Over the last couple of months I've been nurturing an idea to hold a sort of casual attempt at a game design round table where I (or another organizer) would present a topic (and some basic information and potential arguments) to a group of independent or professional game designers. Everyone involved would then be let loose to sound off on their opinions on the topic at hand, argue with each other, and so on. The goal was to provide a format and location that would encourage designers to discuss topics in game design intelligently and thoroughly throughout the duration of the seven weeks that the topic is left open. Once the seven days of discussion were finished, the person who initiated the event would both write a piece which compiled original thoughts and combined them with arguments and perspectives of everyone who contributed to the discussion. Originally, I was planning on doing this by setting up a mailing list and slowly getting more and more people involved but then I realized that the Game Design forum at GameDev.net would be a near-perfect locale for this. With this in mind, I started up the first attempt at the Game Design Round Table (it's a metaphorical table) entitled No More Health.

Round Table Topic -- Regenerative Health

Regenerative health systems are actually a pretty simple one that have radically changed the way that first-person shooters and a number of third-person action games are played. The idea behind regenerative health is that players can take a finite amount of damage in a short span of time before they are sent into a "dying state" -- which is typically indicated by a pulsating red screen -- and if they take more damage beyond that then they will die. If a player does not take any more damage when they are in a dying state, though, and instead seek cover and avoid enemy confrontation and fire, they will slowly return back to their normal state. The concept of player health is now entirely dynamic and up to player interpretation via some sort of interface cue, red tinge, or other full-screen indicator. The effect of this mechanic is that it abstracts the older method of requiring players to manage their health and pick-ups, something that most players inherently understand (mortality), into a very streamlined and intuitive experience.

GiantBomb has the first occurrence of regenerative health in Wolverine Adamantium Rage (Genesis, SNES; 1994). The mechanic's first mainstream appearance, though, was in its devolved form in Halo (Xbox; 2001) which had fully rechargeable shields which absorbed most of the player's damage. Once the shields were out of energy, though, Halo still relied on a more traditional health system which included requiring players to find health pick-ups. Halo 2 took this concept a step further with a fully regenerative health system, tasking the player with only managing the ammo and type of his/her weapons. The net result of the widespread adoption (in games like Resistance 2, Killzone 2, Call of Duty 2-4, and so on) of this mechanic into modern action games of all types is that players are no longer thinking of their health some arbitrary number or percentage in the middle of a heated combat encounter.

Does this mechanic simplify action games in a good way? Is the reduction in manageable resources a boon or detriment to players? Are the hit-and-run (to cover) tactics that regenerative health systems not only encourage but often demand beneficial to most of the games that this mechanic is employed in?

No More Health

A health bar is one of the most iconic interface features in the history of video games. Children of the 1980s to the early 1990s have that grown up with a concept of health as being integral to their gaming experiences. For those of us with that gaming history, unless a game designer attempts to trick us, we know what a red bar on user interface means or what a numerical value next to a cross or heart or similar icon represents. That interface feature tells gamers one thing: how safe they are. A low health value or a slim portion of a life bar means that a player is going to play things as carefully as he/she can; no chances taken means that a player can reach his destination, doing something brave or stupid means that a player has little to no chance of surviving a given stretch of gameplay. A full health bar gives a player the sense that they're safe, they can screw around -- maybe get a little explosion happy -- and still have room to breathe.

The concept of the health bar is one that the "core gamers" -- the kinds of video game players that have been around for years -- all understand perfectly, but the attempt to concretely display an abstract concept isn't a particularly elegant solution for conveying player stability and well-being. And along those lines, the necessity to replenish a health bar yields some realism-breaking gameplay conventions: floating health supplements (medicine packs, wall-mounted rejuvenation centers, etc.) that games generally have players walk over to either be added to an inventory or be instantaneously consumed for additional health. Putting aside for a moment the lack of realism inherent to this gameplay practice, such a system also tasks a player with an additional layer of resource management. This is typically an additional resource to whatever weapons, ammunition, and other items that a player has in action games, though as Aaron Miller points out this is not necessarily a bad thing:

The search for health can be a very interesting diversion in an FPS. It can flavor encounters, making some situations more desperate or requiring stealth or diversionary tactics. It can also, as it was in Doom, be a source of meta gameplay in and of itself, requiring players to risk environmental hazards or rewarding them for quirky exploration.

And as Jason Adams adds this can create for a sort of exigent mid-combat tactic on its own:

[...] a system with health pickups can lead to interesting situations where a player makes a mad, basically suicidal run into a group of enemies with the goal of reaching a health pickup just before death, or where a weakened player can be tempted into navigating difficult environmental obstacles.

While both of these points are absolutely valid reasons to adhere to the more time-tested approach of a concrete health-based method of conveying player well-being and sustainability, are any of the aforementioned points actually necessary? While the presence of in-world health gives players a reason to move about the battlefield when they are near a death state -- which is an inherently tense gameplay scenario -- the same behavior can be demanded of a player through weapon and ammunition scarcity and management. More than just the management of health, though, a concrete health system enforces a certain type of play style, one of which is that of a strategy of long-term survival as gxaxhx points out:

However, in a game were resource management is part of the experience, I believe regenerating health would be a detriment. Imagine if Left 4 Dead used the health generating mechanic. Most likely the zombies and special infected would have to hit much harder to be any threat to the players. So, rather than experiencing maps where the players literally limp across the finish line, pleased with the way they managed to pull through as a group, most maps would be either the players getting shredded by the infected or flying through the map with no problem.

While specifically referring to the style of gameplay that a non-regenerative health mechanic promotes or enforces, gxaxhx's argument actually goes back to the treatment of health like an in-game resource. So long as health is a player-managed resource, the player's actions and carefulness, based on current game events, will be directly related to the scarcity of health in the game world.

If a concrete, or non-regenerative, health system is some form of resource management, then a regenerative health mechanic is a means of enforcing certain player behaviors and game flow. As Josh Petrie says it:

[Regenerative health systems] almost necessarily cause gameplay to be sliced up into reasonably short, controlled encounters of at least some minimum intensity. A regenerative health system basically gives the player a damage threshold: get through an encounter under this threshold, you live and can move on to the next encounter. Otherwise you die. Killing the player is accomplished by overloading this threshold in a short period of time, which means you can't build suspense in the player via prolonged needling little encounters that keep them at low health [...].

KnTenshi argues the drawbacks of such a system:

You're in the middle of a skirmish with a group of enemies. You are blasting away at them as they quickly bring your down to the 'dying' stage. You dive for cover and a few moments are back to full health. You now return to the blasting away part and them subsequently returning you to near death. This cycle continues until either all of the enemy are dead or a lucky shot from the enemy drops you to 0. There isn't much suspense unless you willingly run into the midst of the enemy. [...] For me, if a game designer wants to implement a regenerating health bar, it should a a [sic] large chunk of time to fully recharge. Using a slow health regen, enemies are more likely to start hunting you down rather than just using suppression fire, which would increase the pressure for the player to do more than just pop in and out of a hiding place.

And Luke Parkes-Haskell writes a superb post that begins with praise about regenerative health systems while bringing up what he feels are complications when the system is carried over into the competitive arena:

Personally, I feel that the system has it's merits. In a game that warrants quick encounters with equally matched opponents (e.g Call of Duty 4 multiplayer), wherein the player has the potential to be eliminated reasonably quickly, giving the player the ability to regain their health prevents them from suffering immediate game-impacting permanent health loss if they are glanced early after spawning - and thus they are also not immediately forced to predictably move to locate health pickups. It serves in some respect to assist in leveling the playing area in an often sprawling, maze like arena, and prevents lucky or random shots from killing outright, but with a lower player health, a misjudgment or tactical error can easily lead to death.

[...] However, such a system certainly has no place in some competitive arenas. Consider Unreal Tournament 2004, and the common one-on-one death match. Ultimately, a regenerative health system could only detract from the nature of the game. Experienced and veteran players are very much familiar with the strategic and tactical nature of determining one's path through the chosen environment, denoting that damage and health critical pickups are only available at certain intervals. Failure to adopt an appropriate means to control these power-up points and deny them to the opponent is a very prominent emphasis in what would otherwise be a much more bland and imbalanced game; since otherwise the player could not be encouraged to move; since they can find a suitably strong position to sit in, and remain within it, given that they are able to always regenerate their health. Without the ability, they can occupy this position, but will be forced to abandon it should they take damage - and in the case of many a good level design, health pickups are in the more vulnerable and frequently passed through areas of the map.

My favorite contribution to the thread comes from Drew Marlowe (a designer on Full Spectrum Warrior: Ten Hammers, Mercenaries 2, among others):

I think that regenerative health systems are definitely a step in the right direction when it comes to simulating firefights in action games. This is because in many circumstances those systems are fantastic at simulating the real life feeling of being fired AT but not being hit which makes up the majority of 20th and 21st century combat. [...] In a real firefight (disclaimer - I've never been in a real firefight) many more bullets are fired into walls, the ground, the air than actually hit a target. The oft quoted statistic is that in the current US wars the US is firing a quarter of a million bullets for every insurgent they kill. All these bullets are getting fired in order to scare the shit out of the targets, and get them to not move or fire back while the US figures out how to safely kill them. It usually works because getting shot at is FUCKING SCARY. They also fire so many bullets because it is really REALLY hard to shoot someone. If you've ever shot a gun, you know that it's a little bit harder in real life than it is in games to get an accurate head shot at 50 yards.

However, getting shot at (not hit) in a game is a non-issue. It doesn't really phase most gamers because it's just a game. With their music up, they may not even know that a bullet just zipped over their head. It doesn't feel scary because you're not punished for almost getting hit. So how can a game properly simulate the feeling of getting shot at, of needing to be behind cover, that makes first and third person shooters look and feel realistic? You can do it by make it a lot easier for the bad guys to hit the player, but allowing the player to respond to getting shot without a long term punishment. Against 5 or 6 enemies the player will stick his head out of cover but quickly be overwhelmed at the volume of fire and duck back down - not because he was afraid of the amount of bullets being shot at him, but because his health was getting low due to being hit.

The concept of a player experiencing and understanding the distinction between the danger of shot at compared to the consequences of being shot is fascinating. The influence that a game's health system can have on such an experience is arguable, but the benefits of allowing players to experience a sort of "warning shot" that does enough damage to cause some visual/interface/post-processor effect but doesn't incur any long-term damage (that would be almost inherent to a more concrete health system) seem worth looking into (in theory, if not practice). Drew goes on to illuminate one of the most notable difficulties of simulating such a experience by saying:

Additionally, and perhaps more importantly, it is extremely difficult to communicate to the player exactly what circumstances will cause him to be shot. If I stick my head up will I have 3 seconds to fire, or will I be shot immediately? If I am shot immediately, does that mean I am never allowed to poke my head up at that spot, or is it just the randomness of the AI's fire that caused the shot to land so quickly? The player has no idea, and because he needs to find a health pack every time he makes a mistake he is discouraged from experimenting in order to find out more about the combat systems.

Stroppy Katamari went back to the days of Action Quake to illustrate the benefits of a sort of hybrid health system that merges some aspects of a concrete health system with those of a regenerative one:

Action Quake, an older NRH game which everyone interested in FPS game design should check out, has a bleeding/crippling/bandage system which could easily be adapted to a RH game. When you are shot (anywhere but on armor), even a very minor hit like 5hp from grenade shrapnel, you start continuously bleeding health - slow or fast depending on how serious the initial wound was. To stop the bleed, you must bandage yourself. This takes several seconds, cannot be cancelled and leaves you defenseless. So there is an awesome trade-off and mindgame that follows from one player being wounded; they have to bandage, eventually, and if you catch them at that moment you win. But they know this, so they can wait (and bleed) long enough to ambush the other player following their blood spatter tracks, and then bandage. There is also a leg damage mechanic which cuts the runspeed to half or so when you recieve a hit to the legs, also curable by bandaging. This makes for viable tactics like taking one fast shot at the enemy at long range, inflicting the bleed and leg damage, and then stalk the slow, bleeding enemy while staying out of sight. You can even use a shotgun for this as the initial shot only needs to hit the legs, not do real damage. [...] You could do regenerating health the same way, requiring the player to go defenseless for a while, etc.

As a number of the people who contributed to this first attempt at this kind of project and discussion noted, the usefulness and enjoyment that can be derived from a given system are entirely dependent on the specific game which employs it. This is, of course, an incredibly valid point, but one I'd like to encourage further round tables to worry less about. The goal of these discussions is to attempt to get developers and designers to communicate with each other about certain issues in game design; this requires a knowledge of gaming trends, for one, but more importantly it requires designers to make intelligent arguments with one another. Making an argument isn't as simple as posting a quote from a famous designer or citing an excellent game which had a certain design but, rather, making an argument and supporting it with a combinational of practical examples/experiences and general theory. A discussion about games is generally impossible to boil down to pure empirical data and concrete facts, so I would encourage contributions to think a bit about their own arguments and, if necessary, feel free to generalize into theory rather than relying on an obvious fact like "this all depends on implementation" or "this only works in this specific game." There are lessons that can be extracted from specific examples and that's, ideally, what I'd like to have people take from their discussions with one another in the future.

There is no correct way of handling the concept of a player's life and longevity in a video game; there is no genre in which this is more true than in the fast-paced action of a typical first-person shooter (or a similarly-designed third-person action game like Gears of War). The varying means by which shooters achieve intensity and encourage players to adopt certain play styles is, in a lot of ways, entirely dependent on the way it handles its health system. As a number of the discussions and points made throughout the round table discussion illustrate, there are benefits to both an abstract health system where a player's life is largely up to his interpretation of interface/screen cues (regenerative health) and that of a more traditional, concrete health system that relies on life bars, numerical values, and in-world health replenishment. There were some absolutely great contributions to this sort of test run of the Game Design Round Table -- a name which I admit sounds incredibly pompous and pretentious but I think reflects the goals of the discussion pretty well. Before I wrap up, though, I want to point to a post made by Nick Halme which I couldn't figure out a way to integrate into this article but, also, made a number of very poignant design points through Halme's own past design project.

The next Game Design Round Table will be posted to the GameDev.net Round Table, with some revised guidelines, on Tuesday, April 21st. If anyone has comments regarding my first attempt at handling this whole thing that they want to direct personally to me, feel free to e-mail me trentp@gmail.com. I plan on doing this every two weeks (this may turn to three, as preparing this took a decent chunk of time) with the first week being a discussion of the posted topic and the second week being the eventual posting and discussion of the aggregate/argumentative article written by whoever organized the current topic. I'll probably keep doing this for a few more times and refine the guidelines and eventual article format as I go along, but if anyone would be potentially interested in spearheading a later iteration of this feel free to e-mail me.

Thanks to everyone who contributed to this, the discussion was great, and I'm sorry if I couldn't work every post into this piece.