Posts Tagged ‘task homogeneity’

The State of Cooperative Play

March 21, 2011 10 comments

I have never played a co-op game in my life and neither have you. If you think you have, you are mistaken. Let me explain.


I said that to provoke you and to warm you up. I really want to argue for these four claims:

1.’Co-op games’ vary massively in terms of the way in which they are cooperative.

2.Because these differences are very important, we need a way of understanding and describing them.

3.Distinguishing between ‘loose’ and ‘close’ (as in the adjective ‘close’) cooperative play is a good way to do this.

4.Close co-op games are the best sort of co-op game, which is a shame, because there are almost none of them in existence.

I will start by introducing the idea of a spectrum of cooperative gameplay that ranges from ‘loose co-op’ to ‘close co-op’. These terms refer to the group dynamics involved in that particular sort of gameplay.

Loose co-op gameplay: your actions and thinking need not take any account of what your partner/allies are doing. You can happily move things forward by acting independently of your group/team and without having to coordinate your activities or respond to what everyone else is doing. You can play like this without communicating with your team-mates or discussing plans with them. This is a bare minimum of ‘all being on the same side’.

Close co-op gameplay: your actions and thinking have to take account of what your partner/allies are doing in a direct, sustained and (at least relatively) deep way. You constantly have to read the situation you and your partner/allies find yourselves in, take stock of what resources and abilities you have between you, think about how you should all respond to take care of things, carry out your part in this, and then monitor the situation to see how it develops. This is not a solo activity – group communication and quick group decisionmaking are crucial here – usually frantically improvised.

The spectrum has ‘loose co-op gameplay’ at one end and ‘close co-op gameplay’ at the other. Note that these are two extremes, and probably no game is a pure example of either. One game can contain lots of different sorts of gameplay: it might have some sections or modes that were closer co-op than others; it might have certain mechanics that were loose co-op whilst others were close; and at certain moments – for instance if a player is knocked down or incapacitated – one game might suddenly become a lot more closely cooperative. The gameplay will also vary depending on the players involved and how they approach things. With this way of thinking in mind, we will now have a closer look at why different sorts of gameplay fall at different parts of this spectrum, and then we can move on to some examples.

Identifying the foundations of loose and close co-op

What makes a game play like a close co-op game? What leads to this sort of team play? We can look at the question from two perspectives. Firstly, we can looks more at the enemies, puzzles and obstacles faced by the players. Secondly, we can look more at the distribution of powers and abilities (required for overcoming obstacles and enemies) among players. Whilst they pretty much describe the same thing in different ways, they are both helpful angles to consider. You can measure how close the co-op gameplay of a game is by looking for the presence of the following two features:

i.Tasks (enemies, puzzles) that no one person can deal with alone. You find this when tasks are sufficiently complex or require multiple simultaneous actions or application of skills that one player simply cannot provide alone.

ii.No one player having the necessary powers/abilities/tools to contribute everything necessary for progress past the tasks encountered in the game. This is effectively the flip side of (i).

The more strongly you find (i) and (ii), the closer the co-op gameplay. An example from Lara Croft and the Guardian of Light will demonstrate this. Many of the obstacles encountered by the players are too tall to be climbed unassisted – so one player must provide the other with either a boost-up or hook up a rope for them to climb with. Any time you get to such an obstacle, you cannot deal with it alone, so you need to think about what your partner is doing and how you will get past it. The more strongly you find #1 and #2, the more frequently you have to work together in this close way, the closer the co-op gameplay is.


A closer look at combat and co-op gameplay

It’s worth having a closer look at combat, because this is a large part of most co-op games and is a central part of the genres that most co-op games originate from: action, FPS and RPG. I want to show that combat is a largely homogeneous task, that it therefore does not display the features of (i) or (ii) and that an emphasis on combat results in a loose co-op experience.

As an example ‘combat task’, lets take a room full of monsters in Diablo that have to be butchered by our brave party of adventurers. Each enemy is effectively just a ‘deal x amount of damage’ task for the group: after taking x amount of damage, the enemy dies and is no longer a problem. Now because a point of damage done to an enemy is the same no matter its origin, anyone in the group can contribute the damage and help with the task. And anyone can do the entire task single-handedly if they are able and willing – it is no requirement of the task that more than one person do the damage. Dealing the damage to this monster is a completely homogeneous task, anyone can do it on their own and everyone has the necessary skills to carry it out. Multiply the ‘damage dealing task’ involved in one monster by a room of monsters. Then (ignoring complexities like boss monsters, resurrection and healing), we can just see the room clearing combat as one big task to munch through a stack of monster health. Combat at its most simple and homogeneous: one stack of hit points that has to be munched through; co-op at its loosest.

This is obviously an extreme example – most of the time there will of course be damage prioritising between enemies to ensure they get killed more quickly, enemies and allies will be throwing around healing spells and other magic effects, some allies and enemies will have special abilities and immunities. But the point is this: the combat task is a largely homogeneous one that anyone can contribute to on their own and without having to think about their teammates; often if you just look out for yourself, keep doing damage and try not to die, you will contribute positively to the group and to its combat-task. This is very loose co-op.

Take Borderlands as another example. You spend most of the game chasing some quest or other and that usually involves either killing someone, retrieving something, or both. This means lots of combat. In Borderlands, there are four players classes that seem to play very differently. The Siren can ‘shadow-walk’, phasing out of normal space and into the spirit world, going invisible and invulnerable for a few seconds before phasing back to the real world in a blast of energy. The Soldier can make turrets; and so on. Despite these superficially very different abilities, they never come to anything particularly deep in terms of how the game has to be played because functionally they all deal with the same core task in the same way. They all deal damage, all chip away at that hitpoint stack and because this is a homogeneous task and the central part of the game, there is no real distinction between the classes in terms of their co-op gameplay function.

In Borderlands, all the classes are ultimately dealing with the same homogeneous task in only superficially different ways. All of them deal damage, all of them have ways of dealing with heavily shielded enemies, all of them can do the different elemental damage types and so on. There is no need for people to work together because despite their very different abilities, each player is set up in a such a way that they can perform all the tasks the game could ask of them. This is to be expected from a game that is designed to be just as playable for one person as a four person group.


Loose co-op has all the player roles functionally homogeneous, interchangeable and inter-substitutable. No need for teamwork here, or even a team. The tasks are sufficiently simple or homogeneous and the player capabilities sufficiently comprehensive relative to them that co-op play can easily be replaced with solo, or quasi-solo play – where players operate largely independently despite being in the same game.

What sort of combat would be conducive to close co-op gameplay? It would have to require that the players worked together to get through it, and this would mean either their individual powers not being sufficient to defeat the enemies alone, there being more demands and problems to deal with at once than any one player could cope with, and more tasks/simultaneous inputs than one player could perform or input alone. Some examples: Boss battles where you have to each focus on one of a boss’s weak spots to kill it – No one person could deal with both the weak spots at once. In L4D and L4D2, any player unlucky enough to get covered in boomer bile loses the ability to see effectively, and all the nearby zombies will swarm at them. This creates a situation where the other players in the game have to help or else they will lose a teammate.

An enemy that could not be defeated by one person alone would promote close co-op play: with one person having to lure a particularly vicious enemy in one direction, exposing it to a sneak attack from their partner (like with the Hobbits in Lord of the Rings: Return of the King who could not win many straight-up fights with the larger monsters). In Resident Evil 5, there is a section in a dark cave system where one player holds a lamp and generator to light the way which takes up both their hands. This player has to shine it round the caverns to reveal the zombies in the dark corners and lead the exploration – at the same time, the other player has to kill whatever gets revealed and keep their partner alive. Resident Evil 5 has a section where the players are on a boat – one drives, and the other takes charge of the shooting. Imagine a game where each player could only use one element of magic, so a certain enemy has to be frozen by one player, then shattered by another; this would be another example of close co-op combat.

So, more complex combat tasks and asymmetry/limiting of player skills can foster close co-operative gameplay even in a heavily combat focused game. But this sort of gameplay is not found for any sustained period of time in most co-op games.


Why is close co-op good?

Close co-op gameplay requires uniquely tight group communication, planning and teamwork. When it all works smoothly (or near enough smoothly), you can see the sum being greater than the parts – insights and ideas can be transmitted amongst the group members, collective brain power and gaming skills applied to any task or problem that is encountered. This provides a way of dividing human attention and focus among more complex tasks and challenges than we could deal with alone, so we can have more demanding and interesting situations to deal with, which has to be a good thing.

Then there is the social aspect of close co-op gaming. Because everyone is focused more intently on what everyone else is doing, it allows for more recognition of when people could work together or provide someone with some support. Perhaps more importantly, because everyone is focusing on similar things, it allows for greater opportunities to spot the same things, comment or joke about the same things – because you are all confronted with the same raw material in front of you, more or less. I had some issues with lag in Lara Croft and the Guardian of Light which led me to kill myself repeatedly in a spike trap; this became funny because my co-op partner I could both see my suffering and struggling – if their attention was elsewhere, this opportunity for humour would not have happened. Close co-op has a lovely scrambly improvisation-y quality to it, with confusion, panic, rough plans and botched executions. There is a drama to this sort of gameplay that more isolated and individualistic play can never have.

This is why I say that close co-op games are the best co-op games and are some of the best games, full stop. The problem is that there simply are not many games that any significant degree of close co-op gameplay – I’d love to hear some suggestions and recommendations of some. We will now move on to look at some case studies – I won’t be able to say everything I want to about these games here, even in terms of how they work as a co-op game, but I will draw out some of the most important points. I will consider the two best examples of close co-op gameplay I have encountered first: SWAT 4 and Lara Croft and the Guardian of Light.

Case Study: SWAT 4

In this game you play as a member of a SWAT team. This is a game about working carefully and methodically as a team – and then trying to deal with things as best you can when it all goes horribly wrong. You start off with a vague idea of a situation – hostages, gunshots, bombs – and a map of the building involved. Your job is to make the area safe, securing hostages, defusing bombs, and arresting or incapacitating any hostiles. The cooperation starts with the reading of the floor schematics and intel on the situation. How are we dividing into fireteams? Which route shall we go in through? Who will take which equipment? If I go for the breaching shotgun to open the doors, you can take the optiwand (a fibreoptic camera you poke under doors to see what is going on on the other side); as you are loading up on flashbangs, someone should take something more aggressive as a backup. Etc, etc. No one person can take all the different bits of equipment that you will be called on to use in the mission, so you have to divide up the responsibility for the equipment and associated tasks.


In SWAT 4 you are forced to divide up the combat tasks between your team because you will be slaughtered otherwise. As a squishy human being, you can only take a very limited amount of damage, even with full armour on. On your own, you will die quickly when facing any group of enemies. There are more angles and tasks than any one person can cover – you cannot see all the different doorways at once, you cannot throw a flashbang out there and be covering this person’s back with your assault rifle. You cannot take out the gunman before that hostage can be killed and check behind that door. Unlike in most other games, the combat in SWAT 4 demands close co-op play. The combat has to be divided into discrete tasks, and that is why it encourages this sort of approach.

Procedure is important too. You have to shout appropriate warnings to anyone you encounter, try to detain rather than kill people, report evidence, report back to control when an officer gets killed, and so on. It really helps to have everyone working closely together on this as you have to perform these tasks in a very tense and dangerous environment.

Looking at how an unexplored room is dealt with should highlight most of the relevant features of SWAT 4. Someone gets out the optiwand to sweep under the door as the rest of the team take up positions to cover. The optiwand provides a view of what is on the other side of the door. You look through and read the situation. Movement; feet; one -no, wait – two gunmen. And a hostage off to the side – is someone in that side room too? You communicate and quickly work out a plan between you. As I detonate the C4 on the door you have a flashbang ready to throw into the room. Then you run in first and drop the gunman on the left, you through second and take the guy on the right. And so on.

After working out the plan, you turn to actually executing it. When it works it can take a matter of seconds and is tremendously satisfying. This sort of closely cooperative unit of action is a large part of SWAT 4 games and what makes them brilliant. And when it goes wrong – which it will do sometimes- it can be great fun too. A hostile you were not ready for, slip-ups on the part of the SWAT officers, a hostage running into the line of fire. Improvising through these is great fun too – the other end of the spectrum to the calm and calculating decisionmaking that happens out of action. Either way, when you’ve done the room entry, you all assess the situation: Any doors to cover? anyone hurt? Anything to report? Anyone to detain? What now?

Case Study: Lara Croft and the Guardian of Light

In this game, one player takes control of Lara, and the other Totec – they spend their time running through tombs, killing enemies and solving puzzles to save the world. They work together to get through the puzzles, challenges, and enemies that the game throws at them. One half of this game exhibits very close co-op gameplay whereas, the other half is more traditional loose combat fare. Comparing their abilities – and how they function relative to the sort of tasks you have to deal with – in both halves of the game will highlight the differences between the close and loose co-op side of Lara Croft and the Guardian of Light.

Guardian of Light

I will start with the combat side of things first, where the gameplay is very loosely cooperative. The characters have the same health and ammo bars, pick up the same health and ammo, take the same damage, deal the same damage,use (almost entirely) the same weapons, both drop bombs, both have a primary weapon with infinite ammo and both can roll around insanely quickly to avoid most attacks. Lara moves a tiny bit faster than Totec, and Totec has a shield that can deflect certain kinds of attack, but these differences are negligible. The combat in this game can be dealt with without needing to communicate or plan – if you both just get on with running around and fighting as best you can, you will be fine. The closest co-op elements in this mode are the following: dealing with the occasional enemies who carry a big shield that protects them from ranged damage – this either needs to be blown up, or they need to be attacked from behind. Some enemies will resurrect if their corpses are not blown up. Both of these can be dealt with individually. Of course, there is the quick checking of your partner’s health and ammo situation to make sure that you do not steal some health or ammo pickup they could make better use of. Overall, is its clear that the combat side of this game is very loosely cooperative.

The environmental/puzzle/exploration/challenge side of the game is the complete opposite, requiring a high degree of co-op play. Most of the puzzles require the action of both players to see how the various parts work, and then to solve them. One person runs over here and try out that lever, one person goes to see what that pressure plate does. If you hold that open, then can I use that platform to get over there and put an explosive -etc. The same goes for collecting the powerups: want to get that ammo upgrade that is over the edge there? One player will have to lower the other down on a rope. Want to get that special Red Skull artefact? You’ll have to work out amongst yourselves how you’d even get onto that ledge and past that trap. Etc.

Lara and Totec have distinct abilities that force them to work together in this half of the game. Lara has a grappling hook/rope device which can be used to lower either herself or Totec off the edge of ledges, grab onto certain hook-y objects, create a tightrope for Totec to run along, and she can even perform wallruns herself by swinging with it. Totec has a shield that he can put over his head for Lara to jump on top of and he can also throw spears into the wall for Lara to jump on. Every puzzle, obstacle and bit of exploration will require a combination of these different abilities.

Challenges and bonus objectives push this further. One of the challenges is to complete a level in under a given time. This requires joint planning of an optimal route through a level (in terms of speed) and working out how best to divide up the tasks between the two of you so that you can do things quickly. Usually this requires a few runthroughs to work it all out and practice it. Actually carrying out the speedrun is fantastic fun; it is tremendously satisfying to see it all come together – or to make it work even when it doesn’t quite. Some of the bonus challenges task you with doing things like rolling a massive boulder through a bit of a level really quickly, whilst enemies are trying to kill you, and traps are being triggered all over the place. Again, this requires both thinking and acting together.

Some hopes for the future

Portal 2 does not look like it will be very focused on combat and it seems that the puzzles will require the two robots to use their different portals (and their actual body mass) together to progress. It should be clear by now what I mean when I say that I think Portal 2 could well be the first ‘close co-op’ game yet, and why I think it will be so special in light of this.


Monaco looks like it might be a great game to play multiplayer co-op. Whilst it seems to scale from singleplayer to multiplayer, if it manages to allow genuinely distinct class abilities to be used in combination in multiplayer, and keeps the emphasis off functionally homogeneous tasks like straight-up combat, this could be a really exciting co-op prospect. Either way, this is one to watch.

I may never end up trying Artemis, but it looks like a very closely co-operative game and I would love to see how it actually plays in real life. It is described on its website as follows: “Artemis simulates a spaceship bridge by networking several computers together. One computer runs the simulation and the “main screen”, while the others serve as workstations for the normal jobs a bridge officer might do, like Helm, Communication, Engineering, and Weapon Control.” It divides up the tasks on the Ship’s Bridge among the players; they each have to work closely together and each have something different to contribute. I would love to see this sort of template used in radically different ways.

My other main hope for the future is Shogun 2. I’ll be starting off a co-op campaign of this with a friend soon, and – worries about how long it will take notwithstanding – am intrigued to see what possibilities it presents. I would love to hear of anything that I’ve overlooked.

Co-op beyond co-op games and co-op beyond the limited conception used so far

By now it should be clear how this way of understanding cooperative gameplay rightly highlights it happening in all sorts of places that are not usually referred to as co-op games – team play in FPS, MMORPG and RTS games, for example. But the sort of cooperative play I have been looking at so far is rather limited – I have been focusing on fairly mechanical examples of puzzles and combat. There is so much rich and rewarding territory beyond this that I haven’t touched on – and most games have neglected too. Collaboration and shared creativity are two types of cooperative play that are massively under-explored at present. Consider the following examples:

Collaborative storytelling and acting in a roleplaying game. Imagine two actual players having to take up the bluffing in this sort of scenario: [Dragon Age Spoilers]

Whilst I’m familiar with musical collaboration outside of computer games, I’ve never seen it explored that thoroughly in any gaming setting. I’m not expecting it to translate across perfectly, but there have to be unique and novel possibilities for collaboration here that have not been explored yet.

Playing with ideas in a group can be great fun: this is how in-jokes develop and live, how some of the richest humour is created, and it is also how some of the best ideas and arguments get explored. I have seen some of this in many games (without the game system directly fostering it): the in-jokes we developed whilst playing Trine, the nicknames we gave the enemies in Lord of the Rings: Return of the King, the way we mocked and satirised Resident Evil 5’s infantile plot as it unfolded infront of us. What excites me is the idea that a game could place more of a central emphasis on this sort of play.

Improvising music together or a piece of drama, working on a painting or sculpture together – the gaming analogues of these have not been fully explored yet. I will openly admit that the more creative and improvis-y side of co-op gaming is something I am not very well acquainted with at the moment, but I’d appreciate some thoughts from anyone with more familiarity with these areas. I’m thinking – off the top of my head – about things like DnD, roleplaying, any kind of artistic/musical collaboration, MUDS, Little Big Planet, Second Life, Minecraft Multiplayer, Gmod, Sleep is Death.

Gaming has so far only realised a small amount of the possibilities offered by cooperative play, and I hope that in the future we can see continued innovation and experimentation along these lines.

Conclusion and recap

‘Co-op games’ vary massively in terms of the way in which they are cooperative; because these differences are so important, we need a way of understanding and describing them. Distinguishing between ‘loose’ and ‘close’ cooperative play is a good way to do this. Close co-op games are the best sort of co-op game (for my tastes at least), but I do not know of many games that fit into this category. I would love to hear of anything I may have overlooked.

Beyond looking at the sort of games that are common at the moment, I have highlighted how many more forms of cooperative play remain largely unexplored. I hope that innovation and experimentation continues to drive things forwards, and I would love to hear from people with more familiarity in some of these areas than me.