We’ve been working on Falling Star(working title) for around 4 months now, and in that time the game has evolved quite a lot from its early prototype build (which we still have somewhere!)
Very recently we’ve started getting near-final unit artwork and animations, and it’s surprising how much a good piece of art can improve the game. Despite some teething problems with the animation plugin we’re using it’s all starting to come together.
While waiting for assets I’ve been able to focus much more on the gameplay implementation and to an extent tweaking the look and feel. A lot of this was pre-emptive, given there weren’t assets for it at the time, but one of Unity’s great selling points is the ability to quickly include new assets into the workflow quickly.
The focus on gameplay means we’ve iterated through a number of changes, even over the last few weeks. The combat system has changed from accuracy based “hit or miss” gameplay, through to a range based “damage dropoff” mechanic. This feels much better and works well with the game. Changes to the actions units can perform have evolved as we’ve test played. Some things just didn’t work out because they didn’t mesh well with the feel we’re trying to get. Cutting them now lets us make each unit as unique and attractive as possible, and gives our artist time to make any required changes!
Having time to focus on the coding has also shown up a few issues with the implementation over the last months too. The way commands are processed has changed, as has the network management and format of requests. But being able to handle these issues as they come up, without worrying about sound or sprites, has led to much more robust code.
Advice on what to do when the code gets unwieldy is mixed; some say rewrite it all from scratch, some say just run with it. I guess it depends on how much time you have, and for me having that time to work with the codebase meant analysing the bits that weren’t working and would take a lot of time to maintain in the future, fixing them up and reaping the benefits later on. So I guess it’s contextual, but certainly don’t let bad code ruin the game!
The next few weeks will see the game rapidly expand towards looking and feeling complete. There are definitely a few months to go but with artwork and sounds coming in thick and fast it’ll start to feel a lot more like a full game!