Ghost Loop - What we did and what we didn't do

GHOST LOOP - POSTMORTEM

Hi everyone! Kim from PureLeeks here.  Hope you all had a nice holiday and that the new year is treating you well.

In this post I want to share  a little of the development process of "Ghost Loop" with you and what we learned during that time. 

PureLeeks:

PureLeeks is a group of three friends consisting of two artists and one programmer/developer. We have all worked on games in some capacity and wanted to make our own spare time projects, mainly in Unity.

We met up and planned our first project in October and through talking and some brainstorming we decided that we wanted to make a game for Halloween (spoiler: that didn't work out). 

We quickly found that this was way too ambitious and decided to make it a game for the coming holiday instead. We then contacted a few friends and asked if they wanted to do music and sound.

We're all working people and knew that we had to make a lot of compromises. We also wanted to make short games.

WHAT WE LEARNED:

• Plan everything - We made a timeline for the project and tried to follow that. As we all are working people, time issues arose quickly (we expected this) and pushed our release date back. We also decided on what game engine and software to use here.  We also forgot to include production time for sound systems.
• Make early prototypes and share them with others - We really didn't do this at all. Feedback from developers and non-developers alike, would have helped us fix gameplay at an early stage and made the game more fun in the end. Greyboxing is king!

• Stay organized - We used a combination of Google Drive and RealTimeBoard and it worked like a charm. Here we posted work logs and even finished assets. 
• Communicate! - Communication is key. If everyone know what they're supposed to do at any given time, development will go much smoother. If all team members are on the same page, you will also avoid disagreements later down the line. 
• Make a asset production pipeline - This one is really important. Figure out the pixel size of everything before starting to produce assets. Are the humans, 128x256 or are they 256x256? How big are the NPCs in comparison to the environment? What colors do we use? What format is best for sound files? A poor production pipeline will result in loads of extra work. We had to fix several assets (rescaling, colors, file conversion etc) because we failed to settle on a strict set of guidelines before starting production.
• Make sure everyone has access to the project during development - We failed miserably at this. During the whole development cycle, I was the only one with the project. This resulted in me using loads of time I could have used on programming for implementing assets and the artists waiting for me to finish this process up. This also resulted in lots of uneccasary bug fixing at the end of the development cycle. We also found that we had different views on how the game should play.
• Start testing early - This will help catch bugs and performance issues, but maybe more importantly, you will find out if your gameplay is fun and engaging. We started handing out test builds during the last week of development, which was way too late. We found that our core gameplay had some critical flaws and tried our best to fix them before release. If we had planned better, started testing earlier and everyone had access to the Unity project, we would have ended up with a more fun game.
• Stay motivated! - This one is quite hard and highly subjective, but we found that milestones helped us stay motivated. We eventually realised that "Ghost Loop" would never be perfect and settled for what we had. If we had continued trying to fix the core game, I think we would have lost motivation/interest and never released the game.

The points above are all things we will keep in mind as we start working on our next project, which will be announced soon. "Ghost Loop" was our first game as a group, and we learned tons of stuff about managing a project and seeing it through all the way to the end.