580 days ago hailstone wrote:
Hi again,
I've had a chance to look at some of the other games and I'm impressed. Well done everyone.
Now that I've had some sleep I'll add some more about what I did and what I will try next time.
What went well?
The game is playable. I started off with a clear concept of what I wanted - An Asteroids clone. Having that in mind was useful for when I was cutting features; should I implement the menu or work on the end condition for example.
I didn't try to get everything done perfectly (this goes for assets and levels too). Which means I can make a note of it and come back later when I've done other features that are needed to make the game playable; I think this is the prototype stage where you want to have your concept shown really quickly. Then those notes could be used to make a schedule for a more polished release.
Make use cases for all your features. I didn't do this at first but it really helped to get me thinking how to break up features and get the parts working together.
As I finished parts I created a backup and a screenshot in case I wanted messed it up and needed to revert back.
What didn't work?
I think I spent too long on the design. Fortunately, I was able to use parts of another game I had made and I completed most of it in the last couple of days. Starting earlier might have allowed me to do some testing and fix the bugs.
I also spent some time testing IDEs for python but I didn't end up using any of them. Instead I used Kate text editor. This is part of pre-production so I had set time aside for these sort of tasks, I just shouldn't have set so much time aside for it since I already had tools that I could use to achieve what I wanted.
I had created a schedule but it was too high level. Things like create beta and finish design document. What I should have done was break these down further but not get into how it will be done just what needs to be done, eg. spawning asteroids (this is where the use cases are helpful). I sort of did this but didn't schedule it from the beginning.
What will I do next time?
Definitely start working on the prototype earlier after I have the core design and tasks that need to be implemented worked out. Create a schedule for the parts of the features, low enough that I know what needs to be done but not so low that to say how. Create the use cases early.
I will try to get tasks done each day. Also write about the process and progress at each stage might have been useful (I only did a changelog and screenshot).
To close I'll say that I spent around 20 - 30 hours on this.
Till next time...
PS: It might be useful to have a post moretem check box for posts just like there is a final release check box for files.