iDevGames

Background Info

After growing up wanting only to make videogames, and making a few small games with Pascal and later Hypercard in grade school, I released my first game for the uDevGames contest in 2004, called Kill Dr. Coté. It won the award for Best Gameplay, was a fan favorite, went on to be published by Freeverse, and got me my first job in the industry as a programmer.

Since then, my output has been scarce. I quickly followed up with “Arachnoid: Predator of Worlds” in July 2005, but after that, working full time in the industry drained me and thwarted any progress on independent work. Several years passed, and during that time I first yearned for the energy to work on a project, and after that failed, I started to even doubt whether I would be capable of such a project and still keep myself fed. When I heard that uDevGames 2008 was starting, I figured it would be the best time to find out once and for all. While in 2004, my purpose was to prove my talent and ability to the world, this time around it would be to prove it to myself!

About Laserface Jones

Laserface Jones vs. Doomsday Odious is a dual-analog, omni-directional scrolling shooter. Setting-wise, it follows in the trend of science fiction B-movie style of my previous games. An alien known as Doomsday Odious is a planet-sized ameoboid that devours entire worlds whole, and it has set its sights on Earth. You play Laserface Jones, a robot that knows only how to destroy, and you must fight your way inside the beast and destroy it from the inside.

It’s definitely an old-school, arcade shooter. The difficulty is off the charts, but done so in such a way that teaches you how to beat it, not just to kick your ass and laugh at your suffering. There’s no doubt, though: it’s a quartermuncher, and it’s proud of it!

Laserface Jones draws inspiration from a lot of sources. I take a lot of joy in hiding little homages to other games and media. If you pay attention you’ll catch references to Gradius, Life Force, Contra, Axelay, LOST, Battlestar Galactica, and Emerson Lake & Palmer.

Flight of the Jones

There were four distinct phases to the development of Laserface Jones:

1. Idea selection
2. Gameplay Prototype
3. Level Editor
4. Content Creation

Idea Selection

When uDevGames 2008 started, my first task was to choose an idea. This was a bigger step than it would initially seem. I had three ideas that had been swirling around in my tempestuous mind for some time. The first was TANK, which was germinated on my podcast with Blake Buck and William Miller as a game we wanted to make, and never did. I do, however, have an entire design laid out, again, in my brain. The second idea was the dual-analog scrolling shooter. It was to take the winning mechanic of Kill Dr. Coté and address the main criticism, which was the lack of variety. The third idea was an experimental, abstract shooter. It would have been an experiment in adaptive difficulty, and the gameplay felt solid (in my head) but it lacked in ambition.

Narrowing down these three ideas to one was a difficult process. TANK was easily the favorite idea at the outset. It had style, it had frantic (but untested) gameplay, and it had a lot of potential. The problem, however, was that it required a LOT of art assets, which I knew from the get-go would be my weakness. That alone was enough to eliminate TANK as an option.

So it was between the shooter and the experiment. I knew that having competed before, plus what I was capable of then versus how much more I know now, that I could easily complete the shooter in three months. The problem, however, was that I also now had a full-time job to contend with. Would my ability be able to offset the loss of five days a week to work? I “decided” on both ideas several times, settling on one and then see-sawing to the other, over a period of about two weeks.

It was the formation of a title that cemented the shooter in place as my idea. The original title was “Gunkaiser” but a quick google search revealed that people were already using it as the name of a Gundam or something like that. So a quick brainstorming session turned “Gunkaiser” into “Gunface” and then to “Laserface.” A hop, skip, and a “Jones” later, and I had my character and my game.

Gameplay Prototype

This phase went remarkably fast, maybe about a week or two. I knew from Kill Dr. Coté that the mechanic of using WASD to move and the mouse to aim and shoot was a winner. I just needed to make sure I added to it, and provided enough content to keep the game fresh.

Level Editor

I knew off the bat I would need a level editor. I got away with having no level for awhile, and just having a constant rightward movement and automatically spawning enemies, but that was not to last.

The level editor is build into the game. Hitting TAB in a debug build brings up the Edit Mode, where you can click to place obstacles and enemy formations, and drag them around.

The editor took about two weeks to create to the point where it was functional enough to make levels, and another few weeks afterwards to polish up to the point where it was 100% satisfactory. So about a month altogether.

In addition to the editor, the game can read in .svg files that define the “path” of the level. I’m super proud of this, as it means I could draw the levels out in Illustrator. Something didn’t work right, or a turn was too sharp (or not sharp enough?) Easy. Open it up in Illustrator, tweak a few control points, and resave. This also allowed me to very easily do crazy things with the scrolling, as opposed to straight right or straight down. For instance, the first level has a section that waves back and forth as you navigate the rocky shell of Doomsday Odious, and level two actually does a loop-de-loop.

Content Creation

This phase was the final part of the dev cycle, and was the most rushed. After dilly-dallying with a few in-engine features that I ended up cutting out of the game completely (enemies with build in obstacles, obstacles that would rotate in-game) it became imperative that I actually make the damn game! I had a functional-but-bad first level that needed to be redone, and levels two and three didn’t even exist yet!

Thankfully, the Level Editor made it a lightning fast process. After trashing what I had done on the first level, recreating each of the three levels took about 48 hours, not including the time required to create new enemies to populate those levels. The first level took the most time in that regard, requiring a few weeks to get all enemy types in. The second level was much faster, and the third level just reused enemies from the first two levels. (If you couldn’t tell that I was strapped for time by the third level, you may also notice there’s no crazy curve in level three, just a straight line!)

What Went Right

Accepting help from others

If you take any advice from this postmortem at all, let it be this. ACCEPT HELP FROM OTHER PEOPLE. When it comes to game design, I have a monster ego. I very much subscribe to the “auteur” school of design, and believe that games should have a director, much like film. For Kill Dr. Coté, everything in the game was done by me and me alone. The problem is, I’m not the best musician in the world, and my art blows.

After mirroring my Developer Diary on my personal blog, I got an email from Ben Coello, who was a fan of Kill Dr. Coté back in the day, and we had talked game design back and forth for awhile. This time around, he attached a concept sketch of Laserface Jones, and it was absolutely amazing work. I immediately asked him if he wanted to jump on board as character artist, and he agreed. He also suggested his friend Dan to me, who is a sound designer and ended up providing much better gun sounds for the game. I later got help from Bruce Morrison and Steve Tze, who provided some level art and enemies, respectively.

Aside from that, I realized quickly that I would not have time to compose all the music for the game. Out of six pieces, I composed four, but had to license the final two tracks. I was pleasantly surprised that my own work held its own against the licensed stuff, and they actually complemented each other very well.

No burnout

I’m not sure exactly why, but my patterns of burnout for this project did not match previous projects. For uDevGames 2004, I burned out for a period of about three weeks, during which absolutely nothing got done. For 21 Days Later a three-week contest, I burned out completely during the entire second week. From that I concluded that the middle third of the cycle was burnout time and tried to prepare myself for it. But it just never happened.

There were periods of burnout, where I felt like I was sick of the project, and couldn’t work on it anymore, then decided to rest, then it became imperative I get back to work, but from looking at my entries on the Developer Diary, each entire burnout cycle happened over a period of 24-48 hours. I am not sure exactly what the reason was for this.

One theory was the frequency with which I alternated tasks during development. I shifted often between coding, art, and music, which kept the project fresh. Music especially was rejuvenating. Composing the music was arguably the most fun I had during the three months. I put early game music from Konami and Capcom on my iPod and made it my commute music, studied the HELL out of it, and tried my best to emulate the sound and orchestration. Even just banging around on a MIDI keyboard for a night with no results was enough to change gears and keep things fresh.

Timely triage

In the beginning of the contest, I wrote a uDevGames Survival Guide in which I stressed that ideas must be “scalable”, that is, if you are ahead of or behind schedule, how easy it is to “scale” your project to match. I had to scale Laserface Jones down several times during the contest, and each time it was a breath of fresh air, and took very little away from the final product.

I knew right off that I would only have time for three levels, but in the beginning there were to be tons of guns. I wanted DOZENS. Each one would level up and have an “Overload” which was like a supercharged alt fire that used up stored gun experience. Dozens quickly became five: the original three plus a machinegun weapon and a melee-range blade. Over time, I realized that the machinegun was redundant, and the blade was just unnecessary. The Overload concept itself was also pruned away later, just on the basis that it wasn’t a crucial, high-risk game element.

Additionally, there were ideas for tentacles in the game, which used verlet integration for the segments so they could slither around, and could also be severed at any point. They would reach in from holes in the walls, and enemies could also have any number of tentacles. This simply ended up being too much.

The best part though is now all this stuff can be put back on the plate for later versions of Laserface Jones. And the more that version has over this free version, the more viable and enticing it becomes as a commercial product.

What Went Wrong

Waffling on an idea for WEEKS

I used a lot of time deciding on an idea. Frankly, this could have been decided on prior to the contest. I went back and forth between ideas almost half a dozen times, and it cost me valuable time.

Trying to make an ambitious game while keeping a full-time job

This time around, compared to 2004, I knew I would have significantly less spare time due to working full-time. I work for Freeverse in New York, as a game designer and programmer. While most people will look at that and quickly say it’s an unfair advantage, I assure you it was not completely so. While yes, working full time doing programming and game design kept me sharp and mentally “in-shape” I also resolved early on that Laserface Jones must be done on my own machine, on my own time. In the beginning, I figured I would just work on the weekends, and scheduled out tasks for the 12-13 weeks that I had. However, as the project picked up steam, it quickly ballooned to fill up every last bit of free time I had.

To give you an idea of my routine during the height of the contest:

• 8-8:30AM: Wake up for work. Much design was done in a Moleskine
• Notebook (called the “Assbook” in reference to its storage location in my back pocket) during the subway ride to work.
• 8-8:30 PM: Get home from work. Again, design was done during the subway ride home in Assbook
• 9:00 PM: Eat dinner
• 9:30 PM to 4AM: LASERFACE JONES.

Rinse, repeat, for three months. That said, it paid off in the end, but it took everything out of me. By the end I was counting down the days until the end of the contest. Anything longer than three months and I would have simply exploded.

The Future of Laserface Jones

The future looks bright for Laserface Jones. It’s got fans that are begging to see it on Xbox Live Arcade, or the iPhone. I have experience creating for XBLA (from working on the port of Marathon) and agree that’s where the fans of this type of game are. XNA is a similar option, with the advantage of not having to go through the greenlighting or certification process of XBLA, but with the disadvantage of requiring a complete code port to C#.

The iPhone is another very likely possibility. The market is huge, and many hardcore gamers own iPhones (even though it isn’t their primary gaming device.) And not to mention, the code is very similar to Mac. The porting is trivial.

As for the game itself, I have always intended Laserface Jones vs. Doomsday Odious to be free. It is a sort of short “pilot” episode to see how interested people would be in more. After the reaction it has gotten, it would be a shame to not follow it up. I’m already laying out the groundwork for “Laserface Jones vs. The Bullet Legion” which will be a commercial title, and I hope to turn it into a series, similar to a comic book or television show (as opposed to the “movie trilogy” approach many games seem to take toward episodic content.)

Stay tuned, because the world hasn’t seen the last of Laserface Jones!

• Developer: Justin Ficarrotta
• Title: Laserface Jones vs. Doomsday Odious
• Genre: Action
• Development Team: Justin Ficarrotta (Game Design & Programming, Level Art, Music), Ben Coello (Character Design & Art), Dan Mazur (Sound creation), Bruce Morrison (Level Art) & Steve Tze (Level Art)
• Development Cycle: 3 months
• Source Code & APIs: Xcode (C++), OpenGL, SDL (+ SDL_mixer and SDL_image)
• Critical Software: Adobe Photoshop & Adobe Illustrator, Cheetah3D, Garageband, Audacity
• Hardware Used: MacBook Pro 2.4 GHz Intel Core 2 Duo, M-Audio Axiom-61 Keyboard, Korg Kaossilator, Korg DS-10 Synthesizer

Further Resources

• FIDRIS Postmortem
• Snowball
• Boston - Mouse in the Sewer Postmortem
• Magic Stones Postmortem
• Kiki the Nanobot