Single-Shard/Server Gameworlds – EVE Online

Single-Shard/Server Gameworlds – EVE Online

EVE Online

EVE Online

Title: EVE Online
Developer: CCP Games
Publisher: CCP Games (originally SSI)
Release Date: 06 May 2003

Single-Shard Universe

One of the problems that seems to be regularly faced by MMO developers is that of how many servers to have at release. This is difficult to assess and, from what we’ve seen with many releases, very difficult to get right even by a team of veterans in the industry. The biggest hurdle seems to be that first few months when the flood of new players come in. In that group, you have a percentage that bought the game and, for a range of reasons, aren’t going to be playing past that first free month. This seems to be a consistently large number, and I’d venture to guess it’s at least 20-30% of the people that buy the box who do not play past the first month. You also have the majority of the players ingame for play sessions that are more frequent and much longer than normal. After that initial period, most MMOs start to see a sharp drop in concurrent users. The players begin to notice this and usually results in the Doom-n-Gloom prophecies on message boards about the game dying a premature death. This is sometimes exacerbated by the noticeable (but not necessarily steep) reduction in new players coming in, resulting in the earlier towns/areas appearing to be relative ghost towns.

Compounding the problem in many MMOs is the ‘big world’ design. Players ask for big worlds to explore and the devs go out of their way to accommodate that. The problem, however, is that it’s not enough to just have vast expanses. They need to be filled with something substantive and that is very hard to do. That, however, is a topic for another article. :)

One direction to go is the single-shard universe. While this design can create some major technical hurdles for the developers, it seems to present a solution to several of the major issues MMO devs face throughout the lifespan of their game. It is most beneficial to virtual worlds that have a focus on community, and several MMOs have taken this approach. One of the most notable ones being EVE Online by CCP Games.

It definitely resolves the issue of ‘dead’ servers. Unless your game completely sucks, you’re not going to have a dead server. Everyone in one game world means that people are not spread out across multiple shards. Even if  you have 500 people ingame on one server, the game is more alive than if you have 500 people spread out across five servers. Currently EVE Online has daily concurrent user counts of over 38,000 players but in it’s first few months, 5,000 was a lot more common. Had that 5,000 been spread across several servers, especially with the massive size of the EVE universe, the game would have certainly lost players to the ‘dead’ server perception.

It allows for more flexibility in controlling and shifting the story arc. With multiple servers, offering players a choice of options is very hard to do when it comes to major quests and plot events unless you are looking to have very divergent paths for the servers. The best that can be done there is either offer the appearance of a choice  or offer meaningless choices – neither one is really all that fun for the players. With a single  server, the option to present meaningful choices to the player is there. You still are following one path on one server, as opposed to the possibility of multiple story paths across several servers. For example, if the Evil Foo is guarding the Crystal of Safety, and the players need to kill him to get the crystal, what happens if a server decides not to kill him. Do you halt the story arc and leave him standing there forever on that server until he does? That doesn’t seem fair to the servers that killed him. Do you create a separate path for that server based on their decision? If so, then two or three major quests down the line and you will have your hands full managing divergent plots. Do you go in and use a character to kill the Evil Foo so that server moves to the next step of the arc? If so, then you have told the players their choice is meaningless.

There’s also a better chance of player notoriety. While only a select few ever rise to any significant level of fame, in a single server game world, a famous (or infamous, for that matter) player becomes known universally. With multiple game worlds, it is uncommon to gain notice beyond your specific server. Say ‘Chribba‘ ‘Winterblink‘ or ‘Goonswarm‘ to any of the quarter million EVE players, and it is very likely they will not only know the name, but will be able to relay a personal anecdote or two related to it.

If you’re building a sandbox, a single-shard universe seems the best way to go. The community builds stronger since it is not divided. Emergent behavior, fashion, ingame culture, and trends have a more widespread effect because the viral nature of them do not have to work against the hurdle of a divided (by servers) community.

The Council of Stellar Management, a commitee comprised of  player elected members of the EVE Online community, is much more effective because of the single-shard universe. Had EVE players been spread out among several servers, each with its own cultures and issues, addressing any one issue to any reasonable degree would have had a lot more hurdles.

This has been more of a ramble than anything else, but the overall point is this: A single-shard universe seems to offer several benefits to both players and developers alike. It helps prevent the ‘dead’ server issue, strengthens the overall game community, and offers developers more leeway to offer players meaningful choices ingame.