Minimizing Tedium: Not Always Straightforward

A couple weeks back I talked about how you can pick two combat roles at once in my MMO, and switch them out in town. A comment from David Grundy:

Interesting on the role switching. But, why only “whenever they’re in town”?

Why create an artificial barrier to a player doing something that you as a designer want them to do? I think the question to ask yourself is; do you want players to switch roles pretty easily so that they can contribute in different situations; the answer you’ve put above sounds pretty much like an emphatic yes. Next question, do you want to create an un-fun, time-consuming and ultimately pointless mechanic for the player to be able to do this?

That’s a fair question. Who on earth will enjoy running back to town to switch classes? Nobody. So why make them do it?

The reason I make it hard for players to switch classes is so that players can’t super-min-max every scenario. I don’t want players switching classes every few seconds between each fight. It would be tedious and time consuming, and you’d have to carry tons of extra equipment around to make it work.

Sometimes “Less Structure” Actually Means “More Tedious”

But it would work: you’d level faster by switching classes constantly. So people would feel competitive pressure to do it. And since I expect to have a fair number of WoW-gamers who have been weaned on achievement-minded gameplay, they will fall right into this trap. They’ll min-max the game until it’s not fun to play, and then they’ll complain bitterly about how tedious the game is. And at the same time, the person who plays the “right” way, not changing classes constantly, will compare themselves to the people doing the tedious thing and will feel dumb for not doing that too.

This design pitfall is called “balancing through tedium.” It happens when the designer goes “nah, we don’t have to worry about that scenario, it’s so tedious nobody would do it!” That kind of thinking works in board games sometimes. It works in children’s games sometimes. But it doesn’t work in traditional MMOs.

Calculating the Price of Tedium

Of course, I wouldn’t let that happen. I’ve seen firsthand how damaging it is to let tedium be the limiter of your game design. So instead, I’d homogenize and simplify my skill designs, watering down each one so that you could switch classes constantly and it wouldn’t mean as much.

Let me give you a really simple example of what I mean by watered down skills. Each combat skill has an “emergency ability” on a 5- to 10-minute cooldown. Normally you’d have access to two of these powers. But if you can switch classes constantly, you’d eventually have easy access to dozens of them! And whenever you used an emergency power, you’d switch to a new class in the next fight, so you’d always have one ready to go.

Assuming these emergency abilities are powerful (they are), this would do bad things to the difficulty of the game. People who had mastered dozens of combat skills would be able to clear dungeons far faster than people who only had two combat skills mastered. So in time it would be seen as “mandatory” to know twelve combat skills and switch between them constantly. And then I’ve failed and everybody hates the design.

Of course, I wouldn’t let that happen since it’s so easy to predict the problems with it. Instead, I just wouldn’t give out abilities with long cooldown timers. Or else I’d create some complex system of shared timers. This would require me to homogenize the skills: they’d need to be on the same timer, so different skills couldn’t have different cooldowns.

In the end, what did I do? I lowered the overall fun level to avoid some (very occasional) tedium. It’s a net loss in fun.

Minimizing Overall Tedium

The above example is actually one of the easiest ones to fix, but it’s also the easiest to explain, so I went with that one. There are actually a bunch of places where switching classes too easily would suck the fun out of the game.

I want to reiterate that I’m not being punitive for the sake of being punitive. I don’t think that mentality has much place in modern games. I’m actually doing the opposite: I’m saving players from being “forced” (through competitive pressure) to min-max their characters in tedious fashion. And to a lesser extent, I’m saving players from having boring homogenized classes.

It’s kind of a lose/lose for me, of course: players will say I’m being a dick, even though I’m doing it specifically to be nice to players in the long term.

Feeling Clever Is Good… But Don’t Overdo It Or It Dies

It’s also important to let players feel clever while they play. And I admit that if you could switch classes constantly, you would feel pretty clever for a while, switching to just the right situation for each battle… but you’d only feel clever for a while, because it would be more-or-less mandatory (due to the social pressure to keep up with your friends). Then it stops being “clever” and starts being “work.”

I want players to feel clever. I also want them to min-max, if that’s what they like doing. Remember that there’s a third set of abilities in the game, along with the two sets of class abilities you get. These abilities can be cherry-picked at any time, and doing so will let you min-max boss fights (and regular fights too, but not so much). I think it’s fun to carefully plan out your boss fights. I just don’t think it’s fun to have to plan out every fight.

And as another commenter ‘Expert Novice’ mentioned, maybe I should have special items that let you switch classes in mid-dungeon, but have a price attached. Or maybe there are “class-switch stations” outside of dungeons. That sort of stuff is easy to add later, if it turns out not to be a problem. When everything is up and running, I can go back and minimize the pain to just the amount that’s beneficial.

And I’ve Got Nothing Against Pointless Mechanics

Going back to David’s earlier question, do I want to create an “un-fun, time-consuming and ultimately pointless mechanic?”… well, even though this mechanic isn’t pointless, I do have a lot of other pointless mechanics in the game. (Even more so than a typical MMO.)

This week while doing skill designs, I added a completely pointless prerequisite for the high-level skill Necromancy. Necromancy is mastery of death, so naturally, the prerequisite should include Dying. So now Dying is a new skill you get by… dying. Ideally in new or epic ways.

“Why!?” If you have to ask why, then I don’t think I can explain it in a way that would convince you. But I’ll try: I want to add a lot of pointless details to my game even though they don’t enhance the minute-to-minute experience of the game. In the big picture, I think they do enhance the game, and perhaps even define it in a way.

The biggest reason they matter is that they are the gears that make emergent behavior possible. But another important reason is that they’re memorable. Interesting. Quirky.

Nobody talks about how tight the reward-loop is in their favorite MMO. What captures imaginations is the fringe stuff. The details. The crazy bits. At least that’s true for me. And it’s my ridiculous fringe project, so there!

This entry was posted in Design, Project Gorgon. Bookmark the permalink.

18 Responses to Minimizing Tedium: Not Always Straightforward

  1. Eric Heimburg says:

    And I definitely didn’t mean to pick on David, whose many comments were insightful. I just wanted to give a longer answer to this question.

  2. The problem with balancing by tedium is that you haven’t actually prevented the player from changing classes to prepare for a fight – you’ve just increased the amount of time required in order to do so. I don’t know what you’re going for in terms of travel times versus mob respawn rates versus groups holding an instance open while the DPS all switch from ranged to melee because the next boss requires everyone to stand on top of him, but players will do these things if the difficulty is high enough to be worth doing so.

    (And yes, you’ll be blamed no matter what the outcome is. Players suck like that, sorry.)

    Quoth Eric: “Or else I’d create some complex system of shared timers. This would require me to homogenize the skills: they’d need to be on the same timer, so different skills couldn’t have different cooldowns.”

    Speaking of balancing by tedium and long cooldowns, I’m not a fan because I remember multiple raid nights where 40 players sat on their rears for 10 minutes because the raid leader felt we’d be wasting time and gold by pulling without at least one member’s cooldown available for things like combat res, divine intervention, etc. That said, why do the emergency skills need to have different cooldowns based on how they implement the function of saving your rear no more than once every 10 minutes? Why not have all of them share a single cooldown and let players who land in emergencies more than once every 10 minutes die?

  3. Mavis says:

    I was wandering about the town thing….

    But I agree unless certain content resets if you go into town – a sub set will still do that.

    I suppose it would reduce the use of that behaviour.

    Why not just a long cool down timer on te class change? Unless in town?

  4. Aaron says:

    @Green Armadillo

    The case he’s trying to avoid is where it’s more efficient to switch classes all the time than to pick two for the specific dungeon you’re trying to do. Presumably heading back to town, switching classes, then going back to the dungeon is less efficient than just pressing on with your current class setup.

  5. Bronte says:

    I know you are convinced of it, but I am not. I think it would be a lot more convenient to carry around the extra set of equipment than it would be to run back to town. It sounds like part of the objective is that people should not carry so much equipment in the bags. Well then the process has two steps now, go back to town, respec, go to storage unit (bank, whatever), switch out gear.

    I support the original commenter, I think it seems a tad arbitrary.

  6. Pingback: Quote of the Day « Procrastination Amplification

  7. Brian Miller says:

    Let me provide you with a counter-example. Rifts allows players to switch among many roles on the fly, at any time and yet does not suffer from the sort of A.D.D. class hopping you seem to feel it should engender. In Rifts players typically fall into playing whichever role they are most comfortable with and stick with it even when switching to another role might be somewhat more efficient. Unless you plan on making your encounters require tuning to an exponential degree in order to be overcome you will not have players constantly switching from one class to another to make it just a little bit easier.

    Rifts role switching feature is one of the best examples of how to make a game far more accessible and enjoyable. There is a bit of gear swapping to go from DPS to tank but having done a bit of it myself I never found it too cumbersome nor did it require all that much bag space. There was still plenty of room to pick up more loot and carry around a bunch of tokens / crafting materials to boot.

    If you have not played rifts at all I strongly urge you to do so if for no other reason than to see for yourself firsthand how easy, accessible role changes can enhance a game. Do it before you make the mistake of creating a frustrating and ultimately pointless game mechanic based on groundless fears.

  8. Jeremy Thornhill says:

    Would people really switch classes all the time?

    Brian mentions RIFT, a game which I played and enjoyed. You can switch classes any time, anywhere. I can’t speak for everybody, but I almost never switched class “mid-stream.” I picked what my group needed at the time and would usually just stick with that through the encounter (or if solo, I would fall back on my “kill stuff solo” class 90% of the time).

    Aside from just plain momentum to keep playing “your class,” there are plenty of mechanical ways to prevent overuse that don’t impose the tedium of a trip to town on everybody else. Wipe out resources, for example, like WoW: when you switch classes, you lose all your buffs and class mechanic resources. It’s still adding tedium (to regain resources), but it’s (presumably) not as bad as going to town. And it’s presumably enough to make class-flipping useless as a min-max strategy.

    If you don’t want to do that, how about a cooldown, or a period of reduced effectiveness after changing classes that is multiplicitive when you switch frequently, slow down experience gain for X monsters after switching your class, or any number of more subtle ways to discourage only the frequent class switchers?

    “Going to town” just seems like the nuclear option to me. You surely prevent tactical class choosing, but you also disincentivize the kind of class switching that you’d (presumably) want to facilitate in the process (like, say, changing class when group composition is altered).

  9. Adam Bienias says:

    Eric, I’m completely behind you with this idea and with everything you wrote above.
    I mean WoW doesn’t have switching classes so it is not a problem. Now we give a player possibility to change a class and suddenly time become his problem ? So “the only solution” is to make this insta ? No, no, no… Many modern MMO’s fall into this spiral of doom with bad decisions, mostly because this is not a problem about “do you want players to switch roles pretty easily” but about a design as a whole. Switching classes is serious design decision and does not fall into a category of “yes” or “no”.

    1. As Green Armadillo pointed, there’s travel system you need to take into consideration. Teleportation anywhere, anytime, will kill this idea. So if you want to limit switching classes, remove teleports. Personally I think that easy teleportation systems killed/removed a lot of fun from games, so removing it would benefit not only ‘switching classes’ (but that’s a whole different topic ;) )
    2. You need also to think about social aspect. If anyone can play any class any time, you will not find (or very rarely) groups asking “Looking for healer”. Player stop to care about “with whom they’re playing”, as they will start playing with completely random persons. The same problem has Dungeon Finder in WoW. Some players hate DF because now it simply became grind fest with random avatars, instead of social play/grouping with real people. But that’s also a problem with instancing in WoW. They created instancing and “forced” people to go there, and now they had to fix “LFG” problem with Dungeon Finder. Another spiral of changes that with simple “yes or no” solutions, changed large part of the game.

    3. Tactic/balance factor. In both PvP and PvE people stop adopting their gameplay to current situation. Now instead of tactics, all it matters is numbers and gear. Group tactics has been replaces with another solo I-Win button.

    4. Creative/fun factor. In Dark Age of Camelot there was a fun event in which hundreds of players created lots of low level mages, and went to High Level PvP areas to zerg high level players. With instaclass switch it will be something that won’t have this fun/magic factor attached to it. Making it harder will at least force players to create this special effort to create such events.

    5. Creative/balance factor. Also in Dark Age of Camelot there was a tactic in PvP named “ninja jump” in which bunch of rogues gathered togheter in large group of 30-50 players, and this group was attacking lords in keeps (lord is a main boss in each Keep, that must be killed in order to take ownership of the keep). So instead of pouncing a keep doors and fighting guards in long battle, they invented this tactic to go directly to lord. If you give players possibility to instachange classes, this will became another “only way to do this – got borred after 5 times”. People will lose interest in taking keeps.

    I could go on and on, with many more examples of how insta-class would affect a gameplay. For some players, insta-switch would save their time, but at the same time it will make game more dull, repetitive and grind oriented. This is the issue many other MMO’s must face. If you change many tasks to be insta, ‘suddenly’ players will get borred after shorter time. The best way to enchance game time is to give players many different ways to be creative and to achieve one thing in many different ways. Most instant actions are depriving players of creativity, and multiple choices.

  10. Aaron says:

    I love the “Dying” skill. Here’s hoping it’s an active ability, not just a passive skill.

  11. Really interesting post – thanks.

    It seems to me that a failure to consider the value of tedium is one of the things that’s really biting World of Warcraft right now. The design team there have been very public about their desire to streamline the game and ensure “fun” is first, and it seems to be backfiring.

    I’ve featured this post on the MMO Melting Pot today, btw – as always, it’s fascinating reading.

  12. kdansky says:

    Oh, the part about ridiculous and pointless mechanics is so true! WoW once did have a lot of these and it was a fun game, but in the process of perfect streamlining, it lost a lot of its charm.

    Another point you could talk about: It is perfectly reasonable to have a few things where the players can break an encounter in a creative way, as long as the greater game stays intact. If someone figures out how to solo Patchwerk (level 73 raid boss) at level 70, that’s fine, if it’s not the mandatory way to do it. Case in point: That fight of a rogue doing that did take multiple hours of perfect play and it is insanely hard to reproduce. But Blizzard still removed the abuse that made it possible. Or mobs that EVADE whenever they can’t reach you. Instead mobs should just get murdered, and write a message into a log file, so the dev can figure out what went wrong, and improve it. But don’t band-aid by patronizing the player.
    Of course, if every fight starts to get exploited, the designer has to do something about that. But please leave some quirky things in the game, and don’t be WotLK.

  13. Eric says:

    Part of the disconnect, I think, is that I haven’t been able to fully describe my game yet. For instance, traveling back to town isn’t a death sentence — it’s rarely going to be more than 3-4 minutes, if you know what you’re doing. And then a team member can return you with portal magic. In my head I’m equating “go to town” with “use an item on a cooldown” or whatever other mechanic. I just mean “you can’t do it all the time.”

    @Green Armadillo – I don’t actually have instances like that, at least not at first. Instead I have EQ2-style mega-dungeons where multiple groups can hunt at once in shared worldspace. I always liked that a lot better, it lets groups disband and re-band in the shared world in a natural way. (Plus, of course, it fits my engine limitations better… each instance neesd a hefty chunk CPU power to do physics/placement checks, so I don’t plan to have a lot of instances.)

    RE: Rifts, I played it a while, but not for too long. I’ll accept that it isn’t a problem there. However, the classes were pretty simplified there, and combat is very simple. I expect this game to feel different — not really requiring tons more strategy, but some strategy. Very simple example: there’s lots of DPS classes in this game, and sometimes a DPS class will have a hard time fighting an encounter where another one will have an easy time. If it takes just a few seconds to switch classes, doing so might end up saving 15-20 seconds per group encounter. That adds up fast if you’re farming. It’s also safer, too, since the shorter the fight, the less danger of death.

    But really, I may well be wrong about this. If the combat doesn’t come out feeling particularly different then yeah, I’ll take out the restriction. It’s always easy to take out restrictions… it’s impossible to add them after the fact.

    And if I’m wrong about the value of switching classes, well then I have a good 50% chance of being wrong about classes at all! If I let players switch classes constantly, I may just be better off letting you pick and choose abilities from all classes you have arbitrarily.

    As Bronte mentioned, it is arbitrary. But really the whole “class” imposition — restricting you to two of your many combat skills — is as arbitrary as anything else. Right now I think those will be necessary, though.

  14. Eric says:

    @Aaron – yep, when one masters the Dying skill, they learn abilities such as Feign Death.

    @Hugh – thanks!

    @kdansky – yeah, that’s an easy designer trap to fall into. You wake up, drink your coffee, and see a youtube video, and then gasp. “He exploited my dungeon!!! Time to hotfix!!!” Well, if it took a couple hours to do, along with a lot of preparation, the exploit is probably not a high-priority fix…

    @Jeremy – that’s fair, yeah, there are lots of ways to limit class changing without causing downtime. I’m definitely not against those… just not sure what will work best yet.

  15. Imakulata says:

    “And as another commenter ‘Expert Novice’ mentioned, maybe I should have special items that let you switch classes in mid-dungeon, but have a price attached.”
    Do. Not. Want. Putting a price on it is an incentive to grind. Putting a delay is better – you’re making the encounters faster by being in the proper spec instead of a hybrid but you need to spend time doing so. It will probably end up on the “do this to be optimal” side anyway but is much closer to being optional than Expert Novice’s solution.

  16. Pingback: When Tedium Makes A Game Better — MMO Melting Pot

  17. Jason says:

    My thought after reading this was “why allow class-switching at all?” A valid mechanic is having players pick a class (or two) and then they stick with that. If they want another combination, sort of like Guild Wars, they create another character. It has side-benefits in that you have more flow through your lower level areas and content.

    An alternative is a respec sort of feature. Let them switch to a different class anywhere they want, but charge them for it. So, perhaps you charge them in XP at some ratio of CurrentXP to NewXP. So if they do a lot of switching they will actually lose levels in that secondary class. Not sure it really works all that great for level-based games. Works better for skills.

  18. darkeye says:

    Gw 1 and 2 would be good case studies for what you are trying to do. I don’t think Arenanet consider the primary/secondary system in the first game a success, they changed it for the sequel because it was difficult to balance, very complicated for new players and they decided that they wanted the professions to have stronger identities. The ideal was for players to focus on the primary professions skills and taking a splash of 1 or 2 from the secondary profession, intended or not classes that have been revamped recently are much stronger when taking all primary skills. I’m probably biased but they seem to have got a good balance in the sequel between tying the player to a balanced build and giving them choice for only a few slots of the actionbar.

    It might help to consider a system where players pick a ‘vocation’ at the beginning of the game, but still keep it open to pick any skillset later. Three vocations such as Soldier (weapon master, strength, endurance, training), Ranger (rogue type, agile, athleticism, guile, wilderness survival) and Scholar (all knowledge based skillsets) should cover all or most possibilities, and give bonuses when a skillset that fits a vocation is picked as primary but not secondary. If there was some way to only allow players to pick a subset of skills (‘utilities’) when a skillset is picked as secondary, and then allow only the secondary to be changed outside of towns, the player would be locked into their primary profession until they return to town, it gives some flexibility and players are more attached to the primary.

    Heh and not to go on too much about Arenanet, the way they handle the cooldown issue on elites is to prevent the player changing them out of the action bar if they are on cooldown, which only works because they have a single slot designated for an elite skill.