View on GitHub

Ri-Karts!

The best Yo-kai racing game (probably the only one)

HOME

Post Mortem

Ri-Karts born as a college project, is in fact a whole subject. Likewise the game pillars was chosen in function of team members and their skills, game pillars chosen:

The Yo-Kais theme was chosen later by us, cause it matches wll with the game pillars and it was the most voted theme proposed. The entire game development elapsed during the 2n quarter of our college (CITM), from beginning of february to beginning of june, 4 months. A very limited time to develop the game and also the engine in parallel.

What went right

  1. We have 2 game designers checking and testing the kart’s behaviour during all the development. Thanks to this, the kart’s control feels very nice: maneuverability, drifting, turbo, etc. We have had a lot of problems with physics during the development and have the security that the kart’s always it would be ok it was great.

  2. Map Generator Tool. We had an amazing tool that allowed us to make new circuits very fast. This was very useful to check maps and for level designers in general.

  3. Simple Art Style. Ri-Karts’ art style was chosen with the idea that our level 3D artists doesn’t have experience. This way we could made props good enough to enhance the map.

  4. Crunch sessions at the end of every sprint. Before every delivery, at the end of the sprints, we always met in a college’s classroom to ensure that everything was ok. This sessions were key and aboslutely necessary in the development, due to we didn’t have an office, the moments when we were all together helped a lot, especially in what a communication refers.

What went wrong

  1. Lack of DoD and breach internal deadlines. A lot of times during the development we stablished that tasks must be done for X day and later if someone breached the deadline there was no consequence. This caused bad costumes inside the team and some people let tasks for the last day, etc.

  2. Not enough gameplay for player 2. As I said before, we focused on the kart’s behaviour and the player 1, hoping that some items will be enough for the player 2. But the fact is that player 2 in our game does not have much to do. We should assigned some designer to ensure player 2 gameplay, like we did with the player 1.

  3. Physics system. We started the development with a realistic physics engine, Bullet. It gave us a lot of problems with shocks and ramps. We finally had to remove Bullet and pass to own physics system, developed by our programmers.

Conclusions

Ri-Karts is the first 3D game that we ever developed. Also is the first time that we have the challenge of developing the game and the engine at the same time. And none of us had worked before in a group with more than 20 people. All this factors caused, that costed us a lot of time to find dynamics to work well together and keep some control.

We are the first to recognize that the final result is not the best, it isn’t what our minds imagined when we started at the beginning. But the process and what we learned is what real matters. We have lived what means develop a game and its engine at the same time. Difficulty of make a lot of assets by different people and keep some artistic coherence. Throw a lot of work to the trash, because lack of communication, some mechanic doesn’t work or whatever.

In conclusion, we have learned, what it means develop a FULL 3D game by your self. And it is very difficult, I must say. But Ri-Karts is just the beginning, in this group there are great people with a lot of skill in what they do. I’m looking forward to seeing what can they achieve, amazing things for sure.

Written by: Jordi Nunes, Producer of the project.