The indies' guide to game making

Tom Francis

Page 1 of 10

This article originally appeared in issue 246 of PC Gamer UK.

You might have heard that “It's never been easier to make a game.” And it's true. But how do you actually make one? What do you make it 'in'? How much does it cost? How long does it take? Can you sell what you make, and do you owe anyone any royalties? Do you need to learn a programming language?

I don't know, but I do know a lot of indie games. And lots of them are made with tools and suites that claim to be beginner friendly. So for each of the most popular tools, I found an indie developer who had made something cool with it, and asked them what it's like to work with.

I'll also cover how much these tools cost, what your rights are when it comes to selling your work, and what platforms they can make games for. If you've ever been interested in making a game, hopefully this will give you an idea of how long it takes to pick up, which tool will suit you, and where to start.

Contents

Game Maker
The built-in sprite editor isn't bad.

GameMaker

What is it? An all-inclusive development suite for 2D games. You can either create rules with a drag-and-drop interface, or write code in its scripting language, GML.

Price and licence: The limited version is free, basic version is £30, a version for teams is £60. You can sell the games you make with any of them, no royalties.

Makes games for: PC and Mac. iOS and Android versions are £120 extra each, HTML5 is £60.

Link: http://www.yoyogames.com/gamemaker/studio

Tutorial: TIG forums tutorials

Spelunky
Consistent rules are simpler to code.

Case Study: Spelunky

Developer: Derek Yu

Get it: for free

How long does Game Maker take to learn?

It shouldn't take more than a couple of weeks to get the hang of Game Maker. It's very intuitive and there is a wealth of tutorials and scripts for it online. On TIGSource Forums we've compiled a list of tutorials that should get an intelligent person up to speed quickly.

What prior knowledge or skills are helpful?

Some programming knowledge and familiarity with C-based languages would be helpful, so that you can take advantage of Game Maker's scripting language GML right off the bat. No serious GM game is made without using it. On top of that, some skill with making pixel art couldn't hurt! In my opinion, Game Maker really benefits generalists who want to do a bit of everything.

What can't you do with it?

Game Maker 8.1 (the version I'm using) is too slow to handle modern graphics and audio. It is strictly for games that look like they came from the '90s or earlier. But I don't know if that's still true of Game Maker Studio, the latest incarnation of GM.

Spelunky
Making levels tile-based saves a lot of time

How long did Spelunky take to make?

Spelunky took me about a year of on-and-off work to finish, which is maybe double what I guessed when I started working. But I also didn't anticipate that the game would get as popular as it did.

How much of the development time was enjoyable?

I'd say it was 90% enjoyable and Game Maker played a big part in that. Given how easy it is to use, you can spend most of your time doing art and design! That's the most fun aspect of game creation for me. If you enjoy programming more, you might find Game Maker's limitations more frustrating.

How much did it cost you to develop?

I don't think I spent any money on the original Spelunky, aside from the £12 registration fee for Game Maker (£30 these days).

If you could give your past self one piece of advice before starting to make the game, what would it be?

I'd be afraid of messing with my past self too much, since things turned out well and I attribute part of that to my naiveté. Maybe I'd just send myself a cookie!

Conclusion

Game Maker is one of the easiest tools to use for an absolute beginner, and it's flexible enough to make almost any 2D game you can think of. The only reason not to use it is if you want to make something in 3D, or you're planning an adventure game or J-RPG. There are better options specifically for those.

Contents

Unity
Unity looks like a level editor, but does everything.

Unity

What is it? An all-inclusive development suite for making 3D games.

Price and licence: The free version has all you'll need as a beginner, and is fine to use commercially unless you're making more than $100,000 a year from your games. The pro version has fancy things like pathfinding, physics, and graphics tricks, and costs £924. No royalties for either version.

Makes games for: PC, Mac and Linux. iOS and Android versions are £246 each.

Link: http://unity3d.com/

Tutorial: Infinite Ammo's Unity tutorial

Game Making feature Aaaa
Aaaaa! is a game about falling.

Case Study: AaaaaAAaaaAAAaaAAAAaAAAAA!!! – For the Awesome

Developers: Ichiro Lambe (Dejobaan Games) and Alex Schwartz (Owlchemy Labs - Unity Version)

Get it: from Steam

How long does Unity take to learn?

Alex: I'd guess you're asking about how long it takes to prototype a game idea, i.e. make some kind of first playable 'proving ground' to figure out if your game is on the right track toward fun. At Owlchemy Labs, we found that a prototype of gameplay should take less than 7 days for us, and if it takes longer than that to find fun, we toss the whole prototype as it's a good sign that we're on the wrong path.

In addition to that 7 day rule, we also follow a "one day of prototype == one month of polish", so a 6-day prototype will take 6 months of full time work to build out as an actual, polished game.

What prior knowledge or skills are helpful?

Ichiro: Knowledge of another 3D engine, Solid C# (or C++, etc) skills, vector algebra, the ability to research ("Google, help me find documentation on fast dot product functions.") and communicate well ("Dear Unity Community, can you help me with X problem I'm having with Y?")

What can't you do with it?

Alex: I'm of the camp that anything is possible with sufficient time, money, and effort. Also WD40 and duct tape. But honestly, we haven't hit development walls that prevented us from fulfilling our creative goals, and anything that annoys us is usually just a minor editor idiosyncrasy.

How long did you think the game would take to make, and how long did it actually take?

Alex: We estimated around six months, with only two full-time developers on Owlchemy Labs' end and one to three part-timers on Dejobaan's end. It ended up being closer to 8.5 months, with the last month of work spanning the course of about 4 months, due to updates and launch timing.

How much of the development time was enjoyable, and how much was unpleasant?

Alex: I've had the unfortunate opportunity to work with some pretty nasty engines in the past. Thankfully the work within Unity was mostly pleasant. The most unpleasant part of the development of the game involved bringing in 3D Game Studio's proprietary asset formats where a source asset was not available, but that was solved early on in development.

Game Making feature Aaaa
Unity is best for 3D games.

How much did it cost you to develop, and what did that money go on?

Ichiro: Most of the project was done via revenue share, so development/marketing costs were well under $20k.

How well has your game done for you financially, on a scale from 1-10? (Let's say 5 is 'just enough for me to make another game', and 10 is 'more than I could possibly need')

Ichiro: Time will tell. Most of our revenue comes about over the course of two years after the initial push (when we do bundles and Steam sales and so forth). The original Aaaaa! was doing pretty well, until we added it to the Portal 2 Potato ARG. That dialled things up to 11.

If you could give your past self one piece of advice before starting to make the game, what would it be?

Ichiro: Add in one more killer mechanic that pervades the entire game and gets fans of the original to pick up the semi-sequel.

Conclusion

Unity is the best combination of approachable and versatile for 3D games. It's more complex than Game Maker, but about as easy as it gets for 3D development. The free version is very generous, and the recent addition of Linux support makes it one of the only noob-friendly tools that can make games for all three desktop operating systems.

Contents

Adventure Game Studio
AGS provides the basic adventure game structure.

Adventure Game Studio

What is it? An all-inclusive development suite for making 2D adventure games.

Price and licence: Free, you can sell your work, no royalties.

Makes games for: PC

Link: http://www.adventuregamestudio.co.uk

Tutorial: Official AGS tutorial

Ben There Dan That
Adventure games require a lot of art...

Case Study: Ben There Dan That

Developer: Dan Marshall

Get it: from Steam

How long does Adventure Game Studio take to learn?

AGS is really good, in that it abstracts quite a lot of the complicated stuff for you and all just works. So you don't have to think about pathfinding, or setting up text stuff. You can pretty much make the bare bones of something simple in a weekend with a lot of perseverance and very little programming knowledge, because it's largely just filling in forms and ticking checkboxes.

If you want to do anything exciting or interesting with the game though, you're going to have to learn some C. Nothing too complicated, but it's all still programming and it'll take some getting used to.

What prior knowledge or skills are helpful?

You really need to understand the basics of C programming to get anything out of it. While the bulk is done for you, like rooms and characters, you still need to be able to tell them what to do and where to go. You can get by with very little knowledge of that, I expect, but you need to understand the syntax and basics like variables and if/else statements.

What can't you do with it?

Lots, I'm afraid. AGS is great for making faux-1990s point and click games, but nothing else. If you want to even very basic visual effects like fading in-and-out, or zooming the camera, you're going to have a rough time.

It's also Windows only, so no iPad or Linux versions.

AGS is also kind of getting on a bit, now. You could probably make an AGS-equivalent in Unity in about a week, and get all the multi-platform benefits and visual boost with it.

Ben There Dan That
...but not much serious programming.

How long did you think the game would take to make, and how long did it actually take?

I think it probably took about as long as I'd expected it would, a couple of months. But then, Ben There, Dan That! has a deliberately... uh, slapdash style.

How much of the development time was enjoyable, and how much was unpleasant?

Ben There, Dan That! was pretty much a laugh from start to finish, largely because I wasn't taking it seriously. 99% of the game design was done down the pub, and the vast majority of the rest of it was done over lazy weekends. It's a funny, stupid game. I think maybe the only unpleasant bit was writing all the dialogue - there's a unique reaction for most things in the game, and so sitting there churning out dialogue did start to become something of a chore.

How much did it cost you to develop, and what did that money go on?

Uh... I don't think I spent anything on BTDT. I can't really remember. The budget is so close to zero that for all intents and purposes it is zero.

How well has it done for you financially, on a scale from 1-10?

Ooh, tricky. On its own, probably a 3 or 4. BTDT was originally free (what was I thinking?) but it's now Pay What You Want, or available as a Double Pack with the sequel on Steam. I kind of see the two as a constant pair, now. From that point of view, definitely a 9. I could always use more money.

If you could give your past self one piece of advice before starting to make the game, what would it be?

HEY! IDIOT DAN FROM THE PAST. Make the graphics nice!

Ben insisted that the graphics should be deliberately awful, and I went along with it. It was a sound idea, technically, it gives the whole thing a rogueish South Park charm. In reality, I think all it does is turn people off who take one look at the screenshots and move on.

Conclusion

Adventure Game Studio is perfect if the cool thing about the game in your head is its story, characters or humour. If you want to make all-new game mechanics, you're better off with Game Maker. But if you're a writer or artist and you want to tell a story with as little coding work as possible, this is where to start.

Contents

RPG Maker
RPGM comes with art to get you started.

RPG Maker

What is it? An all-inclusive development suite primarily for making 2D RPGs. Includes some graphics sets to get you started.

Price and licence: £18 for the older version, £55 for the latest. You can sell what you make, no royalties. 30-day trial available.

Makes games for: PC

Link: http://www.rpgmakerweb.com/

Tutorial: RPG Maker Web's tutorial

To The Moon
If you can produce art like this, make a game.

Case Study: To the Moon

Developer: Kan Gao (Director/Designer)

Get it: from the official site

How long does RPG Maker take to learn?

6 months and 17 days for the average person between the IQ range of 115 – 140, with a catastrophically significant margin of error.

I think a few months should be enough to get the basic technical aspects down, without going into the optional scripting. It's much quicker than most tools of its nature, though it's still like having to work out just to hold a brush before you can start painting. The engine's got the foundation covered, such that anyone could just sit down and get a character sprite to run around on a map with a basic battle system; but to create your own systems and mechanics involves programming logic just like any language, albeit with much more simplified syntax.

It does actually have programming capabilities, which opens more doors than there are windows. And to be able to use it as well as to create something like this (which was indeed created with RPG Maker) takes much longer, and is a lot more in touch with traditional programming.

What prior knowledge or skills are helpful?

On the technical side, programming experience (the engine uses a scripting system based on Ruby) and a general grasp of logic.

Otherwise, every creative skill comes in handy – the engine actually has a rather active community, which is exciting because it also means that there's a lot of bartering for resources going on. If you've something to contribute like drawing and composing music, then not only can you cover those aspects of your game, but also exchange your services for others' contributions in other aspects as well. It's like Burning Man, but with less sand.

What can't you do with it?

As far as 2D stuff goes, there's actually not much of a limitation as far as the program's capability is concerned. You can create an entire set of new systems and make a quirky adventure game like Fleuret Blanc , or operate on the entire engine and create something like U.S.G. .

However, it's sometimes not very efficient (from both developing and computing perspectives) to make some of the things, relative to alternate engines and programming languages. The main technical restriction for me right now is the inability to port to Mac & Linux; but that might (and hopefully will) change in the future.

To The Moon
Get used to drawing every character from 4 directions.

How long did you think the game would take to make, and how long did it actually take?

21 months. We started in February 2010, and it was released on November 1st, 2011. But its initial 'expected' release date was April 2010. Though to be fair, the game was originally going to be just 30 minutes or so; but then I realized I was being an idiot.

How much of the development time was enjoyable, and how much was unpleasant?

It wasn't all prancing around a meadow, but I did purr a lot. A large positive factor was the variety of tasks to break down the monotony of “work”; it's nice to switch between different sets of things to do every day, whether it'd be music, dialogues, or struggling to tinker with the pixel art that some wonderfully talented folks were helping me with.

The unpleasant part was mainly personal rather than work-related. Whereas the blessing of the project was turning crappy things into something meaningful and productive, its curse was being forced to dwell on them throughout its duration. But still, making the game was definitely an 87.3% enjoyable experience.

If you could give your past self one piece of advice before starting to make the game, what would it be?

Plan out the progress in small segments, then spread them out over twice the time so you can actually follow it sustainably. Also, tomorrow's winning lottery number is 08 21 59 37.

Conclusion

RPG Maker is very well suited to a very particular kind of game: Japanese-style RPGs with turn-based combat and top-down exploration. You can certainly stray from that template, as To the Moon does, but the further away from it you go, the more sense it makes to use Game Maker instead.

Contents

Flixel

What is it? a starting point for making Flash games that handles a lot of the tricky bits for you, like physics and savegames

Price and licence: free, you can sell what you make, no royalties

Makes games for: any web browser with Flash

Link: http://flixel.org/

Tutorial: Making a basic platformer

Case Study: Canabalt

Developer: Adam Saltsman

Play it: free, on Adam's site

How long does Flixel take to learn?

This is a really hard question for me to answer objectively - not because of the Great Marketing Potential but more because it's really really hard for me to imagine not knowing how to use it, since it was all built out of my personal tastes, etc. I think if you picked it up and set out to make a like arcade game clone with no prior knowledge you could probably get it up and running in a week or two, without any prior coding experience. To really get flixel singing would take a few more weeks at least, but ultimately there is just not THAT much in it to learn, really.

What prior knowledge or skills are helpful?

The most helpful thing would be C-style programming experience and/or some object-oriented programming experience. Neither are strictly necessary but Flixel is definitely informed by these practices and my experience with them. If you know that stuff then the learning time for Flixel is more like 1 or 2 days.

What can't you do with it?

Making 3D games is more or less out of the question (outside of duplicating the raster math from really old fake 3D arcade games or something). It's not really cut out for some slices of Flash game domain too - there is no built-in support for movie clips, for instance, and doing large levels with zooming camera effects is actually possible now but not really thoroughly intentional in the library's design. Flixel is definitely strongest for games with a relatively fixed perspective and film-strip style art assets.

How long did you think the game would take to make, and how long did it actually take?

Approximately 5 days, and approximately 5 days.

How much of the development time was enjoyable?

It was basically all completely awesome. Mechanically it is such a simple game, there weren't many rabbit holes to go down. The least fun part of the process was a few hours sunk on semi-successfully searching the web for some free sounds to use as fodder for the things I couldn't Foley .

How much did it cost you to develop, and what did that money go on?

Approximately $200, but that was for a plane ticket to Phoenix to attend a little game jam thing some friends were having. In retrospect, that was super important, but I didn't realize that at the time. Usually I spend a lot of money on sound, but I'd pretty much used up all my war chest for personal games at that point in my life, so I did all the sound engineering myself, and Danny Baranowsky wrote the music for free as a personal favor. Thankfully Canabalt did well enough that I was able to repay him for his contribution!

How well has it done for you financially, on a scale from 1-10?

The Flash version was probably about a 2 or so. The iOS version is I think more like a 7 or 8 maybe. We did an Android Humble Bundle which was practically a 5 on its own. I'm not sure if I'm supposed to add those up or average them or what! I give it a 5.2382 overall score.

If you could give your past self one piece of advice before starting to make the game, what would it be?

That is the one thing I've ever made where I think I would have nothing at all to say. Except maybe "go ahead and pay Danny ahead of time, it will definitely all work out." But then GOD who knows what would have happened? DON'T MEDDLE WITH THE PAST.

Conclusion

Flash games are still one of the most accessible ways for a stranger to go from hearing about your game to playing it, and Flixel is likely the best way to make one. Flash doesn't work on iOS devices, though. It's worth knowing that Unity games can run in a browser with a plugin for Firefox, or natively in Chrome. And Game Maker can make HTML5 games, which run in modern browsers without plugins.

Contents

Unreal Development Kit
You can jump in and play very quickly.

Unreal Development Kit

What is it? An all-inclusive development suite for making 3D games in the Unreal Engine 3.

Price and licence: £60, no royalties on your first $50,000 in revenue, 25% royalty afterwards. Free for non-commercial use.

Makes games for: Almost everything – PC, Mac, iOS, Android, Xbox 360, PlayStation 3, Flash (web), WiiU, PS Vita.

Link: http://www.unrealengine.com/udk/

Tutorial: Epic Games official tutorial

Waves
Waves makes the most of UDK's flashy effects.

Case Study: Waves

Developer: Rob Hale

How long does UDK take to learn?

UDK is tricky because you aren't just getting an engine but an entire game framework. You're always building on top of how Epic like to make and structure their games, and this can take a long time to get used to.

The result of this means you'll spend the first few weeks finding out where everything is and how it works. It's almost always the case that if you want to do something then Epic have already done it, and it'll be in the code somewhere, but it's down to you to find it and figure out how to use it. That learning process will continue for years - I'm still finding stuff after working with the engine for over ten years now.

But you can be up and running and making something simple very quickly maybe within just a couple of hours and be relatively competent in a few weeks.

What prior knowledge or skills are helpful?

Any previous programming experience is very helpful, but be prepared to swallow your pride and do things the "Unreal Way". I've worked with a lot of very experienced programmers who wanted to rewrite every line of code in Unreal because they felt they could do it better, and every time the result was worse than how they started. This is Epic's engine and they've been making games with it for over a decade, so if it seems odd or counter to how you would do it, there is always a good reason behind it.

Other than that, you can pretty much learn as you go. When I started in games back in '01, I was modding Deus Ex and knew absolutely nothing, and I've taught myself everything I know along the way.

Any skills in making art, sounds, models or programming are going to be helpful and will get you working quicker but everything you need to use the engine is documented really well (far better than when I started) and the community is very helpful.

What can't you do with it?

Anything that involves changing the terrain in real time, and voxel worlds like minecraft. UDK is designed for very pretty static art-led worlds divided into levels, but that's pretty much the only hard limitation. The scripting language isn't very fast to execute, so anything involving thousands of objects moving around and interacting with each other every frame is also out. But there are ways you can get close to that with clever programming.

How long did you think the game would take to make, and how long did it actually take?

I thought Waves would take a couple of months to make originally, and if you asked me during development how long I thought I had left to work on it, I would always say "A couple of months".

That's got far less to do with the engine, though, and far more to do with my not having a firm idea on what I was making. Waves never had a design document, and was always a case of experimenting with different mechanics, and keeping what worked and ditching what didn't.

I think I was pushing back against my AAA background, where everything needs to be designed up front, and if an idea sucked, you often had no time to figure out how to improve it. Ultimately I worked on Waves for about 9 months in total, but I wasn't doing 40 hour weeks or anything, so the total time completely eludes me.

How much of the development time was enjoyable?

There were a few times when working on the game was a chore. Specifically getting the user interface and menus done was the most soul crushing part, because while you need all that stuff in order for people to play the game, it doesn't feel like you're working on the game. All of those indie game trailers that show the menu system exist because the developer spent weeks on those menus and by god they're going to get some of the limelight because nobody else cares about them.

I will say that if you don't enjoy solving problems and being faced with mysteries on a daily basis, then don't get into games development. The majority of the time at the end of a project is spent debugging, and that tends to be a case of 8 hours of headbanging frustration and swearing followed by a huge Dopamine surge and dancing for 3 minutes when you finally solve the problem.

Ultimately though I was making a game I wanted to make, how I wanted to make it, and for me that's the reason I get up in the afternoon.

Waves
3D engines tend to be better for fancy lighting, even for 2D games.

How much did it cost you to develop, and what did that money go on?

Living expenses and music. I had about £6k saved up when I quit my job and went full time, and I spent every penny of that finishing the game. I was lucky in that I had no dependents or mortgage to worry, about and my parents let me move into their spare room so I could save money while I got the game out. UDK only has an upfront fee of $99, but you don't need to pay anything until you're ready to release, and all of the tools to make a game are available for free - unless you really need Photoshop or Maya.

Hiring in any freelancers is always expensive, because there are very few that will work for a profit share. Generally the minimum hourly rate is around $15, those hours add up really quickly, and they need to be paid before the game is making any money. The only thing I couldn't do myself was the music, and I knew I wanted a good original soundtrack for the game, so I bit the bullet and paid for it. Fortunately that gamble paid off in the long run.

How well has it done for you financially, on a scale from 1-10?

I'd rate it as a 6. I can afford to make another game and I even have a little bit extra so I can pay some freelancers, but if I don't get another game released inside of a year then I'll be in trouble.

If you could give your past self one piece of advice before starting to make the game, what would it be?

Don't call it Waves! Nobody remembers the name and you can't find it on Google.

Conclusion

UDK is the thing to learn if you ultimately want to be making big, shiny, multi-platform games. It's harder to learn than Unity, and you'll need to be very versatile or form a team to make something that looks good. But it's also great experience if you want to join an established developer, since Unreal 3 is the most commonly used engine in mainstream games.

Contents

Around the web

by CPMStar (Sponsored) Free to play

Comments

highlights