What went right
Coding – In past entries I spent far too much time and energy trying to maintain good coding practices. That was completely thrown out the window this time around, and, though we had to deal with a lot more bugs during development, I actually finished a game.
Basecode – I owe a lot of that success to having a solid basecode this Jam as well. In previous attempts, I had a basic idea what I wanted to make and prepared base code for that. More often than not, the theme would completely shatter those ideas leaving me with very little useful code. In this jam, I included a wide variety of libraries, and was able to choose ones useful to my project.
Map Editor – In particular, having the ability to import .TMX files right from the start was a godsend. The ‘export to lua’ feature of Tiled is great, but not perfect.
Teamwork – Despite being in separate locations, it was a very different experience having someone else to bounce ideas off of, complain to, and test gameplay. It’s much harder to justify slacking off to someone else than it is when you are competing alone.
Graphics – This was my first time entering with a partner, and having someone else looking after graphics removed one of the most stressful parts of a compo for me. It also changed my design pattern. Normally I use placeholder graphics, and then create final sprites/textures once the code is solid. This time, Alex had a lot of the graphics done before I even had the code to support them. This eliminated a common trend of my past entries where I would write a feature which later had to be cut due to lack of graphics.
What went wrong
Timelapse – I messed up my Chronolapse / VNC configuration, making my timelapse screenshots completely unusable.
Music – We had planned on being a three person team, with someone else looking after the music. It didn’t end up being finished in time, so we had to use placeholder music I created earlier in the compo. (I used GreaseMonkey’s Autotracker, OpenMPT, and FL Studio 10.)
FPS – I didn’t have much, if any, time to optimize the raycasting engine. While the walls are pretty fast (using quads to draw them in stripes) the floor and ceiling are drawn pixel by pixel, which is painfully slow at anything but microscopic resolutions. Fortunately our graphics were already intentionally low-res, but even so, the game simply doesn’t look as nice scaled as it does full res, and I know a lot of people are not going to be able to play it with those settings.
Theme – We had to completely scrap our original idea due to the theme. I’m still happy with how the game turned out, but it definitely took a very different direction than we had intended.
Tags: ld24, post-mortem, postmortem, video
