Indie game news, reviews, previews and everything else concerning indie game development.


The Game-Maker Archive — Part 14: Laser Light

gireader My association with Recreational Software Designs started early, maybe around the time of my first game. I don’t remember the circumstances. Maybe I wrote in with some suggestions. Maybe I was trying to show off my work. Whatever my motivation, I was fourteen and unhampered with caution or tact. I mailed a letter and maybe a 3.5” diskette, and then forgot about it. Weeks later, the phone rang. Against my normal habits, I picked up. The voice, which asked for me by name, sounded uncannily like one of my friends. Being fourteen and tactless, I told the voice that it was an idiot. The voice was confused. I unleashed more rudeness. The exchange continued until the voice identified itself as the president of RSD, a certain Oliver Stone. Tickled with the oddness of the situation, I laughed for a minute or more.

I’m not sure why he stayed on the phone, or indeed continued contact with me. Eventually we developed a rapport. He would mail me pre-release versions of new Game-Maker updates; I would scour them for bugs and inconsistencies. I would mail in my newest creations; he would introduce me to other Game-Maker users and show me their work. This went on for a few years.

For the 3.0 release of Game-Maker, RSD chose to transition from floppies to CD-ROM. In 1995, this was a big step. It was like having a book or an album published. Within a year AOL mailers and demo discs would render the CD common; in 1995, it was still a magical endless data well. So RSD now had a whole CD to fill, and to justify the leap they needed to fill it.

I was prolific, and able to hide my ineptitude behind polish and an intimate understanding of the game engine, so evidently I was just what RSD needed. They contracted me to design six games, and to sign over another two. My rudeness persisted; when asked to contribute, my first impulse was to toss them a couple of my least favorite games. It was only with later discussion that I twigged their desire for new, flashy, and instructive content. With that goal in mind, a certain inspiration struck me. I progressed at about a game a week. Some of the games served to demonstrate certain design concepts; others spun themselves out of a whim.

At reader request, here are those six games, in the rough order of development. I’ll hold off on the overt criticism, and instead try my best to explain what was going through my head. We’ll just see if a sensible train of thought develops.

Glubada Pond

glubada The best I can do here is reconstruct my motivation. It’s clear that I was inspired both by Taito’s Bubble Bobble and by Novotrade’s Ecco the Dolphin. It’s also clear that I was both fascinated with monster mechanics and eager to bend Game-Maker toward different goals and play structures, beyond the standard inventory-based action-adventure games.

The usual Game-Maker structure involves finding power-ups and defeating monsters as you travel a map in search of an end point. Here, I tried a more classical arcade structure. Instead of searching for a destination, how about we clear the level of enemies to move on. Sounds simple enough, right? The idea goes all the way back to Space Invaders — or Breakout, if you want to get philosophical. I also liked the hop-’n-bop structure of games like Mario Bros. or Tumblepop, where you disable enemies before knocking them out for prizes.

As usual, my ambitions led to wrangling with the engine’s eccentricities. And as usual the wrinkles that I could never quite smooth out determined the game’s identifying quirks. Limits in character idle sequences meant that a character couldn’t just stay put when done moving, so I had the my fish face the audience and wiggle back and forth. The end result is odd and a little creepy, but certainly memorable.

Since the only way for a player to progress was to touch a designated exit tile, I couldn’t directly tie success to monster deaths. My solution was for each monster to leave behind a tiny bubble; collect all the bubbles and insert them in a vending machine (I don’t know; I think I was out of ideas), and the machine would open, allowing access to the next level. A problem was in the power levels of enemies.

Any item left over from a monster death would also, technically, be a monster; it would just be a monster with positive rather than negative qualities. If the monster had a lower power level than the character, it would die on contact, passing to the player its positive qualities — such as increasing a counter. I didn’t want to make every monster of a lower power level, or else the player could simply ram them to defeat them; the point was to shoot bubbles at them to disable them. Yet if one of these higher-level monsters touched the reward bubbles, it would defeat those bubbles and cause them to disappear.

This was a dangerous situation. If there were only so many monsters, and thus only so many reward bubbles, what would happen if some of those bubbles vanished before the player could collect them? Basically, the player would be stuck. One solution might be to overload the level with monsters, or even allow them to respawn, and only ask for so-many bubbles to progress. That isn’t ideal either, as Game-Maker has no option to reset counters either on character death or on leaving an area. So if you were to die, or rack up bubbles in an early level, you would build up a backlog that you could trade in later to zoom right through the levels.

I never really worked out the problems, so in that respect the game is flawed. It is possible both to get stuck without bubbles and to mine bubbles for later. Despite the inelegance, the game finds its own flow and basically works. The faults almost open up a strategic element. It’s a strange game, though.

Crullo: Adventures of a Donut

crullo The premise here is pure whimsy. After Glubada Pond, I figured I’d go with a more traditional platform adventure. Instead of a fish shooting bubbles, I had a doughnut shooting raspberry jelly. I didn’t think really hard about this.

Once I settled on a theme, the rest of the game was me screwing around with tools. This may be the first game where I did all the visuals in Deluxe Paint, and I was anything but subtle about it. I just created piles of random blocks decorated with gradient fills. Likewise for the sound I pulled out an old Radio Shack keyboard that had been gathering dust since the late ’80s. Whether or not the notion was appropriate to a game about a doughnut, I figured all the sounds would be musical, or at least synthesized.

The game, then, has a strange atmosphere. The sound effects give it a cold, mournful, and sterile sound. The visuals are noisy and hard to differentiate. The only thing in keeping with the theme is the monsters; for foes I littered the levels with more savory bakery items: bagels, croissants, English muffins. I’m not sure what they were up to, and I didn’t bother to give them much personality or behavior. They were just sort of there, as obstacles.

I put just as much effort into the level design. I chose a block set, settled on a starting and an end point, and drew random, winding tunnels and passages and rooms to connect the two. I threw in the odd secret passage or geographical feature, but I never really made sure the geometry matched the character’s movements and abilities. I figured it was possible to progress, it was fine.


zark Whereas Glubada Pond gets caught up in mechanics, and Crullo gets caught up in the design tools, Zark gets caught up in genre. After Crullo, I wanted to stretch the boundaries again. Game-Maker really wasn’t made for shooters, especially scrolling space shooters, but I was determined.

Based on some earlier experience, I knew what really didn’t work; I just wasn’t sure what did. I figured that the only way to create a constant scroll was to ensure that the player’s ship always moved right. Even when backing up, it would move at a lower speed than when moving forward. It did little good to prevent scrolling back to the left, so I let the player flit around and explore at will.

I also had poor experience with weapon pick-ups. With Game-Maker you can’t just exchange one weapon type for another on the fly; you can only upgrade from one to the next. That is, if you load all the weapons onto the same key — which I intended to do. So instead, I chose a numbered system. Each number represented the number of shots the ship would emit at a time. They would spread in various directions and patterns, depending on the number. This worked out well, except in that fast-moving monsters tend to skip across the screen rather than moving smoothly. Any shots are classified as monsters. Thus any fast-moving shots have a good chance of skipping past a target even if you shoot it head-on.

This may be the first time that I experimented with large, multi-block monsters. When the player destroyed a weak point, it would unleash a high-power “explosion” monster that would swirl around and destroy all the other boss parts, which would themselves unleash swirling explosion monsters. It worked pretty well, if you could ignore the occasional engine bug that would cause an explosion to randomly spawn a boss segment. Not sure why that happened.

Peach the Lobster

peach Back to the familiar action-platformer. I figured that RSD deserved its own mascot, to help give the company an identity. My solution: throw a lobster in a track suit, and rip off the general design of another mascot game.

By now I was comfortable with importing graphics from Deluxe Paint, and indeed a bit more skilled at it. I even managed to pay some attention to the theme and storyline. Yet I was just as distracted by the process as I was on Crullo, and so made some strange errors of judgment. All of the monsters are one block tall, and Peach is two blocks tall. His only attack is with a claw gun, which shoots from above the waist. This means that, all things being equal, there is no way to hit the enemies. Oh, if you screwed around and fought with the game you could eventually kill them. It was just a nuisance to do.

Again when designing the levels I paid little attention to the character’s abilities. If a jump required the player to hammer on the jump keys and glitch out the game, that was fine — so long as it was possible in the end. If it was possible to avoid an enemy, even if it was nearly impossible to do so, then that was all I asked.

I’m not sure what happened to Peach’s other six limbs. Maybe they’re under his clothes?

The Patchwork Heart

heart Peach was an exhausting project, and it took much longer than I expected. I cooled off by tossing together what I felt was a simple, brainless game. It consists of three maps, one tile set, a character with basically no animation, and no grand plan behind its design.

The character is a golden orb; for variety, and just to dink around with a technique I hadn’t used before, I had it emit motion lines when it jumped. The result was a kind of neat trail effect, which also served as a secondary attack. To set the levels apart, I played around with palette swapping. To further break up the sameness, which I felt made the game confusing in places, I made it so every surface touched would also turn golden. This allowed the player to sort of leave a trail, as well as generally establish a sense of ownership over the level geometry.

I borrowed monsters from Zark; since they were mostly body parts, I had them drop pools of blood when defeated. Thinking back to a secret from Wolfenstein 3D, I had the pools restore energy whenever the player touched them. With the sharing of resources, I gave the game a tenuous story connection to Zark.

Aside from a weird super-jump gimmick, that right there was the game. In retrospect, despite the lack of effort, it’s probably the most playable and progressive of the bunch.

Clyde & Zeke

ducks The deadline was quickly approaching, and I wanted to get in one last game. I figured I’d do a simple demo to show how monsters could also be used as helpful partners. I looked out the back window of my parents’ house, at the lake. The lake was full of ducks. There was a theme: one duck following another duck. Why not.

I found a photograph of a duck, and shrank it down to 20×20 pixels. It was going to be swimming, so it didn’t need much animation. I whipped up some actually pretty decent marsh tiles and imported them from Deluxe Paint. I created a single adversary — a paper boat — and a single pick-up — a school of fish, which restored health. I then quickly threw together a maze level. Find your way from A to B; use your AI partner to protect you. That’s it.

The game only took a day or two to finish, and I sent it off. I know that RSD received it before finalizing the CD, but it arrived late enough that it seemed to skip their minds. Although I got paid for it, it never actually wound up on the disc. When I asked why not, they seemed as surprised as I. No great loss, but it is sort of curious.

The 3.0 release of Game-Maker was both the most influential and the final one. For another year I continued to pester RSD with suggestions for further updates, but the programmers moved on and soon after RSD ceased to exist.

[Read all of our Game-Maker Archive editorials]