This is my second time participating in Ludum Dare, and I have to say, it was quite an enjoyable experience. The first time I participated, I did not finish my game. I learned from my mistakes and managed to finish my game this time around. The game could do with some more enemies, tower types, and levels, but overall, I feel like my attempt was successful. Unfortunately, I got sick on the first day, so I had lost a sizable chunk of my time. Still, I managed to finish my game. I have to say, I’m quite pleased with myself.

After 3 days of hard work, I've finally finished my game:

https://ldjam.com/events/ludum-dare/38/$26457

Thoughts:

I am honestly very proud of myself for completing this. This is my 3rd time participating in Ludum Dare, but it's the first time where I actually felt like I had created something which was fun.

What went right:

I think what I did right this time as opposed to previous times was that I cut out a few of the unnecessary features on the planning stage. Content like homing rockets and complicated bosses are fun to think about, but take forever to implement and generally really aren't worth it. I really got lucky here in that I didn't make things too complicated or this game would have taken much longer to program.

I also spent a lot less time actually writing code and a lot more time testing my game to see if it is balanced and fun. Overall, I think I made the game slightly difficult, but still playable. I spent around 4 hours in total just trying to balance out the game.

What went wrong:

The graphics really weren't that great. I had originally planned to collaborate with an artist friend of mine, but turning his art into digital graphics suitable for games. On the bright side though, I learned quite a lot about GIMP and fixing images. After a few images, we just sort of gave up on trying to make the graphics look nice and I just drew a few quick images in GIMP. Sorta explains the discrepancy in the art style of certain objects, if you've noticed that.

I also tried my 'hand' at making a font out of my own handwriting so the text would look handwritten, but in the end FontForge just gave me a bunch of errors and produced some funky-looking overlapping font. Teaches me to learn to use my tools before an event :P.

The last thing I messed up was actually with the game engine I programmed before the event. Unfortunately, apparently I made some of my collision detection algorithms O(n^2) or something, because my game lags really bad once there are around 50 bullets or so on the screen. I also messed up the preloading of images. Apparently my game doesn't always load images before using them, which sucks because then the player will just see blank spots rather than the correct images. I still haven't figured out how to get images to preload and use loading bars to track the progress... would really appreciate it if any of you in the comments told me how.

Closing thoughts:

I was pretty satisfied with my performance. My friend is going to get a drawing tablet next time so he'll be able to draw effectively on computers. I still have a few things to fix in my game engine and other things, but overall I'm pretty happy about how I have done.