Dunk Dunk devs on their 7-year journey from hobby project to published game

In this guest post, BadgerHammer’s Nick Gripton and Nick Holder share tips, stories, and advice for hobbyist developers on how to turn your passion project into a published game.

You've all seen these kinds of comments or reviews for indies… “Seven years to make this heap of junk?!” – that kind of angry thing.

Well, don't let it bug you. Making games takes tons of time, and finding those hours on a hobby project can be hard sometimes. You might wonder why you’re even doing gamedev in your free time? Gamedev isn't a hobby. Gamedev’s a job!

Dunk Dunk took us seven years of development time from prototype to final release. Here are just a few of the reasons it took us sooooo long.

#1 - Life eats time

We run our hobby studio (and we’re proud to use the word hobby) in our spare time.

BadgerHammer is just two people: one dev and one artist. We’ve got ZERO budget. We live in different cities, and for most of the time on this game we were in different time zones.

The best place to start is to demonstrate how much time we’ve actually got available to put into this so-called hobby.

There are 168 hours in a week. What a lot of hours!

• You’ve got your day job: 40 hours a week.
• Chores, taking the kids to school, doing the dishes, tidying up, etc.: 25 hours.
• Spending time with family and friends: 25 hours.
• Relaxing, maybe exercising if you can fit it in: 14 hours.

It’s already kinda depressing at this point – you’re spending about three times the amount of time you have “working” than you do hanging with your loved ones. Good thing you love gamedev, right? It only leaves about 64 hours a week for your hobby. Oh, except you also need to:

• Eat and sleep: 60 hours.

For us, it basically boils down to about four hours each a week to spend on our gamedev hobby.

Great! So now we know how little time we have. And this doesn't even take into account weird turns of life, like having a baby, or moving your family from one side of the world to the other.

#2 - Technology moves faster than hobby devs

We used Unity to create Dunk Dunk, and Esoteric Spine for our character animations. Both of these tools saw big changes over the seven years of Dunk Dunk’s development. Most were positive changes, but often replaced old functionality or implementations leading to extra work outside of making a game fun (or having fun making a game). One upgrade will often then require another package to be upgraded. Where possible, we opted for stability.

Upgrading across versions of Unity can take time. Improvements will leave older code behind, which will need replacing. We always used LTS (Long Term Support) versions so we could reduce these painful transitions until we saw real benefits or had to meet a minimum spec for a platform. Dunk Dunk was using the 2021 LTS at launch.

Spine got loads better (we’ve been using it since beta), but upgrading any library can take time, and it can cause issues. For the most part we resisted upgrades.

Working on a project for so long means you can see the history of our Unity experience a bit like the rings in a tree (this was our first Unity game; previously we used Haxe/OpenFL/Flash). The most obvious rings in the dev tree are how we approached messaging in the game over time. For example, we used delegates for messaging events in gameplay in one place, and called a function on instances in another, and poll shared game data variables in others still. There’s no one way everything is structured. We had to constantly resist re-engineering what wasn’t pretty, but worked, in order to keep things moving.

Another change that happened during development was gaining the ability to nest prefabs inside other prefabs. This was a real win and one we were happy to upgrade for, but a bit late in the day for some of our UI, and caught us out with localization of UI elements that were added before that update.

#3 – Features are always creepy

Something we both learned early in our careers was, “It takes 10% of the time to get 90% of the project done.” It nearly always holds true.

Dunk Dunk went through three complete art changes. The first was for usability reasons: It wasn’t easy for the players to tell where the ball was going to connect on our original tall characters, so we made them small and chunky. Suddenly it played right, but then people struggled to understand which team they were on, or even which player they were. NickG had done some illustrations for a music video that got canceled and wondered if the game could look like that. It worked, but it meant redoing everything (again).

We also remapped our controls. Our original concept for Dunk Dunk used a twin stick system, because we’d always loved the immersion of games like Fight Night and Skate on Xbox 360. But when we demoed our vertical slice at Rezzed, most people just didn’t get it. NickH spent the time it took for pizza to arrive in a restaurant on the first night (three hours, no joke) to rewrite the system to use simple, platformer-style controls. The next day it was an instant hit. The legacy control system still remains though (we call it Pro Controls).

Demoing is one of our favourite things to do. There’s something about the energy of people enjoying something you’ve created. But demoing (unless you do it at your local indie meets) comes with a cost in both time and money. Did we say zero budget? We meant sub-zero. Demoing always comes with feedback which has to be processed, filtered and possibly actioned. There are only so many people you can please, and one of those people has to be you – the world’s pickiest client.

We decided that it wasn’t enough to just create a local multiplayer experience and made the choice to create a bunch of single-player content. Some really great things came out as a side effect, like CPU players and modifiers (or “mutators,” as we call them).

We went down blind alleys of things that were interesting but impractical given where we were in development. Online multiplayer, which would be great for the game, was a nine-month struggle that never made it to release, as the game wasn’t built with that in mind. We approached this twice, once on our own (across the world with terrible internet on one end), and again when a publisher was only interested in Dunk Dunk if it came with online multiplayer. We got something working, but the latency of the methods we were using didn’t really feel good enough.

Another downside is that making late changes has an exponential knock-on effect. Small changes might affect many systems, create new bugs or require updates. Suddenly, they’re not so small.

#4 – Hobbies are a "Serious Business"

Time to get your business head on (or,“BidNiz,” as we call it at BadgerHammer).

With Dunk Dunk pretty much done, it took us about a year to get a publisher, and then another year to get to release.

We spent that year of our gamedev hobby making social media posts, pitch decks, and filling in forms, and it sure felt like a job sometimes.

After going through different levels of due diligence with a number of publishers, we were on the verge of giving up and releasing ourselves. But a collision of coincidences meant all the stars finally aligned and a publishing deal emerged from the spooky mists. It was grueling, but we learnt a lot about the process and also a lot about what we want from our hobby.

Feel free to use our pitch doc wireframe – we got it pretty honed down.

A publisher comes with their own set of requirements for assets, content, more single player content and extra modes. We signed up for another year of dev to our “finished” game. Remember that exponential knock on effect – we really felt it here.

It’s also worth mentioning that meetings can really eat into your time when that hour is a quarter of your allocated gamedev hobby time.

Accountants appear. Companies get set up. Lawyers hover around contracts that need to be reviewed, negotiated, and signed.

Maybe it's time to reclassify your hobby if you need a lawyer and an accountant.

Don’t give up!

The funny thing is, if you spread those four hours a week over seven years, it's actually less than a year of full-time gamedev. Wait, we made this “heap of junk” on our own in less than a year?! Not bad!

So, it's okay if it takes seven years to make your game. Tell you what: Take as long as you like. There are loads of legitimate reasons why it can take that long. Just try to make sure you’re enjoying the experience, learning from it, and making something you’re proud of. Even though it might not always feel like it, it's okay for gamedev to be your hobby.

Dunk Dunk is out now on Steam and Nintendo Switch™ – get it now and support hobby devs! Explore more Made With Unity games on our Steam Curator page and get more insights from developers on Unity’s Resources page.

Nintendo Switch™ is a registered trademark of Nintendo.