I took a while to get to this but here goes. I’ll keep it short since I don’t have a whole lot to say, and we’ve all got a rather large number of games to vote on.

This is now the third LD48 I’ve participated in, but only the first time I’ve been able to finish a compo entry. Usually my lack of staying power is either due to time constraints, or in the beginning poor knowledge of my tools. This time around, I was firing on all cylinders. On that note, let’s analyze what went right.

1. Theme. This worked for me. I had been mulling over the idea of an ant colony game a few months ago, and took the time to do a bit of research into the little critters. This led to the central mechanic of the game, not actually controlling the ant, but rather controlling one of the “standard bearer” ants using pheromones to lead the other ants to an objective. This was in no small part influenced by the units in Darwinia.
2. Tools. This could have gone either way, and fortunately it worked out. Reason being, about 3 weeks ago, my computer died. Harddrive started to die and fortunately I managed to rescue my important files before anything happened to them. But the thing was already a write off (kind of a franken-netbook, NVIDIA ion mobo with a dual-core Atom. Looked great in the shop, but got her home, not so much) so I reallocated some of the hardware budget this year for a Mac Mini. Trial version of FlashBuilder took care of my IDE since I couldn’t get Xcode to do it properly in time (not giving up on that one just yet) and found a nice little pixel editor in Pixen that was nice and easy to work out so no impediments there. I used Flixel, and while there are still many aspects of the the framework I don’t use, my knowledge of it is now adequate to streamline everything nicely.
3. Confidence. This is a big one. From the very beginning, I maintained that I would finish and deliver a game. I went in with the knowledge that I may have to cut features that I didn’t have time to complete, but it wouldn’t detract from the overall product. I think testament to this is the fact that I hit my first roadblock about 13 hours in, decided to go to bed, but came back after a few minutes away and figured out the problem. From that time on it was a steady journey.
4. Interaction. This could have been better, but it was not bad. I maintained a journal of sorts on my blog, complete with progress videos of what I had achieved up to that point. I think this also helped with the confidence, since it was something I could see as concrete evidence that I was progressing, plus there was a sense that other people were watching too.

There’s probably a few other plusses in there, but I’m going to venture that they fit into those main categories.

Now for what went wrong, and how the game failed to deliver a quality product.

1. Style and Polish. I’m making this purposefully vague, because for me it encompasses a number of deficiencies. I don’t like to think of myself as “not the artistic type”. The mindset implies that the inability is out of my hands and nothing will change this. <geek> Neji vs Rock Lee for the Naruto fans out there </geek> Ok, got that out of my system, but it’s true. Where I’ve failed on this front is to actually do something about it. I should be making pixel art in my spare time, I should be playing around with synthesizers and music generation software to get my skills up. I’ve done it with my programming skills (and there’s still room for improvement there, to be sure), now I need to do it with my creative side.
2. Fundamentals. The comments on my entry are pretty consistent as the to the level of Feedback in the game; basically there’s none. What is there was hastily tacked on in the last minutes before I submitted. Due to time-zone realities, I lost the last 6 hours of the compo; I live in Japan and start at 10 am saturday, but finish at 10 am monday. And a scheduling incident meant I didn’t have this monday off. So, I wrapped up at 4 am and tried to get a bit of sleep. I don’t think staying up any later would have made a difference, but it’s possible that if I had slept sunday night and did some work early monday, it *could* have been different. None of this of answers the core issue, and that is that the feedback and readout mechanisms were not included from minute one.
3. Time management. Somewhat related, but there’s enough here to make a separate point. At a couple times, I spent too long working on a non-essential bit only to have to rip it out and toss in some jury-rig to let me move on to my next objective. Most of these incidents came later as I got too tired. Having a somewhat flexible plan before sleep-dep kicks in would have worked better.
4. Too much coffee. I love coffee. I really do. You can have your red bulls and monsters and whatnot (although I do love me an Oronamin C every once in a while..), give me coffee. And I’m not talking Nabob or Nescafe or any other that, or even Starbucks triple-dopio-mochi-whapaninos, but a french-press-made-to-the-consistency-of-battery-acid brew. This had two major (negative) side effects; caffeine crash and burn is nasty, and my stomach has had a passionate loathing of me for the last two days or so. There’s no way I’ll be ruling this out of future endeavors, but pacing. Pacing.

So, there you have it. My first post-mortem for my first complete LD. If you’ve made it this far, thank you for your patience. If you’ve read all this and played the game, you are a champion. Pat yourself on the back.