How to Design a Video Game, by Mangcho

Game development is divided into 3 stages:

1. Pre-alpha: Organizing graphics/data in human language

2. Alpha: Developing on computer until working build finished

3. Beta: Testing Alpha build, fine-tuning features, promotional material

When the beta stage is finished, the game goes gold and is put up for sale.

Some people are good programmers. Others are good artists and musicians. Others are good writers. Decide which of these gifts you have the most of, and leverage it to put the game's foundation together. Whatever knowledge you lack, either learn yourself (Youtube is great for this) or find someone else with it. The second option is faster, but riskier and more difficult than the first, and the cheaper the help, the more likely it is the helper will flake. Often this is because: 1) helpers overestimate their abilities but don't want to admit it; 2) they said they'd do it because they wanted you to think they could. The reasons for this come down to immaturity and delusion on their part. Bottom line, choose your helpers carefully, and assess their abilities BEFORE you tell them you're a game developer looking for help. Once you start moving into the alpha stage, give a helper a deadline of 2 weeks to work on a small piece of your game. DO NOT reveal everything about the game, just enough information to get them started on the little nugget you've given them. Your game is a collection of trade secrets that others will poach if they can. Don't give them the opportunity.

When a helper turns in their work, but it's not lining up with your needs, ask if they can revise some elements to bring them in line with your ideas. Unless they work for you full-time or have an equity stake in the final product, playing this card more than once can get tricky. Communicate as much as you can before the helper sets to work, so you don't have to constantly manage them. If you reach an impasse, you can take the result they give you as a baseline and sculpt it into something usable.

Depending on the genre, your pre-alpha stage may consist of drawing levels on graph paper, or documents full of headings and basic instructions of what things you want to appear and when. If you can get this done, congrats, you've made the alpha stage infinitely easier for everyone. Especially you.

There is no need to understand coding at the pre-alpha stage. You just need to know what games do, which is, loading certain bits of media in a sequence, when certain things happen. The game also needs to know how to make those bits of media behave. Of course, there are all kinds of tricks you can use to make those bits load faster and smoothly, but that's for the alpha period.

If your game is in a 3D world, whether 1st or 3rd-person perspective, you're building maps with exits to other maps, and furnishing them with objects (assets). Take out paper and start sketching. Think about what items are needed for the player, why they're needed, and where. Same goes for enemies and NPCs.

If your game is a visual novel, you're organizing dialogue, music and graphic effects (animated or not) in a detailed way. Load a word processor and make a nice numbering system for branching dialogue topics, so you you can build a "library" of conversations out of them. Then the computer will choose a conversation from the library, based on a dice roll. Or just make something linear with fewer options. Either way, you're gonna be writing and organizing with headings. RPGs have a similar pre-alpha stage, only with more focus on the battle system and player stats.

If you're doing a 2D game, top-down or side-scrolling, buy a roll of graph paper from your local art supply and pencil in terrain, enemies, items, vehicles, whatever's needed. Use distinct shapes for each element, and create a "legend" for the symbols that includes behavior for each thing. To keep your levels distinct, blend in a mild gimmick to each one (unique environment, focus on jumping over shooting, etc) without compromising the core gameplay, to keep players interested. (Gimmicks are not bad in themselves, just take care not to lean on them in a single way, so the well doesn't run dry. Build on the gimmick, don't coast on it.)

If you're making a simulation, study the real-world mechanics of whatever it is you're simulating and figure out how simple or complex you want your sim of that thing to be. What features will the player need to keep their interest? What won't they need? Work it out in plain language, in a way your neighbor could understand. That last sentence goes for every genre, but when you're dealing with sims, your system needs to be clear and sensible. Anything that doesn't deliver a real payoff to the player's involvement needs to be cut.

This process may take a year or more if you work alone. Finding people to help you is difficult, since so many "devs" get overwhelmed and give up, and few people want to invest in a doomed project. Bottom Line: YOU CAN'T GIVE WHAT YOU DON'T HAVE. Get the fundamentals in stone and spend 90min+ each day chipping away a little bit on your game. You should at least have something to show for it by the end of a year, at least in the pre-alpha stage. What's more, you'll have everything so organized and solid, you can jump into your game after a long day of shift work and generally know what to do next.

Things are different if you're already versed in programming and design, especially C# and software like Blender. You'll want to take the Japanese approach and focus on building as much as possible on the computer, rather than physical paper. Use placeholder graphics/sound and if you do anything as a separate draft, let it be the dialogue or basic map layouts. Writing your overall design, in broad strokes, can be done in a notebook, but you've got such a leg up with the computer you can do most of your work there and save time. Time is the friend and the enemy. 90min+ a day and you can make (slow) progress.

It took 32 months for Toby Fox to make Undertale, from start to finish. That's 2.6 years, full-time. If you're struggling with a job that demands heavy hours, figure out what can be used in a demo, grab some gameplay footage when it's done, edit it into a nice little pitch video using Vegas/Premiere and kickstart it, put it on GameJolt, or indiegogo, or make a .io page, set up a patreon, all of these at once--whatever it takes, and see what happens. Maybe you'll get enough money to take a leave of absence from work, and spend months grinding away at the finished product. Just remember, the longer you wait after a demo drops, the grumpier your investors and the more likely it is you'll have your idea ripped off before you reach the beta stage. Especially if you come up with something really unique, appealing, and simple to make.

Never allow crowdfunders to have any real say in the game's content. This includes giving rewards like "your character in the game" . The investors will never be satisfied, their ideas will conflict with your vision, and you'll just have work and stress for yourself, and your team, for a result that only weakens the finished product. Keep the game content under lock and key, listen to feedback but don't implement anything that will hinder your efforts. This work is stressful and tedious enough without somebody else sticking their finger in it.

Unless your game is a robust 3D adventure of sorts, or a 2D game with high production and content (eg. Cuphead), $20 is asking too much. $10 is a healthy figure but the game has to provide a solid 10-15 hours of entertainment for $9-$12 to be reasonable.

Eat well, stay active, organize your time. Whatever game you want to make will take shape if you follow these principles.

--M.