Digital Forensics, Dizzy Knight
Everyone in the industry should have an idea about what a Post Mortem is. Really any development field is going to have the exact same mechanic or something quite close to it.
In short, it’s a necessary and healthy function to endcap any development cycle that lets the developer themselves spell out what went wrong and what went right with the goal being to be able to better identify those things later and improve upon them.
Generally, post mortems are an internal or personal thing. A lot of devs these days are posting their post mortems out to the general world, none more so than indie developers.
In this series we will be looking at some of the more interesting indie post mortems that have popped up and identify both common and unique mistakes to each.
This development cycle for themodulus was a labor of love over the span of several years. Having come from a studio background that had left them feeling burnt out and unwilling to submit to the crunch and overwork associated with the that, they chose to strike out on their own and create this game.
That is a sentiment that I am afraid I will find to be one of the unifying factors in most of the successful game post mortems that I will eventually cover.
Overdoing the Prototype
One major takeaway early on was the developer’s use of a ton of free assets when fleshing out the prototype of the game. As you will read in the actual Post-Mortem this helped them in some ways by giving them some knowledge in UI design and implementation but everything was eventually remade as the art and design of the game evolved.
Testing, Testing, Testing
This one was a surprise to me at first, but upon reflection it made perfect sense.
TheModulus had a lot of trouble finding testers. With the market they were trying to reach and the amount of language compatibility they were striving for, they would need at least a tester for each and every language the game was going to include.
Suffice to say they had to pay for some testers and had a lot of trouble getting effective test results.
Automation
This is something I hope will be seen in some future post mortems I cover, at least the more recent ones. TheModulus was able to set scripts and bots to run fast iterations of sections of the game and its code to find bugs that a human user may never see just by testing as a normal player.
While the bots found issues and were certainly helpful, they were not sufficient enough to replace human testers.
Free time isn’t Free
Work-life balance seems to be the first victim to the inexorable hunger that is game development. I imagine it holds true for most other forms of development as well. TheModulus saw their free time as Dev time and when you are holding down 9–5 job that can be outright damaging.
They eventually found their way into a balance that worked for them, even if it was just the time to eat with their SO. It can be the little things that help keep you going.
Please go and read their actual Post Mortem here:
https://www.reddit.com/r/gamedev/comments/9t0ity/finally_launched_my_iosandroid_game_dizzy_knight/
This article was only a brief overview and I cannot reinforce how important reading these are to developers of all kinds. These articles will always serve to be a highlight to these post mortems. They are invaluable to us all and everyone should go read them! These minefields have already been swept by those that came before and while you will likely find some on your own, you can at least avoid the ones that have been shown.
Dizzy Knight can be found on the App Store and Google Play.
