Devlog 1 - God thinks I'm funny

I’m starting a series of developer logs where I just write down everything I’ve tried to do over the past week, and share a little progress. Largely I’m doing these for my own self-reflection but I hope there’s a little humour and maybe even something useful in there too.

We plan, God laughs”. Oh boy, I went into this way too optimistic. This was the first week I adopted an agile methodology (after watching the excellent Lychee Game Labs video on making games fast) and to say I need more practice at this is an understatement. Which is funny, because it looks like overstating my capabilities is my raison d'etre.

The goal of this week was to prepare the code base for the upcoming upgrade feature. This means;

• Refactoring the build and HUD UI so there’s room for it visually
• Adding a way to modify player stats that actually tracks who/what is modifying them
• Breaking towers 

First off, credit where credit is due, I actually got a lot done! During this week I;

• Developed a game statistic system
• Refactored the game building menu into sub-menus (cleaner code)
• Drew a bunch new user interface elements
• Replaced the wave/life ui elements in-game
• Started (oh boy) working on a tower component system

But I did not get everything planned done. The component system is less than half finished, and the majority of the interface stuff works but isn’t actually connected to the hud or replacing the old interface elements.

Maybe I did start on Tuesday (Easter Monday) but the real problem here was just planning. There’s a reason the common software developer adage is ‘however long you think it is going to take, triple it’. I was way too over confident that certain things were going to take a lot less time than they did.

For example, in my head the stat system was a single day job, tops. I’ve written similar things before so obviously I can power through this really quickly - I can even look back at old versions and copy what was working. Except I didn’t do that, I just wrote it from scratch again; at the time I was telling myself ‘it’ll be faster to write it from scratch, those old versions didn’t really work anyway’ but really I was scared to go back and have to read a bunch of messy code from a year or more ago. That could have saved time.

All told the stat system took almost three days. That single delay led to me playing catchup all week and when the component system also ballooned in workload it just wasn’t feasible to get everything I planned done. Luckily (luckily?) I have a cancelled holiday this month so I’ve actually gained a week extra to work. I’m going to use this week to catch up a bit and then hopefully not fall behind so quickly. I still need to prepare the new build (Indev 11) for release after all, but more on that next week!

What I hope to take away from this week is to plan a more realistic workload, plan tasks before I get to them (so I have a better idea of scope) and to reuse old code where appropriate. 

Have I learned from my mistakes? I want to tell you I have, but I’m looking at this week’s workload…

“Insanity is doing the same thing over and over again and expecting different results”.

The Great Train Defence is a tower defence game where you’re often on the move, and it often feels like that’s my life too! Drop a comment below or chat to me on discord, and save me from just having to talk to myself!