tldr: Tools Matter, Sleep, Limit your Content Scope. Great Success!
The Results!

I went into Ludum Dare with a certain nervous excitement. I’ve been poking at indie development for over a year at this point, with very little to actually show for it. I’ve got a bunch of concept documents, a couple prototypes, and a single 2D ‘action’ game to my name. Ludum Dare is talked about in tones of respect among the indie community; participating successfully seemed to be a right of passage. Being in way over my head didn’t stop me from signing up for MLG Raleigh as a Platinum Zerg, so it sure as heck wasn’t going to stop me from mingling with the much more accepting indie scene.

Of all the topics, the one I didn’t want to see was Minimalism. Seriously, how buzzword can you get? And if you look at the voting, nobody really liked anything… I think Minimalism won with 330 positive votes, out of something like 4k voters.
After a quick search, I latched onto the idea of ‘trimming the excess’, and also the literary definition that attempts to leave critical parts of main characters featureless so the reader can mold it to their preference. So, simplification and user choice. Not a bad set of concepts. Explicit branching paths is time consuming, so I went with interchangeable component sets. Then I went crazy and decided to use audio as the core component. Derp.
I’m not a musician. Sure, I can sing a capella choral, and jazz. I play a bit of piano, guitar, and clarinet. A very little bit. So that was my first major mistake. Still, creating audio loops can be less time consuming than other forms of Content, and it did make the game feel unique, so not a terrible mistake… just an awkward one.

Having good tools makes a world of difference. I built my game using Unity3D, 3DSM, Photoshop, and Reaper with NI VSTs. This allowed me to inherit a lot of polish with minimal effort. The down side, of course, being that you have to put a lot of effort into making it look unique. I spent that effort on the core mechanic (audio tracks), and let everything else retain most of its workflow heritage. I’m not saying tools are everything, but they made my job easier. The content generators provided are a godsend.
Learned a lot about 3D object animation; I haven’t used it very much in the past and there seems to be a big difference between objects and humanoids. Worked out smooth scene transitions within Unity. In general, about 10-20% of my time was spent learning how to do what I wanted with the tools. 20-30% was spent on geometry, and another 20-30% on the audio content. The remaining time was spent on code, logic, textures, and polish. That means I should be able to do more next time, since I’ll know the tools better.
Even with frequent commits to SVN, I lost a couple models and audio tracks that later needed to be recreated and tweaked.
I got so wrapped up in the base content, I forgot some details that would have made it more immersive. There’s no ambient audio, and no interactive sound effects like footsteps, button clicks, door open noises, birds chirping outside, etc. I had a couple hours to spare at the end, but I was so fatigued I just called it a day.

Near the end, I realized I had overshot my scope, and had to trim off about 40% of the content. It removed a little too much content for a final version of a real game, but it was the right choice for this specific event.
I slept enough. That made a huge difference. I was loath to see the hours go away, but this is a creative endeavor and I have zero creativity when I’m sleep deprived.
I considered stocking up on food and drinks beforehand, so I wouldn’t have to leave at all. The drinks were a good call… enough coffee and beer to keep me sated. I skipped on the food, and while having it on hand may have helped a little, the enforced trip outside for grub cleared my head.
Overall, I think this was a great success.
