on Aug 18, 2008
In response my article on the decline of Survival Horror, Luke Maciak suggests a different kind of SH game, and while I always love discussions on gameplay mechanics, this one really scratched my particular itch. I think it would make a tremendous game.
What he’s proposing is a Survival Horror game that focuses on the strengths of the genre (being in frightening situations) and de-emphasizing the things it does poorly. (Combat, which also undermines the scare factor of a game.) Read his bit for the full set of ideas, but the short version is that the game should focus on hiding from monsters as opposed to fighting them.
Allow me to join in with the armchair game design…
The drawback here is that you will run into the same problems faced by other stealth-centric games: You can build a game around sneaking, but sneaking is a pass / fail activity. (As opposed to a combat game, where you can have degrees of success – you can be partly injured, but you can’t be partly discovered.) You have to allow that the player will make a mistake and be spotted from time to time. If foes are unbeatable, then the game becomes very difficult and unforgiving. If foes are beatable, the player might just realize that it’s more expedient to blast their way through the game instead of fiddling around with all that sneaking.
I just sneaked past nine guys. Oops! I walked right into number ten. Combat ensues, and the noise ends up bringing the other nine. If I’m going to fight them all anyway, then those twenty minutes of observing patrol patterns and careful movement were a complete waste. I call this the “Metal Gear” effect.
One way around this might be to give the player a weapon that kills in a single hit, but with limited ammunition. Unlike other survival horror games, in this one you can headshot zombies (or whatever) and drop them with one round. But ammunition is very scarce, and using the gun is a sort of penalty. The number of bullets you have determines the number of mistakes you can make. If you blunder into a zombie, you have to use a bullet and then find a hiding place until the trouble blows over. Like Assassin’s Creed, those hiding places should be iconic and easy to identify for a panicked player.
This is much like how the Prince of Persia: Sands of Time series worked. As a game designer, you give the player a way to correct a limited number of mistakes, and you throttle that supply of “bailouts” to keep them from being reckless or cavalier. You replenish that supply at carefully controlled intervals, to make sure they have enough to get through a given area.
The trick here is that you’d have to train the player very carefully at the beginning of the game. If you hand a player a gun, they are going to want to Wolfenstein their way through the place. It’s human nature. That gun makes them feel safe and powerful. Letting them learn via repeated failure (death) is not fun and would be a disastrous way to begin the game.
But all-stealth-all-the-time is going to get monotonous. You can’t take that one moment of “oh noes the zombie is coming this way I hope he doesn’t see me!” and stretch it out over ten hours. Thief and Metal Gear have other types of gameplay mixed in there to keep things interesting. Let’s build on a few established ideas:
Once in a while games will give you an area where you can block a door or a window, but I haven’t really played any that brought this idea to the forefront. I can envision a game where there would be large set-piece conflicts where you have to prepare for a coming onslaught of foes by erecting and maintaining a series of barricades.
Although once again, this creates a binary pass / fail situation. If the player does a poor job in building their barricades, or if they overlook an entry point into the room, then they’ll get swarmed and die. (Or deplete their ammunition, which is effectively the same thing.) That’s no fun.
You can fix this by making the “barricade” sections of the game have “fallback” rooms. So, if you get overrun in the first room, you can bolt to the next, where you’ll have thirty seconds or so to seal the place up before your enemies begin assaulting the new room. (Let’s assume that between these rooms we have heavy steel doors that foes can’t penetrate, and that we’re worried about them swarming in through the windows.) You can give the player a few rooms like this, and then simply require them to survive for a limited time. (If we’re talking zombies, then perhaps they must survive until dawn. Or until the power comes back on. Something like that.) They have to survive for (say) five minutes, and if they botch the barricade in one room they can try again in the next.
The risk here is that they would barricade the safe escape route. Again, it would all come down to player training. You have to explain the rules via an NPC, a narration, or a soliloquy. This could be really fun if you can clearly communicate the goals to the player. They would run around the room, repairing damaged barricades before they collapsed entirely. The smashing and rending of the enemies as they tore their way into the room would make for some tense moments.
The final gameplay mechanic could be a nice monster chase. Present the player with an unbeatable foe and then have it chase them through an area. This is exactly the mechanic of the Guardian chase in the Antlion lair from Half Life 2: Episode 2. Now, that part of the game wasn’t a huge thrill for me, but I think it’s a solid idea and many people loved it.
As with Episode 2, this would need to be a set piece, with the player crawling through windows or darting through doors, with the monster having to crash through walls in pursuit.
Again, you run into the pass / fail problem, but I think the gun could be used to solve this. If the player shoots the monster, it will be knocked down for a second and allow the player a chance to get a few steps away. Once again, you need to carefully communicate how this works to the player, or they will see that their bullets “hurt” the monster and simply try to kill it outright.
I don’t want to get too hung up on plot here. A good plot is crucial for making the game interesting, but the above gameplay is flexible enough that there is no need to marry it to any particular story. Certainly you begin with the premise that the player is trapped in some sort of horrible place infested with zombies or other monsters. You isolate them, and then you provide them with some motivation for going forward. The most important thing is to make sure the story doesn’t undermine the gameplay.
But for the sake of fleshing this out, I’ll toss out an obvious plot to act as the backdrop for all this. I like Luke’s idea of a MacGyver-style protagonist, so let’s start from there:
The player is a scientist at some sort of research facility. (Those troublesome corporations / governments, when will they ever learn?) An outbreak occurs. Amid the chaos, one scientist darts into a large test chamber and slams the door shut. She cowers in the dim light for hours as snarling horrors pound endlessly on the door. Eventually the pounding subsides. She doesn’t have any food or water. She waits as long as she can, but it’s clear she doesn’t have any choice but to leave her safety vault in search of help. After scewing up her nerve and giving herself a character-revealing pep-talk, she opens the door and emerges into the ruined labs.
As she exits, the game begins. As a scientist, she’s far better with tools than weapons, and has to rely on ingenuity and environment-based puzzles to survive. You can introduce the player to game mechanics one at a time as they “find” useful items. “Oh look! The screwdriver: I can use this to remove vent covers and move between rooms.” Or perhaps, “A lighter? I can use this to set off the fire suppression system!” And so on.
One good object to give the player would be a cell phone. They can call for help, and an NPC can guide them through the game. A little icon will appear letting the player know the cell phone reception is good in a given room, and that they can call out. Doing so will give a hint on what their current task is or deliver a quick a tutorial on an upcoming area. You’d have to be very careful with the writing to avoid reminding the player of the unintentional comedy of the Metal Gear codec sequences, but if done properly you should be able to train the player without completely wrecking immersion. As a bonus, the cell phone could double as an ersatz flashlight as well.
This would make for an interesting game. There would be sneaking sections, barricade sections, chase sections, and puzzle sections. Mix in a few “cutscene” rooms to give the player a reward for progressing, and you have the ingredients for a game that can offer a lot of variety and can dish out some genuine thrills. From hold-you-breath suspense of the sneaking sequences to the adrenaline-pumping barricade and chase sections.
Sadly, what I’ve just described is, without a doubt, a mainstream AAA game. I wouldn’t attempt this in 2d, and a 3d game with this many elements would be extraordinarily difficult to pull of on a tight budget. Heck, 3D avatar-based games in general are tough for indies, even when dealing with established and well-understood gameplay. Something like this would require a lot of careful playtesting and note-taking on the part of the developers to deliver a fun and polished experience.
Still, it’s fun to dream.
Shamus Young is an old-school OpenGL programmer, author, and composer. He runs this site and if anything is broken you should probably blame him.