Tuesday, October 13, 2009

Put Newbies in the Best Possible Community

For big MMOs that cater to all the myriad playstyles, let’s ditch arbitrary server selection and instead come up with a model that works based on preferred styles of play.

Timmy likes PvE, Jimmy likes crafting, Joan likes PvP, Edward likes crafting but hates PvP, Dave thinks that PvE is a waste of time but loves to participate in the PvP and economy. Out of these five players, zero of them can make a meaningful decision about what server to join when they’ve first logged into a modern MMO. PvP-flag indicators do far too little to help people decide where to settle if flagging is even implemented—most games don’t bother—and roleplaying designations have even less meaning. Given only a list of server names there’s no way to make a meaningful decision of what server to join—a decision that can make or break a player’s in-game experience.

Let’s find ways to make server choice work better for the new player.

If the player has friends already in-game on their first log-in, let them provide the names of their friend’s characters and/or what guild their friends are in, and the player will be shuttled over to the appropriate server.

When an experienced player signs up for your MMO, give them a list of playstyles and ask them to rank them on a hate-dislike-neutral-like-love continuum (or have the player rank the playstyles from most preferred to least). With responses to five or ten such questions, you can gather enough information about the player to put him or her in a server with people with whom he or she will actually enjoy playing.

For inexperienced players, ask them questions about what kind of activities they would want to do in game with language that is newbie friendly. “Do you want to make the weapons that your friends use as they charge into battle?” instead of “Do you like crafting?”

Use this information to put people on servers where they can easily meet like-minded people and have fun. This doesn’t mean segregating all the various playstyles, it means integrating them where they are compatible and separating them when they are incompatible. If someone really hates getting ganked, don’t put them on a PvP server. If someone wants to craft and doesn’t mind PvP, put them on a server with other crafters—or put them on a PvP server that’s short on crafters.

A small amount of time spent gathering player preferences can lead to huge dividends when players find friends faster and get hooked sooner. There’s no reason to make player’s fire meaningful decision in game be a blind decision.

EDIT: Dblade, in the comments, suggested a better alternative to a Q&A with the new player. He suggests a Oblivion-like tutorial that allows the player to show their play-style through actual play. I think this is a great idea, though it can't sufficiently judge if the player likes PvP and crafting. A tutorial where the game learns the player's preferences coupled with a few simple questions before the player dives into the full game will greatly increase the chance of the player dropping into a community in which he'll want to take part.


Hirvox said...

Somehow this reminds me of the questions that serve as the character creator in Ultima games. Perhaps we should include a Bartle test in the server selection screen? The UI could be similar to the UI in Windows Wizards: A radio button set with "Help Me Choose A Server" as the default selection, "Join My Friends" as the second selection and "No, I'll choose my server manually" as the third.

Also, if you signed up from an invitation from your friend, your friends list should get pre-populated, including your friend and any/all people from his friendslist that he selected.

Glyph said...

I'd had the idea that maybe they could not only separate the players based on whether they preferred PvP or PvE, but also balance the player's abilities to either. That might be too much work to handle though...

Brian 'Psychochild' Green said...

I like the concept, but the disadvantage I see here is the potential to overwhelm a true newbie. It can be hard enough for someone to pick the role they want to play via class, now you're demanding more choices from them.

You also have the possibility that a group of friends new to the game might have even more dissenting opinions about which server to hit. "Nosepicker says it's better for PvE!" "No, Boogereater is where I can engage my PvP fetish!" "Fuck you guys, I'm going to Toejampicker because they need crafters!"

Instead of, you know, the usual intelligent method friends use. "I picked Firebelcher because it has the word 'fire' in it!" (True story, except for the server name.)

Jason said...

Rather than separate players by server, separate your world by continents and let players freely travel between them. At this point in my life, I refuse to play any game that separates people into servers and makes them pay to transfer. If I find out someone also plays the same game as me, I simply want to be able to play with them and not jump through hoops or have to say "Sorry, different servers, guess we can't play together."

motstandet said...

I also think sharding is an antiquated concept.

evizaer said...

Servers aren't an antiquated concept; Almost every MMO uses them. It's a de facto standard for managing populations above what one server can handle.

I'm also in the camp that wants one world for all the players to share. Servers create unnecessary boundaries in communication and play between groups of people that shouldn't necessarily be separated. They're completely arbitrary.

motstandet said...

They are a thing of the past; a RELIC OF THE OLD AGES.

At least they should be.

Dblade said...

I don't know if a single server with no channels can host hundreds of thousands of concurrent players. A big problem is that later on in your lifespan a single server game will have tremendous problems if it uses gated instances like raids for endgame. Endgame players can overwhelm the instace capacity-we had problems in FFXI on bahamut server where the pop was high enough that people couldn't get into them. I think most of the single server games don't really have real organized events like that, so its less of an issue.

As for the point, Brian is right, newbies wont know right away what they want to do, they need to play the game to discover it. Unless you want to segregate all newbies on an initial newbie server, than parcel them out after they get some experience and know what they want to aim for.

evizaer said...

In the article I mentioned that you could give newbies different questions to see what they'd most like to do. If you ask a newb 3-5 questions about what they're excited about doing in-game and you phrase the questions in an easy-to-understand practical manner, I don't think newbies will be at all confused.

motstandet said...

Dblade, I don't know if you are speaking of technical or design limitations, but the technical need to shard the game has pretty much been eliminated. EVE has been out since 2003 and is single-shard; Darkfall is single (regional) shard; the instance-based Global Agenda is single-shard.

Project Darkstar is a free, open-source Java solution to virtual world scaling.

Zoneless and Shardless You can have zones and shards if you want to for game design reasons, but they are no longer required for scaling.

Dblade said...


I don't know about that, EVE's concurrent player limit was 45k, and it has what, maybe 300k subs? Could a single shard system handle 90k+ concurrent users? Darkfall is low population by choice, so it is kind of moot. Until we see a mainstream, high population traditional MMO embrace that model I think it's much too early to assume the technical limitations of single sharding are moot.


yeah i read that, but I don't think asking players questions works, even general ones. You have to let them sample gameplay for a bit to give them an idea of what they like to do. Even in general. you may like crafting in one MMO, but in this one may find you don't like the implementation of it.

If you let players try stuff first, you'd have a better foundation to assign people, and you also just gave them a tutorial in the various aspects of your game. Like elder scrolls: oblivion, where you escape the dungeon.

Hirvox said...

Eve also does it's fair share of instancing: every star system is a separate instance. Without prior warning, a star system can handle around 800 players duking it out, with reinforced nodes it can handle around 1100.

A highly scalable, seamless MMO simply doesn't exist yet, and for good reason: as the number of participants in an event increases, the number of logical status updates increases exponentially. It's simply not feasible for the server to provide blow-by-blow commentary on every combatant, and anything less could lead to desynchronization. Throwing more systems at that problem doesn't really solve it, because they'd have to talk to each other in addition to talking to the players.

Bronte said...

I agree with you on some points.

1. What of games like Champions Online with one server and several instances. One good thing they do is that they list for each instance how many of your friends or Supergroup members are in that instanced version of the world. In my opinion that is a step in the right direction and makes the choice a lot easier.

2. The Q&A system before picking a server has two basic flaws. First, if you just got the game, you want to get to the action ASAP. Second, this system has the inherent problem of grouping all the newbies together. One of the cornerstones of MMOs is veterans helping out newbies. This system threatens that. How about that questionnaire is filled as the player goes through a tutorial and then ends up on the server that he or she is most compatible with?

motstandet said...

You guys are citing issues that aren't even issues.

Eight-hundred users fighting in the same area is absolutely astonishing, and definitely a feat of engineering. Albeit I don't even think players can process 800 ships at once. Asking for more computer scalability is superfluous if humans can't even keep up.

300,000 players is a significant statistic, and so is 45,000 concurrent users. That's more than a vast majority of games ever see in a lifetime. Could EVE handle double that, or could a behemoth with WoW-like numbers handle all its concurrent users? I have no idea. Keep in mind that WoW's shard clusters already share databases. The reason those shards aren't merged is simply because the game world is too small (and my grandma can process more close proximity users than WoW's servers).

evizaer said...

The question isn't "Will the Q&A idea lead to a perfect community?" The question is "Will Q&A lead to a community that's a better fit than a list of meaningless names?"

A quick Q&A while the download manager patches you up to speed before you first log-in is not going to drive players away. It doesn't have to be mandatory, either.

I think the Oblivion-style tutorial that determines what you might want to do is a great idea for MMOs. It's most likely better than the Q&A idea. I'll add it to this post so people notice it without having to jump to comments.

xenovore said...

Excellent ideas, all!

I think something like this would be very helpful (and possibly necessary in the near future as MMOs become more mainstream). However, it needs to remain optional (or skippable in the case of an Oblivion-style intro area); the more experienced players should not be forced to deal with anything like this.