The Case For Co-Op Games

I am running an experiment. My company's first game, AI War: Fleet Command, is an RTS game that is entirely cooperative. Not that you can't play it single-player -- you can, and it's completely fun that way -- but there is no player-vs-player mode in a genre that is known for its pvp skirmishes. Was this a smart design decision? Time will tell.

Why Co-Op?
I don't have any focus group data, survey results, or market studies to back up my feeling that co-op is important. However, it seems like there are more and more sites like Co-Optimus around, more and more forum posts from frustrated players wanting to know of good co-op games to play with their dates, spouses, or children, as well as a neverending flow of articles and comments in magazines about how there are too few co-op games (seems like Game Informer has something almost every month).

All of the above is anecdotal at best, but there is at least one fact that cannot be disputed: gamers, as a group, are aging. There are an increasing number of us gamers who have been buying and playing for ten, twenty, or even more years. For those of us who are married, having the time to sit down and play single-player games is a luxury; for those who are parents, I imagine it's a downright rarity. Playing online against strangers is not much different from playing solo in this sense.

Worse, when the only modes to play are competitive, any of your family -- spouse, children, or otherwise -- who are on the bottom end of a skill gap are not going to be too excited about playing against you. And, frankly, you won't be that excited about playing against them. The days when my buddies and I would spend hours competing against one another in Mario Kart, GoldenEye, or Counter-Strike are long gone; half of us live in different states now, and a lot of my friends have fallen somewhat out of gaming with the years, anyway.

Co-Op Is For Hardcore Gamers, Too
Oh, no! Is this evidence that gaming is dying? Or that gamers are destined to forever be a self-renewing population of adolescents and college students? Of course not. Anyone who looks at the sales of current titles can see that the market is expanding, and not just in the casual games space. What I do think this means is that there is a new market segment for companies to explore. And on paper it seems like a potential goldmine.

I'm speaking, of course, of hardcore gamers who want co-op. It solves all of the problems I mentioned above: you play it with your family, instead of alone or against them. If there is a skill gap between you and other members of your family, no problem! In a game designed properly for co-op, there will be difficulty settings that are appropriate for this. There will also be mechanics in place to make sure that the weaker players can be assisted by the stronger players, and for insuring that the weaker player always has something to do and a way to feel useful, without the stronger player having to play below his/her true skill level.

This sort of functionality is hard to tack on at the end of a development cycle. To create a great co-op game takes forethought and will affect a substantial amount of the game's design. This opens the doors for many new design concepts that haven't been seen before in a multiplayer context.

Unfairness As A Design Concept
Perhaps the biggest design concept is unfairness: in a competitive multiplayer game, everything has to be finely balanced so that no player or team can exert an unfair advantage through any mechanism other than skill. In a co-op game, you're free to have your game be as unfair as you like. In fact, unbalancing the players on a team can even be a positive thing, because it encourages specialization and teamwork.

It can even help to naturally give weaker players a boost, leveling the playing field so that the advanced players take the more difficult position/class/loadout/route/whatever, while the newer players take the easier path. Thus everyone is doing something skill-appropriate, while still contributing to the team goals. The weaker players aren't overwhelmed and the stronger players aren't bored.

Co-Op-Specific Technical Improvements
To cite a purely technical design change enabled by co-op, the networking and threading code in AI War were very much impacted by the decision to make the game purely co-op. In a competitive RTS game where you are commonly playing short matches with one or more strangers, a star network design is required so that if any one player drops out, the game can continue without them. In a co-op game with friends, it is instead more important to be able to save and resume multiplayer games.

When playing a co-op-focused game with friends, the same people generally play through the entire game together, so this means that a client-server network model can instead be employed. This vastly reduces the network load on all of the clients, and doesn't affect the network load on the host one way or another. This is great for players on slower connections.

Another benefit also arises from the client-server model: when there is a single host against which all other game simulations are synced, some processes can be run on the host alone. In AI War, this lets us run a secondary AI thread on the host alone: the AI issues commands via a similar underlying mechanism to how human players would. This vastly reduces the system requirements for all of the computers except the host; quite often there is one power user in a group who has hardware that is better than the rest of the group, and now that hardware can be put to use without requiring that everyone else spend money and upgrade. This also allows gamers to use an older, otherwise-discarded PC as a client for other family members. My wife uses my old PC as a client when we play AI War together, for instance, and it gives a great experience for us both; otherwise that older machine was just wasted.

Asymmetrical Game Design, The Easy Way
Other benefits are mostly in the game design arena: when fairness is not an issue, you are free to make much more interesting opponents. Mario faces off against everything from lowly Goombas, to the slightly-better Koopa Troopas, to giant and powerful bosses. Just think how boring the game would be if Mario's only opponent was Shadow Mario, an evil counterpart that has exactly the same abilities as the player.

Co-op games, like single-player games, can pit the players against overwhelming odds. This sort of scenario rarely, if ever, comes up in competitive multiplayer, but it's something that happens all the time in real life. In AI War, for instance, the human teams are vastly outnumbered throughout the entire game (as in most movies or books in which humans face off against superior aliens or AIs), but it's still "balanced" in the same sense that Mario games are, because the scenario is still quite winnable for the human players (assuming they are playing on an appropriate difficulty level).

Where Does Co-Op Fit In The Current Pantheon?
The point of this article is not to say that I think that there is no place for single-player games or competitive multiplayer games. I'm a huge fan of single-player FPS games, the Silent Hill franchise, platformers, and competitive multiplayer games like UTIII. Plus about a hundred others. However, I think that the "cooperative hardcore gamer" demographic is a growing one, and an underserved one. We need games of all kinds, and at the moment what is most lacking is full-featured cooperative play. I love the fact that Warhawk lets my wife jump in and join me at any time, I just wish it let her have a unique name, her own stats, ranks and advancement, etc. I absolutely adore the co-op in PixelJunk Monsters, but I was perplexed when they made the trophies single-player only. Midnight Club II was immensely fun to play in a quasi-co-op mode, but they took out splitscreen altogether in the current-gen sequels.

In many ways, despite some of the high-profile co-op games that have come out in the past year, it feels like the industry as a whole is trending a bit backwards on the co-op front except when it comes to networked play. That's another thing: when it comes to consoles in particular, splitscreen is the name of the game for most co-op players in the "older hardcore gamer" demographic. When I want to play co-op with my wife, she's sitting in the other chair next to me, not at the other end of an Internet connection. We don't have the money or the inclination to buy two consoles, two TVs, and two copies of every multiplayer game that interests us, and I doubt anyone else does, either.

AI War requires a single computer per player, but that's more the PC game mentality; PCs are a lot more plentiful than consoles and TVs in most households. It also requires a single copy of the game per player, but we priced it at $20 (as compared to the $60 that most other new games in the genre cost) partly in order to be friendly to families that might need multiple licenses. The next game from Arcen Games, Alden Ridge, features co-op on a single PC, using a single keyboard or multiple game pads. There's room for many different co-op models, I think, and the appropriate one depends on the specific game.

The Case For Co-Op, In Conclusion
There are many great ways to integrate co-op into your games, but the key thing is to at least consider it. The question should not be "how can I add this feature with the least amount of effort, since only a few people use it?" But rather, "how can I make this feature really shine, so that more people will buy my game specifically to have it?"

Intuition and observation lead me, as well as many others, to believe that widespread co-op gaming is coming sometime soon. Most likely in the next ten years at the outside -- and it will be big. Now is the chance to be an early mover, to embrace the challenges of this new game mode and come out ahead of the rest of the pack. You might wonder if, since all my own games are co-op, I'm worried about increased competition from other co-op games making my games seem less unique. I say it can't happen soon enough.