About Monsters: Resurrection and Summoning

Almost every shooter has featured monsters with some specific gimmicks, ever since the genre was created. Most of the time, these gimmicks are designed to help the monster itself fit its particular role (or the gimmick is the sole reason the monster fits in the role). Some tanks might have some shield ability or armor; some swarmers might have some leap, dodge or wall-cling so they can swarm better; supports might be able to disable the player to benefit other monsters. There are two gimmicks in particular that I see often, usually to great effect: resurrecting and summoning monsters.

Resurrecting a monster is pretty obvious; there’s a dead monster on the ground, something happens to it, then it’s alive and able to attack the player again. Summoning (or spawning) a monster is where you create a monster that wasn’t there to begin with. The Archvile from Doom 2 is the most famous example of a resurrecting monster. While running around, if it comes across a dead monster, it could revive it with full health again, ready to attack you. Hilariously enough, the Archvile from Doom 3 is a good example of a summoning monster; as an alternate attack, it goes through a short animation where it spawns another monster, again ready to attack.

Resurrection is a classic, first featured in Doom, later used in Quake 2 (its purest form) and forms of it have appeared in Quake 1, Dead Space and others. When a resurrecting monster is sent along with the rest of the encounter, it’s basically a defensive support. Because of what it does along with other monsters, it demands your attention, otherwise you could end up fighting the same heavy-role, for example, twice in a row. You have to kill the resurrecting monster before the others or risk wasting resources. The Archvile does enough on its own without reviving other monsters to warrant singling out, but the Medic from Quake 2 or the Infestor from Dead Space have nothing going for them when it comes to combat ability. Yet in Quake 2 and Dead Space, when you see or hear one of them, you know you’ll target them first. A more major effect it has on the game, however, is that it lets the mapper refresh threats that the player has just defeated (with the added effect that you now have this other monster to deal with). Many times in custom Doom maps, I see situations where the player clears a difficult fight in order to access a switch or a key. After that, an Archvile spawns in and starts reviving things the player’s just killed. It forces him to fight through those same obstacles again, unless he’s fast enough to stop it before it gets too far (which he should know to do, if he knows what the Archvile does already). Of course, there’s also a random element as well, because there’s no guarantee that the Archvile will resurrect particular monsters, and it’s very unlikely it will resurrect all of them.

Summoning has a different, but similar effect. The main difference is that the player doesn’t necessarily have to have been through the area to get the same effect; he knows (potentially) that there is at least one monster there that is spawning other monsters. Like resurrecting monsters, it presents the player with a growing situation that he can sabotage if he moves quickly and understands the threat. There’s also potentially an even more random element as well, because the player can’t really predict how many it’ll have time to spawn or what monsters will be spawned, but that depends on how it’s done. Doom’s Pain Elemental creates only Lost Souls, and only does so when it can see the player; the Teleport Dropper creates a spread of random monsters, but again, only when the player is around; the Doom 3 Archvile can spawn them regardless, but it seems like it’s scripted on each Archvile which monsters it spawns. Because of this variation on how it can work, they might not be able to accomplish the same effects. A potential downside to summoning over resurrecting is that the player probably won’t be familiar with the situation he’s going into, and that may or may not be desired.

These gimmicks tend to have a decent amount of sub-gimmicks, as well. The definitions of these two systems are surprisingly broad, and there is a lot that can modify it. They could also be combined to a degree; Dead Space has a lot of corpses lying around, just as decorations to help the game’s atmosphere. After you’re introduced to the Infector, however, they become a bit more threatening. The Infector reanimates those corpses as new monsters, ready to attack. In very basic terms, it’s resurrection because of what it does, but it’s also summoning because it’s making something that isn’t a monster into a monster. Another point was brought up in Wolfenstein 2009; resurrecting a dead monster as something else. The Elite Guard enemy could revive SS troops as Despoiled; a similar, but faster, stronger and heartier enemy that can’t be similarly resurrected, itself. This has the same benefits of standard resurrection, only the player has a potentially new challenge to work with.

Another idea introduced in Quake is how to prevent these resurrection or summoning mechanics. The most basic form is preventing a monster from being revived if it’s gibbed (blown into pieces) by doing much more damage than it can take. The Zombie from Quake 1 didn’t have a lot of health, but it could resurrect itself after a few seconds if it wasn’t gibbed. This was later expanded in Quake 2, by allowing the player to gib corpses after they were dead (you had to do enough damage with the last shot, in Quake 1) so Medics couldn’t revive them. This gave you the choice of using ammo on a threat you’ve already put down to prevent it from getting back up. In my opinion, it didn’t work as well as it could have, because Medics were almost just as easy to kill as it is to gib a Gladiator, for example, that’s already dead. However, even more effective, to deal with the Infector, if the player is paranoid enough, he could chop limbs off of the corpses he comes across to handicap the monsters that will be created from them, or even prevent them from being resurrected. Summoning could be sabotaged as well. The Teleport Dropper from Quake 4 dropped these beacons that monsters could teleport to (it just spawned them, really), but those beacons could be destroyed. With an explosive weapon shooting the Teleport Dropper, you could easily kill it quickly and prevent it from summoning anything. This is a double-edged sword, as I’ll talk about later.

There is an aesthetic difference, also. When you summon a monster, you’re bringing in something that wasn’t there. When resurrecting, even with the gimmicks that I’ll discuss in a moment, you’re bringing in something that already exists into an environment and turning it into a threat. Take the Infector for example: it takes set piece decorations (that’s effectively what they are) scattered around the area and revives them as monsters. Story-wise, it’s just turning corpses into Necromorphs. Resurrection abilities give full context to what is going on, and in the case of the Infector and monsters like that, it animates non-threatening decorations as monsters. There are times when pulling a monster out of thin air is what you want to do, and also can have proper context, however. For example, the Pain Elemental in Doom 2 is very good at what it does and its summoning is given context because it’s very clear that the Pain Elemental is creating the Lost Souls it’s spitting at you.

As gimmicks, both of these are pretty tricky to balance. Not only do you have to consider the monster itself, but you have to consider the monsters they resurrect or summon. The Archvile in Doom was already a dangerous monster, but then it can revive Revenants, Barons of Hell and other dangerous monsters. However, monsters like the Cyberdemon and Spider Mastermind (bosses) couldn’t be revived at all. If they could have, that is a massive drain on the player’s ammo, at least, even if he’s never damaged by them. For summoning, Teleport Droppers could only bring in certain kinds of weaker monsters, the most dangerous of which was the Berserker (which was pretty dangerous, but rare, and you have some strong weapons by that point). The balancing factor for these gimmicks themselves is the monsters they effect; if too-strong monsters can be revived, the reviving monster is too strong. If too-strong monsters can be summoned, the summoning monster is too strong. This can be especially difficult to balance correctly for less-experienced players. People who know how to deal with reviving and summoning gimmicks know to go right for the monster with that gimmick, but other players might not and will attack the others first.

Of course, if the reviving/summoning monster itself is too weak, there’s the very real risk that the monster will be killed before it can do much of anything. This was the problem Teleport Droppers had. They were very easily killed before they can do anything, and because their summoning could be sabotaged they would almost never summon anything unless the player has no idea how to deal with them. Considering the build-up they Teleport Dropper is always given, it’s very underwhelming and ruins the effect for the player. These was a similar effect for Quake 2’s Medic, where killing the Medic itself was potentially much easier than shooting the corpses already lying around. Thus, you have to consider how quickly the monster can be brought down when you design it and whether or not the player would actually see its effect. I’ve also spoken a little on the glass cannon style of monsters before, and I may have mentioned that I’m not a fan of how those work, exactly. There’s a risk of doing that here, as well. If the Teleport Dropper, as easily-killed as it is, could have summoned monsters like the Light Tank (a very dangerous, hearty and hard-hitting monster), for example, that would make it a glass cannon. This is even worse for reviving and summoning monsters, as well, because the net effect is that either it’s pretty tedious for the player to keep killing these much-more-dangerous monsters, or the encounter you’ve planned out never actually happens.

Both of these concepts are classics, and have appeared in many different shooters over the decades. If the monsters are balanced correctly, they can be the most interesting, iconic or even feared monster in the game, and encounters with those monsters can be some of the most memorable in the game. However, to get to that point, you have to consider balance, how these are used, what these do, what you can use to modify them, and how they can go wrong. When in doubt, look to the examples we can find in these various shooters over the decades.

Roles of Monsters: Hard vs Soft

In the Roles of Monsters series of articles, I’ve often talked about how this role works with that role, or examples of monsters that fit into multiple categories. As you develop your lineup of monsters, you want to consider how these monsters fit into these, because that affects how you design your map. The issue here is that a monster might not fit one role as well as another. So for the purposes of putting together a lineup of monsters, I think of a sliding scale; if a monster fits well into a roll, it’s a hard example of that role. If it fits to a degree, but isn’t that great at it then it’s a soft example. I like to use these terms, hard and soft, to describe how well a monster fits into these roles. And of course, if a monster doesn’t work in a particular role at all, then it just doesn’t fit and you can’t use either of those terms. You can use this system to plan out your lineup of monsters to see what you can use and where you can use them.

Now for some examples of this application. Take the Vore; it’s a really good turret, that much is evident in how its attack works and how it moves. That’s what it’s designed for, and it excels in that role. So, it’s a hard turret. Another thing is that because it’s good at getting your attention because of that attack and because it eats more damage than many other monsters in the game, it could conceivably be used as a tank. However, it’s not as good at being a tank, because it doesn’t have any way of mitigating the damage it takes and because it’s a sitting duck, so it’s a soft tank. The Ogre, as well, is a bit of a soft turret and a soft tank, because its bouncing grenades take your attention and are hard to avoid if they’re placed above you (which they often are in Quake). You can analyze any monster like this (though some, like Hexen’s Stalker, are really tricky to classify).

The one term I don’t really like hearing is the soft shocker. What is a soft shocker? If we look at the terms as we’ve laid them out, it’s a monster that can work to surprise the player back into higher-pace gameplay, but not amazingly well. By that definition, any monster you don’t like to fight is a soft shocker. The Mancubus from Doom is a soft shocker, the Gorilla from Hard Reset is a soft shocker, and any heavy would be a soft shocker. It’s really general, and it’s the most subjective thing I can think of, so I prefer to throw that right out the window.

The term kamikaze is more of a modifier for an existing monster than a role in and of itself. A soft kamikaze can’t exist because the monster is either a kamikaze or it isn’t. You could argue that Red Faction: Armageddon’s Berserker is a soft kamikaze because it doesn’t charge at you and explode immediately, but it still is considered a kamikaze because it has that ability. For the purposes of creating a lineup of monsters, soft kamikaze doesn’t help you one bit; if you want a kamikaze in a specific place, you want one that explodes immediately. If you place a “soft” kamikaze, you’re placing it because of the other talents it brings to the table.

Another major point is fodder. Most of the time, fodder is just fodder. Almost any monster can be made into fodder by placing it in a situation the monster doesn’t quite work in. For example, the Teleport Dropper in Quake 4 was basically fodder because it was placed so poorly, in situations where the player can kill it quickly before it can do anything. Swarmers especially, by just placing individual swarmers and not allowing them to gang up on the player, you make them fodder by taking away what makes them threatening. Because of this, you can’t consider hard or soft fodder. The term mostly indicates that the monster lacks any ability as a tank, support, swarmer, shocker, etc.. However, for reasons I outlined in the fodder article, they’re still very important.

Of course, most of this is without saying. It’s mostly a set-up for another article, where it would be helpful for me to have defined these terms in advance. However, it should still be helpful to those putting together lineups of monsters for the first time, as they’re now given a logical process to follow. A decent portion of the time in the Doom modding community, when new modders add custom monsters to a project, they just take something that looks cool and add it with little regard for how it really in relation to the others in the lineup. Hopefully this and the aforementioned other article will help out with this.

Player Upgrades: Other Considerations

Now that we’ve covered the two forms of progression, there are other elements that affect both in different ways. Additionally, the two aren’t mutually exclusive, and can interact in different ways. In this article, we explore these topics.

As evidenced by games like Doom and Singularity, both kinds of upgrade progression can benefit from hiding certain upgrades, money or tokens in secrets hidden around the map. Doom often hid weapons in secrets (chaingun hidden on the second map, rocket launcher on the third, etc.), along with other very helpful items. Singularity was one of those games that featured both a money and token system, and often hid BOTH across maps. I like this because it can draw the player further into the game, making him explore around and appreciate some of the graphical details he might not notice if he was just trying to get to the end. I think Dead Space could’ve benefited from this to draw the player further into the game (something important for survival-horror shooters like that), but all of the credits and power nodes were somewhat obvious.

Unfortunately, hiding upgrades or some form of upgrade currency in secrets can be hard to pull off correctly. Secret weapons can potentially wreck the progression curve in your game, to the point where there really doesn’t feel like it’s easing you into the progression at all. In Duke Nukem 3D, you can literally get every weapon available in the first episode in the first map. It’s balanced by the fact that ammo for some of them is limited so you can’t use the RPG or pipe bombs as often as you might want, but there’s still not much actual progression there. Quake 2, over the course of the much-longer game did introduce weapons more slowly, but still did have more advanced weapons hidden in secrets. The super shotgun was first available in a secret map in the first unit (essentially halfway through the unit); the next earliest is a secret at the very end of unit 1; the first time it’s obviously available is at the start of unit 2. The first rocket launcher is in a secret at the start of unit 4. The first time you can find the BFG is a secret in unit 6 (out of 8). It gets around that problem of ruining progression by spacing them out quite a bit more, so it still feels like a slow progression.

A potential problem with hiding currencies to upgrades in secrets is that it’s tricky to strike a balance with how much to hide, and how much to keep open. In several games that hide currency in money systems, especially (Singularity, Red Faction: Armageddon, etc.), very sharp players end up with a massive surplus of them, even if they’re used for other things, like Deus Ex: Human Revolution. One game that did this better was Hard Reset, where a certain amount of points gets you an upgrade. However, not many of those upgrades made you more powerful (beyond the obvious two or three that you’ll likely get first), so it didn’t mess with the game’s balance much. Tokens end up working better in this regard because they can be balanced more easily than amounts of credits. When you hide money, you’re hiding some abstract amount; it could be big part of a small upgrade, a small part of a big upgrade, etc.. When you hide tokens, that’s a whole upgrade (or half of a better upgrade, or whatever)! You can more easily plan for the player having this amount of upgrades (what’s in plain sight) while accounting for the player having all of the possible upgrades. This is one of the main strengths of the token system.

Another consideration is how open-world the game is. By open-world, I mean the game has defined objectives, but doesn’t care how they’re accomplished or whether you do them quickly, and even has optional side-missions. Crysis (the first half) and Red Faction: Guerrilla, for example, had gameplay like this. Because the game is open-ended by definition, a linear progression is difficult to pull off correctly. How can you know a player will be at this location to pick up this new weapon? Crysis (and Crysis: Warhead) occasionally had caches of weapons designated as secondary objectives, usually with a powerful new weapon as an introduction. Red Faction: Guerrilla had a combination between player-driven and linear progression; after completing certain missions, new weapons become available to unlock with salvage (money), and occasionally new enemy soldiers show up with new weapons (which you can pick up). Because you can’t introduce weapons in Quake 2’s style of linear progression, you have to tie weapons to particular events, or risk the player just missing the progression.

Player-driven upgrade systems can work better in open-world games, though, because the player might be able to go back and pick up upgrades that might be needed for some part of the game. For example, in missions in Red Faction: Guerrilla where you have to capture a vehicle but you don’t have an Arc Welder (makes vehicle capturing much easier because it kills the driver without damaging the vehicle), you could go looking for salvage to unlock it beforehand. It’s easier to balance around, since the developer can assume the player has some kind of upgrade, because the player could go somewhere else, get that upgrade, then return. This is one instance where I think allowing the player to grind for infinite upgrade currency can work, but I have never liked it for more linear games because it encourages the player to sit in one particular area, often in a style of game that encourages the player to keep moving forward.

One other factor with player-driven upgrades is to prevent the player from getting particular upgrades until he’s progressed to a certain point of the game. In Wolfenstein 2009, you can upgrade weapons and Veil powers with gold hidden in the game. Many more powerful upgrades aren’t more expensive than previous ones, but they require certain levels be completed before you can purchase them. Similarly, Red Faction: Armageddon has certain tiers of upgrades. The first tier is available from the start, but the higher tiers with more powerful upgrades are locked until you reach certain milestones in the single-player game. Another take on this is the schematic system in Dead Space and Singularity; You don’t even know about these upgrades until you find schematics hidden in the levels (in Dead Space, these take up inventory space, so you actually have to decide if you want it). In Dead Space, they were fairly obvious, but many of them were hidden in Singularity. This system adds a tone of linear progression to a player-driven progression system; preventing the player from getting too powerful, too early in the game, and slowly introducing him to new powers.

Something that might surprise you is how upgrade progression and story go together. One way in which Red Faction: Armageddon feels horribly unfinished is that some of its upgrade system isn’t explained in story at all. In the backstory for the game, it says the main character has been upgrading the Nanoforge bit by bit, so this is simply a continuation of that. What isn’t explained, however, is why certain tiers of upgrades are locked, why an upgrade to the Nanoforge lets his guns store more ammo, or things like that. While it doesn’t affect the gameplay, things like this can hurt the player’s immersion if there are widely unexplained and implausible elements like this, which can lessen enjoyment (if the player cares about that). It might not be very serious, but again, worth consideration.

As you can see, you don’t have to commit to either a player-driven or linear form of progression. In fact, I think the best method is a combination of both; mostly linear with some minor player-driven upgrade system. Red Faction: Armageddon, though I think it was rushed, did have both used quite well. As you progressed through the game, you found new, more powerful (or more useful) weapons. Additionally, you found salvage (currency) for purchasing upgrades for your Nanoforge (granting new usable abilities and upgrades for your character). You could customize your weapon loadout to fit your play style, and upgrade how you wanted to fit that; you could take melee weapons and explosives and upgrade your explosive and melee damage, for example. Or pick a longer-range loadout, increase your damage with headshots and the clip size of your weapons. It locks some more powerful upgrades until you’ve progressed a certain way into the game, as well, so you have to progress in power as the game steps up its difficulty, and you are prevented from beelining to the powerful upgrades you want. Additionally, the game also got harder because they actually assumed the player did get some upgrades (and it’s still possible to beat the game if you don’t even get any, though it is much harder), and indeed the game actuallygives the player several upgrades in the form of weapons, which the player uses more often than anything else.

This also isn’t limited to shooters; RPGs, strategy games and some other genres all feature progression. Think of Starcraft 2, where each new mission in the campaign introduced you to a new unit you can use (linear), and can upgrade certain things outside of missions (player-driven). Starcraft 2 also did a very good job working it into the story, as well, with explanations for how each upgrade works in the game’s lore, and the reasons why you’re getting these new units in the campaign. Level-based RPGs also blend the two very nicely, with Baldur’s Gate and Torchlight being perfect examples. Progressing through the game gets you experience and loot, experience gets you levels, which increase your power. Just about everything I’ve talked about about in this article works with these types of games as well.

Upgrading the player is almost essential to these genres of games. Shooters can get dull if the player never increases in capability in some way, and RPGs are entirely built around it. In order to implement a proper system for it, you have to understand all the different ways a player can acquire those upgrades. I don’t believe for a second that this is everything that affects how he gets upgraded (it was all I could think of), but hopefully these articles are a good place to start in how you design your system.

Roles of Monsters: The Heavy

The monsters we all remember the most clearly are those huge, beefy monsters that can both take and deal a lot of damage. Monsters that the encounter usually center around, and are usually a pretty difficult fight even on their own. These just outright try to kill the player (usually through stats), and take center-stage every time they appear. This category is the heavy; of course, named after the Team Fortress 2 class that wraps it up in a competitive multiplayer sense very nicely.


Shambler, from Quake


The first, most archetypal heavy is the Shambler from the first Quake. The first time the player runs into it, he can assume how dangerous it is because of its size, sound set and its overall design. It has the highest health in the game, and can do the most damage. As its ranged attack, it projects a stream of lightning for a few seconds that deals steady damage for as long as it’s connecting (you have to break line of sight to avoid it); dealing heavy damage if you’re caught in the full blast. Unless you have armor, its melee attack can potentially kill you in a single shot, as well. In addition to its high health, it’s resistant to explosive attacks, making it even tankier.

Heavies are kind of simple in how they work. Like the Shambler implies, heavies have to be able to both deal and take a lot of damage. There’s not much beyond that, in how their role works. It also helps if the heavy is competent in both ranged and melee combat; however, I’ve seen heavies that specialize in melee (Quake 4’s Light Tank) and heavies that specialize in ranged (Quake 2’s Tank, hilariously enough). Heavies are supposed to be intimidating, and having a heavy that’s useless at either ranged or melee combat lessens their impact unless they’re placed so that players are unable to take advantage of that (trickier than it sounds. It’s a good idea to grant your player enough respect to assume he’ll pull stuff like this).

Heavies by definition are a little bit like shockers. The player usually hates fighting them because they’re so dangerous and take so much punishment. However, there’s a major difference; heavies only shock by virtue of raw stats, most of the time. Shockers usually have some mechanic like some innate unpredictability, quick movement or things like that, but heavies are mostly just pure damage and high health. Also, once you learn how to fight them, they lose that shock value unless the situation makes its high damage and high health more immediate. Shockers are kind of required to stay shocky after the player understands them.


Tank, from Quake 2

Quake 2 was another game that has a pretty decent lineup of monsters, and of course, it has heavies of its own. The Tank was its primary heavy-class monster. It took a lot of damage compared to other monsters (but some of the weapons almost-literally melted it), but it was practically a weapons platform. One arm is a machinegun, one is a reliable, decently-damaging blaster, and the other is a three-chamber rocket launcher. It was almost exclusively medium-to-long-range, and it had no melee attack of any kind. In fact, as I’ll discuss in a moment, it was really punished when the player got close.

The biggest point of the Tank is that it has so many attacks. It doesn’t just pull the same trick all the time like the Shambler does. Each attack has its own tell, and its own way of dodging. You had to duck under the sweep from the machinegun, dodge the blaster shot, and usually had to take cover from the rocket launcher. More than that, there’s another version of the tank you encounter towards the end of the game called the Tank Commander. It had a different color so the player could tell them apart, and a lot more health. Of course, the Tank was pretty much phased out once the Tank Commander was introduced. This kind of variety is good for heavies, because it keeps them dangerous. The Tank could be harder to deal with than the Shambler because it does different things than just the one attack, and the Tank Commander is good for refreshing it if it got stale (or would have been, if it was more different).

The Tank shows that heavies can work well as turrets, if they have a properly-designed attack for it. The three-chamber rocket launcher was great for bombarding an area if it was positioned correctly, and it was tanky enough that you won’t going to clear that obstacle quickly. The two roles aren’t really exclusive to one another, but I consider it primarily a heavy because it’s so much more dangerous than any other monster in the game (again, by virtue of high health and high damage). It does have one massive flaw that forces it to be used like a turret, however; if you got behind it, you had a few seconds where it needed to turn around to be able to attack you. If you were were quick, you could stay behind it and unload with the super shotgun, basically keeping it helpless. In my opinion, heavies shouldn’t have flaws like these, or if they do, they should be used in such a way that the player doesn’t have much opportunity to take advantage of those flaws. It really blunts the effect this role should have.


Brute, from Dead Space


Heavies can also work as tanks (talking about the role. When I’m talking about Quake 2’s Tank, it’s with a capital T). The Brute from Dead Space is a pretty good example of this. It was a mostly melee monster, but had a ranged attack as well (not many people know about it, but I played around with the first one you run into and noticed it) letting it adapt to those situations. It could also close distances pretty quickly if you let it, and hit like a truck. It makes a good heavy in its own right, but what makes it a tank is that it has several thick plates in the front that mitigate damage. If you want to hurt it, you have to shoot it in the back, or hit one of the weak points near the shoulder (tricky to spot, but they’re there).

Heavies can work as tanks, by definition. Heavies sort of need high health in order to have the impact they want. When you add a gimmick like the Brute’s armor plating, you have a very effective tank as well. The difference, however, between tanks and heavies (and the reason I keep these two roles separate) is that heavies should be perfectly capable of killing the player. Tanks just need to get the player’s attention somehow so they can absorb damage, but they don’t have to actually kill the player themselves. Every heavy I can think of off the top of my head would be a decent tank, but not every tank would be a good heavy. Red Faction: Armageddon’s Monolith isn’t, for example.

The problem with the Brute, however, is a problem I alluded to with Quake 2’s Tank. With Dead Space’s limb-severing mechanic, you can kill them very easily. If you know where to shoot and are fast enough, you can take advantage of the gaps in the Brute’s armor and pick it apart with the game’s first weapon! Even if you know how to fight them, the Shambler and Tank (ignore its turning speed for a second) can still be difficult fights. As I said before, heavies shouldn’t have flaws like this, or if they do it should be done in such a way that the player can’t take easy advantage of it; otherwise the effect the heavy should have is lost.


Hercules, from Binary Domain


The Hercules from Binary Domain lets me make one extra point that I found interesting. As a heavy, it was pretty standard; it took a lot of damage, it had a large chaingun that ripped right through you, and it was very, very obvious in a fight. It wasn’t encountered often (aside from a few rail-shooting segments), as well.

What was interesting about them was another monster entirely. There was one support monster you encounter throughout the game called the Whirler. All it did was ferry individual fodder into the fight. About half-way through the game, though, we see flying Hercules’ being kept aloft by Whirlers, making the Hercules a pretty different fight, though we immediately know what to expect because we have seen both monsters before. While it was a concept that was only really hinted at in Binary Domain and I haven’t seen it fleshed out anywhere, heavy-specific supports (supports that only exist to support heavies) would be very good to see somewhere.

Heavies are pretty straightforward in how they’re balanced. You’re trying to make a monster that absorbs and deals a lot of damage, so make sure it absorbs and deals a lot of damage. When you’re making a monster that tries to actively kill the player, there’s a trap a lot of people fall into where they design it in such a way that it does it too well. The attack does too much, it’s too difficult to dodge, etc. You always have to give the player some way of defeating it. If your goal is to kill the player, put him in a crusher as soon as the level starts. The goal here is to challenge the player and provide an enjoyable game. Once you have that in mind, balancing them is easier than other roles, like shockers and supports.

One more thing I have to say, however, is about heavies with specific vulnerabilities. The Strider from Half-Life 2, for example, was only vulnerable to explosive weapons, and even then it took a lot of rockets to bring down. What ends up happening if it’s outright immune to your other weapons is that you have to make sure that the player has more than enough of that high-power ammo to bring it down. Otherwise it gets really frustrating for the player (and indeed, that first time you fight them in Half-Life 2, it was). In order to balance the large amount of ammo the player needed, they made sure the rocket launcher couldn’t carry much ammo and placed crates that give you infinite rockets, so you had more than enough for that boss fight, but didn’t have enough to outright break other encounters. However, if you go through all that trouble for just the one monster, it’s not really a heavy anymore, it’s just an outright boss encounter.

The heavy is a role that has been around since the beginning of the shooter genre (the SS in Wolfenstein 3d), and is one of the central roles you can find in most of them. Even modern-military shooters could feature a heavy in some form because of how simply they work. They’re also one of the easier roles to balance correctly. Like shockers, they have a tendency to evoke that reaction from the player, and convey a feeling that a big battle has just started.

Roles of Monsters: Shockers

I did say in the kamikaze article that they can provoke some emotional response (really, just panic or fear) from the player. Shockers are a class of monster dedicated to making the player panic, through some mechanic or because of how it works, or scary to the point where he gets a little afraid from just hearing a sound from one of them. For example, a kind of famous moment from my playthrough of Quake 1, a friend of mine signed in on chat and it played the Doom 2 Archvile’s sight sound (that I set to play when she logs on). Despite that monster not being in the game, I still reacted and immediately dove for cover to avoid it.

Now, this is probably the most subjective article I’ve written on monster roles. How much panic a particular monster inspires widely depends on the player, his skill with the game, and his play style. If you disagree with what I say about a particular monster, I hope you do still see the point that I’m trying to make.

Fiend, from Quake

Quake’s Fiend is the first monster I consider a shocker (and honestly, what prompted me to write this article). It has some of the highest health in the game, and is very quick besides. It attacks by slashing with its claws when up close, and will leap at the player if it’s not quite close enough, doing pretty high damage if it connects. Several times in the game, the player walks through a doorway or opens a door, hears the throaty cry of a fiend and has to react quickly to avoid getting a face- or side-full of spikes and claws.

The player’s first encounter with a Fiend is one of the most frantic spots of the game; it’s leaping everywhere and you’re doing your best to avoid it, but you’ve probably already taken one or two good hits from it. If you’ve fought it enough times before, you know what to expect and how to deal with it, but not that first time. Because it’s so in-your-face with its mobility and damage, and because it won’t die easily, this thing makes the perfect example of a shocker.

Spawn, from Quake
Another example from Quake (seriously, the last episode of the game on Hard mode is one of the most frantic and difficult games I’ve played), the Spawn fits beautifully into this category. It keeps a low profile, and announces its presence with a quiet growl, but then starts bouncing all around trying to crash into you while making a very distinct clicking/slapping sound. It does lower damage than the Fiend, but is harder to avoid, and around the end of the game they appear in high enough numbers that they can be considered swarmers. Its real danger, however, is that it explodes when killed, doing as much damage as if you fired a rocket at your feet. Worse still, because its attack requires that it gets close to you, it’ll probably be right next to you when it explodes.

While the Fiend is shocky by virtue of its mobility and damage, most others are because of some innate mechanic. The Spawn is a great shocker because of a mix of both; it’s so in your face, but also has its kamikaze secondary ability. The best qualities of shockers are either high mobility or some stealth, and usually some high damage that’s tricky to avoid. It’s not an exact science, however; you’ll probably have to play around with a few different designs in order to achieve the desired effect.

Hunter, from Half-Life 2
Like with most dangerous monsters, good introductions are helpful to have. With shockers, it can be helpful to inform the player that this is something that should really be feared. Half-Life 2, famous for introducing its monsters and gimmicks fairly gently, does it very nicely again. In Half-Life 2, Episode 2, towards the beginning of the game, you enter a deserted area and hear odd sounds from an unknown enemy. After looking around for a way to continue, you and your sidekick Alyx are ambushed by a Combine Hunter, which knocks you unconscious and mortally wounds Alyx (Sorry for the spoiler, but it’s ten minutes into the game and you should be playing it anyway). The next part of the game is looking for a way to help her. Later on, about halfway through the game, you’re given another introduction, where you and Alyx hide from a patrol of Hunters, before attacking and killing them. In this way, the game lets you know that this is a monster that should be feared and taken seriously. It can work pretty nicely.

The unfortunate thing is that when you tell the player that this is a monster that’s worth panicking about, you have to make sure that the monster is worth panicking about. While again, this is pretty subjective, I don’t think the Hunter is as much a great a shocker as the Fiend, for example. It takes and deals more damage than most monsters and can be pretty quick, but it’s not as in-your-face as the Fiend or Spawn. It mostly hangs back and fires its flechettes at you. However, while it’s a great monster and another excellent design to come from Half-Life 2, I’m mostly bringing it up as an example of how your game can introduce shockers.

Archvile, from Doom 2
Some shockers just need one particular gimmick. Doom’s Archvile is pretty quick and takes some heavy damage, but also has one very interesting attack. It sets you on fire (just an aesthetic thing; it doesn’t damage you at this point), and after a few seconds of channeling, if there is still a clear line-of-sight between you and the Archvile, it blows you up dealing heavy damage and throwing you in the air. It requires you to stay on your toes. The main gimmick of the Archvile, however, is that it resurrects other monsters. Depending on the arena, all of these factors combined can really make the player panic; especially if it’s an area the player already went through and left some bodies behind.

It can really be considered a support/shocker (though I don’t consider it a support because of how dangerous it is, even when its alone), and is even used as a turret occasionally because of how its attack works. It’s a great example of a monster that can fill several roles depending on how it’s used. Some shockers just need some basic gimmick, however, depending on that gimmick, to be effective. For example, the Poison Headcrab, despite being a hard support. I have no doubt Quake 4’s Teleport Dropper would be a good shocker if it didn’t have those crippling problems it does, but it just wasn’t dangerous enough for that.

In general, effective shockers are ambush monsters. They work a lot better when they get the drop on the player; think of the Fiend, leaping at you from behind doors or from around a corner. Or Poison Headcrabs jumping at you from the shadows. If you design your shocker to be focused on ambushing, loud, distinct sounds and high mobility are really beneficial. If a player’s startled, he’s probably not going to fight back effectively against a monster that’s already in his face. You shouldn’t do this very often, though; Doom 3 is heavily criticized for pulling so many cheap scares throughout the course of the game. Pulling this as few times as Quake did, though, can make for a few memorable fights (Play Quake 1 E1M3 on hard mode, you will know exactly what moment I’m talking about).

Almost by definition, shockers become less shocky as the player gets more familiar with them. People fear what they don’t understand, so as they come to find patters with shockers, they don’t fear them as much. This is why I think the Fiend, Poison Headcrab and Archvile are so great; even when you understand them, they still require quick reaction, so they still startle you and make you scramble to avoid them even if you perfectly understand how they work. The Spawn still feels really random and unpredictable in its movements, so it’s a bit more stressful to fight as well. Always make sure your shocker can hold up after the elements of surprise and mystery are gone, so it can still fulfill its role as a shocker. One great example is the Wraith from Red Faction: Armageddon. Even without any upgrades, you can tag it with the magnet gun, melee it a few times, and it dies without much of a challenge. It just can’t hold up as a challenge once the player knows how it works.

I feel like shockers would do very well in games with a lot of ambiance, that try to draw the player into the world. Doom 3 was one such game, with beautiful ambient sounds and good detail all around; you occasionally heard gunfire from a surviving marine or the distinct growl of some particular monsters. If you have a well-designed shocker in a game like that, you just need to play a sound from it, and the player will react, usually getting more cautious or jumpy. Doom 3 tried to do this with the Commando Zombie, but like Red Faction: Armageddon’s Wraith, the Commando Zombie just couldn’t hold up to being a shocker after you realize you can duck under its tentacle attack and just blow it away with the shotgun. However, the intent was there, and if it was more dangerous, it would have worked beautifully.

In survival-horror games, you should approach shockers through a certain lens. In most games of this genre, most monsters only make you panic because the game itself is built around it. The environment tries to hide the monster, sounds and stingers play when they appear, and the game mechanics are such that you don’t have an easy time reacting to their attack. If you try to decide if a monster in a survival-horror game is a shocker, ask yourself if the monster would have the same effect on the player in a game like Quake, where the player is more capable. For example, in Dead Space, most monsters try to surprise you, but the only one I think of as a shocker is the Twitcher (because it’s so fast and hard to avoid). The Slasher, Leaper or Brute would be perfectly fine to deal with in Quake, but I would consider the Twitcher a shocker because of how difficult it would still be in that other engine. Again, however, it had a problem; if you upgraded a certain weapon’s damage enough, you could literally kill them in one shot with no effort, making it useless as a shocker.

Shockers can work pretty well in main encounters; Quake proved that a shocker (Fiend) combined with a turret for area denial (an Ogre on a high perch) can work pretty nicely, for example. The potential problem is if those monsters can hurt one another (in the case of kamikazes especially), and especially if the game has Doom- and Quake-style infighting, where monsters will fight one another if they take damage. Many shockers that are just all over the place in combat can provoke this a lot. That’s when shockers could work better as lone-wolves because they can wreck encounters if they aren’t used carefully.

One small problem that both turrets and shockers (more than most other roles) run into is that the player can potentially just retreat into an area he’s already cleared out and is more convenient and fight there. The shocker would probably follow along and the turret might not be able to, defeating the purpose of an encounter with both. Quake deals with this by usually locking you into an arena, preventing retreat, and unlocking. Depending on how often this is done, this can work well or be a complete disaster. Quake allowed the player to retreat most of the time, but there were occasions where the player gets locked into a battle; usually for especially difficult battles, so they don’t lose their impact or challenge. However, this is a subject for some other time; I’m just bringing it up here because it affects shockers and turrets quite a bit and needs some thought from the developer when he designs an encounter.

Like fodder, shockers can help the designer dictate the flow of combat and gameplay in general. After lulling the player into some form of security with fodder, you can throw a shocker at him to bring the level of involvement right back up. The main objective is to allow low-energy segments as breaks while keeping the player engaged, and then draw him right back into a high-energy segment; something shockers can really excel at.

Analysis of a Moment: FreeSpace – Playing Judas

Most of what I’ve written about has been about first- or third-person shooters; a lot of it does apply to certain simulation games like MechWarrior or FreeSpace, however, and these games as well have their own lessons we can learn from. FreeSpace was one of those series’ that had a lot more good than bad, and it’s a beloved classic for almost everyone who’s played it. However, there was one mission in the first game in particular that isn’t always remembered fondly.

Playing Judas was the one and only stealth mission in all of the first FreeSpace. In a previous mission, you go through great lengths to capture a Shivan Dragon; a superiority fighter. In Playing Judas, you get to fly that Dragon (refitted for Terran use) flying a stealth recon mission against a Shivan-held jump node. The mission itself entails avoiding a few point-defense turrets and other Dragons patrolling the area as you scan transports and other ships coming through the area. If you get within a certain distance of the patrolling ships, they immediately know you aren’t a Shivan and they open fire on you.

After we scan a few transports, some bigger ships come in, making things a lot more tense if you’re discovered. They can’t tell if you’re friendly or not; it’s only the enemy fighters and turrets that can. To make things worse, also, for some reason your jump drive goes offline temporarily and you can’t jump out of the area if something goes wrong. Then the largest, most dangerous ship we’ve seen from the Shivans jumps in. We have to scan it, then it throws some fighters out, they spot you, there’s a small, frantic skirmish and then you can jump away, ending the mission.

It should be a pretty awesome mission because you’re flying in a Dragon. Dragons are proven before this to the player to be very maneuverable and incredibly powerful. The mission in which you capture it shows it to you in no uncertain terms; you have to disable its engines by hitting it in a certain spot with a particularly slow beam weapon. Missions in which you dogfight them are almost a nightmare, as well, because they’re so dangerous and so hard to hit. Prior to this, you don’t have a single fighter that’s anywhere close to that strength and maneuverability, so you’re practically jumping for joy at having access to this!

Shivan Dragon
The problem is that your Dragon is horribly gimped, and basically equivalent to the Apollo (the first fighter you fly; not the shuttle). You have poor maneuverability, paper-thin armor and cheap weapons, making for a very disappointing experience. Furthermore, you’re constantly threatened by REAL Dragons, almost mocking you. I can see why they did this, though; they gave you this really awesome fighter for STORY PURPOSES, and handicapped it to make this stealth mission still very tense and make it so you can easily fail if you aren’t careful. If you’re caught, you can be torn to pieces by the other Dragons. But they could have done this without playing on your expectations like this.

To accomplish the same goal while giving you a full-strength Dragon, they could have set one or two Lilith-class heavy cruisers (something you couldn’t even try to kill in a superiority fighter) as a constant threat. Up to this point, Liliths are very scary unless you’re a decent bomber and have support, so there’s still a very real threat even if you have that very capable fighter.

Shivan Lilith
Story-wise, it would have made sense as well because that jump node was held by the Shivans in a contested system (if I recall correctly) and would be guarded. Having that constant axe over your head is important in a stealth mission, even with a full-power Dragon. There are other stealth missions in several FreeSpace mods (Shrouding the Light and the Procyon Insurgency, for example) that use overwhelming odds like this.

Of course, what good is having a full-strength Dragon if you can’t use it? At the end of the mission is that small scuffle and dogfight when they spot you, before you can jump out. That’s a perfect opportunity to give the player a short, intense fight with his souped-up fighter. With the Liliths around, it still stresses the player because he knows he can’t win in the long run, but just like in the original mission, he knows he just has to hold out until his jump drive gets back online.

The other major complaint about this mission is that it’s a stealth mission in the middle of a large, shoot-’em-up space fighter simulator. I didn’t think this was as big a deal because there were quite a few types of missions in the game already (escort, assault, disable, etc.), but I can entirely see the issue here. It’s a “Don’t shoot anything!” mission in the middle of a “Shoot stuff that’s red on your radar!” game. The only way you could improve this is to have another stealth mission earlier, so it isn’t as big of a shock. Story-wise, however, I have no idea how they could have worked that in properly, because there’s really no stealth-capable fighter until at least the expansion (and even then that’s not amazingly stealthy).

FreeSpace is one of my favourite games; nearly every aspect of the game is excellent. Unfortunately, the game that has no flaws whatsoever doesn’t exist (besides Lemmings, because honestly, it’s Lemmings), but we can learn from those flaws and make these games better as a result. This article is mostly academic, of course, unless someone uses the map designer to make a new version of that mission that implements these changes. We can speculate how this would work, but until we try to play through the mission, we aren’t totally sure how this would work. I’ve hopefully done a good enough job illustrating and explaining my ideas and why they might improve the mission, though.

Roles of Monsters: The Turret


Most of the offense-oriented roles we’ve covered have been pretty fast as a rule; swarmers have to move quickly and kamikazes have to be able to keep up with their targets. By definition, they’re usually close range as well. It stands to reason that some monsters, however, don’t have to move around much and are still dangerous from a distance. Turrets are one of those monster roles that isn’t very mobile (though they don’t have to be immobile) and usually attacks from a large distance.

Vore, from Quake
The Vore from the first Quake is one such monster. Of course, the word turret implies something immobile, but the Vore can actually move around. It’s pretty slow, however, and is usually placed far away, across terrain it can’t traverse (across a gap, on a ledge, etc.). Because it’s so suited to long-range, its attack has to be able to hit at those ranges; thus, it has a projectile that homes in on you (famously so, since it’s capable of hunting you around corners) that’s hard to avoid and does a pretty decent amount of damage for what it is. It also has the second-highest health in the game, so it won’t die particularly quickly.

The first problem you run into when designing a turret is what kind of attack to give it. Because they’re usually placed away from your target, projectiles don’t usually have a good chance of hitting. The Vore gets around this by having a famously aggressive seeking projectile. Alternatively, you could use a hitscan attack; an attack with an instant travel-time, like the bullet-based weapons in many older games. Hitscan attacks on turrets do have problems (and indeed, their own class), but I will go into that in detail later. There’s no hard-and-fast rule for these, you’ll just have to plan out how your turret will work and be careful how to design and balance its attack.

Icarus, from Quake 2
The Icarus from Quake 2 was another decent turret, with a twist. It had a quick-moving projectile attack (quick blaster burst), and could take a bit of damage, but the major difference was that it hovered in the air, making it a bit more mobile than the Vore was. Usually they were placed high up where the player wouldn’t expect, also, and it was still pretty slow so most of the time its added mobility was just so it got a better firing angle on you. Its rapid-fire, fast-travel projectile attack also helped it actually hurt the player from far away. It didn’t have any hitscan attacks or homing projectiles, but its attack was so quick that you have a good chance of taking damage from it. They were usually encountered in small groups of two or three in later maps, as well.

Turrets benefit quite a bit from good positioning. If they aren’t placed well, they might not be able to hurt the player, or they can be avoided pretty easily. However, if your turret can fly, like the Icarus, it can reposition itself as it needs. The only major issue with that is that you might not be able to restrict it to a particular area (though if you need it to reposition, you might not want to).

Jackal Sniper, from Halo 2
One more variant of glass cannon is the sniper, and though there are many kinds (Deadeye from Binary Domain, Combine snipers in Half-Life 2, etc.) the best example I can think of is the Jackal Sniper from Halo 2. Like all glass cannons, they died very quickly and had very-damaging attacks. The difference between more ordinary glass cannons and sniper-types is that sniper-types have long-range attacks that are USUALLY (but not necessarily) hitscan. The Jackal Sniper had a Beam Rifle that was capable of wiping out your shields in a single shot, and was very difficult to avoid because the attack instantly hits, so you have to depend on its random chance to miss.

The major downside to sniper-types is that they’re even more of a luck-based challenge than more usual glass cannons. I’ll use an example from a well-known Doom mod; In Ultimate Torment and Torture, there were one or two particular maps that featured Zombie Railgunners and placed them pretty far away. The big problem was that they could hurt you from across the map for a decent chunk of damage every time they hit. In an effort to balance this they’re usually very weak, health-wise, but you still couldn’t easily kill them. Most of their ability to survive is that they’re so far away, and in the case of those Zombie Railgunners they’re impossible to hit with most weapons in the game. Like other glass cannons, they’re very difficult to properly place without being either useless in function or unfair on the player. The upshot is that they can work very well in accomplishing the job turrets are made for, if they’re well-used and well-designed.

Speaking of which, the main reason to use turrets is as area-denial. Sort of like a support, turrets are designed to back up encounters with other monsters, but they do so by reducing the area in which the player can operate. If he does go into the turret’s line of sight while fighting other monsters, he risks taking even more damage than he would otherwise. Because of this, they should be restricted to a certain spot so they can keep an eye on that area (definition of a turret), and they have to do enough damage that the player doesn’t want to draw its attention. Conversely, they should be tanky enough (or far enough, especially in the case of snipers) that they can’t be easily wiped out by the player. You might think flying turrets go against this, but not necessarily; they still might be denying an area, but that area itself is shifting, depending on how you design the area itself and how that flying turret moves. Different flying turrets can move in different ways to cover an area more effectively. For example, The Icarus constantly moves towards you (it’s the standard monster AI for all the early Id Software games), but the Heavy Hovertank from Quake 4 shifts around its position randomly to try to get better angles on you.

Turrets can also work well combined with other roles. The Ravager from Red Faction: Armageddon is a good swarmer/turret; they bounce around back and forth between walls and ledges, doing ever-increasing damage to force the player into or away from an area. Like other swarmers, if you kill one, it reduces the threat, but the effect isn’t completely gone. Depending on the design, turrets can also be tanks, as well; since they do so much to shape the area the player can fight in, they demand some attention (The Vore does this). And fitting the purpose of these articles in general, turrets work well when placed in tandem with other roles. Imagine an arena with some limited cover from a turret, with a tank and swarmers being the real challenge of the encounter. The tank takes your attention while the swarmers try to deal the damage, more effectively because the turret is limiting your movement.

Battlefield control is one of those elements of design that map-designers like to have a lot of flexibility with. Turrets let them design those more interesting encounters, especially in games where the player is especially mobile. Doom 2 and every (single-player) Quake game have these to balance out how capable the player is. We can use these to make the game more challenging, and more fun for the player.