I first signed up for LD something like a decade ago but this is the first time I actually made the time to finish a game. I’m pretty happy with what I turned out, but I thought I should make some notes for the inevitable second attempt …
Time-management-wise I was happy that I found enough time to finish a game without it utterly consuming my weekend. I think I may have sunk a bit too much time into playing around with the game-in-development when I could have been adding to it – certainly the timelapses I’ve watched don’t seem to involve as much, let’s say, “testing” as I did. The one regret is I didn’t have time to add audio. I know that I could have thrown some appropriate sound effects together fairly quickly using sfxr or one of its variants, but I basically had no plan for creating music. That’s something to figure out in advance of next time.
Also on the topic of knowing my tools, Game Maker was pretty quick for throwing together something that got sprites onto the screen, and some of the things that bug me about it (like not being able to create functions with named arguments) were a lot easier to ignore when I knew the limited scope of the project. I spent quite a while during the compo figuring out how the tile and particle systems worked, though, which I could have done in advance. I think the game probably still contains a particle type called SMOKE_LEFT which actually goes to the right, and vice versa.
I think I made a sensible choice in the game idea since it was easy to get the bare bones working but had a lot of scope for adding stuff. I might have been a bit laissez-faire about this, though; I had key content such as tiles missing until late Sunday evening because I figured “there’s time to get that done, putting a dragon in it is more important”. (Anyone who looks at the project file will notice how chaotic the tile palette is, which is the result of figuring I wouldn’t need most of the possible configurations and only drawing the ones I needed when I needed them.)
As far as gameplay goes, I think it’s fun, but having not played a lot of LD games previously there’s definitely more of a trend in the more polished games towards ones that can be finished in some sense. I went for a score-attack based game because I figured that with limited time to create content, it’s easier to produce more fun for the player if the challenge is repeatable. Having watched a couple of streamers play my game, though, I suspect some people were disappointed by the lack of a “you win, stop playing and rate it now” point. If I do an updated version (hopefully!) I’ll add a medal system giving you various scores to aim for, and maybe a gallery showing how many of the possible forms you’ve seen.
I think I also needed to telegraph the objectives a bit better. When I was picturing the game initially, I intended there to be a big arrow on the screen at all times pointing you in the direction of the Power Matrix. In the end I didn’t add the arrow because I wasn’t having trouble finding it anyway, but that’s probably because of knowing the map a lot better than most players. As a result I think some people were confused that they didn’t get any points for killing enemies, because the main objective of tracking down the matrix in order to transform wasn’t flagged clearly enough.
Oh, and the platform physics are too slippery. No idea how I didn’t realise this while testing but the tank in particular is way too hard to bring to a stop. Fortunately, the player can fly a lot of the time anyway.
Anyway, it’s been great fun, and I’ve enjoyed playing and rating the other games in the compo too. Now to wait (with a little trepidation) to find out how I scored …
and a terrible display (default UE4 level) of the camera angle to be used 
.