IF AT FIRST YOU DON’T SUCCEED… Blow it all to hell and start again.
Refactor after refactor…I’ve been down this road before. It feels like a war of attrition, rebuilding one script, to fix one issue, and breaking sox other scripts in the process. …Fix the six broken scripts and now repeat the process 6x times whatever the hell, and eventually, BOOM - everything works.
When I used to build and code system simulators, the number of scripts would climb pretty high, so if a refactor touched 100 scripts, it was better than the alternative of starting over. On the heels of that memory, i completed something like ten major refactors before i saw it: I know exactly what I want to do with this project. Contrary to my integrated system simulator days, i started this project with no idea how i wanted to structure the project - so it was with some small heartbreak and very little surprise that i conceded it was time to start over. I would bring the art and audio files, but no code would be copied - the framework and concepts would be rebuilt - not from what i had done before, but from what I should have done.
At the time of writing, I’m pretty deep into the rebuild. I haven't gotten back all the functionality I had before, but the pieces i have rebuilt, are built stronger and so much more modular.
I’ve adopted a modified process for building out the scripts, which runs somewhat counter to best OOP practices, but is suited to my status as a full-blown-newbie…Instead of creating scripts which contain all the parameters and functions of a given parent concept, I'm creating shared variable classes and a new script for each function. This allows me to keep the system adaptable and easy to modify without ruining other aspects, but it adds a layer of complexity in terms of accessing parameters.
In the end (or really, the pre-beginning), it has occurred to me that, if I do this right, I will have built out 80% the game before I really start the first level. It is an arduous and unrewarding path up front, but I think when I reach that point it will be fun to build levels and skills into the pre-existing framework. I have to assume this, or I’d quit.
For the rebuild, I started a github repo before the project even began. From the bottom up, you can follow the commits and see the project developing, starting with the creation of the new project.
The git repo and commit process has made it so much easier for me to pick up where I left off, even after a few days of not working on it.
Having come into this new build with a much fuller understanding of how things should interconnect and where they should be broken into subcomponents has already made the development process much clearer. Although it is a setback to start over and copy no code, even from the current standpoint (which has less functionality than the prototype that preceded it), I can easily say this was worth doing, as it has helped create very clear functional boundaries and far more scalable modules.
There isn't much to show right now in the way of actual gameplay, but many of the systems that will be needed to develop the game have been implemented, tested, and interconnected, which will make the game development side of things much smoother as I proceed.
I took the last week or so off to spend time with family and clear my head. I haven't resumed the effort yet, but figured I'd start getting back into the flow of things by reviewing what's been done and writing this post. I think tonight, I will sleep...actually, I think I'll sleep every night this week - but next week, I will plan to get things moving again, and I'm very excited to get this thing off the ground and share an early release version on Steam.
I will try to keep a running list of all devlog entries
Thanks for checking out some more of my work! As always, I hope you enjoyed witnessing as much as I enjoyed creating!
© Photos and words by 
