Daniel Benmergui madeToday I Die. Messhof made allthis stuff. When these indies talk about game design, you’re able to probably bet they’ve got something interesting to say.
Both designers organized an IGS talk entitled, “Control Inspiration” which, according to the official proposal, is about “their design processes, and why Jesus didn’t tap.”
Essen opened by talking about his own personal history of game design – he started out making clones ofJoustandCrystal Questbefore moving onto more abstract games likeBoolandBooloid. The designs of each game shared a common mechanical simplicity that continues to characterize Essen’s current work.
Essen showed a “not very fun” old prototype ofFlywrenchwith “too much crap…there’s a lot of fat to be cut.” The game was too hard and not fun enough, in Essen’s opinion. He put the game aside for a few years and started watching short films likeInterkosmosfor inspiration.
Interkosmosmade him think about controls and how weird the keyboard is, and whether or not the keyboard is really the best way of allowing players to control games. He showed a video of a really complex, really simulation-y chair/controller/interface made for helicopter games. “I can’t decide if this is the way to go,” Essen admitted, or whether it’s better to stick with the keyboard, dealing with the simplest form of control and mapping more complex mechanics onto that.
Robots. Essen thought about robots, at one point. You program robots and let them figure things out via trial and error, which is kind of like how games work via their instructions. This was illustrated by a video of a starfish-shaped robot trying to learn how to walk first with all of its legs functioning, then with certain legs disabled, and so on.
He showed a robot trying desperately to get up, trying and failing until it finally turned itself over and began to move. “Imagine that’s a game, and this is just you dying over and over again.”
Coming from this new perspective, Essen went back toFlywrenchand heavily simplified it, making it the highly enjoyable, satisfyingly punishing game it is today. Even after simplifying it and making it more enjoyable, however, Essen felt it was important to keep some element of randomness and total junk (henceFlywrench‘s discordant soundtrack).
Essen’s interest in randomness extends to visuals, as well – he experimented using weird textural feedback loops to create harsh but random (and kinda cool) visual effects, but they were too intense to use on a large scale. He then tried using it as an explosion texture before deciding to mute the feedback intensity and use it for all ofJetpack Basketball‘s graphics. “The fat is trimmed,” Essen stated, “but it’s all still noisy and messy.” To him, this was a good thing.
Benmergui didn’t really know what this talk was supposed to be about, but he assumed it was about creativity in some way. With that in mind, he talked about how painful it was to makeToday I Die. Not because he’s a bleeding heart, who feels some sort of artistic torment, but because he didn’t know what the hell he was doing.
Why make the game? Benmergui wanted to make a game about how hard it is to get up in the morning. You have to transfer between a stage of sleep and awakeness, and how you don’t want to, because you know what you’re going to have to face throughout the rest of the day. He then had an image of a girl sinking in stale water, before struggling back up into shining water and conquering her fears.I Wish I Were the Moongave him the confidence to makeToday I Die, sinceMoonwas similarly metaphor-laden and players seemed to like it.
After playing a little poem-builder program made by someone else, Benmergui thought he could use the interactive poem idea asToday I Die‘s driving force, as a means of expressing the character’s worldview. The next version ofTiDallowed the player to mix and match key words within a poem. Putting one word in a different place would chagne the word entirely: “die,” when moved to the front of the poem, became “defeat.”
This idea failed, howevver, because you had no idea what would happen when you switched words – it felt arbitrary and random.
“I started freaking out,” Benmergui said, because he’d already done a shitload of work and was now worried he’d have to abandon it. Which is exactly what he did, at least for a little while.
The poem thing definitely wasn’t working, so he just gave up and started making a tactical space shooter. This got a laugh. It was okay, Benmergui said, “but I started feeling like this was wrong.” This also got a laugh, for some odd reason.
Benmergui initially felt that when you’re stuck in a project, you assume that the next project will be much easier. That keeps you going. In reality, the “next” project is never easier to make. According to Benmergui, “if you have to solve difficult problems, you better solve them for games you feel are important.”
Benmergui changed his mind and decided he wouldn’t give up on TiD completely, but would attempt to circumvent the problem by removing the poem mechanic completely. The next version of the game simply involved a jellyfish who, in response to the player’s mouse movements, could shine and swim and avoid mines as he swam upward.
“It was okay,” he said, “but this was wrong.”
The word-switching mechanic Bwas interesting, Benmergui realized, but he couldn’t make it work. He clung onto it and got burned, but then he learned that sometimes your ideas are bigger than yourself and you have to let them go.
As a result, the final version ofTiDdoesn’t allow palyers to switch words around within the poem – you can only move words from outside of the poem back into the poem, which is much easier to understand and feels less arbitrary.
He came up with the idea for another game,Today I Die Again,which isn’t a sequel so much as a revision. He wanted to fix the problems of the previous version, like how people got confused about what certain words were meant to do.
At this point, ti was time for audience questions.
How long didTiDtake to develop? Six months.
Early in the talk, Essen mentioned that the old version ofFlywrenchhad “no soul,” and just felt like a series of challenges. One audience member asked exactly what gives the newer version ofFlywrenchthe “soul” its predecessor was lacking.
The original, Essen said, was just like a bad flash game with distinct levels and score ratings. After completing a level, you ask yourself why you’re still playing – “it’s not a compelling challenge.” Conversely, it’s easier to lose yourself in the new version ofFlywrench, because it’s got its own aesthetic style and the challenges are more interesting.
The final audience question came from a dude who wanted to know why Daniel Benmergui’s hair looked the way it did. Benmergui responded that he only answers that question once every six months, and he’d already answered it a month ago.
