Help me develop my game idea (really long and lame)

[Gukumatz]

Hey guys,

When the fall comes, I’ll be starting my bachelor in game development and basic coding, and following it up with either an in-depth programming/animation education or a economic/software project leadership education and starting my own company. I like telling stories and I love the game as a medium.

The game I want to make is a story-centric one, not a gameplay-centric one in that it’s a singplayer-only game and it’s going to be non-linear in a structured fashion. I find most “non-linear” games out today, like Oblivion, to be very abstract and am a bit frustrated by the generic, generated content, so I want to make a very designed, living world.

Working title is Underground. The sub-title is Occupation.

[Warning: Longass text ahead. Apologies in advance for my English and lack of entertaining writing skills.]

The story is centred around a fed-up office worker loosely based on Ed Norton’s character in Fight Club; your typical cubicle worker who resents his life. After the loss of his wife and child in his late 20s, the person loses most of his touchstones to reality and joins the Army as a way to ground himself. We’ll call him Leonard, Leo for short - from Leonard in Memento.

The tutorial part of the game will come through a few hours’ worth of boot camp exercises and will root the player in the basic FPS fare of character control, use of cover and weapon handling. The game will be unofficially set in the 90s, because I don’t want to be handling futuristic prototype weaponry and the like.

I’ve got some ideas for the cover that’ll be developed further, based on my own amateur experiences with in-doors paintball and Home Defence wargames. Basically, if you put a guy in an urban or in-doors setting against an enemy he doesn’t know, he’s going to be using whatever he can as cover, even if it’s just a door, and I want the character to be able to both interact dynamically with the enviroment - as one would in real life - and to be able to create cover for himself in a fully destructible enviroment. An explosive charge should do more damage than just creating black burn mars on the floor; a nice, little crater, for example.

Leonard finishes his Army stint after a few years. This will mostly be told in non-interactive cutscenes, though I want to give the player a few minutes of balls-to-the-walls action, to keep him hooked and to play up the FPS part of the game.

The real story starts six or seven years after that, though.

In year x+10 (x being the year he joined the army) a neighbouring army invades and occupies the country. (It’s not the US; a fictional country probably based on Europe, mostly because I don’t really know the US and because the US doesn’t really fit the bill) It’s a flash invasion à la Red Alert and Freedom Fighters, with occupying forces being deployed very quickly over the entire country. The standing army faces several quick, ruthless defeats and the character isn’t given much of a chance to join them before they’re effectively out of play.

The occupying army puts on a benevolent face for the sake of public opinion and generally treat the population well, although details will surface during the game that the country is being used for strategic staging purposes. Much like Norway during WWII - and of course I am taking a bit of the game from that.

Leonard is upset and figures he’s got nothing to lose. The real game starts with Leonard watching the invasion in disbelief at home (think broke-down apartment, cheap TV-set, large book collection) and grabbing his Home Defence kit and getting ready to deploy to the army. His door’s kicked in by either uniformed hostile combatants or local, revolutionary insurgents who knows he has an army background. You gain control of him behind his couch with a handgun and a stock H&K G3A3. (I debated using the HK416, but it’s a more recent development and definitely not viable for HD-troops for a few years yet. Also, I want to give the player better weapons later)

Whether or not you shoot your way out or GTFO through the fire escape behind you is up to you. Considering you’ll be unarmoured and there’s three of them, the fire escape may be made the more viable route. Afterwards follows a chase/hunt part where you’ll want to either clear out the building and get back to your room to get the rest of your gear or pick down some easy targets and get their gear. When you’re geared better, you’ll be contacted or drawn to a place close by, where some of your friends are holed up. If you’re in your house to get the gear, you’ll be called up. If you’re outside, someone will come for you in the street. This is a mild form of the conditional story advancement techniques I’ll be using later in the plot.

I’ll skim the plot forwards a bit, to get to more of the core gameplay. You meet up with your friends in an apartment and there’s four of you. Two of them are already equipped; if you picked up extra weapons or if you are willing to relinquish any of your own, you can arm the other two. Considering there’s no organized resistance yet, you now have a squad of up to five men to command à la Freedom Fighter.

You can issue them basic commands - take cover, cover me, fire, retreat, on me and charge. Intuitive commands that are hotkeyed for easy access - but for the most part, their AI will take care of them. One of the features here will be that one of them will be a more-or-less permanent sidekick character or bodyguard, who another player can come in and control by just picking up another controller. I want to create the feeling that the soldiers are at once unique and personable, as well as mortal. If one of your guy gets shot up and dies, he’s dead. Not coming back. However, he can also be wounded and taken off action after a map and come back later. The scarceness of people availible will force you to be careful with your guys, and I’m hoping to back this up by creating empathy for them. It’s a hard thing to do, but I hope it can be accomplished.

After you’ve pulled a few operations - raided a few weapons caches, demolished a few outposts, picked off some soldiers or officers - you’ll be contacted by a larger, organized resistance group who’re having a gathering meeting later that night. You’re encouraged to explore the city until then. In this phase, and many times later, I hope to implement much of the same “social stealth” system as Assassin’s Creed is working on, where you can hide in the masses and blend in, as long as you’re not carrying openly.

When you get to the meeting, you’ll be able to share the intelligence you’ve gathered for bonuses, to encourage further exploration. Say you’ve found a weapons’ cache - the next day, you may be set up to raid it to provide weapons and ammunition for the resistance. Or transports, or life-saving medical supplies which means you don’t have to send your hurt to the hospital (which’ll be monitored, of course) or a lot of things.

The second thing that happens at the meeting is that you’ll be instated as a Rank X -3 (where the leader is rank x) as given recognition for your military experience and that you’ve led the small squad so far. You’ll be assigned one squad of twelve X -6s with one x-4 so you can split your squad into two for tactical purposes. X -6s is the basic unit in the force you’ll control, the untrained resistance fighters whereas x -5s are either veteran resistance fighters or fighters with military background. They have a significant health bonus, are more likely to have their own gear, shoot more accurately, are more likely to use single or burst-fire and are generally less prone to do stupid things.

On a sidetrack note here, I intend (or rather, wish) to introduce an AI element called Artificial Stupidity. You know the one, where your trooper suddenly gets an angst attack and has to have a sit-down, or talks really loudly during a stealth-operation or blows his cover by opening fire on the street.

The third thing that happens at the meeting is the real introduction of the element of “I don’t have the right fucking answer.” Basically, in real life, you’re going to have to work on guesses, assumption and doubtable intelligence. It’s the same here. And we’ll be randomizing it, so you can’t just memorize the sequences: some times they go your way, some times they don’t. I’m not saying they’re set 50/50, though, things are more likely if they’re obvious, but more on that later.

Here’s how it goes: At the end of the meeting, the group is dismissed with orders to start activating each their own cells. (Basically telling people to keep their eyes open and if they see anything, report to them) The people leave, being told to meet the day after next at the same place, same time. As they’ve left, you’re held back by the leader of the resistance who gives you access to a safe-house to bunk in. At the end of the conversation, he points out that one guy, Mike, was looking really shifty, nervous and sweaty the entire time and refused to meet his eyes. The implication is that he’s a collaborator.

You’re given two real choices;
1: Have him followed.
2: Don’t worry about it, it’s a bad time all over.

Now, if you go for option two, two things may happen. (And this is randomized) He can be clean, at which point nothing happens. Or he can be a snitch and the next meeting has it’s door kicked in by the occupying army, at which point you have to defend yourself and change your location.

If you go the safe path and have him followed, two things can happen;
1: He’s a snitch and you get advance warning that something’s going down, and you can bail house or prepare accordingly.
2: He can be clean, but notice that he’s been followed - inexpertently - and get all outraged and pissy, at which point you may lose trust in the group. Trust will be important, though I don’t know quite how, yet.

I feel this will add a good portion of immersion and realism to the game, as you know there are some choices which are made on bad intelligence and there aren’t really any “right” decisions. You just have to go by your gut feeling.

At this phase in your game, this HQ will be the place where you go to get missions, although you’re free to roam the city otherwise and any intelligence you get and any damage you do will be rewarded and appreciated back at HQ. The missions will be toned by requirement, but will basically be of the martial type with two subsets.

Subsets one is direct action against the enemy. Destroying or raiding supply dumps, ambushing convoys, raiding outposts, taking down hostile troops, sabotaging the infrastructure, etc.

The second subset is the more interesting one. Public opinion and trust in the local populace will aid you in many small and big ways. You can gain new safehouses, hide in people’s shops, be aided financially, use social stealth and be offered help and intelligence by civilians and so forth, if your public opinion is high. These missions will go to raise these; they’ll include saving journalists to write propaganda pieces for you, set up illegal printing presses, shake down informants for information, infiltrate enemy bases, defamation, setting up illegal radio stations and providing security for closed public rally meetings.

All of these will require a considerable effort - for example, if you’re safeguarding a meeting you’ll have to post civilian and armed watchposts for early alert, provide backup and guard the escape route. Your public opinion will also affect the odds of someone snitching on you, so don’t underestimate their worth.

Having a high public opinion also increases the pool you can draw resistance fighters from, as well. Of course, things like unnecessary civilian casualties and collateral damage will decrease public opinion.

The last facet of the game I’ll go into today is the relative non-linearity of fixed storyline events. For instance, at a certain phase of the game you’ll get bumped up to a X -2 rank because of the previous owner vacating the post in one of the following manners;

1: He gets shot in an encounter.
2: (If you keep him safe in 1) He gets shot on the way home from the encounter, in an ambush.
3: (If you keep him safe in 1&2) He dies during a raid on the HQ.
4: (If you keep him safe in 1-3) He abdicates his post to secure his family during a phase with heavy resistance crackdown.
5: (If he’s alive at point Y after 1-4, OR if his family is killed as collateral) He abidacates his post to serve as a X -1 at another place.

I want this more non-linear approach to build up an illusion of a proper real-life dynamic and to keep the game’s pace smooth and unbroken.

The Oblivion Kingkilling is a major pet peeve of mine. [spoiler] When the Emperor dies at the beginning of the game, I was left with the feeling that this is fucking stupid. I’m standing right here in front of him, paralyzed-like, even though I’m pretty damn sure that I could’ve saved him. I could’ve flung a fireball in the assassin’s face or stunned or chilled or insulted his mother, but I’m left helpless to further the plot because of essentially lazy design that breaks immersion.

Right, that’s all from me today. If there’s any response, I’ll post further info, but I know hardly any of ya’ll have reached this point It’s cathartic to have it down in writing though, so I guess I’ve found my satisfaction with this thread already.

Love,
Ebster.

[Omi_no_Kami]

Disclaimer: I’m not a professional designer.

Unpleasant observation: This isn’t going to happen.

It’s good that you’re working on design, as you can always use the experience. The padding it gives your resume doesn’t hurt either. However, I think you’re honestly better off starting with small projects that emphasize your design abilities; a game of this scope requires, at the very least, several dozen experienced programmers, artists, writers, and between seven to twenty million dollars in funding. Long story short: as a newcomer to the industry with no substantial design experience or track record, nobody’s going to trust you with the money. Or much of anything, to be brutally frank. Design jobs are kind of the holy grail of the industry, and before anyone trusts you with their money you’re going to need a lot of experience (traditionally this would be in either coding or art), additional experience in lower management and assisting designers, and a lot of people willing to vouch for your ability. At least that’s my understanding.

So my honest opinion: overenthusiastic, poorly planned, and not gonna happen.

But enough of the depressing crap. You asked for help developing your idea, so let’s ignore the corporate stuff and look at the concept.

First off, you need to figure out what market you’re trying to appeal to. It sounds like you want to focus on the “FPS RPG” genre, like Deus Ex, in which the player is plunked on a levelling treadmill from day 1. If that’s the case, you need to figure out the economy of your game: what statistics will the game will need to keep track of for each PC/NPC, and how will they hook together? Which statistics will be transparent to the PC? Which will the PC be able to change? Is progression a one-way street, like in most traditional RPGs, or will players be able to undo earlier changes if they decide their current build isn’t working, like Planetside?

You’ve mentioned that you want this to be non-linear, and yet you seem to be interested in emphasizing the storytelling. This seems a little contradictory to me, so have you thought about how this is going to work? It sounds like you’re taking an industry catchphrase and haphazardly throwing it at a genre… it seems to me as if you’re just taking a bunch of procedurally generated missions (and for them to feel truly random you’re going to need a *boatload * of content so that people don’t start recognizing the same sets), and breaking them up with decision trees. Decision trees aren’t nonlinear, and they aren’t a good way to do it. For one thing, it limits the set of possible outcomes to the collective imaginations of the designers, and several million players are going to be a hell of a lot more creative than a room full of designers. Second, if the entire plot is just a series of branching nodes that urge the player through the game, it won’t feel nearly as open-ended as you might think. Rather, my guess is that most savvy players will instantly spot the ruse, and look for a way to do something else. When they find that the tree forces them to grind through your plot, any illusion of nonlinearity instantly disappears. Because remember: one of the greatest joys in a nonlinear game is the feeling that you’ve done something to outsmart the designers.

If you’re stuck on this idea you could probably try something like the FF Tactics approach, in which the player is herded through a pseudo-infinte number of procedurally generated filler missions which are broken up by intermittent linear missions meant to drive the plot forward. The obvious drawback there is that filler missions have a depressing tendancy to feel like filler missions, and they get boring really quickly.

On that note, what is driving people to keep playing? Progress or the illusion of progress is absolutely key here. Think about tetris: you did the exact same thing throughout the entire game, but every time you scored X number of points you’d go up a level, the speed would change, and the music would change. None of those things really matters, but because it felt like you were being rewarded you always had a sense of progress. For an FPS RPG there are two obvious rewards, both linked directly to the economy: you get better abilities, or you get better equipment. Either way, the rewards are there to create a feedback loop; you shoot people, and you’re rewarded by getting things which make you better at shooting people, which instantly motivates a desire to go out and test your reward by shooting more people. Repeat ad nauseum.

Now that we’ve looked at some big questions that need to be addressed, I should point out that there are already a lot of games which do something similiar. Rainbow Six: Las Vegas is an outstanding squad-based shooter, and since the main focus of the game is conducting squad-based tactical assaults, that’s where they put their focus. You aren’t going to beat that unless you have an insane budget and a 48-month development window. Gears of War is another outstanding member of the genre… the thing is, FPSs in general tend to be extremely focused in their format, and so it’s going to be very difficult to compete based on the action component alone.

On top of this, you need to be thinking about how your AI is going to be structured. Since FPSs tend to be twitch-oriented you’ll have to create a very simple engine, since it won’t have very much time to execute; anything over 5-7ms is going to be really, really noticeable. “Shoot the bad guys,” “Don’t shoot the good guys,” and “Minimize your exposure to the bad guys’ attack vector” is about all you can hope for. This leads us to our next concept: AIs are extremely stupid. There exist some extremely clever AIs, but you’re setting yourself up to fail by designing a system which relies on the AIs driving themselves. This sounds like a game in which the NPC squad constitutes a significant portion of the character’s “wealth”, but to make this succeed you need to treat NPCs as particularly intelligent guns: they’re one of many tools the player can use, but the player *must *be the one driving them. Otherwise, you’re going to end up with tons of players who are angry because the game led them to believe that the NPCs were capable of “doing what they need to do,” which is a concept so nebulous and so variable that any designer who tries to accomplish this with AI deserves a firm kick to the rear end. (On that note, it’s worthy noting that a good deal of the art in implementing AI in a game isn’t making the AI smart, but rather concealing how stupid it is.)

Also, I’m a little confused: you mention that this should be single-player, but then you refer to someone else being able to possess NPCs on-the-fly to join in; is there supposed to be a framework to support multiplayer campaigns built in?

Side note: your “artificial stupidity” idea is interesting, but it’s also a really bad idea. In a tactical FPS a good deal of the player’s success depends on his reflexes and on the decisions he makes. Now obviously, you want the player to have fun, and in an FPS you have to assume that a good deal of players are going to die. A lot. Now I would claim that one of the big reasons this isn’t instantly and unforgivably frusterating is because in 99% of the cases, the player can say “Oops, I messed up”. The player was able to do absolutely anything the game allowed him to, and his death was (hopefully) a direct result of a mistake he made.

Now, take that control out of the player’s hands and put him at the mercy of random events that he has no control over? He’s going to get right pissed, right quick. The only way I could think of to mitigate this would be to give the player partial control over this events: assign NPCs an “Experience” or “Frostiness” statistic which the player could pump points into by sacrificing other areas (marksmanship, strength, technical expertise). This way, if one of his guys messes up and gets him killed, he can console himself by saying “Shoot, I should’ve emphasized Experience more” and adjust his tactics accordingly.

While we’re on the subject of player frusteration, you’re not going to get away with a multi-hour long tutorial. Or an hour-long tutorial. Or a twenty-minute long tutorial. Tutorials are boring, preachy, and they don’t involve blowing things up. If your learning curve in an FPS is shallow enough to merit a long tutorial, that means there’s something wrong with your design. The best way to handle this would be a three to five-minute tutorial introducing basic concepts of gameplay, and a thirty to fifty-minute (bypassable) first mission that gradually introduces additional concepts during play.

Also on the subject of gameplay, cut scenes are on their way out. They’re extremely disruptive, boring, and repetitive.

Finally, this is going to sound counterintuitive but in most cases you’re going to want to write the story last. You start by designing the world the game will take place in, and figuring out what the creation that world is going to entail on the technical end. Basically, you hold a crapload of brainstorming sessions with your team (and you want a team of designers, preferably with as diverse a smattering of backgrounds as you can get. Collective effort almost always trumps individual vision in these cases). Start with something simple: “This is going to be a ‘realistic’ FPS RPG with a seat-of-your-pants feel in which the player will start with a ragtag group of untrained rebels and end up as high-ranking participants in an organized revolution”. Keep talking until you’ve gone through the basics of the world, and then sit down and figure out the engineering behind implementing all of this. How will levels be generated? How will they be populated? What does ‘realistic’ mean in terms of the player’s experience? And so forth. Actually writing a story should come after you have the world working, and shouldn’t begin until you’re 40-70% of the way through the development.

And that’s yet another reason why they need people with a good deal more experience driving these projects: as a head designer you would end up making a lot of these decisions, and making good decisions in this context is as much a function of experience as it is of ability.

So my advice? Start by finding some games you like and taking them apart to figure out what you like. Then see if you can implement what you like in an extremely simple context. And then work up from there.

I hope that helps!

[Yag_Rannavach]

As Omi no Kami said, you aren’t going to get a company willing to give you money without some experience in the industry. What you’ve described takes alot of money and sounds overambitious.

My suggestion is to break it down to core gameplay elements. Cooperative multiplayer, ‘realistic’ indoor environments, and a sort of tactical gameplay map, or some such thing. This will help you focus on trimming the features that aren’t 100% required in order to bring the scope down to something managable.

Also, as Omi no Kami said, as part of the previous step, you should decide who you want to focus on. What type of players do you want to appeal to? Ghost Recon, Deus Ex, Traditional FPS, RTS/TBS? Deciding this will help you figure out what the core features you need to support are.

As an example,
If you want to support strategy gamers then something like X-COM might not be a bad idea, but you’ll cut the ‘urban hide in crowd’ dynamics. If you want Deus Ex, then you’ll need to come up with an advancement system for you and your troops. Traditional FPS and you’ll probably want the ability to buy/earn bigger guns and can probably cut the squad.

If you want to make something like GTA then you can probably cut the ‘randomly generated assault/defense’ missions and stick with a few storyline missions that can alter future storyline missions. Instead of making “assault the weapon cache” a mission that breaks up continuity, just randomly place weapons caches and let the player assault it when they go out for the night.

I would caution against having a ‘stupid ai’ unless you have an advancement system that players control (put points into bravery or some such thing), or let players choose who to take on a mission and make the smarter AI cost more. (1 smart to 10 stupid or some such thing)

[Carnick]

The game industry is like hollywood. You can’t get anything to your name unless someone trusts you, and you can’t be trusted unless you’ve something to your name. The game industry is saturated. Everyone wants in, for good reason. You won’t be seen by potential employers unless you have 5+ years experience under your belt. How do you do that when you can’t get a job in the first place? Good question. Here are some potential ways in:

1. Learn a trade people need. Everyone has big ideas. Not everyone is a programmer or a graphics artist.

2. Jump on an internship. These are pretty rare now because the industry is so bloated with professionals in the pool. Search for small, newly opened game companies and grab an internship if you see it.

3. Start small. Very small. Learn Flash and make a small, addictive puzzle game. Upload it to a portal like newgrounds.com. These projects are small enough for one or two people. This will show people you can make successful designs.

4. Start at the bottom. Get jobs as a tester. Yeah, testing sucks, but at least you’ll have forward momentum. I’ve known a few designers that started this way.

Good luck. If you can make it you have more resolve than me.