Prototype Postmortem

Keep Warm!

Keep Warm is a solo created prototype that was made as part of Fam Jam, a small community Discord event. Submissions were open from November 20th, 2020 to November 29th, 2020. I started development work on November 26th, in the evening, and concluded with the event a few hours before deadline. The options for theme were 1) Previous Jam Remix 2) Holiday Themed Game 3) Both.

The Details

The goal for the creation of Keep Warm was to create a COD: Zombies styled game, with snowmen that aggressively pursued you to freeze you, and the ability to create and throw snowballs. The design was intended to have a multi-staged world, with multiple difficulties, and powerups to buff the player for a short amount of time. It was intended to have a slightly more lighthearted style and tone, with it having a sort of fantasy envisioned experience. I would be using my engine of choice Unity 2020.1.6f1, along with assets taken from the asset store and Kenney's own. I ultimately did not achieve the goal of having a COD: Zombies experience, but did great work on making the world seem whimsical and fantasy like.

Finding Experimental Success

As work began, one of the first things considered was the core mechanic of the snowball. It was important that this was done right. Thankfully, resources were available on trajectory projectiles, and that helped boost the project along significantly, and the code was to a limited extent reusable by the UI for projectile prediction. Another thing that helped the project significantly was the powerup design. By the time I was creating the powerups, time was quickly running out. I considered creating a script to handle each powerup, but then architected a "Powerup" script that handled all the actions of collisions and logic. The custom parts of the powerups visually and mechanically were handled by their respective components, rather than the powerup directly acting on gameplay.

Time was quickly running out...

One of the biggest reasons that there were issues in the creation of the game boils down to time available and time used. I ended up wasting the afternoon on a custom lighting engine that would have looked amazing paired with the right models. Sadly, this went unused as the low poly models did a disservice to the cel-shading and lighting. Another significant hang-up was starting way too late. This meant I had way less time than most of the other participants, and as a result, there was a crunched effort to ship the project out. This meant a major drop in sleep (only 6 hours sleep in 48 hours) and a loss in productivity due to brain fog. Add in trying to maintain a new relationship, and things quickly got out of hand.

The risks that were taken!

There were some major gambles that were taken during the jam. One of the risks were that the projectiles weren't going to work. I had initially thought to do physics based snowballs, but I wanted more control on how the snowballs operated. It was a bold move to add the snowball mechanic, arguably one of the biggest parts of the prototype working. Following a tutorial turned out to save the day here, as I would have been farther behind with an inferior project had I gone with the physics route. Another big risk was that I wanted a really stylized world. Sadly, an entire afternoon was taken on a gamble that a technically amazing looking cel-shading and lighting system would help the low poly art to stand out on it's own. It was only after the entire system was made that it was discovered not to look great, and the whole system lays abandoned in the project files, not to be seen at this current point in time.

I had to make a choice...

At one point, I was falling so far behind, that I worried trying to create multiple scenes and complex UI work would take away from trying to polish up the pickup system and combat. As a result, I needed to make a choice. Do I go ahead and try to force what I have into a game? Or would I go ahead, ignore the idea of a fully realized game, and polish up what I had going well? In the end, I was forced to drop most of the waves system and game menus, and focused solely on the experience I had going. Overall, this was the wisest pivot for the prototype as many of the things people praised would have been more incomplete/unrealized had I continued course for where I wanted to be.

How did it go?

Reception post release was generally positive, considering that I opted to basically brand it as a prototype (of which it is!) As for the comments, many noted the lack of feedback on enemy locations and the lack of music. In addition, there was a couple issues with the FPS controller that caused a cascade of issues for level design that then proceeded to impact snowball collisions and spawning, impacting the experience in a minor way.

Taking Aim!

Going forward, there are some changes that need to happen in my personal work:

1) I need sleep. 

Being well rested as a developer is essential to any project, and crunching is never a recommended technique. It's only a sign that you weren't prepared, or that you prepared poorly, or that powers above your own did not want you to succeed. I simply started late on the jam.

2) I need to better weigh what actually needs done and what appears needing to be done.

If I had not tried to work on the aesthetics of the prototype, maybe I could have had a better user experience for the player by making it an actual game rather than broken up executables. 

3) I need more time.

Because the jam was a limited time, I couldn't add much more time at all. If I had had the full week, it would have become more of a game, and less of a prototype.  Time is actually one of the most precious commodities you have, and if you don't use it appropriately, it will run out. As a couple of deceased YouTubers once stated so often: "Momento Mori. Unus Annus." (Remember Death, One Year - Your time fades quickly, and death is inevitable.)