first big problem in getting a multi-player game going, from the user's
perspective, is gathering the team. This can, as all such things
can, end up much trickier than it first appears. There are a number of
different gathering strategies and each should be considered for its pros
and cons when designing your game. Here are four common ways to
gather a team:
1) Next on the Bus -This strategy is basically first come, next served.
The game is like a ride in Disneyland. Games are filled by people in order
of appearance in "line". If it happens to be friends, great, but more
likely it is just a shared experience by a bunch of strangers. The advantage
is there is minimal waiting for the next game to start, and no hurt feelings.
The disadvantage is that friends can't chose to play together and there
is no balancing by skill level.
2) Pick-Me Style - The far more common approach is the Pick-Me style.
Most of the games, including Diablo work this way. Unfortunately it brings
back unfortunate echoes of schoolyard horrors everywhere, with a few people
waiting to be picked for what feels like their entire lives. Th Net tends
to bring out both the best and worst in people, so small kind impulses
are usually translated into inclusiveness and slightly jerky tendencies
seem blatant. (More on on-line chat psychology in the next installment.)
The advantage of this method is that friend groups can easily get and
play together. The disadvantage is that games often take a long time to
get going as people continually drop in and out of the group. This is
not as bad with games that can start with a flexible number of players.
But, it was a horrible problem for games with a set number of players, for
example, with the old Imagination Network's bridge games. Just as you'd
get the fourth person to commit, the second would drop out to join another
team. When you'd finally get started, one player would be dropped from
the Net, and the whole game would be screwed. This would definitely result
in a high frustration factor.
3) Wander and Team - Some of the persistent environment games have introduced
the concept of wander and gather. That is, just start playing, and if
you run into someone you like, play along with him or her. This works
much better in some form of RPG. AOL's old D&D game, Neverwinter Nights
was like this as well.
4) Once more into the Breech, Dear Friends - This is most common in the
Doom-like games. Just walk into a room filled with gun toting bastards
and shoot anything that moves. If you die, you come right back in to play
I wanna play NOW!
One of the big issues that affect the lobbying experience, obviously,
is how long do you have to wait to get into a game. There are two solutions
to this. Either have so many people ready to go and such a smooth system
that a train is leaving the station every 30 seconds, or, more likely
have a game that allows players to join mid-stride.
Persistent gaming is a different beast. Its lobbying issues are much more
of finding people to act as good partners or guides. Most gaming is more
a question of getting a quorum so the game is fun. One of my least favorite
things in the world is playing a game of StarCraft, getting stomped by
the one computer player in the first 5 minutes (because the sucker decided
to go rampaging with 6 Space Marines), and then having to wait an hour
for the rest of the people in the office to finish the damn game so I
can rejoin in the next round.
This is a big issue, particularly with local multi-player gaming. How
do you let the losers have something to do while the winners fight to
the death? When I was working for PF.
on the modem for the Sega Genesis, the most fun part was designing
multi-player game experiences. One of the things I worked on was a four
person chess variant called Chaos Chess. The board was two squares larger
on either side, and each player started in a corner position. The basic
rule was that once your king was taken, all your pieces changed color
to be part of the army of the person that defeated you. But then the game
would go on for a while without you, and you had to just sit there twiddling
your thumbs while you waited for the end. So we introduced the concept
of the Chaos Knight. When you lost your army, you came back as a randomly
placed Knight on the board. Basically, you got to be a pain in the ass
to everybody left. When you'd get taken, you'd lose a turn and then just
pop up somewhere randomly again. If you happened to take someone else's
King in the process then, boom, you're back in the game.
So the question is, how do you design something for the losers to do that
keeps them engaged with the game while the winner is still being determined?
Why? Because you want to keep them engaged in their current game so that
when the winner is determined, they will be excited to start a new game
with a chance to win again. Otherwise the winning becomes hollow
for the winner and winning becomes less desirable for the losers. When
you get to the top of the hill, don't you want to wave to all those at
My Buddy. My Target.
And finally, to end this first installment of four columns on multiplayer
game design, the issue of team games vs. world full of targets.
In a world full of targets, like most of the Quake variants, the purpose
of the Lobby area is threefold. Strut, Taunt, and Group. Strutting is
the cowboy who walks into the Wild West Saloon and saunters up to the
bar with the "Yep I'm bad, don't screw with me" attitude. Taunting is
the guy that's already sitting at the bar with his hat pulled over his
eyes and who says, "You supposed to be some sort of tough guy?" to our
friend walking in. And grouping is when the lesser gun fighters realize
that maybe there's a fight bout to break out and maybe they'd better pick
sides. The interactions you're going to get are, and should be, much cockier
and nastier. That's the flavor of the game. And you probably need to set
up some magic bar doors that don't let the inexperienced into a place
where they're going to get their asses kicked, or the tough into the newbie
areas to go on giblet melees.
The issue of gathering teammates is much trickier. Here you need to match
a blend of skill levels, so someone in the group can act as guide. (It's
much more fun for all involved, and cuts down on tech support big-time)
Here you need to create an environment that lets people gather together,
and makes it clear who wants to play what for how much time. (It's no
fun to gather a team only to find ten minutes into the first game that half
the team needs to go.) And you want to give people a chance to chat with
each other so they can get to know the other players. Chat is where we'll
start again next issue. The wonderful world of getting to know your opponents,
and letting them learn all about you.