guy's going to get shot at some point...


Interestingly, I've seen two games that mechanically reward balanced characters: Sufficiently Advanced (http://www.rpgnow.com/product_info.php?products_id=51719) and A Wanderer's Romance (http://www.stargazergames.eu/games/a-wanderers-romance/).

Can you elaborate on how they do it?

Grinner
2013-10-19, 08:17 AM
Can you elaborate on how they do it?

I can do my best.

A Wanderer's Romance literally has a Balance trait. When you first roll up a character, points are assigned to one of the character's four main traits: Earth, Fire, Water, and Air. When you make any roll, it will rely upon two of these traits. The Balance trait is equal to the highest number of elemental traits with the same rating. For example, having 3 traits rated at two each would give you Balance 3, while 2 traits rated at three each would give you Balance 2.

Balance itself is used primarily to determine initiative, but it also factors into a number of combat techniques.

Unfortunately, the game does have its share of observable flaws. For instance, if the GM lets you pick your traits (as opposed to rolling for them) and you dump all of your points into Fire, you'll end up with Balance 3 for having three traits rated at zero.



Sufficiently Advanced is a sci-fi game about the distant future. The problem with games about the distant future is that games about the distant future have distant future technology, and distant future technology is far superior to previous eras' technology. As such, high tech characters will tend to outclass low tech characters.

Not here.

Since immediate victory by the virtue of superior firepower would work at cross purposes with the game, the game has two different sets of mechanics. The first is a set of traits which grant the character simple competence in the form of training and technology. The second is a set of traits which allow the players certain narrative powers at cost of inflicting dramatic twists of fate upon themselves. Then there's the Import score, which determines how efficient their narrative powers are.

Anxe
2013-10-19, 09:52 AM
You seem to be focused primarily on the statistics side of the game. Will my character be useful in enough situations or not? I've always thought that skill based games were more about role-playing than roll-playing. You have an idea of what your character has done in the past and then you can exactly create that. As for feeling useful in game, you just have to work to create situations that match your character's skill set.

Example: Maybe the ex-baker's skill set won't be useful in a combat situation, but it'll be useful when the party has to pose as a catering company to get close to their assassination target.

jedipotter
2013-10-19, 10:29 AM
one problem I've noticed in particular is the maximal skill problem, in that you often have to get your skill up to maximum competency just for it to be effective at all. there is no room for generalists, sure you can make one but these are generally considered useless.

I always have a problem with players that think this way: I must maximize my skills to ''to effective'' and ''have fun''. It's another way of saying ''I must auto do all skill checks'', and it sure ruins games quick. This type of play quickly has a huge plus and can make most checks. But this makes problems as the character's can do anything. ''Oh, a 20 feet flaming pit..I leap over it and knit a sweater while I do that'' or ''I sweet talk the guard to let us pass and have him put me in his will''.

So I see the problem as the Auto win player way of thinking. As if they were to fail a skill check it would end game.

Scow2
2013-10-19, 10:48 AM
I always have a problem with players that think this way: I must maximize my skills to ''to effective'' and ''have fun''. It's another way of saying ''I must auto do all skill checks'', and it sure ruins games quick. This type of play quickly has a huge plus and can make most checks. But this makes problems as the character's can do anything. ''Oh, a 20 feet flaming pit..I leap over it and knit a sweater while I do that'' or ''I sweet talk the guard to let us pass and have him put me in his will''.

So I see the problem as the Auto win player way of thinking. As if they were to fail a skill check it would end game.
Well, the big problem I've seen, especially in D&D, are Scaling and Opposed DCs. The biggest offenders in D&D are Stealth, Perception, Intimidate, and, in-combat, bluff, and Pathfinder added Acrobatics to that list.

Jay R
2013-10-19, 11:10 AM
You've made some important observations, but I don't agree with all your conclusions about them.


The result is that skill-based rpgs seem to be in many ways, just unbuilt classes, with far more potential to choose bad choices that screw over the character, ...

Yes, absolutely. Free choice includes the ability to do something stupid, or you're really limited to the choices that the designers wanted you to take. The solution is not to make stupid choices. Or to make them, discover the mistake, pay for it with your character's life, and learn from the experience. Either way, have fun with the game.


...one problem I've noticed in particular is the maximal skill problem, in that you often have to get your skill up to maximum competency just for it to be effective at all. there is no room for generalists, sure you can make one but these are generally considered useless.

The use for a generalist, in my experience, is to be the one person who goes on any solo piece of the mission. My 2E mage/thief is not as powerful as the party's other magician, but is ideally suited for sneaking into the enemy encampment for spying and sabotage.

But yes, a generalist is overall weaker than the rest. That's what makes those characters a bigger challenge to play, and therefore more fun.


Sure its more customizable, but that just means there is more room to make bad choices and holes to fall into. I say this as someone who likes skill-based systems in theory, but find myself disappointed in how they work practically.

The ability to make bad choices is automatically included in any system that allows you to make choices that the designers didn't plan for. Therefore it's a feature, not a bug, since it is the necessary risk that accompanies your ability to come up with something completely new.


I don't know, I guess this is just a realization I've had and just wanted to share it. I guess what I'd want, going forward from this is to somehow find a way to make a skill-based system live up to its ideal that you can take any skill you want and have a viable character concept and not just fall back into classes-but-without-the-classes. Don't know if this problem is in other skill-based rpgs so I don't know, my experience might be biased.

"a skill-based system live up to its ideal that you can take any skill you want and have a viable character concept"

It depends on what you mean. In some games, I can develop any skill I want. But I had also better develop good defense and attack skills, to make a viable character. If you mean that, then we're in agreement. But if you mean a system in which any design is equally viable, I actively do not want this. Part of the game is character design, and it should matter. It should be something I can do well or poorly.

If I design the ultimate fighter pilot for a jungle crawl game, or the perfect wilderness survivor for a game of palace politics, it's my fault, not the system's. (If I design a character without knowing what kind of background I'll be in, it's the DM's fault. I give my players a multi-page world background before character design, with specific design hints.)

And it's important to realize that the game system isn't what's in the book. It's the combination of the book and the GM. I consider it part of the GM's job to make scenarios in which the characters can do well and survive and conquer (or do poorly and fail and die). If the characters can't breathe underwater, only a poor GM would design an underwater adventure. But by the same token, if all the characters can breathe underwater, it's a poor GM who runs all their adventures in the desert.

The most flexible system I know is Hero System (Champions, Fantasy Hero, etc.) Lots of people don't like it because character design includes a fair amount of basic math. I enjoy that, but many people don't.

For your purposes, it has the huge advantage of complete flexibility to design any kind of character - the system doesn't restrict your options. But it has the complementary drawback. You have the potential to make bad choices that screw over the character - the system doesn't restrict your options.

Slipperychicken
2013-10-19, 11:21 AM
I find that point-based systems are excellent when you already have a clear concept in mind (complete with strengths, weaknesses, personality, etc) and aren't so concerned about optimizing.

valadil
2013-10-19, 11:25 AM
Really? I've noticed the opposite problem.

When I play a specialist in a an a la carte skill game, I usually start with my specialty up pretty high if not maxed out. It might advance by a few points, but there's very little room for upward mobility when you start at the top.

In the meantime, the generalists have time to catch up with me. They'll pick up on the things that effective and buy them up. They'll get to 90% of my score and cut into my niche.

That's whinier sound than I meant it to be. What bothers me is that my specialist doesn't get better at his specialty, while the characters who aren't specialists do get better. It seems backward.

FWIW, I've mostly observed this in GURPS.

If you haven't already checked it out, Dark Heresy is an interesting compromise between class and point buy skill systems. Simply put, your class determines how much each skill costs and when they become available. You can still buy whatever you want to represent your character, it's just the price that will vary.

nedz
2013-10-19, 02:06 PM
Really? I've noticed the opposite problem.

When I play a specialist in a an a la carte skill game, I usually start with my specialty up pretty high if not maxed out. It might advance by a few points, but there's very little room for upward mobility when you start at the top.

In the meantime, the generalists have time to catch up with me. They'll pick up on the things that effective and buy them up. They'll get to 90% of my score and cut into my niche.

That's whinier sound than I meant it to be. What bothers me is that my specialist doesn't get better at his specialty, while the characters who aren't specialists do get better. It seems backward.

FWIW, I've mostly observed this in GURPS.

If you haven't already checked it out, Dark Heresy is an interesting compromise between class and point buy skill systems. Simply put, your class determines how much each skill costs and when they become available. You can still buy whatever you want to represent your character, it's just the price that will vary.

This was a feature in the old chaoism style games too. The higher your skill the harder it is to improve, consequently in a long running game everyone has high points in the skills most commonly used. You could only retain your niche if it really was niche, in which case it rarely mattered.

RochtheCrusher
2013-10-19, 02:43 PM
Sure its more customizable, but that just means there is more room to make bad choices and holes to fall into. I say this as someone who likes skill-based systems in theory, but find myself disappointed in how they work practically.

Well, how much satisfaction do you feel when you buy a chair, versus buying tools and lumber and building one? Sure, the one you buy is more likely to work (or win the game), but the one you made yourself makes you feel good when it works.

Now, obviously, regardless of the system your party will want everything. Someone has to pick the lock, and someone needs to be able to lie about what the first someone is doing. These may or may not be the same person, but you want to do both things well, because it sucks to not get the lock open and get caught doing it because you have too many ranks in swimming.

Unless your thief is in Venice... hmmm...

Anyway, I mostly see this as an issue for the DM to rectify, to make choices interesting. If you have a knight who is good with a sword and shield, but he's away from his friends and needs to figure out where they went... well, is he decent (not exceptional) at gathering information? Or tracking? If not, why not? What was his plan for when he fell out of his wheelhouse? Did he have one backup skill, like lying (specialization all the way down) or did he put five points in everything he thought was useless, just so he could tap dance really badly if that ever came up?

Or, did he give all that up for an extra point of swordsmanship, which is bloody useless now that he's lost?

Of course, if your DM doesn't challenge you like this, I think he's missing an opportunity in a skill system. After all, a D&D Paladin can't train in Bluff, so it seems unfair to punish him for neglecting to... but in a skill based game? It's much easier to say, "Whadda you mean you didn't take Freefall?! I don't care that you're a hacker, the game's in space for crying out loud!"

Lord Raziere
2013-10-19, 02:44 PM
ok ok, cool, all of that is good information to know, glad to know that I'm not the only who thinks this BUT.

I'm also glad that this isn't the only problem, and that there can be an opposite problem as well, all good to know.

as for Dark Heresy and other WH40k games….Eh…..I consider them a class system. thing is class and skill based systems often are not "pure" they often have both, so I mostly consider any game that has very strong class-like stuff to be a class-based one even if they have skills attached to them. (Exalted I don't count because its castes are VERY broad and open-ended in concept, and not a very linear progression of power, and can be vastly different in how they play, even in the same caste.)

and yea I sort of dislike auto-win to. its not about the destination that is victory, it is the journey towards it...

SassyQuatch
2013-10-19, 03:23 PM
And here I thought that we were going to have a discussion about a Beast Wars RPG.

Disappointing.:smallfrown:

Ravens_cry
2013-10-19, 03:24 PM
Any game you can customize meaningfully is going to suffer from this to a greater and lesser degree.
'Tis the nature of the beast.
While some games might have mechanics that make this more difficult, for example making it cost more to level up a skill the higher it progresses, a specialist in their chosen field is going to be more competent in that field than a generalist. Frankly, the strikes me as only fair. In fact, I would find a game that didn't reward specialization to at least some degree to be downright unfair. I spent these points/levels/feats, shouldn't I get something out of it?

Scow2
2013-10-19, 03:27 PM
And here I thought that we were going to have a discussion about a Beast Wars RPG.

Disappointing.:smallfrown:You're not the only one.

Ravens_cry
2013-10-19, 03:35 PM
You're not the only one.
Surely there has been a Transformers RPG . . . right?:smallconfused:
Right?:smalleek:

SassyQuatch
2013-10-19, 03:51 PM
Surely there has been a Transformers RPG . . . right?:smallconfused:
Right?:smalleek:
The license has never been signed out. There's a few fan-made games and a couple of attempts to work them in to other games with the serial numbers filed off, but never anything official.

Ravens_cry
2013-10-19, 04:00 PM
The license has never been signed out. There's a few fan-made games and a couple of attempts to work them in to other games with the serial numbers filed off, but never anything official.
There was a a innovative Ghostbusters RPG, a TMNT (based on the original comic) RPG, but no Transformers RPG?
What is wrong with the world!?

Rakaydos
2013-10-19, 04:04 PM
In Edge of Empire, it is pretty easy to make a dedicated specialist- just be a Droid. Given the way their character buildin rules work, only droids get the build points to really specialize heavilly in one thing- while every other race gets "free stats" that automatically diversifies them more than a droid ever could, even if they tried spending their greater number of points for it.

Which, really, just means that they're purpose built machines. A battle droid can shoot... but they've got the Intellect, Cunning, Willpower, Presence and Brawn of, well, a battle droid. Find me an organic who knows 6 million forms of communications, and I'll get on threepios case for waddling awkwardly around with a Brawn, Agility, and Willpower of 1.

And if the specialist is causing problems with encounter balance... "We dont serve your kind here, Droid."

The Grue
2013-10-19, 04:46 PM
In response to those who have argued that specialization is an inherrently good thing:


A human being should be able to change a diaper, plan an invasion, butcher a hog, conn a ship, design a building, write a sonnet, balance accounts, build a wall, set a bone, comfort the dying, take orders, give orders, cooperate, act alone, solve equations, analyze a new problem, pitch manure, program a computer, cook a tasty meal, fight efficiently, die gallantly. Specialization is for insects.

-Robert A. Heinlein


I tend to apply this to RPG characters as well. For a D20 example, look no further than the Factotum. :smallcool:

NichG
2013-10-19, 05:07 PM
The big culprit behind this is systems that assume that boiling everything down into a single roll is sufficient to make the game a rich experience. That basically has a couple strong consequences on the kind of play that materializes, and its why specialists are so important for those games:

- Any opposed check, the person who has put more into their numbers will win that contest. If winning that contest means they get to do something bad to you or your own plan fails entirely, you have to put a lot of resources into winning that one critical roll.

- Even for unopposed checks, that skill number is basically the thing that decides 'how well will the party do' in the particular situations dominated by the skill. It becomes a game of Go Fish. If there's a DC 30 door, you just want someone to have 20 ranks in Lockpicking, rather than everyone having 5.

Now, specialization is good in a lot of ways - it protects player niches and prevents people from being marginalized. But at the same time, 'I want bigger numbers' isn't very interesting game design.

My suggestion is that rather than have skills be primarily about the roll, they should primarily be about unlocking special distinctive abilities at different intervals. That way if these abilities interact with one another, there's a choice that the player must make about 'I want at least 3 ranks of this so I can get this one power, but then 5 ranks of this other thing, etc'. Thats kind of how most of the special powers work in World of Darkness games, for example, and it makes those powers a lot more interesting than the base skills.

Thats easy to do when skills are magic - you're creating new powers for magic anyhow. What's really needed is a set of powers for common skills. What distinctive thing can you do if you're 'this good' at lockpicking? Skyrim's perks are actually a decent model for this.

I tried to do this in the system for my current campaign, with 'waza' that you can get access to at every 10 ranks of a skill. For example, at 10 ranks you can learn the key patterns/combinations of things you pick; at 20 ranks you can pick locks without tools; at 30 ranks you can pick locks that aren't actually locks, such as timer mechanisms, a person's emotional walls, or a prophecy about the special characteristics of the person allowed to enter a certain area. For perception, at a certain number of ranks you can still avoid attacks or threats when you're asleep. For etiquette, at a certain number of ranks you can take back something someone else said. And so on.

jindra34
2013-10-19, 07:05 PM
NichG: All that is true assuming its a die+modifier versus situation. For roll under (e.g. roll less than or equal skill+/-modifiers, and margin mattering) or roll xdy count successes it tends to alter how far one pushing one skill will get you. And add in the person running it knowing how to throw a curve ball (having players not get a chance to decide who makes the roll is a good example), and when, and you can have hyper specialization can end up meaning disaster.
So yes its very much an issue with system simplicity and the person running it playing to the party all the time, than it is a facet of the over arcing scheme.

nedz
2013-10-19, 07:25 PM
In response to those who have argued that specialization is an inherrently good thing:



A human being should be able to change a diaper, plan an invasion, butcher a hog, conn a ship, design a building, write a sonnet, balance accounts, build a wall, set a bone, comfort the dying, take orders, give orders, cooperate, act alone, solve equations, analyze a new problem, pitch manure, program a computer, cook a tasty meal, fight efficiently, die gallantly. Specialization is for insects.

I tend to apply this to RPG characters as well. For a D20 example, look no further than the Factotum. :smallcool:

That quote is normally used to define the Trope of Competant Characters, which most of Heinlein's protagonists adhere to.

It is also a style of play which is well represented on CharOp boards, perhaps over representatively.

Basically it assumes that the Protagonists should succeed at most tasks. I find Characters with weaknesses to be more interesting; YMMV.

The Grue
2013-10-19, 07:32 PM
That quote is normally used to define the Trope of Competant Characters, which most of Heinlein's protagonists adhere to.

It is also a style of play which is well represented on CharOp boards, perhaps over representatively.

Basically it assumes that the Protagonists should succeed at most tasks. I find Characters with weaknesses to be more interesting; YMMV.

Your response assumes a false dilemma; namely, that a character competent at a wide range of tasks has no weaknesses. "Competent at a lot of things" is not the same as "competent at all the things", nor is "competent" the same as "succeeds all the time".

TheCountAlucard
2013-10-19, 07:32 PM
Luckily for my Exalt, most of those default to Intelligence, and Exalts don't take the usual penalty for untrained skill rolls. :smallamused:

NichG
2013-10-19, 09:54 PM
NichG: All that is true assuming its a die+modifier versus situation. For roll under (e.g. roll less than or equal skill+/-modifiers, and margin mattering) or roll xdy count successes it tends to alter how far one pushing one skill will get you. And add in the person running it knowing how to throw a curve ball (having players not get a chance to decide who makes the roll is a good example), and when, and you can have hyper specialization can end up meaning disaster.
So yes its very much an issue with system simplicity and the person running it playing to the party all the time, than it is a facet of the over arcing scheme.

If the margin matters that also adds to the 'high numbers are the only point' situation. You can play with diminishing returns and such, but I think that tends to just be frustrating to people who really do want to specialize.

I still think the real answer is to make it so that if you want to specialize in swords, the optimal way to be good at swords involves forays into other skills that complement your swordplay in subtle ways. For example, again in my current campaign:

There is a Knife waza that lets you basically get a free reroll on attacks when attacking an unaware target. There is a Trickery waza that lets you get a surprise round at the start of fights by winning an opposed roll. If you want to be 'the best knife dude ever' its still in your interest to take a side-trip and boost your Trickery. I could make a Tactics waza that also complements the knife tricks, in which case you might decide to be a Trickery-Knife guy or a Tactics-Knife guy. If there are enough ways to do each thing, you end up with a rich system.

Look at what makes character building so rich in D&D 3.5. All the various classes interact in weird ways when you get to the PrC level, and the feats interact with all the classes and so on. Even though its a 'class' system, you're basically encouraged to dip this or that, not because you're shoring up weaknesses, but because the dips synergize in particular ways. That makes the character building process a lot more interesting.

erikun
2013-10-20, 12:30 AM
I think it really depends on the game. Earlier skill-based games had pretty much unlimited freedom, and as such you could easily end up with the blind one-armed deaf mute crippled sniper who could headshot anything. Some newer games took this and made it dysfunctional in the opposite direction: You are so limited in options during character creation that you either fit into one of the few identified "good" builds or you end up in unable to accomplish much. Class-based games have the same problems, with early games running into problems where the classes were limiting and giving you few options, while newer games give high versatility, but with a lot of those versatile options being weak to useless. D&D3e is a good example, where a spellcaster should either stick with just being a single-class spellcaster or not bother. Multiclassing kills the ability of spellcasting to stay relevant.

That said, I'm not really running into many skill-based games that have your problem. HeroQuest, RuneQuest, Burning Wheel, and World of Darkness don't seem to really run into the issue. They all have their own way of dealing with it, from character generation rules to making such extreme specialization a hazard.

nedz
2013-10-20, 06:04 AM
Your response assumes a false dilemma; namely, that a character competent at a wide range of tasks has no weaknesses. "Competent at a lot of things" is not the same as "competent at all the things", nor is "competent" the same as "succeeds all the time".

You haven't read much Heinlein have you ?

Also Most != All

Tyndmyr
2013-10-20, 09:19 AM
I think I've noticed something in skill-based rpgs. Specifically, they might claim to be more flexible about the character creation process, but all the ones I have tried so far (Exalted, Shadowrun) seem to default to a class-like arrangement. you often end up with a bunch of hyper-specialized characters anyways despite their talk of having any skill you want and being open-ended.

Try 7th Sea instead. Yeah, a degree of specialization does exist, in that not every char is going to be equally good at everything, but literally anyone CAN pick up a sword and fight to some degree, and costs are structured such that hyper-specialization is very expensive. The first rank in a skill costs a mere 2 xp, while the fifth costs 10. So, yeah, you can choose to specialize, but in practice, many people don't, because not specializing works well and can be advantageous.

In this particular respect, it's also somewhat realistic. A day of training will advance someone entirely new to a given topic more than the same length of time will advance the master.

Aasimar
2013-10-20, 10:19 AM
I think the problem is that DC are often arbitrarily decided to provide the challenge, and they, by necessity, go by the maxed-out standard.

So, by the time you are say, level 10, someone with a maxed out skill will succeed at the really difficult stuff on a pretty regular basis (sometimes only 50/50, based on the system) while someone who only dabbles reaches a point where he cannot succeed or only succeeds 10-20% of the time.

Instead, we need to make it a tad less gamist and a tad more simulationist.

Sure, maxing out your skills that are only judged by opposed comparison to others, such as combat skills (in a skill based system) will probably always need to be maxed, at least by most players, in most imaginable systems.

But other skills, say, computer use, Spellcraft, knowledge, etc. They should reach practical limits.

At some point, more ranks should face a diminishing rate of returns, say, at 15 ranks in a knowledge skill you know everything that's truly relevant in the field and more ranks will only make you more aware of fringe theories, stuff that's widely rejected, etc.

At some point, your skill with computers has reached the limits of how useful it is to your spy character, he'll become as good as possible at on the run hacking and usage of the tools of his trades. More ranks would be useful if he were going to go work in an A.I. lab or designing his own operating system, but not really relevant to an action hero career.

jedipotter
2013-10-20, 10:33 AM
So, by the time you are say, level 10, someone with a maxed out skill will succeed at the really difficult stuff on a pretty regular basis (sometimes only 50/50, based on the system) while someone who only dabbles reaches a point where he cannot succeed or only succeeds 10-20% of the time.

The problem I see is: How much of a chance of failure should an action have. Most players like the ''less then 10% failure''. They will admit they want to fail ''kinda sometimes'', but never when it is important. So it is ok to fail jumpping over a mud puddle, but when you jump attack the lich king you must automatiacly make that one.

DM's like me(haha, there are none...) like to keep a more 50% chance of failure.

Rakaydos
2013-10-20, 11:05 AM
The problem I see is: How much of a chance of failure should an action have. Most players like the ''less then 10% failure''. They will admit they want to fail ''kinda sometimes'', but never when it is important. So it is ok to fail jumpping over a mud puddle, but when you jump attack the lich king you must automatiacly make that one.

DM's like me(haha, there are none...) like to keep a more 50% chance of failure.

That's the gamist/narrativist approach.The simulationist approach is "The DC to climb rough stonework is X. Make your characters appropriately."

Rhynn
2013-10-20, 11:30 AM
At some point, more ranks should face a diminishing rate of returns, say, at 15 ranks in a knowledge skill you know everything that's truly relevant in the field and more ranks will only make you more aware of fringe theories, stuff that's widely rejected, etc.

This is why I love 3D6 resolution (and use it in my homebrew). GURPS exemplifies its use best. You roll 3D6 trying to get under or equal to your skill. (17-18 is automatic failure, although IIRC after skill X only 18 is.) That means that past skill 11, every +1 to your skill is worth less and less. The difference between 10 and 11 is +12.5% chance of success, but the difference between 15 and 16 is +2.78%. However, high skill makes you more resistant from negative modifiers (i.e. makes you better at harder tasks). At skill 10, a -1 modifier reduces your chance of success by 12.5%. At Skill 16, a -2 modifier reduces your chance of success by 7.41%.

Combine this with the constantly rising cost of increasing your skills (doubling up to a certain point, then rising linearly) and GURPS strikes IMO the perfect balance between rewarding specialization and limiting it to reasonable levels.


That's the gamist/narrativist approach.The simulationist approach is "The DC to climb rough stonework is X. Make your characters appropriately."

This is what I prefer. This way, the PCs' skills describe their ability to affect the world, and the players can make deductions and predictions about their chances (and when those predictions are inaccurate, they can intuit there's a reasonf for it).


Incidentally, Legend of Five Rings does a few of the things mentioned earlier to balance things out. Your attributes are divided into rings (two each in Fire, Air, Earth, Water; Void is different); the value of your ring is equal to the lowest attribute in that ring. Many special abilities, etc., use the ring value, rather than an attribute value. Also, in 3rd and 4th edition, skills give you Mastery Abilities at certain levels (e.g. Etiquette 3 increases the Target Number of contested Social tests against you by 5; Defense 3 lets you add your Defense skill to your TN to be Hit in almost all circumstances, etc.). So, balancing your attributes is rewarded, and skills serve functions other than flatly determining your chance of success at something.

jindra34
2013-10-20, 12:24 PM
This is why I love 3D6 resolution (and use it in my homebrew). GURPS exemplifies its use best. You roll 3D6 trying to get under or equal to your skill. (17-18 is automatic failure, although IIRC after skill X only 18 is.)


Nope 17 is always a failure. If your (final, adjusted) skill is less than 16 its also a critical failure. Though to people worried about degree, In Nomine (also [at least in english] published by SJ Games) uses a similar 2d6 method (12 [except in 1 situation] always failing 2 [again except in 1 situation] always succeeding) with a third d6 thrown in to compare how well you did, though in some cases (such as skill of 13 and 14, and songs [essentially spells] that you know all three forms of) you can add a +1 or+2 to it.

Slipperychicken
2013-10-20, 01:08 PM
The big culprit behind this is systems that assume that boiling everything down into a single roll is sufficient to make the game a rich experience.


I think that game-writers do this more to make play simpler, rather than richer. Often times, a gaming group will want to resolve their characters' actions quickly so they can maintain excitement and interest. They will want to keep the number of OOC actions and calculations to a minimum, since many gamers would rather not be bogged down in the tedium of excessive rolls and number-crunching. To this end, they will try to make any given resolution simple: see a single number on your sheet, make a roll based off of it, then compare to another number which represents the task's difficulty.


I guess what I'm trying to say is that there's a tradeoff between ease-of-play and the "richness" we describe here. A simpler system might not model reality as much as we'd like, and a more accurate ("rich") system might be too cumbersome.

jedipotter
2013-10-20, 01:16 PM
That's the gamist/narrativist approach.The simulationist approach is "The DC to climb rough stonework is X. Make your characters appropriately."

I don't understand the simulationist approach. Are you saying you make a new character every time you wish to take an action in the game? If the GM says ''to jump over the pit is X'', you make a character to do ''X''?

Or is it like the GM says ''well there are trees in the world, and you might want to climb one. So it takes ''X'' to climb a tree, make your character appropriately.'' Or ''some of the water in the world is more then six inches deep in places known as 'lakes' and 'oceans'. It takes ''X'' to swim, so make your character appropiately."

Rakaydos
2013-10-20, 01:20 PM
Or is it like the GM says ''well there are trees in the world, and you might want to climb one. So it takes ''X'' to climb a tree, make your character appropriately.'' Or ''some of the water in the world is more then six inches deep in places known as 'lakes' and 'oceans'. It takes ''X'' to swim, so make your character appropiately."

Correct. You make your character with the understanding of what they can do, and that sometimes "good enough" is all the specialization you need.

Lamech
2013-10-20, 01:38 PM
This is why I love 3D6 resolution (and use it in my homebrew). GURPS exemplifies its use best. You roll 3D6 trying to get under or equal to your skill. (17-18 is automatic failure, although IIRC after skill X only 18 is.) That means that past skill 11, every +1 to your skill is worth less and less. The difference between 10 and 11 is +12.5% chance of success, but the difference between 15 and 16 is +2.78%. However, high skill makes you more resistant from negative modifiers (i.e. makes you better at harder tasks). At skill 10, a -1 modifier reduces your chance of success by 12.5%. At Skill 16, a -2 modifier reduces your chance of success by 7.41%.

Combine this with the constantly rising cost of increasing your skills (doubling up to a certain point, then rising linearly) and GURPS strikes IMO the perfect balance between rewarding specialization and limiting it to reasonable levels.

I've become a fan of the 3d6 system actually. Bell curves are cool. The one thing I have noticed about GURPS is sometimes when I build a character if I scan for point optimization I get "I should drop all my skills a level and raise IQ/DX/HT". Which can end up forcing you towards a generalist build skill wise. Which is a bit weird.

ANyway, plugging GURPS: you can get countless types of characters with there system. There is a lot of freedom.

Rhynn
2013-10-20, 02:24 PM
I've become a fan of the 3d6 system actually. Bell curves are cool. The one thing I have noticed about GURPS is sometimes when I build a character if I scan for point optimization I get "I should drop all my skills a level and raise IQ/DX/HT". Which can end up forcing you towards a generalist build skill wise. Which is a bit weird.

That will happen if you're going for a very wide spread of skills at high levels, yeah. If you just want basic competency (˝ or 1 points, depending on edition) it doesn't happen (since the minimum bought level is usually 2-4 higher than the default level, if the skill even has a default).

I find it happens less with 50-150 point characters, but with high point-level characters with wide competencies, minimum points in skills and high ability scores are much better. Although in 4th edition, for instance, IQ and DX cost double what ST and HT do; so you'd need 20 skills at 1 point for IQ to be a better deal than increasing each skill by 1 point (or 10 skills, since you're also getting 5 points worth of Will and 5 points worth of Per, assuming you care about Will and Per).

It makes sense to me, too: if you want to be Leonardo da Vinci and great at everything intellectual you try, then a high IQ and low points in IQ skills are the way to go, both mechanically and thematically; with your prodigious brain, you learn everything quickly.

The Grue
2013-10-20, 02:37 PM
You haven't read much Heinlein have you ?

Also Most != All

I have, but I personally am not Heinlein, so the specifics about how he writes characters do not apply to me. I expressed agreement with the sentiment that "specialization is for insects"; nowhere did I say that the only interesting character is a Mary Sue.

As to your second point...that was kind of my point. I'm not sure what you're getting at with it.

EDIT: Ah, I think I understand now. I seem to have misunderstood you earlier; when you said this,


Basically it assumes that the Protagonists should succeed at most tasks. I find Characters with weaknesses to be more interesting; YMMV.

I assumed it to mean that you believed a character who is good at a finite number of things has no weakness, and is therefore good at everything. But by your more recent post, you would seem to not actually believe this. So we're actually in agreement here: a character who is good at a finite number of things does have weaknesses and can be interesting.

jindra34
2013-10-20, 02:41 PM
I find it happens less with 50-150 point characters, but with high point-level characters with wide competencies, minimum points in skills and high ability scores are much better. Although in 4th edition, for instance, IQ and DX cost double what ST and HT do; so you'd need 20 skills at 1 point for IQ to be a better deal than increasing each skill by 1 point (or 10 skills, since you're also getting 5 points worth of Will and 5 points worth of Per, assuming you care about Will and Per).
1. Thats assuming your improving from the 1 point level, at the 2 point level it halves, and at the 4+ point level its a quarter that.
2. Honestly most people I've played with have removed Per and Will boost from IQ (and everyone in magery based games 10(magery up cost)+5(per)+5(will)=20... and a mage will most definitely have more than 5-10-20 IQ based skills, seeing as every single spell is a seperate one).

Lamech
2013-10-20, 03:26 PM
Not having disadvantage point caps screws everything up. HT is a particularly bad offender. "Okay, I buy HT, and I add easy to kill, -1FP, and -.25 Speed, and now I have +1 to sex appeal, also pretty much every form of resistance ever in a mundane setting." Will isn't quite as bad, but its cheaper to buy will+fearfulness than increasing will skills.

Skill generalization goes off the deep end in supers games when someone buys "modular abilities: cosmic reduced time." and now they have every skill ever. (Of course its a supers game so its not as bad since your powers separate you not your skills.)

Rhynn
2013-10-20, 03:33 PM
Not having disadvantage point caps screws everything up.

Disadvantage caps are pretty essential, and counting attribute penalties/negatives (at least past -1 or -2 to any one) against them is pretty necessary for balance, yeah.

And yes, building a basic GURPS mage in 3E required pretty much making sure ˝-point got you 15 in Very Hard spells (or 20 if you were in a higher point-level game).

NichG
2013-10-20, 04:55 PM
I think that game-writers do this more to make play simpler, rather than richer. Often times, a gaming group will want to resolve their characters' actions quickly so they can maintain excitement and interest. They will want to keep the number of OOC actions and calculations to a minimum, since many gamers would rather not be bogged down in the tedium of excessive rolls and number-crunching. To this end, they will try to make any given resolution simple: see a single number on your sheet, make a roll based off of it, then compare to another number which represents the task's difficulty.


I guess what I'm trying to say is that there's a tradeoff between ease-of-play and the "richness" we describe here. A simpler system might not model reality as much as we'd like, and a more accurate ("rich") system might be too cumbersome.

Its not really about accuracy though. Accuracy doesn't automatically make something richer. Its more about the mechanics creating interesting tactics/strategy versus those things being primarily in the hands of the DM to evaluate.

Even in your post, there's a perpetuation of this fallacy. The idea that 'rich' must mean 'number crunching'. Richness isn't about having a final number that comes from more places before you do the binary comparison, its about having mechanically meaningful results beyond just 'did I succeed or fail?' and mechanically meaningful enabling of this or that course of action.

Compare the bare bones of D&D combat with something like FATE.

In a game like FATE, mechanically it doesn't matter at all what you do. Any aspect is mechanically identical, from 'I am the President of the World' to 'I like knives' to 'Who says the buddha can't fight?'. Any differentiation between these things is due to the GM saying 'okay, this is appropriate/inappropriate for the situation' or calling out extra things that happen because of the logic of what you did.

In D&D, even putting aside things that alter your numbers, there are a lot of things that 'matter' above and beyond your final roll:

- Your position on the field matters so things that move or enemies or control your movement have mechanically important effects. Flight, burrow speed, ability to avoid AoOs, ability to bypass barriers or rough terrain, reach, etc all become mechanically meaningful hooks that have nothing to do with the binary pass/fail moment.
- Spellcasters are sort of the grand example. They have laundry lists of abilities that all can do different things than just 'attack the enemy'.
- Abilities that have some form resource requirements (from immediate actions to power points to spell slots) enable manipulation of those resources as a mechanical hook - you can refresh your maneuvers, steal power points from an enemy, etc.

None of these things really require complex number crunching or long series of rolls. But they make the game tactically richer, which means that a decision to 'max out swords' no longer becomes the best choice since there are steps that you have to take to be able to use your sword effectively (get up to the enemy, pierce their defenses, etc).

The idea would be to have the same sort of thing in play with other skills as well. Avoid things where the answer boils down to a pass/fail. Instead of 'do you manage to climb the tree or not?', you can instead ask things like 'How fast can you climb the tree? Can you do other things while climbing? How vulnerable does climbing make you?' and focus on the consequences of climbing the tree poorly or well, rather than on whether you can actually do it or not.

Slipperychicken
2013-10-20, 05:25 PM
Even in your post, there's a perpetuation of this fallacy. The idea that 'rich' must mean 'number crunching'. Richness isn't about having a final number that comes from more places before you do the binary comparison, its about having mechanically meaningful results beyond just 'did I succeed or fail?' and mechanically meaningful enabling of this or that course of action.

I had tried to get at the idea that adding more stuff like that (i.e. more rules for different skills in more situations) increases the time and energy a gaming group needs to spend to figure it out, whether it's a table of modifiers or a paragraph of text.

I'm not trying to say it's bad to include more detailed rules, like a degrees-of-success kind of thing (a number of games already do that). It's just that one ought to take into account that it might make action-resolution too lengthy or complicated for gaming groups, especially those which don't want to worry about how seemingly-inane in-game details impact events.



The idea would be to have the same sort of thing in play with other skills as well. Avoid things where the answer boils down to a pass/fail. Instead of 'do you manage to climb the tree or not?', you can instead ask things like 'How fast can you climb the tree? Can you do other things while climbing? How vulnerable does climbing make you?' and focus on the consequences of climbing the tree poorly or well, rather than on whether you can actually do it or not.

I briefly played Shadowrun 5e, and it has things like this to an extent.

The closest thing to "how fast can you climb the tree" in there (obviously, you don't have much tree-climbing in a dystopian future-city) is the sprinting rules, where you make a roll and the number of successes/hits is used to add to meters ran that round. I could see a similar roll being used with Gymnastics to determine how fast one moves up a tree. Similar rules exist for other skills like first aid and medicine. It also has the potential to roll "glitches" on a skill, which basically give the DM the chance to add complications (i.e. jammed guns, damaged equipment, police showing up to black market transactions, slipping and falling, dropping important things, etc) which make the PCs' lives difficult.

Rakaydos
2013-10-20, 06:10 PM
One of the reasons I like the Cardinal ruleset used in Ironclaw is that a single simple roll has a lot of information in it.

It's a dice pool system with various sized dice based on proficency levels- so a weakling would be rolling a d4 for anything that uses strength, in addition to skills.

So a swordmaster might be rolling d6 Body, D8 melee combat from a carear of Swordmaster (which also applies to two other skills, like Dodge and Presence), A d6 from two skill points in melee combat, and a bonus d12 for taking the Aim action with the Veteran gift. Roll them all at once (2d6,d8,d12), either opposed to the enemy' skill (whoever gets higher gets successes equal to number of dice higher than the other person's highest) or vs 3 (rolls higher than 3 are sucesses)

This can lead to situaltions where your 2d12 and d8 all roll 1, but your D4 succeded- remembering where your dice come helps you describe the situation.

NichG
2013-10-20, 06:30 PM
I had tried to get at the idea that adding more stuff like that (i.e. more rules for different skills in more situations) increases the time and energy a gaming group needs to spend to figure it out, whether it's a table of modifiers or a paragraph of text.

I'm not trying to say it's bad to include more detailed rules, like a degrees-of-success kind of thing (a number of games already do that). It's just that one ought to take into account that it might make action-resolution too lengthy or complicated for gaming groups, especially those which don't want to worry about how seemingly-inane in-game details impact events.


I think the answer there is to make things bite-sized, conceptual rather than numerical, and mostly in the players' hands. That means its the player's responsibility to know specifically what their abilities can do, rather than something where the situation makes everyone stop and look things up (like 'oh, what are the modifiers for fighting underwater ?').

Since I enjoy giving these examples, I'll give more from my current campaign. There's a pool called 'Leverage' that anyone can take to apply to any roll against an enemy. Leverage can be used not just in a fight, but also in any sort of conflict - negotiation, trying to find someone who is hidden, etc.

So for example, one of the 'waza' for the Psychology skill is 'you can manipulate the enemy's psychoses in order to generate Leverage'. It won't come up unless there's a player or enemy with that waza, in which case its more like a 'power' than a random consequence you have to look up. There are other such 'waza' for skills like Trickery (perform a feint), Perception (notice a weakness), etc, that can also generate leverage, apply status conditions, etc.

I don't think this makes the game incredibly more complicated, but it does make tactics a bit more complex (though not to the degree of D&D combat) since in some cases, the DC to achieve something could be impossible to hit under the skill caps in the system without applying points from the Leverage pool.



I briefly played Shadowrun 5e, and it has things like this to an extent.

The closest thing to "how fast can you climb the tree" in there (obviously, you don't have much tree-climbing in a dystopian future-city) is the sprinting rules, where you make a roll and the number of successes/hits is used to add to meters ran that round. I could see a similar roll being used with Gymnastics to determine how fast one moves up a tree. Similar rules exist for other skills like first aid and medicine. It also has the potential to roll "glitches" on a skill, which basically give the DM the chance to add complications (i.e. jammed guns, damaged equipment, police showing up to black market transactions, slipping and falling, dropping important things, etc) which make the PCs' lives difficult.

Yeah, these are all mechanical hooks you can use to make things a bit more rich. The trick is to make the questions 'how far did he sprint? how high did he climb?' meaningful. Its pointless to ask for a bunch of sprinting rolls when there's nothing that cares how far the player moved this round (e.g. if you're doing a 20 round race, by that time you might as well take average unless there are complications).

Similarly, if you're climbing a tree to get an overview of the countryside, there are no penalties for failure and how fast you do it is irrelevant, so why even have the roll? In those cases, what I would say is a 'you must be this tall to do X thing' would be a good mechanic. If you want to be able to climb a tree, you can do so if you have at least 1 rank of Climb. If you don't have the rank, you cannot ever succeed; if you do have the rank, you can auto-succeed when there's no risk.

Slipperychicken
2013-10-20, 10:02 PM
Yeah, these are all mechanical hooks you can use to make things a bit more rich. The trick is to make the questions 'how far did he sprint? how high did he climb?' meaningful. Its pointless to ask for a bunch of sprinting rolls when there's nothing that cares how far the player moved this round (e.g. if you're doing a 20 round race, by that time you might as well take average unless there are complications).

Similarly, if you're climbing a tree to get an overview of the countryside, there are no penalties for failure and how fast you do it is irrelevant, so why even have the roll? In those cases, what I would say is a 'you must be this tall to do X thing' would be a good mechanic. If you want to be able to climb a tree, you can do so if you have at least 1 rank of Climb. If you don't have the rank, you cannot ever succeed; if you do have the rank, you can auto-succeed when there's no risk.

There are plenty of situations (even in Shadowrun) where it matters how far you move in sprinting. For example, when you're running from an enemy (like the cops, or a robotic guard-dog), or trying to run someone down, those sprinting checks matter a lot.

As for climbing a tree, there are a number of risks. A Shadowrun-style glitch might result in your grip sliding, requiring another check to avoid falling or dropping your stuff. A critical glitch might result in damage as you fall out of the tree, hit every branch on the way down, and maybe piss off a nearby hive of wasps to boot.

Rhynn
2013-10-20, 10:25 PM
There are plenty of situations (even in Shadowrun) where it matters how far you move in sprinting. For example, when you're running from an enemy (like the cops, or a robotic guard-dog), or trying to run someone down, those sprinting checks matter a lot.

As for climbing a tree, there are a number of risks. A Shadowrun-style glitch might result in your grip sliding, requiring another check to avoid falling or dropping your stuff. A critical glitch might result in damage as you fall out of the tree, hit every branch on the way down, and maybe piss off a nearby hive of wasps to boot.

This sort of thing is either relevant or irrelevant depending on the style of the game. If I'm running HeroQuest with Big Damn Mythic Heroes, climbing a tree is trivial and not worth a roll; climbing The Tree in the Otherworld would be (or climbing a tree trying to pick the moon out of the sky, for instance...).

In the type of game I'm more partial to, any task that involves risk is worthy of a roll, because big failures create stories: "we're walking through some wilderness" becomes "we're in the middle of the wilderness and one of us needs medical attention, bad" without the GM declaring it out of nowhere. It's sort of an attrition/resource management deal, where accidents during mundane tasks can make important tasks down the road harder. (It's also pretty basic stakes-raising or tension-increasing seen in movies, TV shows, etc.)

But I don't think NichG was really disagreeing with anything you said: if the task is trivial or the degree of success irrelevant, why bother rolling?

NichG
2013-10-20, 11:05 PM
Hm, I'm realizing that my point is very hard to condense down here. Basically I'm not agreeing or disagreeing, I'm trying to point out a distinction between games where the question is phrased 'did you succeed/fail' and games where the question is 'how can you adjust the overall situation using tools in your belt?'.

So for sprinting, rather than ask the question 'do you get away from the cops or do you get arrested?', it feels like it would be better to do something like create a scenario where there's a sniper on a building 500 feet away, and you need to get into the building. With good sprinting, you can close that distance in two rounds instead of three or four, thus taking fewer risks of being shot. It modifies an aspect of the scenario (time exposed to fire). You could, at the same time, resolve the scenario by using existing cover, making a lot of smoke, or just shooting back from 500ft away.

The thing is, this is hard to connect back to the original question of 'the maximal skill problem'. For that I kind of need an example of how skills can synergize with one another to help in a single task. A very simple example would be something like an attack that gives you a damage bonus based on how far you moved that round; it basically means that to optimize an 'outcome' (dealing lots of damage), you can't just go all-in on the 'hit things with swords skill'; instead, by using a combination between skills, you can be better at damage than the guy who just put all his xp into 'hit things with swords'.

I guess my point is that step 1 for constructing those synergies is to create other things in the game environment that are meaningful to the success/failure of the primary question. So if Sprinting just determines 'do you get away or not?' then its only meaningful when the question in the scenario revolves around sprinting. But if Sprinting creates a concrete 'thing you can do' in the world, then you can use that thing as a player to address the entire scenario.

Is that more clear? I'm still not sure this is the best way to put it, so sorry for any misunderstandings.

Slipperychicken
2013-10-20, 11:50 PM
The thing is, this is hard to connect back to the original question of 'the maximal skill problem'. For that I kind of need an example of how skills can synergize with one another to help in a single task. A very simple example would be something like an attack that gives you a damage bonus based on how far you moved that round; it basically means that to optimize an 'outcome' (dealing lots of damage), you can't just go all-in on the 'hit things with swords skill'; instead, by using a combination between skills, you can be better at damage than the guy who just put all his xp into 'hit things with swords'.

I guess my point is that step 1 for constructing those synergies is to create other things in the game environment that are meaningful to the success/failure of the primary question. So if Sprinting just determines 'do you get away or not?' then its only meaningful when the question in the scenario revolves around sprinting. But if Sprinting creates a concrete 'thing you can do' in the world, then you can use that thing as a player to address the entire scenario.

Is that more clear? I'm still not sure this is the best way to put it, so sorry for any misunderstandings.

I feel like the minmaxed swordsman should get the upper hand in terms of stabbing people (after all, he put more resources into it), but he pays for that by suffering in other areas (i.e. he's not very useful once the bullets stop flying), and he might only need so many points in swordsmanship to effectively deal with combat encounters. I think the swordsman's player would just see synergies as another way to optimize his character to be an awesome murderer.

I think the best solution would be to cap out the number of points that can be put into a given skill (at character generation at least). That way it forces the player to diversify somewhat, even if he really wants to make a swordsman who is useless outside combat. Also, diminishing marginal returns in terms of skill increases will mean that Hack N. Slasher will quickly realize that he will get much more utility from raising other skills instead.


As for needing the maximum skill to succeed, one could simply design the system so that one or more of the following apply:


One doesn't need the maximum skill to succeed and do cool stuff with it.
Maintaining the maximum skill has diminishing returns; it requires almost all character resources devoted to it to maintain that status, such that the "overspecialist" lags behind in other areas. Also, the highest possible skill isn't much better than having it at the second highest -it only raises your chance to succeed by like 1% at that point, and the options it unlocks become gradually less widely applicable.
The game presents enough variety in challenges that overspecializing in one area hurts your performance in the rest, putting you at a clear disadvantage.
Explicitly forbid the player from excessively specializing (i.e. you may only put so many of your points, or a certain portion of your points, into a skill before you must allocate points toward others).

NichG
2013-10-21, 12:08 AM
I feel like the minmaxed swordsman should get the upper hand in terms of stabbing people (after all, he put more resources into it), but he pays for that by suffering in other areas (i.e. he's not very useful once the bullets stop flying), and he might only need so many points in swordsmanship to effectively deal with combat encounters. I think the swordsman's player would just see synergies as another way to optimize his character to be an awesome murderer.

This is kind of the problem with skill systems though. The idea that there is one skill that if you max it out, covers 'this set of things that could come up' turns the game into a question of 'did I max the right skills?' rather than 'what can I try to do in this situation?'.

The idea of making skills dependent on eachother is that it greatly enriches the build space with respect to particular goals. If you say 'I want to build a character that can kill things really well' then now, instead of there being one obvious answer 'max swords and dex' there are 10-20 different ways to go about it, each with strengths and weaknesses in various different scenarios that could come up.

Its even better when its not about numbers at all, but is entirely about a set of qualitatively different abilities. Its really obvious that '10 dice' is a better pool than '8 dice' if you're attacking someone, but its not obvious which is the 'best' answer between, say, being able to attack someone at range, versus being able to get a second attack when the person is off guard, versus being able to return one attack a round at the attacker, versus being able to do double damage against someone who has injured you this fight. They all make you better at 'killing things' but all in different ways that are not strictly better or worse than each other, but instead depend very highly on how exactly they're synergized with other things.

Aasimar
2013-10-21, 06:12 AM
I kind of need things to have a set DC. If I make a burglar character, I want the dc to climb stonework to be something specific so that I can become good enough to reliably climb walls of houses I intend to rob, but then stop advancing it (or at least slow it down a lot). I don't want the DC of climbing to arbitrarily rise as I level so that I'll always need to max it out, only to 'keep' my current level of competence.

As I rise in levels, I accept that I might face more difficult circumstances. "The Lich King can only be reached by scaling that ice wall...oh, and btw. you're being attacked by flying spider-monkeys" where a higher climbing ability might be useful.

That's ok, whether or not I choose to raise the ability that high is my choice and I don't feel like my character concept is weaker if my character fails.

But if climbing into peoples houses to steal their stuff requires the same roll from me at level 1 and level 10, something is very wrong.

Slipperychicken
2013-10-21, 07:22 AM
But if climbing into peoples houses to steal their stuff requires the same roll from me at level 1 and level 10, something is very wrong.

Do you mean having the same difficulty at levels 1 and 10, or the difficulty scaling up so your chance to succeed is the same?

Aasimar
2013-10-21, 07:27 AM
I mean that the difficulty of the same task should not scale with me. (though in some systems, particularly d&d 4th ed. the difficulty of any given task is a combination of my level and whether the GM things it's an easy, difficult or heroicly difficult task, instead of being a set dc for set tasks)

Segev
2013-10-21, 07:50 AM
Many systems that want to "solve" the hyper-specialization problem have skills get more expensive the higher you raise them. This doesn't prevent hyper-specialization, but it does start to make a point or two of lower-ranked skills more attractive because you'll see results from raising those other skills.

The real problem that I tend to see in games is that the specialist does always beat out the generalist. This is in part due to the dichotomy of what a generalist is: he is not a master of all things, but is competent in them. In theory, a party of 3 specialists and a generalist will have the specialists be "the go-to guy" for their specialties, and the generalist will be a competent back-up, while the other two specialists are mostly sitting out or throwing dice just to see what little they can do to help. In areas where nobody's the specialist, the generalist is "the go-to guy."

In practice, this party winds up having 3 people who are good at certain things, and then one person who isn't good at anything. His "generalist" competence level is usually too low to be comparable to the difficulties the party faces. Moreover, the larger the party, the less he has any chance to shine, as he becomes "second best" more and more often.

One approach that I like, though its implementation still faced problems, was the way Exalted tackled it. Some quick background: Solars are humans turned up to 11,000, and are the iconic specialists. They can become super-capable at anything, but not (easily) at everything. Lunars are shapeshifters with all the power and flexibility that entails.

The White Wolf system's resolution mechanic takes one of 9 Attributes (stats) and one of 25 Abilities (skills) and sums them to form a "die pool."

Solars are very Ability-focused, with their magics getting better the higher they raise them. They specialize very easily, with magics that act to enhance their already-generally-high Abilities.

Lunars are very Attribute-focused in much the same way.

Attributes tend to be much broader than Abilities because there are only 9 of them while there are 25 Abilities.

Lunars, however, are overall weaker than Solars. Still, they can specialize in Attributes in particular, and they can focus their shapeshifting on specific categories of tricks, but they all come with a lot of extra breadth to them. That bird form that you got to let you fly is also small and somewhat innocuous and can hop through smaller spaces, for instance.

Solars will always be better than Lunars in a direct contest of specialty focus. But Lunars, in raising their specialty, generally improve around the edges. Lunars will have areas they're better in and areas they're worse in, and they won't have the high peak of a Solar, but a Lunar who is filling one of the main party roles will fill it in ways that actually DO stand in for a Solar specialist, but will be broad enough to be effective in other areas. Solars, instead, may be focused in more than one area, and will be hyper-competent at them, but will have sharp drop-offs outside their area of expertise.

When the party expands and a new Lunar joins, he can have a different area of specialty from the Lunar already present, making him distinct, but will also be generalized around that area, as well. Even if a Solar joins in with a specialty that steps on a Lunar's specialty, the Lunar remains competent around the edges where the Solar's drop-off starts.

Like I said, there are a number of problems with actual implementation, but that's how it overall works in theory.

Slipperychicken
2013-10-21, 08:11 AM
I mean that the difficulty of the same task should not scale with me. (though in some systems, particularly d&d 4th ed. the difficulty of any given task is a combination of my level and whether the GM things it's an easy, difficult or heroicly difficult task, instead of being a set dc for set tasks)

I think that's partly a problem with 4e, and partly a problem with individual DMs (who for whatever reason feel that a success should only be when the d20 rolls high, regardless of PC skills).


I heard once that in 4e, ordinary untrained peasants have a 1/4 chance of succeeding if they try to bluff Asmodeus.

jedipotter
2013-10-21, 11:34 AM
I mean that the difficulty of the same task should not scale with me. (though in some systems, particularly d&d 4th ed. the difficulty of any given task is a combination of my level and whether the GM things it's an easy, difficult or heroicly difficult task, instead of being a set dc for set tasks)

You don't want the difficulty of a task to scale with your character? Why? Once you have a bunch of skill points you will automatically do even 'hard' things. Is that fun: knowing the roll you are making is pointless? You want actions in the game to be so easy they are pointless? Where is the fun in that?

Aasimar
2013-10-21, 12:18 PM
You don't want the difficulty of a task to scale with your character? Why? Once you have a bunch of skill points you will automatically do even 'hard' things. Is that fun: knowing the roll you are making is pointless? You want actions in the game to be so easy they are pointless? Where is the fun in that?

I want to face more difficult challenges.

I don't want whatever challenge I do face to automatically be scaled to my level.

If I choose to climb a rough rock wall at level 20 with maxed out skill points, the roll should be trivial and likely skipped unless the situation were precarious enough that a roll of 1 might be dangerous.

Likewise, if I want to scale the cliffs of insanity at level 1 in the rain with minimal skill point investment the difficulty should be impossible and also not a roll.

So, I want the game universe to be internally consistent, requiring the same rolls from everyone for the same tasks under the same circumstances.

An adventure SHOULD lead to more difficult tasks, as you level, but the stuff you could do at level one should not advance with you.

Scow2
2013-10-21, 01:39 PM
I mean that the difficulty of the same task should not scale with me. (though in some systems, particularly d&d 4th ed. the difficulty of any given task is a combination of my level and whether the GM things it's an easy, difficult or heroicly difficult task, instead of being a set dc for set tasks)Actually, that's a misinterpretation of trying to read an ad-hoc guideline. It assumes characters are doing character-level-appropriate things. What it doesn't give is what level a given task should be.


I heard once that in 4e, ordinary untrained peasants have a 1/4 chance of succeeding if they try to bluff Asmodeus.That was D&D Next, which they've fixed... somewhat.

NichG
2013-10-21, 03:38 PM
I want to face more difficult challenges.

I don't want whatever challenge I do face to automatically be scaled to my level.

If I choose to climb a rough rock wall at level 20 with maxed out skill points, the roll should be trivial and likely skipped unless the situation were precarious enough that a roll of 1 might be dangerous.

Likewise, if I want to scale the cliffs of insanity at level 1 in the rain with minimal skill point investment the difficulty should be impossible and also not a roll.

So, I want the game universe to be internally consistent, requiring the same rolls from everyone for the same tasks under the same circumstances.

An adventure SHOULD lead to more difficult tasks, as you level, but the stuff you could do at level one should not advance with you.

This kind of thing is sort of why I think there shouldn't actually be rolls for a lot of things, but instead it should just be that 'at some threshold, you gain the ability to do X' where 'X' could be 'climb a tree, climb stonework, climb ice' at three different thresholds.

Because there are a lot of skills where there's conceptually no penalty for trying over and over again, but at the same time you don't want to forbid retries. D&D's 'Take 10' mechanic is actually pretty good for this.

jedipotter
2013-10-21, 06:05 PM
This kind of thing is sort of why I think there shouldn't actually be rolls for a lot of things, but instead it should just be that 'at some threshold, you gain the ability to do X' where 'X' could be 'climb a tree, climb stonework, climb ice' at three different thresholds.

Because there are a lot of skills where there's conceptually no penalty for trying over and over again, but at the same time you don't want to forbid retries. D&D's 'Take 10' mechanic is actually pretty good for this.

So how do you have a game with no chance of failure? What is the point of playing if you will just automatically win? To be clear we are not talking about every tiny thing: so no ''roll to see if you can drink the cup of coffee without spilling it''. But any time that failure might equal something happening, there needs to be the chance of failure.

TuggyNE
2013-10-21, 07:15 PM
So how do you have a game with no chance of failure? What is the point of playing if you will just automatically win? To be clear we are not talking about every tiny thing: so no ''roll to see if you can drink the cup of coffee without spilling it''. But any time that failure might equal something happening, there needs to be the chance of failure.

You do harder things.

It's really that simple. That is all there is to it: challenges need to be inherently and by nature more intense and closer to epic the higher-level you get. Fighting random goblins is not a challenge at level 20, so why should climbing random trees be?

jindra34
2013-10-21, 07:17 PM
Because there are a lot of skills where there's conceptually no penalty for trying over and over again, but at the same time you don't want to forbid retries. D&D's 'Take 10' mechanic is actually pretty good for this.

See this is where the one of those nice rules for running a game comes into play: If nothing interesting can result, DON'T ROLL. Just say they did it and move on. Or if you really insist on a roll have it represent the time, based on degree over/under (a more skilled person should climb/run/swim/craft faster no? And same with someone who puts in a better effort). Though if its dramatic (very time crunchy like mid combat) rolling is a good idea.

NichG
2013-10-21, 09:29 PM
So how do you have a game with no chance of failure? What is the point of playing if you will just automatically win? To be clear we are not talking about every tiny thing: so no ''roll to see if you can drink the cup of coffee without spilling it''. But any time that failure might equal something happening, there needs to be the chance of failure.

Well, 'rolling' is not the only source of success or failure in a game. Take something like chess. Each piece can 'do' certain things, absolutely. The queen doesn't have a random chance of not being able to move in every direction. However, its how those abilities are used that determines success and failure.

Randomization serves a particular purpose in games (abstractly, it narrows the planning horizon and imparts value to otherwise redundant character options; more emotively, it tends to amplify excitement and mask the fact that most games have to be stacked in favor of the players), but its not the only way that failure can happen.

Jay R
2013-10-22, 11:15 AM
Ideally, this is solved in the experience system. An untrained climber can earn points by walking up a mountain path. Once he's good at that, he needs to work up a rocky slope. At mid-range, he can't improve unless he can climb large rocks. When he's a superior climber, he needs to go up a rock face to get any experience. Finally, only a flat wall will help him improve.

I don't want the DM to guarantee that all possible challenges are on his level. I want him to ignore trivial ones (for him), and run from deadly ones (noting their location so he can come back when he's ready).