Savior Postmortem

April 25th, 2015 4:44 pm

I was pretty happy with my entry for LD32, so I figured I’d write up a more formal postmortem this time; partially because it might be interesting, and partially because I’ve learned a ton from this compo and want to get my thoughts down. So, without further ado, here’s what worked and didn’t work in Savior: On the Advantages of Unconventional Weaponry in Combating the Zombie Apocalypse.


How It Started

I’m not the best programmer and am somewhere between average and solid when it comes to art, so for jams like this one I like to try to come up with a kind of clever gameplay idea that can be interwoven with story (in fact, I just finished my senior thesis on this very subject at university). I think games can be very effective when story and play go hand in hand, informing each other, and I like to play around with that on a small scale for Ludum Dare.

I’ve been sitting on an idea for a game where you play as a scientist curing zombies for quite some time, but never figured out the mechanics of how it would work. With this theme, “an unconventional weapon,” I first wanted to make a sniper-style game where you take pictures instead of shoot people, but I realized I wasn’t familiar enough with the coding I would need to do to make it work. I decided a few hours in to give the idea of saving zombies a go, using a zombie curing serum as a “weapon” that actually helps the zombies. It quickly became a stealth action platformer, and then I was off to the races.


Level 0


How It Changed

My original idea was to do a somber, gritty zombie story. I was going to rotoscope animations in Flash so that my characters would look realistic and the animations would be fully fleshed out. As soon as I finished my first walk cycle it became clear that this wouldn’t be feasible within the time frame; I can animate fast, but not THAT fast, and I’m new to Unity’s animation system anyways. So I took a long break and struggled to come up with a new art style.

The result was this guy: a sort of bean like thing who I grew attached to over to course of the compo.


Some animations in sprite-sheet form for the Player Character


The thing is, this guy looks silly. You can’t tell a gritty zombie story with a bean as your protagonist. So the goals mutated. I started playing up the humor, adding the bloody messages and dopey signs to the backgrounds of levels and trying to entertain myself as I drew. If I could make myself chuckle, maybe I could make other people. I took a lot of inspiration from games like Portal and tried not to take anything I did too seriously. The result, from feedback, seems to have been pretty good!


Stuck with a blank white wall in your level? Write a dumb joke! In blood!


The End

(spoilers after the break — play it now if you don’t want the shocking ending ruined!)

Despite the shift to comedy, I still wanted to play with the theme a bit more. The healing syringe is an “unconventional weapon” in name, but in practice it plays very conventionally. So I wanted to play up the difference in the idea of saving vs. killing zombies with an ending that is somewhat shocking or unexpected. The idea was to still be funny while also making something of a point.




The result, I think, worked well. The player is turned into a zombie and then the care that they showed zombies while they were alive is not shown to them; instead, they are shot dead and the game abruptly ends. Its silly and a bit sad, but also wraps things up nicely for a compo game.


Those damn conventional weapons…


What Worked

  • The game is pretty polished, with a bunch of levels and a clear beginning, middle, and end
  • Art game out nicely in my opinion, both in terms of animations (of which there are many) and background art
  • Really happy with the ending
  • You might think its funny!

What Needed Work

  • The physics are super slippery. This is a thing I got used to after hours and hours of testing and stopped seeing it as a problem, but its clear that it needed to be tighter. I’ve put up a post-compo version that should alleviate some of the control issues.
  • The rules aren’t clear. Another issue that resulted from playing the game as a designer and not as a player. I know its supposed to be stealth-based, but I never tell the player that they have to attack the zombies from behind. Oops.
  • I may have too many levels. I got worried about it being too short, but unfun levels that just serve as padding do not a good game make, particularly when I want players to reach the end.


In the end, this Ludum Dare was a huge success. I made a game that I enjoy and people seem to be liking, but more importantly I’ve learned a ton; about programming and animating in Unity, about making an effective asset pipeline (I was able to very effectively create assets in Flash and move them into Unity), and about anticipating and designing for how players will play your game having never seen it before. I had a great time and hope to be back again in August. In the mean time, I’m playing tons of your games and hope to keep playing more.

Finally, here’s a video walkthrough of the game if watching suits your fancy:

Tags: ,

Leave a Reply

You must be logged in to post a comment.

[cache: storing page]