Why the hate for complexity?

Celebrim

Legend
What good can you say is inherent to complexity? What positive trait can a game not possibly have, without also increasing the number of rules involved?

All of them.

I feel I have to break out "The World's Simplest RPG" again. It has one rule:

Rule #1: Whenever any proposition is made, flip a coin. On heads, the proposition succeeds. On tails, the proposition fails.

By your argument this game now contains all the positive traits that a game can have.

Reading you statement just above this one, I again insist we need a definition of complexity because when you use the word you are tacking on the idea of "needless complexity".

Before trying to define complexity, let me define a related term: "Elegance". Elegance is the property of being the simplest system necessary to achieve a certain desired result. Games should strive to be elegant. They cannot avoid complexity and still be good games. Complexity is good.
 

log in or register to remove this ad

All of them.

I feel I have to break out "The World's Simplest RPG" again. It has one rule:

Rule #1: Whenever any proposition is made, flip a coin. On heads, the proposition succeeds. On tails, the proposition fails.

By your argument this game now contains all the positive traits that a game can have.
No, by my argument, this game lacks sufficient detail in its resolution (for most people). I would prefer a game which did (much) more to differentiate between character abilities. The game is lacking almost all of the positive traits that you would normally gain as a trade-off from adding complexity.

The game could stand to add complexity, not because complexity is inherently beneficial, but because doing so will allow the system to do more of those things that we typically ask an RPG system to do. In modifying this game to increase its depth of resolution, we should mind to increase the complexity by the smallest amount necessary to achieve those goals; no more, but also no less. For example, we shouldn't use multiple dice and charts unless we actually need them to achieve the probability distribution that we want (unless our goal is specifically to increase player engagement with the dice and charts, because that's fun for you).
 

Morrus

Well, that was fun
Staff member
I'm not saying that complex games are necessarily bad. I'm just saying that complexity, itself, is bad. You need to weigh the trade-off, for all of the awesome things you get in exchange for it.

Are you suggesting that some people might enjoy complexity, for its own sake? That someone might prefer to roll seventeen dice, and cross-reference the result on three different charts, because rolling dice and referencing charts is inherently fun for some people? Because I honestly hadn't considered that before. If there are people like that, then my blanket statement might be a bit over-reaching, and the whole topic just comes down to a matter of preferences.

Of course there are.

And there are people who like fine-tuning the engines of their cars, and who like arguing the minutiae of TV continuity, and who memorise stats of sports players, and who painstakingly build model railways, and who like the arduous process of computer coding. I can't imagine doing any of those things.

The things people have fun doing are many and varied. I enjoy spending time designing starships using fairly complex game systems, but I wouldn't enjoy it if it were a simple task which took me 20 seconds. I enjoy the *process*.

(I mean, your example is deliberately bad to make the POV look ridiculous, but the point stands).
 
Last edited:

Celebrim

Legend
No, by my argument, this game lacks sufficient detail in its resolution (for most people).

Detail in resolution is, unsurprisingly, a definition of complexity or not a bad pass at one.

I would prefer a game which did (much) more to differentiate between character abilities. The game is lacking almost all of the positive traits that you would normally gain as a trade-off from adding complexity.

The positive traits are the complexity. You can't separate the two things. Complexity has to do with emergent properties of a system. You don't attain complexity without having at least some properties that occur across different scales.

The game could stand to add complexity, not because complexity is inherently beneficial, but because doing so will allow the system to do more of those things that we typically ask an RPG system to do. In modifying this game to increase its depth of resolution, we should mind to increase the complexity by the smallest amount necessary to achieve those goals; no more, but also no less.

Absolutely. We should strive to make this game elegant. We should not strive to make this game uncomplicated.

For example, we shouldn't use multiple dice and charts unless we actually need them to achieve the probability distribution that we want (unless our goal is specifically to increase player engagement with the dice and charts, because that's fun for you).

Yeah, so chess is a good example of a complex game. It has a simple set of rules, but the interactions of these simple sets of rules create a game that has very surprising emergent properties and tons of agency in the person playing it, in that they have meaningful choices to make that have difficult to determine outcomes that therefore require a lot of analysis.

You probably won't be surprised to find that some people might enjoy analysis of complex outcomes as a thing in and of itself.
 

Tony Vargas

Legend
Before trying to define complexity, let me define a related term: "Elegance". Elegance is the property of being the simplest system necessary to achieve a certain desired result. Games should strive to be elegant. They cannot avoid complexity and still be good games. Complexity is good.

Necessary Complexity is good, I suppose, Needless Complexity, bad. What's necessary vs needless largely depends on the purpose of the system. I don't think the purposes of TSR & WotC era D&D, for instance, were tremendously different...

Um, there's more to overall system complexity than the task resolution mechanic.
Nod, it was just one example of needless complexity in the classic TSR era vs the comparatively clean/elegant and consistent complexity of the WotC era.

For example:
Questions - how much text does it take to describe the OD&D Fighter class?
Heh. Do you include the sections of the DMG describing the fighter's potential followers & that Keep he gets to build at 9th level? What about all the fighter-only magic items, those are essentially fighter class features?

But, another way of thinking of it: could you, from memory, easily re-create the 1e fighter? It's exp chart, combat matrix, saving throws, attacks/round, etc?

What about the 3e fighter? Exp/level, same as everyone else. d10 HD, 2 skill points. A bonus feat at first, one at every even numbered level. Full BAB, Good FORT, Bad REF/Will. Hardest thing'd be remembering which are his class skills (Climb, Craft, Jump, Ride, Swim?).


How many choices does one have to make in character creation, leveling up, and play? How many decisions do you make or elements do you have to line up and know before you can engage whatever resolution mechanic is used?

Then, same questions, but for 3e, 4e, and 5e.
The numbers can be HUGE. Years ago I did calculate the theoretical number of unique single-class 3.x fighter builds, for instance, and it was just ridiculous...
...yet the class description, itself, is stunningly simple. You don't experience the complexity implied by the whole universe of choices because you automatically narrow them down quite a bit based on concept.

I did take a stab at a more complete answer, it got really long:

Chargen:

1e: Choose chargen method, the most common, Method IV, IIRC, 4d6-and arrange, so make 5 choices of stat placement. Choose Race, Gender (because it matters, mechanically), sub-race, Class, depending on race, choose a multi-class combination. Depending on class Randomly roll your gp and buy a bunch of equipment, any bit of which may be critically important. Depending on class randomly determine spells known and prepare 1 to as many as 4 spells for the day, or pick an all-important 'optional' weapon specialization, etc..
3e: Assuming point buy, make a /lot/ of choices about your stats. Choose Race, Class, distribute anything from 6 to 44(?) skill points, equipment.
4e: Assuming elite array, make 5 choices about stat placement. Choose Race, class, & PH 'build' - everything else is chosen for you, right down to gear. ;P (No, really, you could cut to the chase like that in the PH.)
Essentials: Choose Race, Class, Background, Theme, Alternate Class features or Sub-Class, Powers, a feat, and skills.
5e: 5 choices about stat placement. Choose Race, sub-Race, Class, Background, skills from background, & skills from class, and spells known and/or prepared. Depending on class, you might also choose sub-class (like wizard tradition) at 1st. You can let background & class choose most of your equipment for you.

So, yeah, in all eds there's a lot of complexity in chargen. In 1e, there was less choice and more randomness.

Level up? In 1e, add up your exp, at the end of each session, possibly add a 10% bonus, divide it by your number of classes, check your next level goal for each of those classes to see if you leveled up, then go looking for training and flush a bunch of gold to actually level up, then look to see what your class gives you at the new level, because every class & level is different. Depending on class, gain new spell levels, special abilities, etc, possibly gain followers, maybe even fight a duel to keep the new level.
3e: No bonus, no dividing among classes though you might have an MCing penalty (rarely), just check one level chart. Pick a class (or PrC! or next level of a class you already have) to take for your next level, determine if you can actually take it without penalty, and look up what that class/level gets you because they're all different. Distribute your new skill ranks to skills, minding the cost/limits of in-class vs cross-class. Pick a feat every 3rd level, or if a class-level gives you one.
4e: Add up your exp, check it against one chart, unless you miss sessions you all level up at the same time, and pick a power from your class or feat and/or stat bump based only on your level, because there's just one progression for everyone.
5e: Add up your exp, check it against one chart, unless you miss sessions you all level up at the same time, check to see what the next level of your class gives you, because they're all different, pick a feat or spells or other feature if you get one; optionally pick a new class or next level of an existing class if you're using the optional multi-class rules in the PH.



While the WotC era has brought in a lot more customizability the net price in complexity hasn't been that high, because needless complexity, like different exp charts and different THAC0 progressions and the like have been done away with.




In play... OMFingG…



1e: Depending on class, memorize some spells from either a short list of known spells or your full list of spells for your level, or just wait, bored, while everyone else does the same, maybe break camp while they're studying books and praying for 10 min/spell level/spell. Tick off rations, light sources, &c used. Declare weapon-in-hand and marching order - if you want any chance of sneaking up on anything, get your halfling or thief well in advance of the party. Wander around in the wilderness or dungeon, carefully asking the DM for every conceivable detail of the environment that might drop a hint about treasure or danger. Either roll % dice (Thief) or d6 (surprise) to determine if you sneak up on anything (or vice versa). Each side (or player, depending on your DM, the rules aren't crystal clear and some make more sense one way, some the other) rolls a d6 to determine surprise, what determines surprise depends on the character and the enemy and can result in either, both, or neither side being surprised, or some of one side or the other being surprised or not. Each segment you're surprised, the enemy beats on you like it was a whole round, 'cept for spell casting. Probably. Again, the rules were neither consistent nor consistently applied by DMs. Once you're in combat, either attack, or move and attack, or charge, or set a weapon to receive an opponents charge, or cast a spell, or draw and maybe use a different weapon, and roll initiative, again, maybe on a d6 for the whole party, or maybe a d10 for each (it was a super-common variant), and apply casting times and weapon speed and factor in attacks per round to try to figure out what order things happen in, and, maybe even, in some cases, get a bonus attack for tying but not winning initiative... if you attacked, roll a d20 to hit, apply various modifiers based on the weapon you're using, the circumstance, the armor they're wearing, and tell the DM, who will add any secret modifiers and consult a matrix to see if you hit, if you did and the monster could be harmed by your weapon, roll damage based on the weapon and the size of the enemy you hit, and add more modifiers... if you're casting a spell, and you didn't happen to get hit in the middle of casting, look up what the obtuse Gygaxian prose of your spell says, while the DM looks up the extra notes about how it works in the DMG and everyone else settles in for a long argument about what happens... If your next challenge isn't a combat, describe everything you do in excruciating detail, including every argument you can think of about why it should work (and why your character should know it should work), unless you happen to have a 'special' ability that lets you just solve it with a pass/fail % check, or a spell or magic item that you can make an argument (usually involving middle-school/PBS-special/comic-book science) will obviate the challenge in some way.

...oh, and I almost forgot, do the same for your henchmen and hirelings! (but be careful with them, because the DM is constantly adjusting their Loyalty and Morale...)



3e: Depending on class, memorize prepare some spells from either a list of known spells (wizard) or your full list of spells for your level (Cleric/Druid), or just wait, bored, while everyone else does the same. Tick off rations, light sources, &c used, unless you've found/made/bought some infinite-resource magic item. If you want any chance of sneaking up on anything, get your highest-stealth character well in advance of the party and have him make a lot of stealth checks. Wander around in the wilderness or dungeon, declaring every skill check that you think might reveal a hint about treasure or danger. If any challenges come up, everyone roll every relevant skill until someone succeeds (or just take 20). Or just cast a spell, probably from a scroll, that obviates the combat. If the challenge was an unexpected combat, roll contested stealth & perception to determine surprise. Then roll initiative. The highest initiative caster picks the right spell from the list of every spell he knows (spontaneous) or every spell he has prepared (which might be dozens at higher level) /and hasn't cast yet/ or magic item (from everything you've collected or made) for the fight to bring it to a swift conclusion. If you know about a combat in advance, layer pre-cast spells until the challenge is obviated before initiative is even rolled. If anyone gets hurt, pull out the Wand of Cure Light Wounds and heal them back 1d8+1 hps & six second at a time.



4e: Decide on something you'd like to do, or just board the railroad of a linear published adventure. If there are difficulties on the way, the DM may make it a skill challenge that you each take a turn and try to apply something you're good at too until you succeed or fail ("forward"), if it involves a fight, and you want to sneak up on 'em make a group stealth check (DC of their passive perception), if half of you succeed, you gain surprise (one standard action each), then you roll initiative, and, in initiative order, decide where to move, maybe take a minor action, and (mainly) which of your several attack powers to use - if you attack or cast an attack spell, roll to hit (with your predetermined attack bonus) vs the targets defense, and roll the power's damage dice, maybe with resistance deducted. After, everyone takes 5 (minutes) and spends healing surges @ 1/4 max hps each, to get back to full.


5e: Depending on class, prepare some spells from either a list of known spells (wizard) or your full list of spells for your level (Cleric/Druid), or just wait, bored, while everyone else does the same. Wander around in the wilderness or dungeon, declaring every action you think of that might reveal (possibly with a successful check) a hint about treasure or danger. If the DM does call for a check from a character and he fails (with no particular consequences), everyone else piles on and tries the same thing until they succeed, if it involves a fight, and you want to sneak up on 'em make a group stealth check, if half of you succeed, you gain surprise, then you roll initiative, and, in initiative order, decide whether (there's often not much point) & where to move, and declare an action, which may or may not also entitle you to a bonus action, and, if you attack, roll a d20 + bonuses vs AC, or if you cast a cantrip or spell either do the same or call for a save from the target(s) which is also d20 + bonuses but vs your save DC, or do whatever else the spell says. If you hit, roll weapon damage, or if your targets fail their saves, do whatever the spell says. After, if you have an hour to rest, you can spend (roll) some HD to get some/all of your hps back.
 
Last edited:

Reading you statement just above this one, I again insist we need a definition of complexity because when you use the word you are tacking on the idea of "needless complexity".

Before trying to define complexity, let me define a related term: "Elegance". Elegance is the property of being the simplest system necessary to achieve a certain desired result. Games should strive to be elegant. They cannot avoid complexity and still be good games. Complexity is good.
It seems to me that you're conflating cause and effect, when they are really two different variables. You're saying that the cause is good, just because the effect is good. I'm saying that the effect is good if the effect is good, and the cause is something we have to deal with; if we could get the same effect, without investing as much into the cause, then that would be preferable.

Doing work is good, because it gets stuff done. If you could get stuff done without doing work, then that would be preferable. Work, by itself, does not justify its own existence (unless you're into that sort of thing). The real value is in the results.

In order to maximize the elegance of a system, you should minimize the amount of complexity that's required to achieve your desired result. Complexity is bad, in the same way that cost is bad when you're designing a computer. If you could build your super deluxe computer without spending any money, then that would be ideal, although reality will probably fall short of that ideal by a considerable ways.
 

Detail in resolution is, unsurprisingly, a definition of complexity or not a bad pass at one.
I would find that definition to be fairly surprising. I wouldn't expect anyone to conflate the number of steps in the resolution process with the variety and distribution of possible outcomes, unless they were very new to these sorts of discussions.

Nevertheless, for the purpose of this thread, I'm using the term to refer only to the former. By using the same term to describe both parts of that equation, it makes it hard to discuss the elegance factor. The reason why some people have moved away from complex games is because they no longer need that level of complexity in order to realize a given level of resolution, and the resolution is all they're really after.
 

Celebrim

Legend
It seems to me that you're conflating cause and effect, when they are really two different variables.

No, you are persisting in using a poor definition. I've offered up better definitions, but you've neither quibbled with them, nor offered your own, nor accepted them. Instead, you are continuing to talk in a circle without even so much as acknowledging the argument I'm making. So in short, this conversation is probably coming to an end.

You're saying that...

I'm not going to get into refuting that because what I've said is available to see above and it has nothing to do with what I'm saying.

Doing work is good, because it gets stuff done. If you could get stuff done without doing work, then that would be preferable. Work, by itself, does not justify its own existence (unless you're into that sort of thing).

You are back to conflating work with complexity, and they really don't necessarily have a lot to do with each other. For example, the game of chess is complicated, and it requires deep analysis to play well, but there is no "work" in playing it in the sense you mean it. There is no direct relationship between complexity and work, and apparently you've been burned by poorly designed games that require a bunch of work (bookkeeping, math, memorization, table lookups, etc.) to achieve a result and now you think all that work is inherent to complexity but it just isn't. I suppose that there is some inherent increase in the difficulty of learning to play a game with more complexity than a game with less (although, I can think of counter-examples even to that, but no in the sphere of RPGs), but the actual act of playing a complex game does not inherently mean more work in the sense you keep using it.

Not that I even think 'work' is a really great term for what you seem to be talking about, but a better one doesn't immediately occur to me.

In order to maximize the elegance of a system, you should minimize the amount of complexity that's required to achieve your desired result.

Yes, but that's a truism. You are just restating my definition back to me using different words. That's like saying to maximize momentum you should increase mass without reducing velocity. It's something I agree with, but it's also something that is obvious.

Complexity is bad, in the same way that cost is bad when you're designing a computer. If you could build your super deluxe computer without spending any money, then that would be ideal, although reality will probably fall short of that ideal by a considerable ways.

Again, you are stuck on ideas of "needless complexity" or some such. You're continually defining complexity as bad as a tautology. Complexity isn't work. It isn't cost. You can go look up discussions of the (very complex) idea of complexity, and you'll not find them discussing work and cost and such.
 
Last edited:

Celebrim

Legend
I would find that definition to be fairly surprising.

Well, start with a Wikipedia article on 'complexity'.

I wouldn't expect anyone to conflate the number of steps in the resolution process with the variety and distribution of possible outcomes, unless they were very new to these sorts of discussions.

Nevertheless, for the purpose of this thread, I'm using the term to refer only to the former.

*Beats head against wall*

By using the same term to describe both parts of that equation, it makes it hard to discuss the elegance factor. The reason why some people have moved away from complex games is because they no longer need that level of complexity in order to realize a given level of resolution, and the resolution is all they're really after.

Ok, I'll take a different tack. Rules Light inherently supports less granularity in the resolution. The proponents of Rules Light tended to mock the idea that you needed to have fine grained resolutions, and as a result tended to create games that do not as an element of the resolution process give you any degree of granularity or specificity. They are not merely looking for a simpler or more elegant way to achieve what used to be called realism, they are eschewing the need for complexity at all.
 

No, you are persisting in using a poor definition. I've offered up better definitions, but you've neither quibbled with them, nor offered your own, nor accepted them.
Fine, then. What is your word that refers to the number of rule interactions and the amount of effort required to resolve something under a particular game system? Is it just 'work'? Do you want me to say that complex systems aren't bad, but work is bad, and people don't like complex games because they're too much work?

Not that I even think 'work' is a really great term for what you seem to be talking about, but a better one doesn't immediately occur to me.
What's wrong with saying 'complexity' then? The amount of effort required to resolve something is an incredibly important factor in discussing game design. It's hard to believe that we don't have some common jargon for it, yet. I was under the belief that the correct jargon was 'complexity' but I guess that's not widely-accepted.

I mean, if you're discussing a machine that has a lot of moving parts which don't necessarily interact in obvious ways, then I wouldn't hesitate to call it a complex machine. How is an RPG any different?
 

Remove ads

Top