Coincidentally also my first post ever, as it looks like I’ll finally have a weekend free during an LD to be able to participate!

Got home late and got about 4 hours of work in. The elevator pitch for my planned game is a tower defense (sorry, its my first LD), where the enemies evolve between waves, and you can evolve your towers through selective breeding.

I’m planning to assign a list of “chromosomes” to the monsters that cover physical aspects like weight, limb length, intelligence that are translated into gameplay values like health, speed, avoidance. For the first wave I’ll randomly create a set of enemies covering a wide range of these values. After every wave I’ll use a genetic algorithm basing success on distance made into the maze to determine the “strongest” enemies, and breed them to create the next set for the subsequent wave.

Planning a similar system for towers, except they level up during a wave and the player gets to selectively breed them between waves to get the kind of towers they want.

I’m using ShiVa engine for this project, as I’m most familiar with it. Here is the current state of the game, it’s basically just a rote TD at the moment, the interesting stuff will come tomorrow:

I did it, I completed my first ludum dare! It was a ton of fun, but I’m definitely glad there are only a few a year.

Please check out my entry, Mutating Scales:

You can play it on the web, right here.

My first ludum dare, and subsequently my first LD post mortem.

I’m a game programmer for a living, but not in this time frame and not while having to come up with an idea and create all my own content. It was exciting to have total control over the end product, but also much more demanding.

I used ShiVa as my middleware engine, mostly because of my familiarity with it. I ended up up choosing to do a tower defense, but the core gameplay was just a vehicle to show off the genetic algorithm that generated the enemies and the towers you used. Anyway, lets get started:

Right

• Planning – After getting the initial idea, the planning on paper I did ended up being pretty accurate to what was needed for the final product. While planning it, I had an idea of what features would probably not make it in time, and those ended up being the only ones I cut.

• Genetic algorithm – Super proud of how well this worked, and also all the UI work I had to put in to show the player what was happening behind the scenes. This component took at least half of the total time spent on the game, and I really do love it. It nails the theme and makes the game procedural instead of static.

• Art – Since I’m a programmer, I knew this would be my hardest aspect. I ended up making it work for me by making most of it programmatic. I drew a basic grayscale version of the dragons and towers, and cut them up into pieces. I then scaled, tinted and assembled the pieces in code to give the various colors and sizes of the objects in the game. This meant that 80% of the art was done in code which ended up perfect.

Wrong

• Core game – Choosing tower defense is a LD trope, and I wished I had picked something else. The surrounding genetic generation stuff which I was most proud of could have just as easily fit into a twin stick shooter or RPG system. Anything would have probably been better.

• Testing – Didn’t get a chance to do much playing during the process. I had a goal in mind and worked towards it. Turns out the game can be either too easy or too hard simply depending on the initial tower and dragon generation. I should have put more limits and constraints during that process to even it out. Also you can craft comically broken towers, but I would NEVER fix that because it is great.

• Polishing – No time for this, was hoping to give the game objects some basic animation and adjust the maze layout. I spent the only time I did have on what I hoped would be the high impact inclusions, the particle effects and sounds.

Overall I’m happy with what I shipped, a game about dragon sex, and feel better prepared and even more excited to participate in future Ludum Dares. Thanks for reading!

Play and rate Mutating Scales