Monday, March 8, 2010

Defining Massivity

I am so sick of people calling various online games "virtual worlds", or worse yet, "massively multiplayer online (games)". I hope to make a few distinctions, namely the difference between a virtual world and a virtual environment, and when a multiplayer online game actually qualifies as massive.

Damion Schubert noted the selling point of MMOGs compared to regular online games. That difference is "massivity", the potential for hundreds of users to interact in a virtual environment. Since Schubert's comment was in passing during his presentation, he doesn't seem to provide a formal definition of massivity, and I don't know him personally, so I cannot ask him for one.

Ultimately what an MMOG design hopes to achieve is a feeling of massivity. This is an aesthetic experience which cannot be easily quantified nor defined, but I will attempt to do so. Massivity is the feeling that the user is part of a large world which changes without her being there. It is the potential for hundreds of players to interact in some virtual space, all perhaps with different goals.

We can have games without massivity. But we cannot have massivity without a virtual environment, specifically a virtual world.

I like to use a layered definition when talking about MMOGs: virtual world + game = MMOG. Most people can identify a game or at least hazily understand that there are game systems at work when they experience them. But everyone from bloggers to journalists to game designers seem to forget what a virtual world is.

I define a virtual world to be a globally-accessible simulated, persistent environment in which users interact through an avatar proxy. A virtual world is a virtual environment with the following constraints:
  • The environment must contain the concept of location. It must be able to relate entities in the environment to the user with positional information. A chat room is not a virtual world.
  • The environment must persist between play sessions. It must convey the notion of a "living world" which advances while the user is not engaged with it. Any instanced encounter with an end is not a virtual world.
  • The environment must be globally-accessible and consistent, meaning all agents in the environment could potentially congregate at a location. This is technically impossible, but the impossibility must be hidden to the user. Any user understands that as long as he is a part of the virtual world, he can meet (intentionally or by happenstance) any other user in the world.
Channeling is a solution to a technical problem which diminishes the feeling of massivity because the game replicates the same environment, breaking the consistency of the world, and no longer are the technical limitations of the environment concealed from the user. I am not saying that channeled zones are not virtual worlds; they simply break "immersion". As an aside, has anyone ever experienced a channeling system which wasn't annoying or confusing?

Studios are releasing many online games which are breaking molds. They call it a "hybridization"; I call it exploitation of the buzz surrounding MMORPGs. Facebook labels games like Farmville and Restaurant City as virtual worlds. Debates ensue on whether or not anything with character persistence (e.g. TF2) is enough to qualify the game as an MMOG. I find this talk very dangerous because it dilutes the definition of virtual worlds and MMORPGs.

Is Diablo 2 an MMORPG? By Farmville or Global Agenda standards, it would appear to qualify. It has privately instanced virtual environments, character persistence, and public chat rooms serving as lobbies. But there is no global environment where any one user could accidentally interact with someone else. The world does not evolve and move without the user present. Sure items are traded and bands of players do quests together, but is there really a persistent world anywhere?

Any of the recent shooters with character and item persistence, whether or not trading is implemented, still reside in game spaces, not virtual worlds. Games are played on maps which have a beginning and end. They are highly structured games with limited participants. There is no central gathering place where players can interact and put their mark on the world by simply standing around. Character persistence is not world persistence.

MAG or even a hypothetical game with more than 128 v 128 battles are no more an MMOFPS than TF2, i.e. they are not MMOFPSs. If a player logs off in the middle of a battle, his team may lose the match. But the buck stops there. There are no repercussions in the larger game world, because there is no larger game world! His guild doesn't lose land nor is his skeleton a terrain decal for the next week; the players lick their wounds or bask in victory, and they start a new game. Once again, character persistence is not world persistence.

Using today's weak definition, we could classify any online multiplayer game as an MMOG. That would defeat the point of creating genres at all, and we would be back where we started: what is the difference between Everquest and Dungeon Siege? Massivity.


Logan said...

great article. i like your definition better than most i've read recently, the only issue i have is with your views on channels.

lets use WoW as an example... say you have 5 WoW servers, without channels, players from 1 server will never be able to interact with players from another server... but if you introduce channels, now you can merge those 5 servers, and now players who previously couldn't interact now have ability to. (even if it takes a few click to do it) this seems to me like channels can make the game MORE massive.

granted, channel systems in most games seem like an afterthought... and i've yet to play a game with a well implemented system... but that's not a fault of the channel system itself, it's a problem with the designers and the execution of the feature.

also this is assuming that the channels can support a sufficient number of players to feel "massive".. but once again, that's not really a problem with the idea of channels itself, it's just poor implementation.

motstandet said...

I never used player numbers to define massivity, and more players does not mean there is another scoop of massivity.

Using your example, let's say there are 5 Ironforges which combine the populations of five separate Dwarf capitals. If a player runs through one of those channels, they see no more players than they would running through a non-channeled zone. The only difference is that now they have to coordinate with other players or friends to determine which copy of IF they are actually in. "I am standing on the bridge." "Well I don't see you!"

I have yet to see these severe player coordination problems with channels resolved.

Realms are a tricky subject, and I plan on writing a post on them and what they accomplish. Personally, I would rather have a seamless (even with zoning) world than a world with multiple copies of the exact same area. Channels feel dirty to me.

Anonymous said...

what about global agenda (which seems to be the hot topic lately) it has an overworld, but it's separated from the game itself, this is, the game is a set of limited battles, like TF2 or any similar game, but these battles do have an effect.

Personally, i prefer a single world, even if it feels much more deserted, even if it's more complicated, it allows you to ignore any server-related problems

Logan said...

you're right... numbers aren't expressly given in your definition... but i'm still pretty sure that if given 2 identical zones, one with 10 people in it, and the other with 50 people in it, you're going to feel like the zone with 50 people is more "massive"

then again, maybe the number of players deals with the "multiplayer" part of the MMO acronym.

you don't talk about number of players in the definition... but do you really think it's possible to consider a game "massive" without taking the number of players involved into account? (maybe it's not a defining characteristic, but it certainly can differentiate between levels of "massive")

sorry for putting words in your mouth, i guess i just don't understand how you can define an MMO without addressing the number of players in some way.

Logan said...

one more thing... i've met a lot of people that play WoW in RL, but on different servers than i do... if Blizz used channels instead of completely separate servers, i would have the opportunity to play with my RL friends... but without channels, i'm just shit out of luck... i guess i'd rather have the option to play with anyone else with the game installed (even with the slight annoyances that come with channels) than to know they're out there but i can never hope to join them.

maybe that's just me though.

Tesh said...

Do we really have "world persistence" anyway? Kill a critter and it comes back. Clear a dungeon and it's all reset when you go back. The world never really changes, barring a game expansion or two.

What game design elements contribute to "world persistence", and are they marketable?

motstandet said...

There is definitely a population threshold for MMORPGs; it's typically called "critical mass" and happens to align with business and financial numbers. But every game is different, and the world design dictates what this magic number is. 4000 concurrent users in WoW works because the world is designed around this number. EVE would be an empty place with only 4000 concurrent users.

And I suspect you only dislike the Realm system because it is an irreversible decision. What if your character persisted outside of a specific realm, and you could pick your realm every time you logged in? (This is something we plan on doing in our yet-to-be-announced, super-secret MMORPG.)

motstandet said...

The world continues to exist without the player present. That is the persistence.

Other entities in the world may not persist (such as monsters), but that is because they are actually resources, and the world needs to regenerate them somehow. I guess we could wait 4 years for critters to be born and mature, but I doubt any players will be around at that time.

Logan said...

"And I suspect you only dislike the Realm system because it is an irreversible decision. What if your character persisted outside of a specific realm, and you could pick your realm every time you logged in? (This is something we plan on doing in our yet-to-be-announced, super-secret MMORPG.)"

that would be great... but what if you log into one realm to play with one group of friends, then you want to switch realms to play with another set of friends? do you have to completely log out and back in? that would be even more of a pita than channels...

if you're going to be able to swap realms when you log in, why not just use channels?

it seems like you have a personal vendetta against channels... why all the hate?

motstandet said...

You bring up a very good use case that I hadn't considered. Namely coordination/communication between those different groups of friends. With a single-shard, channeled world, the player could communicate with all of them, but have the burden of juggling channels. In realm hopping, the player would have to log on and off, as you describe, unless there is a communication medium between all the friends (and then still they are not all mutual).

I prefer realms because it is part of your character's identity. For example, talk to anyone who played on the WoW server Blackrock (like me), and we will tell you that all other servers are inferior. That is a ridiculous claim, but it is a claim which happens only under a realm system.

No channeling means that realms are smaller, and thus the realm community is smaller. Players have a much better chance of creating and maintaining a reputation with a smaller, closer community. "where everybody knows your name..."

Dblade said...

You are wrong in the main argument although your side argument is good. The problem is you give a great definition for what constitutes a Massively Multiplayer Online Role-Playing Game but the focus seems to be just on what determines Massiveness itself.

That's just a function to players in a playspace. If we someday get a MMOFPS that can have 500vs500 battles or more, it's massive because of the size of player interactions. Otherwise you find yourself claiming a game which supports single battles larger than most MMO server populations isn't massive.

This needs to be qualified though by optimal or expected performance. A game that offers 500 vs 500 battles is still massive in scope even if the actual playerbase is small.

I think your argument is much stronger for games which claim the MMORPG title but fail to live up to all of your criteria. Some games should just be called MOGs or net games. Champions Online to me is closer to what an online Grand Theft Auto game would play like (although its eerily similar to the xbox360 game Crackdown, just without the driving aspect.)

Dblade said...


The channel system actually works better many times over the realm system. I've played a couple of channel-based games and I think I can spot the strength.

Channel or instanced based games center around objectives that are extremely vulnerable to being choke pointed. Maybe it's because their areas are smaller overall, and they simply can't pack enough mobs to support an unchanneled realm. In any case, they channel and instance in order to divide the population efficiently and reduce competition.

It's a false dichotomy though, all modern MMOs make use of instancing to do so in high-traffic areas like raid content, but some are too small in scope to allow a full server so they split it. Champions Online would be impossible with a single server population of 2-3k.


Good article though. I'm sorry to sound negative but it really is a spot-on definition of a mmorpg.

Kenny said...

I feel dumb but what is "channeling"?

motstandet said...

Dblade, I still propose the 'massive' in any MMOG is more than just a numbers game. As I said in the main post, any FPS with 256 or more player battles is still just a FPS unless it has a virtual world component.

Kenny, channeling is a zoning device used in virtual environments to make it appear that more players are in a location than there really are. Multiple public instances are spawned, most of the time dynamically, to accommodate an increasing influx of players to the zone. Sometimes communication channels (not to be confused with zone channels) span the zone channels so players can communicate, but the players cannot see each other or interact in any other way. Monsters and other resources are local to each individual channel.

Players are usually given the choice to switch channels, but when and where this choice is present is dependent on the implementation. Channeling is used in EQ2 (I know it was used at release. I have no idea if it is still there.), Age of Conan, Champions Online, Aion, and Global Agenda.

Logan said...

you're absolutely right that a small population tends to lend itself to a better "community"...

but what if you used channels to help achieve that smaller community? so instead of taking 50,000 players and splitting them up into 10 separate realms arbitrarily, maybe you could use channels to simulate those 10 separate realms... but instead of just having numbered channels that mean nothing, you could give each of the channels a different meaning... so maybe 1 channel is more for "hardcore" players, and another is for "casual" players... maybe 1 channel is more for PvP and another is more for PvE.. maybe 1 channel is for large established guilds that aren't interested in recruiting new members, and another is where guilds and players that are looking for a guild go to meet each other... and these aren't really even the best ways channels can be used (i'm saving some of those for my own MMO).

i guess when i think of channels i see a lot of potential to create MORE community, they just haven't been used in that way before, so it's tough for people (even designers) to see the potential there.

each channel could have slightly different rules, like the PvP and PvE servers in WoW... but you could take that idea and go so much further... and by funneling players with similar playstyles and interests together, you create a better environment for relationships to form, and a better chance for an engaged community to evolve.

channels have a ton of potential... don't make snap judgments about them based on their existing implementation... they could be done soooo much better.

motstandet said...

You have convinced me to take another look into Channels, Logan :P

Logan said...

sweet! i'm glad. looking forward to what new ideas you come up with :)

Kenny said...

@Mot: ah, ok, funcom uses it in anarchy online as well - though only indoor areas like shops and 2 public static dungeons to keep population in sane (not insane) levels.

@Logan: "by funneling players with similar playstyles and interests together, you create a better environment for relationships to form, and a better chance for an engaged community to evolve."

Sure. So you want to make a community ripe for the picking for the right people. Or do you plan to actively police them and enforce the rules? I'm not sure if it's feasible at all. All that asshat Myers showed us was that jerks don't give a flying f-ck about community reactions, they need developer intervention.

Dblade said...

Hey Mot sorry, didn't realize it was you posting. This is what I get for commenting late at night. Now you can have a turn to deal with me. ;)

You are right, but also keep in mind MMO is an adjective. It can still be a FPS and be an MMO. It just needs to be online with a large amount of players in a gameplay unit.

Thing is, MMO used to be only capable in MMORPGs because of network and latency issues. Because of that, a certain style of game existed centered around a persistent virtual world. You are making the mistake of trying to make that game the standard of a more general definition.

The feeling of massivity doesn't need persistence. If you can play with hundreds of people in any genre of game online together, its an MMO. If it's a demoliton derby arena or car shooter game that can have hundreds of people drive and shoot it out, is it really that less massive because it isn't persistent?

Logan said...

@kenny - griefers are going to grief no matter what the developers do... by your same logic we simply shouldn't have the internet because hackers will steal our information and cause us hassles (my account just got hacked a few days ago, but that doesn't mean that blizzard should just scrap just because a small segment of the population are idiots doesn't mean you should design specifically for those idiots.. especially when they're going to find loopholes and creative ways to be an idiot no matter what you do.

personally i think that just suggesting to players how they can organize themselves to get the most out of their game time will be enough to confer a benefit to the community... but maybe that's not enough and you'll have to make some mechanics/rule changes that make designations even more clear... i don't know exactly, it hasn't really been done before... but i definitely think it's worth exploring and has a lot of potential.

yes, something like a "casual" channel would be very tempting for a ganker... but only if it's applied to an open pvp zone... there are so many other ways a system like this could be used... it all depends on the game and the specific zone... maybe for 1 type of game a casual channel would work fine, for another it wouldn't.. maybe in one zone a pvp specific type channel would work, but in another it wouldn't... it's up to the designer to determine what would be best for the game. you're looking only at the worst possible scenario, which is generally a good way to assess a design, but in this case that worst possible scenario can be avoided all together by using a bit of common sense.

hmm... just reread your post and i think you misinterpreted my use of the world "rules" ... when i say rules i mean how the game functions or how the mechanics work... not rules as in what we think of in real life, where they're more or less just guidelines that can be broken but have consequences.. in a game, rules are what define the game, everything operates within those rules and they simply cannot be broken, except by bugs and exploits of course.

does that make more sense now?

motstandet said...

Dblade said...
"It can still be a FPS and be an MMO. It just needs to be online with a large amount of players in a gameplay unit."

What or who dictates "large"? Assuming we maintain the 'map size' to 'number of players' ratio so that relative map presence or occupation is held constant, at what point does the game cross the threshold into massive? At 60 players? 128? Perhaps at 2 sq km?

Or are you saying this is more of an emotional response, "we'll know it's massive when we feel it"? That might be the best we can do with MMOGs outside the realm of RPG.