LD42 Post-Mortem

Posted on August 20, 2018

Play Hovertank 42 in your browser!

Another Ludum Dare has come and gone and here I sit reflecting on how it all went down...

To be honest, it went pretty well - and I'm proud of how things turned out. I continually struggled with time management in past LDs and a lot of that was owed to picking an idea with far too much scope.

This time around I forced myself to stick with something really small with a replayable gameplay loop instead of a set of levels. In Ludum Dare 41, I attempted to make a platforming game that required fairly meticulously designed levels. The time sink that created left me unable to get the game done in time. I wasn't going to let that happen this go around!

The Breakdown

The idea phase is typically one of the major hang ups for me when beginning Ludum Dare. This time around it went swimmingly. I didn't sit and beat myself up over the theme sucking or any of that typical nonsense - I just dove in head first and started working. I talked myself out of making an inventory management game and soon stumbled upon the idea for a shooter where the bullets created walls. I mulled around with what perspective and style would work best and settled on a top-down shooter. I figured it would best illustrate what was going on with the walls.

The initial features started to come together quickly. Within the first hour or so, we had our player moving, aiming their cannon towards the mouse, and firing a basic projectile. This was probably one of the fastest idea to prototype phases I've managed to pull off in a LD - and it made a huge impact on me wrapping my head around the scope early on.

The hang-ups started to occur when dealing with the spawning of walls. I knew early on that the wall would need to spawn in at a direction inverse to the bullet. Initially, I did this partially correct but used the wrong set of Quaternion functions to translate it. Needless to say, after plodding through a couple different maths and tricks to get it working I went back to the original code with a different (and correct) rotation and got it working wonderfully. Unfortunately, this cost me about an hour or so of tinkering and bug fixing to get done.

In the first night I had a moving character and a partially completed main gameplay hook. That was the best first 5 hours I've had in any LD ever - so I went to bed with a huge smile on my face and a litany of features to get coded in on Saturday.

Sadly, Saturday didn't quite go according to plan. I knew I had some important apartment/moving business to attend to and had set aside some time to do so. That time ended up moving further and further back and when it came it ended up taking longer than I had budgeted for it. The whole ordeal cost me probably 3 to 4 hours on top of a fairly listless morning where I got some things done and took a break to play Monster Hunter (hey, I'm only human).

Given all this, by late Saturday night, I had basically completed the core gameplay features and started to implement the enemies. At this point, I had to cut a couple frivolous features from the game, mainly different bullet types and a pick-up that would allow players to temporarily pass through their own walls. All that was left to do was art, sound, and UI stuff.

The bulk of this work took place early Sunday morning before bed and in earnest early afternoon after some sleep! As I was super limited on time, still crushing some bugs, and not really in the mood for 3D modeling, I took the easy route on art and just made my prototype stuff look better. I think this decision was the best in order to get this puppy completed. While the art isn't stellar - it's cohesive, thematic, and not an eye sore. This decision also left me with more time to make a solid UI and background music. The tunes and sounds came together after a little struggle. It's simple, punchy, and thematic. Not my proudest work - but not my worse, by far.

Unfortunately, I still didn't have enough time to get it turned in before the 48 hour mark as I had hoped. By the time the 6PM EDT deadline rolled around I was still patching up some last minute bugs and getting sounds and effects in place. I finally got everything squared away and ready to publish by 10PM that night, missing the deadline by over 4 hours. So, I submitted to the Jam, instead - no biggie - I just finished a LD for the first time in almost a year!

Now for the TL;DR.

What Went Well

  • Idea phase was short and I got to work quickly
  • Pacing was fairly good throughout, a couple hang ups cost me some polish time - but all in all I stayed focused and got the job done
  • I did get to stream a little bit, storms prevented it the rest of weekend and the setup needs a bit of tuning for better stream and sound quality
  • The game has very few bugs reported
  • Got my 20+ ratings - WOOT
  • People find the game fun and challenging; good comments on controls and game feel

What Could Have Gone Better

  • My understanding of quaternions needs improving so as to eliminate the trial and error guesswork of using quaternion functions in Unity
  • I could have had more time to polish the art and add more visual feedback (i.e. player taking damage)
  • I had planned to get controller support implemented, but alas, ran out of time
  • Build process should have gone smoother (didn't have WebGL or Mac build package installed, had to upload those the next day)
  • While my pacing was sufficient, I felt too comfortable taking prolonged breaks and had a couple time management slip-ups that need addressing for the next go around

What's next for Hovertank 42?

I was initially thinking about just letting this one ride post-LD. However, since submission I've come up with a couple ideas on how I could improve and add more content to the game. In that time, I've also implemented controller support into the game. If I manage to take another swing at this game it will definitely feature more modes, more enemies, and perhaps some local multiplayer goodness! 

Of course, PARABOLA, takes precedence over this - so don't hold your breath!

LD43 and beyond...!

For the most part, this was a great LD experience and it was awesome to get back in the swing of getting one done after so many failed attempts. I've had loads of fun playing through everyone's entries so far and greatly appreciate those who have played and rated Hovertank 42!

Already hyped for LD43 in December - just need to brush up on some stuff so it goes even smoother!

Michael


Dig this article? Share it with your buddies: