Picocraft: a bumpy ride

I intend to share information, thoughts and details about the bumpy road this project has taken. This first post focus on project start, socials and organization aspects. Then in future posts I'll share technical details and choice that can explain how to fit this project in pico-8.

The idea comes from Artyom Brullov a pixel artists who once share a cool animated pixel art mockup of warcraft III on twitter and reddit. At this moment I was tweaking 3D modeling code on pico8.

I share the mock-up to a friend telling him "I think I can made it !". He just tell me to contact @brullov_art and ask for a collaboration.

I hesitated because I liked what I was doing, and programming a RTS is not known to be a simple task. I knew it would take time, and I finally accepted the challenge and contact @brullov_art who accepted the collaboration.

@brullov_art took the leader role and took care of the game-design document. He also wrote the roadmap (steps, with no date). Then I began the journey. I knew it would take time, but I did not imagine sooo long.

Social and organizational aspects

Ok, from social and pico8 size project point of view this project take too much time. When working in a team, you need to move fast to not lose your mates. Motivation and pico8 token limit are difficult to handle on a such a project. Token limit is here to help you focus on the essentials, but when the minimum is already big for a RTS, it was a everyday risk not to be able to finish. Token limit is meant to help focus on small things, not make it risky. I think I have twisted the token goal ! I had days, even week, where I did not touch anything, and I finally lost @brullov_art who need to focus on more profitable projects.

The 3 years wasn't a full time job, but a part time hobby. The main thing that I succeed is that I managed to stay focus (if you don't count Ludum Dare 52 ! ). According to pico8 activity report, I spend around 555 hours which is a lot and not so much. For comparison I have around 1500H of L4D2. So in this 555 hours I spent 320 hours on the code editor and 120 hours running (playtest, etc...). At first I though as it will be a long journey, and sayd to myself there is no need to rush. Maybe it wasn't a good idea, anyway I spend some time comparing coding solutions (virtual sprite, state machine, behaviour tree, pathfinding...) before really implementing them. That is not a bad practice but I may have spend too much time on it. Moving forward on the project, refacto and optimization became really important. Unifying data size and structure was easier with editors. So I spend time coding editors and exporters to make it easier to edit and make data compact. Animation editor allow me to standardise all animation and store data at binary level. Map editor help me on multiple level, I wanted to test "auto-border" for terrain, compact data to fit in first bank of map memory, and also place objects. All this couldn't be done with pico8 built in map editor.

But the real danger I think was motivation. Sometimes I didn't touch code for days even weeks, I don't remember well, but I think one time I let it go for 2 months, in summer period.  I also spend 2 months after Ludum Dare 52 tweaking with Godot. The deadly token limit has been reach multiple time. At first I hoped to make it fit in one cartridge. I finally made 3. Token fighting was intense and motivation consuming, the fear of "the impossible" came multiple times. To overcome this, I worked on editors, draft, fresh empty pico8 cartridge with all token free ! After 2 years, (even before I think) I lost my mates direct contribution. I finally gave myself an ultimatum, the game would be out before the year 2025 or will not ! Then I began to share public progression regularly to stay committed. 2025 arrived but it was ready yet ! I noticed a clear difference in pace and the fact that I had 90% help me push it to the end. If I had worked from the start like the lasts 6 month, I think I would have make it in 12 or 18 months. It is not that I stopped sleeping, or work all the day, I was more regular, work almost each day. You can clearly see the progression on the pico8 activity report.

All in all, with the 3 game cartridge, the editors with exporters, I made a equivalent of 6 cartridges ! After all it is like 2 games a year... Ahah !

More seriously, the keys to finish was :

• Stay focus,
• Stay committed to an audiance,
• Try to keep the fun !

In a next project I will add a gamejam strategy, quickly have a working game loop, then feeding it with new features.