Fixing duct tape with elbow grease


So this version is in some ways a postmortem of my experience this year doing Global Game Jam. The theme was Transmission and I (being the crazy maniac) did it solo. The final release is in many ways pretty close to what I was imagining when I was planning out the first concept:

  • You are the operator for an Alert system to warn cities
  • To do that you place Radar systems
  • The Radar systems alert you as there are false positives
  • After that you need to choose which Cities to alert to power up higher defense levels
  • At the cost of power to progress and finish the game successfully

So in theory a neat concept and spin. What ended up happening is a case of accidental over-scoping through displaying and working with the mechanics. This came almost entirely from the generation aspect of the game - generating objects and scripts is easy to implement, but to make sure dynamic generation works as expected is a different matter, especially when I had to deal with scripting basic AI, scripting basic targeting for the defenses, a game loop, and the cherry on top being a text writer for the command prompt style UI I was going for.

The crazy thing is technically, I did manage to write all the scripting for the game, everything I set out to make was made, and stable, just insanely buggy and holding together. In the end I put together more of a rudimentary simulation of a game than anything else.

So what are the lessons and what should be learnt?

  • Get Gameplay first - I followed this rule before and had luck making gamejam games, but with this one I feel into the trap of doing assets first, code later, accidentally forcing yourself into a rigid mindset of "i made these assets, so I should use them", rather than sticking to basic cubes and other basic meshes and sticking to basics
  • Too many mechanics - Turns out trying to make 2/3 interweaving mechanics is kinda hard

But I did get a lot done, so what went right?

  • Foundation Code - Getting the game loop scripting and the singleton behind the command text writer set me up to then script the rest of the gameplay insanely quickly - if a new big of code had to be called during the gameplay state, add it to the switch statement in the game loop manager, and I was set, big changes made easily.
  • Single concept - having one concept set quickly meant it was all a case of making progress to it, the idea was fairly solid, so that meant it could be worked towards without much issue in the way of "will this be ok or won't it"
  • CRT - themeing the game around a CRT and doing just keyboard controls made it easier to make tenfold: the audio could be distant and simple, just keystrokes, and the visuals and printouts could all be done from Unitys most trustworthy and easy to use portion of the Input system - good old KeyCodes. Likewise having it be keystrokes meant putting the input and gameplay was just a case of subscribing to one of a few delegates, simplifying some mechanic scripting such as bunker missiles / radar placement / alerts super easy to mix with rendering a line from the camera and hooking it all up.

That's what I've got right now, and there's probably a lot more that could be learnt from the past few days, the main thing is another Jam game under my belt, another bit of progress in the wisdom of gamemaking. And now I can rest as I get ready to do a longer, slower paced game jam in a week from now :P



Get Cry Wolf

Leave a comment

Log in with your itch.io account to leave a comment.