So I’m finally getting around to writing a post-mortem for Mole Cave Tactics. I know it’s better to do it fresh off development, but I needed a break.

Since it seems to work pretty well for a lot of other people, I’ll go with the time-honored format of

The Good

• The concept: I had my concept in mind for this theme (and the many related themes on the list) since well before the theme voting wrapped up. In fact, I had been bouncing ideas for some sort of tactical rpg around in my head for quite some time prior to the compo.
• I finished: The game didn’t have nearly as much content as I had included in the concept, but it was pretty obvious that that would be the case from the beginning, anyway. Given that this is my first LD, I’m pretty happy I managed to upload a game which runs, has things for the player to do, and has defined ways to win and lose.
• The graphics: Ok, so it could be reasonably argued that my sprites would be better classified as ‘bad’ or ‘ugly’, but, I’ve never done any graphics from scratch by myself before, and I actually rather like how the moles turned out. They need some tweaking, but I think they’re pretty good. The terrain I’m less happy with.

The Bad

• Mismanaged time: I kinda knew from the start that this was going to be too ambitious for a first time LD entrant, but above and beyond that I did a poor job prioritizing different aspects of the game. In particular, I spent way, way too much time on AI. I didn’t really appreciate just how difficult a problem doing AI for this kind of game is, I went for way too much sophistication, and as a result I didn’t pull it off at all. I should have known immediately I was doing something wrong when I started writing code to have the AI units look ahead multiple turns in planning their movement. Random walk may not have been very satisfying, but it would have been more than compensated for by the extra stuff I could code in the time I saved.
• No readme: I didn’t ever get around to writing a readme or even specifying a license, so the player is kinda left on their own to figure out what to do.
• Only one map: This game could *really* have used multiple maps, or a random map generator. This is one of those important things that didn’t get done because I was tinkering with the AI.

The Ugly

• The interface: Oh, God, is the interface bad. I should have taken it more seriously when I started adding undocumented keyboard shortcuts for nearly every mouse action in the game. I didn’t think the interface through very well at all during the competition. Having thought it through after reading the complaints, I think that if I want a mouse-driven interface, a radial menu centered on the unit under consideration (think NWN) would be best. Alternatively, just ditch the mouse and switch to all keyboard.
• The Windows port: for some reason, the Windows port was incredibly painful to make this time. For a long time I actually somehow managed to have a build posted which wouldn’t even run on the machine I built it on. I love Ruby, but packaging Ruby applications is a terrible, ridiculous mess. Next time, I’m going to look into Crate.

Pre Vita

All that said, I had a huge amount of fun doing this compo, and I definitely intend to enter again. I also plan to rework and expand Mole Cave Tactics into a full-fledged game, when I have some time to sit-down and code again. Don’t expect a release soon, though: my other game, Operation Lambda (shameless plug) took half a year to complete; it’ll take a few more LDs before I’m able to crank out games of any appreciable quality in a short time period.

Tags: journal, postmortem