As Explore The Abyss(ETA) has just reached key milestone, I would like to share the history of the project and the journey I took from prototyping to a what is now a demo.

Backstory

I have been playing roguelikes(RL) off and on for the past 10 years. These range from short games like Brogue, Midboss to longer games like Cogmind and sandbox RL like Cataclysm:DDA. I also enjoyed various mystery dungeons games so it is no surprise that when I decided to make a Commercial game, Traditional Roguelike is the first that came to mind.

The start of the journey can be traced back to my Ludum Dare 46 Submission. Although it is a short game and not really a RL, it is the start of some of the engine code that I still use today.

As I did not know which type of RL I wanted to go with, I started by prototyping various ideas that I written down over the years.

One of the first is an attempt to combine cards with RL, and no, I am not talking about roguelike deckbuilder. You can find the prototype here. The prototype is interesting, but I find the combat very tedious, making the game very unfun.

I also prototyped a Mystery dungeon game, i.e. one with long term progression. Knowing that this will not be the last RL I prototype, I decided to make as much of the code reusable as possible, which lead to the start of an RL engine call Ren. I soon find myself losing interest in the project due to it have not having any “fun” elements, mostly stemmed from the fact that it is a big project and a lot of work need to be done to get to the “fun”.

It just so happens that 7DRL is around the corner, so i decided to take a break from that project and do a 7DRL with the engine. The game is call “Abyss” and the post submission version can be found here.

Seeing how a small project can be quickly iterated, combined with the fact that I am losing interest in the previous project, I decided it is better for me to have a smaller project, one with a scope like Brogue.

History

When ETA started, the game isn’t what it is now. It started as room based roguelike, which I had named it “roomrl”. The game had 10 floors, and at the end of each floor, you get to choose what type of room you want to go.

The original idea came from https://studiominiboss.itch.io/chamber, one of the best (if not the best) submission from LD47. I wondered how it will feel like if I replaced the exploration of traditional roguelike with a simple room choosing mechanic, and focus more on the tactical aspect of roguelike.

There were a few types of room; treasure room, combat room, etc. One of the key one is exploration room, where you just have to find the exit. This turn out to be the only room that is fun to play.

I soon realised that exploration is very important to traditional roguelike, and there is no easy replacement for it. I promptly remove the room choosing mechanic and it became just a normal roguelike.

I decided to just make a generic roguelike, throw as much things to it until I find something that stick. What sticked turns out to be exploration and inventory management.

Remembering that I had “Abyss” from the 7DRL, I decided to name the game “Explore The Abyss”, as a sequel to Abyss. Soon, I shift the focus from combat to having interesting terrain and map layout to explore.

Lessons

There are a few lessons from the time I started prototyping to the decision to commit to a long term project.

1. Many ideas are really cool on paper, until you test it out.
2. Getting to the fun as fast as possible is important.
3. Getting out of boring as early as possible is also important.
4. Getting out of Sunk-cost is hard. If not for 7DRL taking me out of project, I would have continue on a project that I am no longer motivated to do just because of sunk-cost.

Conclusion

Getting from prototyping to committing to a project is probably the 2nd hardest thing in game development, the hardest being finishing the game. For devs who are still trying to find a project to commit to, keep at it.

If you are interested in the game, head to the project page and play it. As for where the project will be heading, see this Update post for more information.

I hope you find this post interesting and thanks for reading this long post.