I've spent the last 2.5 months learning gamedev and building PigPonyPocalypse

For the last 5-6 years, I’ve been dreaming of making my own game to capture that feeling of progressing the highest difficulty raid, but combining learning gamedev with a full-time job was impossible for me. So 2.5 months ago, I decided to dedicate a year to solo development.

The first thing I did was put my vision on paper:

• core concept (MMO-style boss fights, where interesting mechanics are more important than reactive bullet hell gameplay, tactics over speed)

• gameplay loop (learn boss -> defeat -> unlock harder version)

• and the emotions (overcoming, tension, solving problems) I want my game to give to players.

Next came the setting. I wanted it to be completely opposite to the stressful gameplay. Also, I knew that in the process of development I would be looking at my visuals 24/7. And I always liked ponies. And pigs. So the choice was kind of a no-brainer for me - cute ponies vs. evil pigs with a metal soundtrack.

With the concept ready, I moved to learning Unity and C#, and honestly, at the beginning it was tough. I spent numerous hours trying to understand OOP concepts (like what should be protected/abstract static/virtual (???) void Something). On top of that, my school knowledge of vectors and trigonometry was kind of rusty, so I invested some time in that too.

After about 3 weeks, I made a simple combat prototype with a green square player shooting moving orange enemy squares and an EVIL triangle boss. At this time it clicked that I really can develop something close to my vision, because at the start I was not confident in my learning skills towards coding.

So after that, I designed my first boss fight with Arcane Corrupted Piglet, drew simple schemes, and described core mechanics in Confluence (but I never tried setting up a Jira board for myself - no no no).

To capture the feeling of raiding with a party, I’ve added controllable allies - a tank and a healer. Then I designed boss mechanics that force the player to move the tank quickly. So to reduce the rush, I added the ability to freeze time - to take a breath and make the right tactical decisions.

Visuals were a pain - trying to make the visual style cohesive and implement Generative AI in the pipeline of visual asset preparation was pretty challenging for me. It was clear after 3-4 days of losing time that gen AI can’t give me the exact visuals I want, so I focused on my art skills and shader tweaking to make the raw assets feel… not so raw.

Soundwise I was a lot more prepared, so I decided to use songs from metal albums I’ve written and recorded in the past few years. As metal songs have a very busy 0.8 - 3kHz range, I had to remix them to be more suitable and clear in the game environment.

I am currently polishing the first Gameplay Demo a vertical slice focused entirely on the combat mechanics.

The game is still very early in development and I will need brutal, honest feedback from hardcore players.

Transparency note: As a solo dev with zero budget, I used AI tools to generate base concepts and raw 2D assets, which I then manually edited, shaded, and animated to fit the game's style.