Background

This is my 5th year of GMTK Game Jams. Last year was my first year of providing a web build of my game, and it really felt like that made a huge difference. My rankings last year were all very good with nearly 4 stars across the rating categories. I was super duper proud of what I had made and it reminded me that no matter how skilled you are there is always room to improve and get better. I’ve always had an affinity for doing things from scratch because I like to get into the nitty gritty, so again this year, I’ve used my own custom built game engine. This is the same engine as last year and it continues to prove itself useful to me.

Prep

Generally, the week before the game jams I do a lot of prep work on my game engine to make sure if there are any new ideas I can get them started and working well enough. In prior years I’ve done things like adding controller support, overhauling the audio systems, making a base project, and creating a bundler. This year the prep work was considerably smaller. I have been using my game engine throughout the year for a hobby project I’m working on so I’ve made smaller changes here and there as necessary, but the major reason there isn’t a huge need to add anything big is mainly now that the engine and platform is the web browser a lot of heavy lifting is done for me. No more do I need to write audio engines that manipulate byte arrays, nor deal with window frames and OS APIs. This has drastically simplified the engine without sacrificing too much control.

Theme

The theme this year was a mixed bag for me. I found it really hard to think of ideas and didn’t really like it as much as prior years’ themes. Usually I’m able to come up with an idea that I like decently well and then execute on it very early on. This year was different and it didn’t feel good for me. I spent the first 5 hours thinking through ideas, how to execute them and also what scope I may be able to achieve in the jam period. I had come up with a few ideas such as: a kingdom builder game, where you start out decorating a house, then a town, then a village, then a kingdom; some sort of Katamari-like game but 2d where sprites clump up together around a central rotating blob; an adventure-puzzler where you’re a UFO and scale up/down to explore different scales to help people with odd things; A Spore cell-stage like game where you control many cells together instead of just one. I started with the Spore-like game and spent a couple hours in that before throwing in the towel on that idea. I spent another hour thinking of other ideas that seemed reasonable. I was thinking about doing a paper-crafts like game where you stack different shapes of paper with the ability to scale and rotate them to make different images. The idea also included scaling up in a way that you could use previous combinations as a group with the same effect. I thought the idea was promising, but I couldn’t think of a good way to produce images and “score” them. The idea I finally landed on was a platformer where you scale your sprite to squeeze through tight places. All in all, I didn’t really get a start on my main project or code until about 7 hours into the jam, but no worries because we had double the time.

Jamming

I felt like the jam this year was a lot more relaxed than previous years. With the extra 48 hours I was able to get more sleep, and do more things with friends and family. I even got to go to a brunch that I accidentally planned. As far as the actually jamming is concerned I felt like I was able to maintain a pretty steady pace up until the final 6 hours where I started to really crunch to eke out as much polish as I could.

Technical Details

I spent the first day making the movement feel good before I started on any real level design. I made a playground to test out the different movement and screen transition before I really started on anything like spriting, level design, music, or sfx. Afterwards I worked on improving things in slices where I would try to make the overall feel of the game seem more complete with each iteration. So I would work on things like adding just enough sprites to add some overall coverage, a little bit of animation, a level or two. It wasn’t until near the end of the jam that I added all the levels I could squeeze in and then fleshed out the rest of the sprites to polish it up some more, I added the world select, the credits, the title screen, and a world select with saving.

Handling Scope Creep

When I first got started on adding levels, I was thinking I wanted to have a hub world that you could traverse through to get to the different chapters. I realized this wasn’t going to be feasible because I would need to either reuse a bunch of sprites or hastily scrap some stuff together. I scrapped that idea for just simple chapter sections that would cover the basis of the progression. With that in mind, I was going to have 4 real stages per chapter, but after spending at least an hour per stage I realized I wouldn’t have enough time to get a fully fleshed Chapter 3. So instead I made the first chapter teach the basics, to act as a sort of tutorial, and then tried to scale up the control and difficulty of the later chapters to compensate. Thankfully, I was able to get my wife to do some play testing and realized that there were some extra difficult parts that didn’t quite fit in with the surrounding difficulty so I was able to make those parts a tad bit easier. After playing my game for nearly 50 hours I couldn’t quite tell which parts were too difficult or not because as I was designing each section I was also figuring out how to overcome the obstacle in the easiest and most reliable way. I also got a lot of great ideas from my wife about different ways I could incorporate scaling, but unfortunately there just wasn’t enough time to include any of them.

Rating Period

I don’t know what happened this year, but in the first 4 hours or so of the jam I got about 20 ratings without really advertising my game. Afterwards, the trend had gone down each day despite some light advertisements on the discord and through playing other people’s games and leaving comments. A couple days into the rating period I had a couple of friends play with various levels of feedback but no ratings. At the time of writing I’m currently sitting at 48 ratings which I absolutely cannot complain about. This puts my game in the top 5% of games by number of ratings.

Given my luck early on, I didn’t need to do much in the way to really push for my game and I’m very grateful for that.

Expectations

I think it’s a bit of a mixed bag. I think I’ll have a better overall rating last year than I will this year. I think what I made for this year’s jam feels a lot better to play, but is more niche and not as accessible for people. Platformers take a lot more skill than what was essentially a walking simulator + puzzle box. Since I used prepurchased assets the year prior, I also expect to have lower scores on visuals despite this year’s visual scoring to be “style”. I don’t dislike the graphics, and I think they’re quite fitting for what the game is, but I can’t quite tell how that will translate to scoring. My ratings from last year look like this, and quite frankly I think it’s a hard act to follow up.

Overall I’m very happy with what I created this year despite my initial displeasure with the theme. I’m excited to have another shot at this in the years to follow. I’d like to thank Mark for taking his time to host these jams and provide a great community for sharing and growth. I’d also like to thank my wife for dealing with my obsessive grind where I practically only paid attention to my computer for 4 days.

Stay tuned for a follow up postmortem for my thoughts on the rating results of the jam, and perhaps a second postmortem for the final results of Mark’s top 20.