Onboarding and the 4 Fs rule

I recently took part in the Juniper very serious game jam and out of the 3510 games, who knows how, I ended up #1 in the Enjoyment category and #36 overall. I also had the pleasure of winning the Best Web Game prize awarded by CrazyGames, who were sponsoring the jam.

All of it with a slightly janky game that I had a lot of fun developing: Rocky rocket

Rocky rocket is a physics-based 2D platformer where you control a little character (an astronaut or robot, if you use your imagination). It’s made of two distinct parts, a head (the white ball) that slides without friction and a body (the red square, yeah okay I did say imagination) that, for its part, sticks to everything.

I’d wanted to develop what is, to me, the spiritual sequel to Six Bit for a long time, but without the guns.

In several Six Bit levels, I liked the idea that you could move around by sliding on oil puddles. That’s what gave me the idea for this dual controller (sliding/sticking).

For movement, I replaced the shooting with two new mechanics:

• the plunge, or the ability to make yourself heavier, to speed up the fall and gain speed.
• a jetpack, for more mobility.

I had such a clear idea of the game when I started the game jam that I quickly ended up with two or three working levels. Simple levels, to understand the game’s mechanics. I exported a version of the prototype and had the people around me test it to get some first feedback.

Between the observations during the playtest and the feedback I got, the verdict was simple: “uh yeah why not, but mostly it’s confusing and hard.”

It’s a bit of a cold shower, players spend more time stuck in a dip trying to get out of it than doing the nice smooth runs I was expecting. And so, well, it just doesn’t really seem fun to play.

Onboarding

Okay, since I’d set up the prototype pretty fast, I still had a lot of time to fix things.

And the first problem I had to solve was making players understand the movement options. Then again, I’d never taught them to them. In that first playtest, nobody used the controller physics or the plunge mechanic (making yourself heavier). Players focused only on the jetpack, and since it’s limited in how much you can use it, it made the game choppy and painful.

So I made new levels as a tutorial. Super simple, so as not to muddy the waters and to control what I wanted to teach the player. I removed all the advanced mechanics, so you could get familiar with the basics of the game.

In the first level for example, there’s nothing to do other than slide. Worst case, you’ll slow down if you don’t stay on the head and that’s all the better, it’ll explain in advance what’s expected of us in level 2: Using the character’s body to move around.

But, a new “problem” just emerged. It’s actually more fun without the jetpack.

The 4 Fs

The 4 Fs stands for Fail fast & follow the fun, and to understand the concept a bit more if you don’t know it, Game Maker’s Toolkit talks about it really well in one of their videos: The games that designed themselves

Okay so, for the fail, that’s fine, it was indeed pretty fast. All that’s left is to follow the fun. And there, really on these first levels, the controller comes into its own. I find myself spending way too long optimizing my route, doing controlled jumps, discovering fun strategies I hadn’t thought of for a second.

And so I spend the next 2 days creating a dozen levels without a jetpack and often even without the plunge, which on top of being fun, really let you understand the logic of the game and learn, level after level, to handle the controller better.

In the end, the onboarding is no longer a simple explanation tutorial, but a real part of my game, just as fun as when you end up with all the mechanics.

Once it was done, I export the game one more time, I send it to my regulars. And then, no more replies. But! After a few hours, I look at the leaderboard on each level, and I see I’m not first anywhere anymore. 🥲 I guess they understood the game better.

Those first “simple” levels that hadn’t been a hit will just be the hardest ones in the final version of the game. I didn’t even modify them. It’s just the context around them that changed and made them interesting. It also let me avoid falling into the classic trap of developing a game that’s too hard, for people who have 5min to spare for me, while they test 50 other games back to back.

In Juniper’s YouTube video, when she presents my game, she actually talks about the jetpack as a light element, added in the last levels to bring a little twist. It’s wild to think, in hindsight, that it was the heart of the game at the start.

If you want to try the game it’s here:

Rocky rocket by godDonutSpin, wiggle, plunge, jetpack your way through all the levels in this physics 2D platformeritch.io

And if you want to discover my favorite games from this game jam, I just made an article too about my favorite games from the Juniper very serious game jam. That way you can make up your own mind about how much I should or shouldn’t be first in enjoyment.