Bunny Bunny Dig Dig: a Compost-mortem

Hello! As usual after a compo, I like to do a short post mortem. Here is how Bunny Bunny Dig Dig was made!

I live in France so the LD starts at 3 AM. I like to wake up at around 6AM to start the Compo with a good bowl of cereals.

🎲 Finding a Concept:

I always give myself at least half a day to find and define a concept that I like and -most importantly- is achievable. I feel it's worth taking a couple extra hours to carefully plan what's to be made and check if it's feasible instead of going straight into the engine.

For deeper & deeper, I explored ideas around exploring oceans' depths, ants digging galleries and a detective game about going deeper & deeper in the psyche of suspects.

I ended up going for a simpler, more litteral take on the theme, as I wanted to challenge myself into making a complete roguelite game for the Compo.

🗓 Planning breakdown:

Day 1 - Morning:

When making games, I always start with a mockup. I wanted to define the aesthetic quickly so I assembled references, mainly from @canalbooks1125 and @Gaziter whose art I like a lot:

With these as references I drew a little mockup. I reused most of the art in the final game of course.

Day 1 - afternoon + evening:

I had a very precise vision of what I wanted in the game:
- A very simple core mechanic about mining with straightforward controls
- Spice it up with a combat twist and random monsters encounters (3 monsters were planned, only 1 made it)
- a procedurally generated mine
- A skill/upgrade tree to incentivize replayability
- a final boss at the end of the mine (didn't make it to the game)

I immediatly started with the mine generation. I know making it full random wouldn't be fun, so I wanted to keep some level of control over the situations the player would encounter. I lost a bit of time figuring out how to create a level design tool, that was efficient for me to use, and I'm not satisfied yet, but here is what I made:

By creating patches ("Mine Segments") with my individual block prefabs, I can design situations such as a Gem Block surrounded by Monster Blocks. The Mine Segments are grouped in collections from which they are drawn from according to the depth you're at and their rarity.

Once the generation was done, I made the player controls and mining mechanic. It's all grid based (player can't move freely, constrained to a grid), so it's really fast to do, you don't have all the issues you might get with a platformer controller for example.

I managed to squeeze the combat, which was reusing most of the mining code before going to bed around 11:30PM.

I knew I would have to rush the next day because I usually try to finish all the mechanics Day 1 to focus on polish & art on Day 2, but I was still missing all the upgrade stuff and the macro loop.

Day 2 - morning:

Woke up at 7AM and completed all the art with a fresh brain and integrated it right away. It took longer than anticipated and I finished around noon. Most of the animations. For game jams I always make a lot of animations directly in code (with tweens) so there aren't that many assets.

Day 2 - afternoon:

After a bugfix pass, I proceeded to make the upgrade tree. The upgrade themselves were quick to make since they're only passive bonuses, so it's just integers to move around. What took longer was deciding which upgrades to make, writing all the texts and thinking of the layout.

I then realized the mining phase was way too easy and boring, so I added the torch mechanic, which is basically just a rechargeable timer at the end of which you die. This immedialty improved the pace of the game and allowed the players to make decisions about their path down the mine and make sure they would aim for the Torch Blocks to stay alive.

Day 2 - evening + night:

I was a bit worried because I had no audio and there was still a lot of polish to do and bugs to fix. I took 1h to make the main musical theme + the village theme (by basically getting rid of the drums and pitch shifting the main melody). I made a list of elements that needed audio feedback. With BFXR and a quick rework of some samples I had in Ableton, I manage to nail the important ones, but I would have liked to add sound to many many more stuff in the game.

I squeezed in some Mine Segments & a Bat variation to add some diversity to the Level Design at the last minute. I knew it'd be a challenge, but the scope was definitely too large. This was the first time I had to rush that much to finish a Ludum. In the end it turned out functionnal, but it misses a lot of work to balance the mine generation, upgrades and combat. I'm still happy the game turned out almost exactly into what I had in mind.

💡 Key Takeaways:

• The challenge of scoping large was fun, but I don't think I'll do that again. I didn't even have time to browse the ldjam website while working, and I missed a more relaxed jam atmosphere. Number one rule is to have fun, so I'll go for a smaller scope next time.
• Sleep is never a waste of time, even if you're late on your planning. You work twice as fast and make less mistakes when your brain rested.
• Picking a simple core mechanic that is visually understandable in images or GIFs on your ludum page is always a good idea. You save time because you don't need to implement a tutorial and people don't get frustrated because they don't understand how to play. I like to think that the game starts on the ldjam website page, so I always put some GIFs that'll help players understand how the game is played before they click on the link.

I hope everyone enjoyed this game jam. Congrats to all the participants and I hope you'll have fun playing Bunny Bunny Dig Dig!

Sorry for the long post, have a rabbit: 🐇