So after 48 of what I can only describe as brutal crunch time I’m a little disappointed that I couldn’t submit my game to the Ludum Dare. About an hour from the end with nothing but tea for food and having coded for nearly 32 hours straight I was faced with serious debugging problems and realised I couldn’t ship the game on time especially since I still had to package it. I could have opted to code for another day and submit to the jam but having stopped I lay down to sleep and woke up nearly 18 hours later so here is what the Ludum Dare community lovingly calls a Post Mortem:
The Good:
Bulk First Coding. I got a lot done. Even though it wasn’t enough I got a lot done. Maps, combat, inventory and all of it was defined using static arrays, loadable from text files. Rather that my usual thinking about the most efficient path I just coded and it was good.
The Bad:
Having to work in the middle was bad. It disturbed my sleep schedule and drew time away from the Dare, next time I will take 2 days off work.
Sleep. Oh god sleep. Usually when I code overnight I remain or increase in coherency but over the course of the Dare I found myself becoming less and less productive. I slept an average of 6 hours per night the week before and even less on the day before and during. It was horrible, I reached all sorts of existential dilemmas and it was really bad for morale. Partly it was because I scheduled preparation for the Dare really poorly and ended up going overtime most nights but also just plain stupidity. Next time there will be a rigid sleeping schedule: Rest before, code as much as I can, Sleep, Do the final stretch.
Planning. Perhaps the biggest mistake. Not planning my code was good but not planning the development properly was the killing blow. My planning process was literally: I have an idea, lets make it. Unfortunately what I didn’t realise was that my envisioned concept was built mostly around a Fire emblem/Advance Wars style combat and map system which, as a package, seemed smaller than the sum of its parts. And more importantly I failed to identify and focus on the core elements of my game or what tied it to the theme. I started programming face first: The player sees the map first lets make the map but what was the core was the combat and inventory, if I had no map I could still get a game of just increasingly tougher fights with an inventory and have it ship but instead I was stuck writing all of that in till the last second to make the game feel relevant. So next time. Plan. Proper plan.
The Ugly:
Perhaps my most rookie mistakes were in the preparation. I wasn’t quite comfortable with my tools and it was painful.
I don’t think the moment where I found I couldn’t use TrueTypeFont or Textures in unison with simple colours the way that I thought you could is on the stream but it was horrible. Without text its very hard to convey to the player any sort of context or meaning and I let it slide but I should have tried to fix that more. I guess at the time I knew that I wouldn’t get to make textures and I decided to focus on the rest of the game rather than text but in hindsight text was incredibly important. Even if I managed to bug fix without text nothing would make sense.
The other thing was packaging. After I woke up and ran around for a bit I tried to actually package the game and it took me two hours to get it working which, if I had to do it in the dare, would have killed me on the spot.
What Next:
I’m probably going to bug fix and finish the game. After the dare I also figured out that the idea that I was interested in completing wasn’t so appealing to me but a friend of mine gave me some inspiration to actually have a think about it and I might actually get it finished.
Despite not finishing I still enjoyed the dare greatly and there’s no doubt in my mind about the next one. I learned a lot about Java, programming, the game dev community and myself and even if it drove me half into my grave I can’t wait to go again.
Having looked at some entries in the dare its good to see what people are working with and its comforting to know that I’m not so out of my element as I felt. I saw some guys make a lovely looking dungeon crawler with the LWJGL for the jam and it made me think about working in a team for the next one instead, even if I do like the idea of finishing a game on my own
And it also made me realise how stagnant my own projects have been. I’ve made more of a game in 48 hours than I have in 2 years and I feel kind of ashamed, daunted and inspired seeing what I can do and what those who I participated beside can do.
So whats next is only the beginning.
For those actually interested here is my game as it was at the end of the dare including source. It’s horribly buggy right now so don’t be surprised if it takes a couple of restarts to get working. Its technically ported to mac and windows but if you don’t have 2 mouse buttons you can’t do much:
http://www.mediafire.com/download/pxyh2s7pjokoxo2/LD28.zip