Postmortem.

It's time to share experience.

Therefore, I post this postmortem.

Defend My Crystal! was launched on October 30th, 2020 - on its 13th day of development, as a Devtober 2020 gamejam submission. The rules of the jam require to write a postmortem, not to mention it would be wise to do so - and therefore I offer you my experience.

Note: to best understand the post mortem it's recommended to first play the game itself.

What went wrong?

First and foremost, let's look at what didn't go as planned.

As any project that inspires you, in the beginning of development of Defend My Crystal! I had ideas that were out of scope. Among such ideas was making several types of enemies and having 5-10 levels with manually customized maps and enemy waves. In release state game had only one enemy variation and 3 arenas with an endless amount of enemy waves. What were the problems?

• As for enemy types, the biggest problem in case of having multiple enemy types would be balancing everything in time. The game already has 3 Mercenary types, each with different functions and stats, what already complicates the game in terms of balancing quite hugely. For me it took a few hours straight replaying my game over and over to make sure the game requires using all types of Allies with only ONE type of enemy - I don't dare to even imagine how long would it take to make the game's random be challenging but fair if there were multiple enemy types.
• Honestly, I decided to switch from levels with finite amount of enemy waves to arenas with infinite ones only a few days before the release. Making the game level-based would make the experience fun, meaning you have some ultimate "goal" to achieve - but it would require hours of design so the experience would always be new and interesting. Briefly, sometimes arenas aren't as bad solution as they might seem.
• Another significant problem is overestimating the amount of work that can be done in terms of one day. With other commitments in life, I sometimes had no time or energy to work on the game in the evenings, what resulted in the fact that a few of these days I've been working solely in early mornings.

What went right?

Apart from negative moments, there surely was a lot of great stuff!

• I dedicated time on game design and pure planning, what actually saved me a lot of time - it possibly even was the reason why I could finish the game in time at all. With defined goals and planned schedule you know the real scope of the project and you are much less prone to waste time and/or energy. Regularly making a plan of work comes especially handy when you can not allow yourself work 16 hours a day and you have other commitments. What you now can play was being developed for 1-2 hours for 13 days everyday, except the last two during which I spent 3-6 hours on preparing the game for the release state. The lesson: plan your work and projects!
• I enjoyed the process. Probably the main reason that keeps a human doing something is satisfaction which can come from many ways. For me this satisfaction came from 1) developing and seeing the game growing 2) sharing the experience with friends.
• Seeing the progress and being supported helped me overcome mental hurdles. I think it is worthy of a separate mention considering how many people struggle. I can not say I always had great mood, but remembering about my work and thinking of its possible success helped me staying in an optimistic way of thinking.
• I opened more art for myself. As a solo developer I can not do everything by myself, so I needed to find some pre-made assets and materials for my game. I found out about more content makers on OpenGameArt.org who make amazing products (music, visuals etc.).
• Experimenting with new mechanics and in-game systems wasn't easy, but it was well-worth it. Briefly, confronting new problems allowed me to learn and grow, allowing me to develop bigger and better projects in future.

What could be improved.

Based on my own thoughts of what went not quite well, here's a small list of things that can be :

• Leave some time for emergency debugging. By that I mean not solving the issue on the last hour, but at least testing the final version of the game on all the platforms a few times 6+ hours before the deadline. I had a significant problem that occurred ONLY in Windows and HTML ports, which I happily resolved in a few minutes.

I suppose you've got a few more ideas from reading the previous paragraphs, because I can say nothing more, haha.

Thanks for reading this postmortem! I hope you enjoyed it and it was useful to you. I'm very glad I participated in the jam - special thanks to its host, Ellian!

Best regards,

Ivess