Whew. With that out of the way, let’s get on to the Post-Mortem.

But first, for the obligatory action picture!

Despite what the Professor may think, 48 hours to develop a game may be one of the best ideas I’ve ever encountered. I learn more in those 48 hours than I do in whole months.

What went well:

• My dialog writing. It was Saturday night, I hadn’t slept well the previous day but I was feeling full of energy because I had a ton of coffee a bit earlier. In this weird hazy sleepy/alert state I got on a roll and wrote pretty much ALL of my game’s dialog in the space of an hour or two. It started from a little seed where the professor made fun of the protagonist because he couldn’t even jump. I realized that the mechanics of a pretty dumb protagonist + annoyed professor could be hilarious, and I ran with it.
• My graphics. Finally. I’ve churned out a lot of crap graphics in past Ludum Dares. I’d always tell myself “I’ll come back and finish it later…” and we all know how that turns out I am very satisfied with these graphics, with the exception of the clouds on the upper background. I obsessed over almost every tile that I drew, going through many iterations and tries until they looked good.
• Prioritization.  I’ve had big issues prioritizing what needs to get done in the past, especially LD23. I came up with a really great solution about halfway through. The way that I prioritized was that I kept a huge list of things that I needed to fix in order of priority. Top was most important, bottom had crazy stuff like “KILL INNOCENT CREATURES”  (don’t ask).  Whenever I had a new idea or something that I should fix, I added it into the list at the correct priority. On the final day, I went as fast through the list as I could, from top to bottom. This REALLY helped – it probably saved the game from disaster.
• The level design. That was a surprise. I’ll get back to this.

What went poorly:

• My sleep. I slept a total of 11 hours during LD: 7 on the first day, 4 on the second. I just can’t function without sleep, and I’m sure that I could have done better with more sleep. I’m not sure how to fix this though. I woke up going “OH MY GOD IT’S LUDUM DARE!!!!!!!!!!!” and was unable to get back to sleep. Too excited.
• Scope creep. Holy crap, you should see all the stuff I had planned. Lava, monsters, death, multiple soundtracks, different areas… what was I smoking?
• Scope creep. I know I just said this. It was that bad. At the 24 hour mark I was seriously considering giving up. I knew that I had absolutely no hope of completing my game. Instead, I totally changed my game, trashing TONS of features that I had planned. (A good example of this is how, when you switch to Ice, you make grass icy. This used to have a gameplay effect, but then I decided it was too much work and tossed it out. I kept the ice because I thought it was cool.) I tossed all my crazy and epic level plans, and instead made levels incredibly simple. I decided to toss any hope of making this a serious game and wrote tons of silly dialog… which was actually a huge success.
• Music. This is depressing, because my production skills have really improved since LD23, and I wanted to know what people thought. But I forgot to schedule some time to do it on the first day or early second day, and we all know what happens to things that we put off in LD.
• Sound effects. I tossed these in last minute, and it shows. Pretty much everyone hates them.
• Animations. These are really important for a game to appear polished. I totally forgot to do them.
• UI. My God, EVERYONE hates the card selection screen. I remember thinking to myself while writing the game, “wow, this screen is annoying to use.” Why didn’t I fix it? I think what happened is I designed it before I had my priority system in place, and then after that I had a learned blindness to how annoying it was to use.

What I learned:

• Levels you make to be hard are impossible. Levels you make to be medium are impossible. Levels you make to be easy are perfect. In LD23, I made some really hard final levels. I don’t know how many people beat them. I think most people just gave up. This time, I made tons of REALLY simple level designs… and most people are saying “wow, the level design was just right!” That’s truly eye opening.
• Have a very high sensitivity to things that seem wrong. If something ever seems wrong, immediately jot it down so you can fix it later. If you notice the controls are slippery at one point and don’t fix it, you’re going to get used to it. But people who play your game wont – they’re not going to play your game long enough to get used to it. Fix it ASAP!
• Your first test of a gameplay element is the one that matters. The first time I tested my card selection UI, I was like “Wow, this blows!” But after dealing with it 50 times I was like “Huh, this isn’t so bad.” Of course, no one other than me is going to deal with that screen 50 times.
• Get at least one beta tester. If you can’t see the problems from the upper two points, this is the 2nd best solution. One guy who tested my game afterwards immediately suggested a fix to the card selection UI – put arrows below to mark which ones were used. It was so obvious that I facepalmed. That is the power of a second pair of eyes.
• Implement the biggest stuff ASAP. Later  you’re going to be bogged down in minor bugs, and you won’t have time.

If you read all that, I am truly impressed! If you just skimmed the bold parts, well, that’s pretty good too. Thank you for reading. Here’s the game!

This entry was posted on Sunday, September 2nd, 2012 at 9:29 pm and is filed under LD #24. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.