How many Tools do you Need?

[howandwhy99]

Lizard posted an interesting point (I thought)

Is this a good way to design games though? Do we really need mechanics that cover 99% of the situations, when we can design simpler mechanics that cover 90% of the situations? How much more complexity do we add in order to cover that last 9%?

What do you think?

Sincerely, I think you are asking the wrong question. 100% of situations could easily be covered by coin tosses. That's simply an unsatisfying method. I think KM is more on track. It's more about abstract vs. specific and breadth vs. depth. In no way is only one rule necessary for any one given situation. The more rules the better IMO.

For my part I use OD&D, so my examples are drawn from it. The core rules are abstract and don't go into great depth. But they do cover great breadth. It includes variety of systems: 1:1 combat, wilderness travel, magic, aerial flight, naval combat, mass combat, dungeon construction, monsters, etc. The point here is the game has great breadth, but very little depth. You don't need depth when you are just starting out at the game.

Historically, published rule depth came from the many house rules and additional elements which were made "core" in AD&D. That's why Advanced feels like someone's vast store of house rules were made permanent for everyone who played it. Some were good, some were mediocre, and some were just plain bad. But the idea that they were the only option somehow stuck in people's heads. They needed to break out.

The best part about OD&D was there weren't 100% of situations covered, but more like 1000% - maybe even more. That's because, instead of multiplicity of single rules addressing single circumstances, you had abstract systems with tons of redundancy. By the time all five supplements were out in '77 the game had 3 combat systems and 2 mass combat systems to choose from. It's not a toolkit with a 1000 different rules for 1000 different situations, but one with a variety of tools for shaping the game to address those elements you wanted it too.

My suggestion is to keep whatever initial rules you use abstract/shallow. And when the players tell you or show you they are interested in some deeper, specific element, then dig out that massive tool chest of vast and redundant rules. Use the specific/detailed rule that work best for the situation at hand. Whether it be combat maneuvers or varieties of grassland and their growth speeds. This way play remains quick and the rules necessary for the ref to remember remain small and always directed towards the individual group's interests.

This is not to say you needn't simplify rules for easy referral. Like the games out in the past decade which use pretty much the same random mechanic for everything. But it allows you to choose when you want to stay simplified and when you want to use a complex rule that enlivens the situation at hand.

By offering a variety of very abstract subsystems an initial published game can allow for a variety of ways to explore the world and ease buy in to the rules. Of course, the breadth can be expanded (like in Birthright), but most players I've known want specifics. With publishers and shared house rules offering a variety of tools to dig deeper into those specific, more detailed areas of the world, Refs can pick and choose which of the multiple options for their own games customizing each to players' interests. And they need only to remember these exception rules without having to memorize a massive number which have never come up.

I think if publishers designed rules for all varieties of situations / areas of play, whether they go abstract, detailed, maybe even some new, interesting mode, then they could sell products for rules mechanics already covered in the core books. Or even their own.

 

[Hussar]

Maybe. My "common sense" tells me that in the following situation, a Cleave would be ruled out because of blatant physical impossibility:

...
.1.
FX2
.X.

Where F is the fighter, X is a wall, and 1 and 2 are his initial and secondary Cleave targets.

There are definitely people on this board - somewhat respectable pro-4e people, not just terrible rules lawyer types - whose "common sense" would tell them that said cleave is blatantly legal, and I'm supposed to handwave something about 1 staggering into 2 and then back into his square in order to inflict my Strength in damage on 2, or whatever.

This, in my experience, creates a clash of expectations and a possible at-the-table argument. A clarifying explanation, set down in plain text, that such a Cleave would be going through a wall and is impossible would help my game, not hurt it, even though I'm certainly capable of adjudicating that 9% of cases on the fly.

But, isn't that the direction madness lies? Do we really need rules text to cover this? And, at what point do you stop? For example, what if your secondary target creature was larger and only half hidden by the wall? Would that make a difference? What if there was no wall but the secondary target was invisible? On and on and on. At what point do you cut things off?

I really think that Ulthwithian has a good point. There is an inherent trade-off between how much rules complexity you add and the actual return for that complexity. Obviously, as Howandwhy points out, some complexity is needed for a satisfying game experience. But, where do you decide to cut things off?

 

[Ulthwithian]

Oh, it's not my point, really. I'm just pointing out that increasing the comprehensiveness of a rules system obeys the law of diminishing returns.

But then, I view everything through the lens of IE and its various fields, so take everything I say in that vein.

 

[VannATLC]

Except that gaming rulesets are programming languages explicitly designed to run on sentient machinery capable of and expected to make judgment calls, including ones regarding basic code execution. In other words, they're not much like programming languages at all.

Disclosure: I don't know LISP. Can LISP make judgment calls?

... Judgement calls are based on the instruction set of the processor. >.>

We're all just code. Some of it is just really, really scary.

You're abstraction a layer further away than the ruleset or the programming language, there.

 

[VannATLC]

What I'm less sold on is the idea that it's a good idea for D&D to do this. Part of that is because I use D&D as a toolkit, and I like using it as a toolkit. I tell it to give me "Postapocayptic orcs riding motorcycles" and it does that pretty well. I tell it to give me "Planescape!" and it does that pretty well. I tell it to give me "Tribal society where the PC's play as gods!" and it does that pretty well, too.

4e is taking some steps back from that, but the extent to which the game retains it will be part of the deciding factor for me. I'm a HUGE fan of D&D's modular nature, and I think that's an edge that D&D has over CRPGs and other table-top games (it hits a sweet spot between GURPS and the Storyteller games that I enjoy!). If all D&D4e can give me is a single world and "have adventure!" I'm going to be unable to use it.

I'm sorry.. What on makes you think 4e is taking *any* steps back ffrom that?

The 4e default setting is no more required than the 3.x default setting of Greyhawk/FR
Its got bugger all to do with the use of the ruleset as a toolkit.

IMO, falling back on the 'implied setting' as an excuse is a major, major strawman.

 

[Celebrim]

Is this a good way to design games though? Do we really need mechanics that cover 99% of the situations, when we can design simpler mechanics that cover 90% of the situations? How much more complexity do we add in order to cover that last 9%

If situations occurred purely at random, then perhaps not much. The problem with edge cases is that when they exist, players (or particular sorts of DMs) often intentionally set about bringing them about. The edge case ceases to be a rare occurance, and becomes something that happens all the time.

I would honestly think targeting 99% cover aims far too low. If you only have 99% cover, then the game is essentially broken all the time in much the same way that a car designed with parts with a 1% failure rate would hardly ever leave the shop. Honestly, I think you want rules that cover 99.9% of the cases that are likely to come up - and takes care to see to it that the .1% that it doesn't cover isn't something people would want to do.

I think most systems already aim that high. That's why rule books are 500 pages long, or why successful system really can produce and use dozens of rule books. And those that don't manage the same thing by relying heavily on GM fiat or player concensus to achieve the same thing.

 

[hong]

If situations occurred purely at random, then perhaps not much. The problem with edge cases is that when they exist, players (or particular sorts of DMs) often intentionally set about bringing them about. The edge case ceases to be a rare occurance, and becomes something that happens all the time.

This is easily solved by only playing with people who know not to think too hard about fantasy.

 

[Celebrim]

It would be a waste of space since there are people with perfectly functioning brains that are playing the game.

The problem with a stand like this is that reasonable people are very likely to disagree. In fact, 'people with perfectly functioning brains' are probably more likely to disagree than not, because that perfectly functioning brain gives them the power to imagine various perspectives and the rhetorical skill to rationalize to themselves the superiority of thier opinion. Besides, they are used to being right, so why not now?

So what you find is that leaving things up to 'common sense' doesn't really solve anything. 'Common sense' is nothing more than peoples unreflected upon opinions, and not only will people hold different things to be 'intuitive' - but smart people are far less likely to hold alot of unreflected upon opinions and generally can be trusted to not have alot of 'common sense'. Leaving things to 'common sense' with a bunch of nerds is just a recipe for spending your evening arguing.

Moreover, even if you can avoid arguing what you find is that every case that you leave up to the GM's (or the players) common sense becomes a defacto 'house rule'. Pretty soon, in addition to your slender rule book, your table is dragging along literally scores or hundreds of pages of informal rules to deal with all those cases that you didn't want to deal with because it would have made your formal rulebook more complicated. You haven't decreased the complexity; you've just deferred it. And not only that, you deferred it to a place where you can't as easily look it up to satisfy everyone what it was or remember how you handled the problem last time. There are a number of board games (Roborally, for example) where our group finally had to write down our house rules that we used every time to play simply because we'd forget exactly what we'd decided on, and someone would do something based off one assumption and have to have everyone else tell them that no, remember, last time we decided on something else. Or else we'd all forget and then remember halfway into the game what we forgot and how much better the game would have been had we not forgotten. After we wrote them down, the games were smoother and more consistant. And we only needed a 3x5 notecard for what had been all sorts of confusion and disappointment.

'Common sense' is a bloody complicated and preeminently confusing rule.

 

[Hussar]

If situations occurred purely at random, then perhaps not much. The problem with edge cases is that when they exist, players (or particular sorts of DMs) often intentionally set about bringing them about. The edge case ceases to be a rare occurance, and becomes something that happens all the time.

I would honestly think targeting 99% cover aims far too low. If you only have 99% cover, then the game is essentially broken all the time in much the same way that a car designed with parts with a 1% failure rate would hardly ever leave the shop. Honestly, I think you want rules that cover 99.9% of the cases that are likely to come up - and takes care to see to it that the .1% that it doesn't cover isn't something people would want to do.

I think most systems already aim that high. That's why rule books are 500 pages long, or why successful system really can produce and use dozens of rule books. And those that don't manage the same thing by relying heavily on GM fiat or player concensus to achieve the same thing.

I don't want to argue semantics over the actual numbers, that's not the point.

The point is, how far do you have to go to cover corner cases? How often does a situations have to come up for to be considered a "non edge" case?

Take falling into lava as an example. Now, RAW is kinda silly really. If you swim in lava, you die. But, by RAW, lots of high level characters could swim a river of lava and come out the other side. So, we have a problem.

Or, do we? How often does lava feature in adventure design? How often do players deliberately act to circumvent the ruleset? That's going to vary wildly from group to group. Should the designers design from the mindset that lava will come up all the time and make detailed lava rules? Or should they say, "Here, these rules will work well enough, go forth and play?"

In my mind, the latter is much better. It would be better to specifically empower the DM to make rulings to cover the corner cases. Put big bold letters somewhere in the PHB tellings players that the rules do not cover all eventualities and sometimes the DM will over rule the rules. Then, include a nice long section in the DMG telling DM's when to do this and why.

Something that was lacking in the 3e DMG unfortunately.

 

[hong]

The problem with a stand like this is that reasonable people are very likely to disagree. In fact, 'people with perfectly functioning brains' are probably more likely to disagree than not, because that perfectly functioning brain gives them the power to imagine various perspectives and the rhetorical skill to rationalize to themselves the superiority of thier opinion. Besides, they are used to being right, so why not now?

This is why you have referees.


Hong "channeling Diaglo" Ooi