Once upon a time, I made a game about a bulldozerlike space ship that fired missiles at boxy objects with glowing orbs in the center.
No, not that. That is a clone of said game.
There we are.
See the original game, the second game I ever made (after a game where you guide Larry the Cucumber across the screen as slushies fall…), and the first game that had any value (though not much), was made in about a week by yours truly when I was a kid (somewhere in the vague mists that lie between 13 and 18. EDIT: I was 17.). So I thought to myself “Self: why not put together a clone of said game? It will only take a couple of days, after all…”
- The original game kept track of three things: the current target, the player’s ship, and one missile. If the missile exploded or left the screen, you could fire it again.
- There is no AI, just randomly moving targets.
- There was little to no high-score business.
A really simple game, then.
I am trying to make a big fancy mecha combat game for my recently deceased (but hopefully soon-to-be-resurrected) game company. The graphics are entirely outside of my role: all I have to do is get a handful of mechs on a screen, in 3rd person view, and, well…
- 3rd person camera.
- Mech’s torso twist is limited within a certain angle of its legs movement.
- The ability to swap between multiple weapons, with a base interface or class so Rusty can code new weapons and particles at will.
- 4 mechs to start out with, with those mechs not controlled by players controlled by AI (press Start to Join)…
- Simple levels with obstacles to use for cover and to make maneuvering meaningful.
In addition to these goals, we wanted to add:
- Simple transformations to a folded-up vehicle mode.
- Eight player default, using XBox Live for multiplayer.
- Procedural levels.
- Dynamic Lighting.
- Normals Maps.
- Tons and tons of weapons.
- Multiple game modes (capture the flag, etc).
The goal is: get the first list done by September, try to get the second list done as well. Then sell the game.
Meanwhile, I went and got hitched.
Well… that was before the game design. But, you get the idea.
So I set up a game plan:
Plan point the first: Build a basic weapon and particle manager for Rusty to exploit. That way he could get working on stuff. Then, I dunno, try and get AIs and stuff working.
Around this time, I got my new desktop, and thought I’d throw ‘Dozer together in a couple of days at home, while I worked on the mech game on my breaks at work. And then my laptop died a horrible wasting death. But not before I learned some things:
This stuff takes time.
I have thrown a grand total of 20 hours or so into ‘Dozer 2. That’s taken me most of the month, less a week for the annual Nerd Posse Camping Trip. If I were working full time on games, it would be a couple of days at the rate I go. But I can’t: I have to eat. I won’t pull a J.K. Rowling and go onto welfare just so I can follow my dreams.
Now, if I manage to squeeze every hour I can out of my week, which I need to be doing to get this done, I can manage maybe four hours a day, five days a week, and perhaps seven or eight on Monday. So, 25 hours total. Except I can’t, because I need to do dishes and take out the trash and scrub the bathroom and catch up on laundry and at some point in all this mess, my wife would like to say hi to me. So, let’s assume 10-15 hours a week. Or: as much as I’d do in a day if I were doing this full time.
The solution offered by one of our dudes? Kickstarter. Problem: we have never finished a project as a group. I’m totally doing a kickstarter for each and every project I do in the future — at the very least, it will allow me to gauge how much interest there is in the project, and therefor how much I need to invest in the project. It’s like a focus group that pays you.
But the Internet is known for vaporware. And I, myself, have produced my share of gas. Fuzzimals, Legend of Anvor, ‘Dozer 2 (the first time), Total Breakdown, The Last Ember… many of these projects I intend to resume when I can, but the fact remains: promised, never delivered.
The big problem with these is they were too ambitious. Fuzzimals and Legend of Anvor were both RPGs. One man doing a JRPG is not impossible, but it really requires a team. And someone over the age of sixteen… When I talked them over with a mentor, he pointed out that normally projects like those have teams. Large teams.
Sonic 4, Episode 2 has a huge team, and it took a couple of years to get out. And they still haven’t figured out how to make a good Sonic game, despite the fact that they are looking in the right places.
Sonic 4 is nowhere near as complex as a JRPG.
Anyway, excuses for my past life of digital sin are beside the point. The point is here, on the blog of game designer (and political/theological irritant) Vox Day:
I’ve seen it happen too many times now, to too many good and smart people who simply fail to understand how difficult it is to build a successful game, let alone a successful game company. If you look at the successful ones, even the ones that seem to come from nowhere, they’ve almost always got a long track record of creating and completing a whole host of little games you’ve never heard of.
I don’t want to ask people to invest in a game company until I have successfully completed several games, so they know I’m not trying to sell vaporware. And until I start completing games, I am not a game designer, but just a producer of gas.
Dreamers dream dreams. Visionaries dream dreams and then inflict them upon reality. I wish to be a visionary. My aim is to become a wizard.
Keeping that in mind…
Dozer 2: What has been done:
- Models made. (easy)
- Moving characters on screen. (easy)
- Camera controls created. (easy, since I took them from a book)
- Missile class and physics made. (moderately easy, but tedious)
- Collision detection implemented (thought it would be easy. Turned out to be a major Pain in the…)
- Particle system for explosions implemented (easy but tedious).
- Target spawning and motion programmed (moderate, but tedious).
- Increasing difficulty implemented (so easy it was hardly work, since it was just adding hit points).
Dozer 2: what still needs to be done:
- Scoring system implemented (easy).
- Missile contrails implemented (easy, but tedius).
- Models textured (moderately easy, but tedious).
- Title screen, story screen, credits, and the ability to switch between them (very easy but very tedious).
- Losing (easy, but has to be done after the screen-change system).
- Input generalized to allow any controller to be used (moderately easy).
- Sounds and music (I don’t even know where to begin yet).
- A little more tinkering and polish (but not too much).
Rough estimate, based on time taken for the first bit? 35 hours. So, about 3 weeks. Pad it out for polish and unforeseen problems, and the game would take about a month and a half to make. So, I can expect to have Windozer done by the end of July, if I work on it.
Now, this article doesn’t exactly say what the average income of an XBox Live Indie game turns out to be, but averaging it out, multiplying by 1.5 to reflect declining sales over the year, and assuming it isn’t really bought after a year, then rounding down gives us a figure that, while seemingly pessimistic, is probably optimistic. And that is around $3,000. Given the utter simplicity of Dozer2, and the fact that I will have to sell it in the lowest ($1 bracket), I would probably go with this rule of thumb for cost/risk analasys:
$1 game: $1,500 (more purchases because it’s below the “this is significant money” threshold for most people)
$3 game: $3,000
$5 game: $4,000.
That means to make a decent salary of $40,000, I’d basically have to make 10 games a year in the $5 bracket. Since my ‘company’ has four guys, that’s 40 games, with me as the main programmer. Not counting company expenses and such.
Let’s assume Dozer2 does well. It’s in the $1 bracket, no doubt, so we’ll say I get a grand total of $1,500 profit from it. That’s $30/hour, so not bad at all, though that’s assuming I don’t give my fellow dreamer/visionaries any of the money. On the other hand, it’s five weeks to do one game. So basically, following the Selfish plan, I’d want to do 4-6 such games to build up the capital to go to part-time at my day job, then another 5-6 to have the head start needed to quit my day job.
Cut the profits four ways, and suddenly, we are talking three years before we go to part time at day jobs as a company. Not counting on Kickstarters. Rough estimates all.
Of course, our inspiration comes from Cthulhu Saves the World, and I Made a Game with Zombies In It (yes, I know that’s misspelled). But each of those games was wildly successful beyond all norms, the Harry Potters of the indie game set.
A model more like RadianGames is more likely. He released 7 games over 11 months, at the height releasing one game a month on the XBL Indie Market. The result? He couldn’t keep up with the bills and feed his family.
And there’s the crux of the matter. I need to pay bills. I have a family to feed. I am politically, emotionally, and religiously constrained to make it possible for my wife to stay home when the kids come, and she wants kids soon.
And my personality type is such that I cannot be effective unless I find a path I can define. So, harsh as the numbers are, the XBL Indie Market is a godsend for me. I just need to find that happy place where I can make a living at it and keep it growing.
I need to make this work. In the next year it has to function, in the next five or six it has to be outright successful, or else it will be stealing time and money from my peeps for no return.
Dozer 2, if I were a one-man operation, is in the aforementioned happy place, since I can build off of it in like manner to RadianGames. My financial needs are not as great as his at present, which gives me the potential to build a foundation upon which to grow.
As opposed to if I worked on the mech game…
- Base particle manager (moderate and tedious).
- Base weapon manager (hard).
- Two or three example particles (moderate).
- Two or three example weapons (moderate).
- 3rd Person Camera (already done).
- Levels (moderate and tedious)
- Animation/fancy model controls (hard and tedious).
- A.I. (Very hard and tedious).
- Controls (easy).
- Mech internal physics (moderate and tedious).
- Mode Swapper (mostly done already in my other projects, so easy enough).
- Physics. (hard times things that need them, including particles).
- Dynamic Lighting (hard and tedious, but needed to look halfway decent)
That gives me a guesstimate of 200 hours of work, or 15-ish weeks (closer to 10 if we get a kickstarter funded, but I have huge doubts). Not counting work done by artists and creating multiple weapons (Rusty’s job). In return, we can probably charge $3 for this four-player game as is. If we get it up to the online multiplayer version (I can’t even estimate the time for that…), so that gives us:
- 5 or 6 months total work, including polishing and stuff from Rusty and artists.
- $3,000 bucks back into the company, barring freakish success.
In other words, it is, in fact, not in the sweet spot.
How to get it into the sweet spot?
Well, I intend to consider that tomorrow morning, but my main thoughts are:
- Work our way up to it with multiple sweet-spot games.
- Create a much less awesome game along the same lines, sell that, and then do a Kickstarter for our planned mech game — I believe it would be funded if people had reason to expect it to be released, and relatively quickly.
As for the stripped-down game, one I think we could do in a couple of months (A third the time) is something a little closer to Zombies: top-down, as per our original design. Two weapons instead of customization. Basically, multiplayer ‘Dozer 2, with each other as targets. No levels or obstacles.
If we do multiple steps along the way, releasing sweet spot games (if, for example, Rusty does some simple/stupid little games at the same time), then we may have the rep needed to get a Kickstarter, and maybe, just maybe the mech game as currently planned onto XB Live Arcade rather than Indie, which would be huge for us.