Ludum Dare 35 – thoughts and experiences from my first game jam

This weekend (15th – 18th April), I took part in my very first game jam – Ludum Dare 35. It was awesome.


If you don’t know, the setup of a game jam is that you have a specific amount of time (usually 48 or 72 hours – they normally take place over the course of a weekend) to make a game from scratch. What ‘from scratch’ actually means will vary from jam to jam, but for the competition part of Ludum Dare, it means that all of the code, artwork, sound and music must be made solely by you, and only within the time allowed (so no using assets you already had lying around). Of course, it’s very hard to fully police these rules, so it does rely on an honour system, but from what I understand, it’s never really been an issue.

Everybody involved starts at the same time – for my timezone that meant 2am on Friday night/Saturday morning. I got home from work on Friday evening, set my alarm for 1am, and went straight to bed. Hurrah for keeping any sort of sane sleeping schedule during the jam! From 1am, it was a case of keeping an eye on the website for the theme announcement. For this jam, the theme was ‘Shapeshift’.


I wasn’t a fan of this one. Themes are chosen by a series of voting stages the previous week, and there were three or four I really hoped would come up. This wasn’t one of them. There was a bit of panic during the first hour or so when my mind would just¬†not work.¬†I couldn’t think of a single good idea. The most obvious idea was a puzzle/platformer in which you need to transform into different states in order to navigate the levels, but I figured that half the submissions would be variations on that.

Eventually, it was Metal Gear’s cardboard boxes that inspired me. I settled on making a stealth game in which you can transform into a variety of mundane objects in order to evade or eliminate enemies. It took me a few hours to get the basic mechanics and enemy AI in, but once that was done, I actually felt like I had something with potential here. Here’s a .gif of the game after about three hours or so:ezgif-3656978035

What surprised me most about this experience was how few problems I ran into. Before starting, I honestly thought the chances of me making anything even resembling a playable game in 48 hours were next to zero, but everything went really smoothly, and by the time I went for a short nap at 9am (seven hours in), I basically had a functioning game. There were still lots of mechanics to add of course, but the foundation was there. I was miles ahead of where I had expected to be at this point.

With the exception of a few frustrating bugs, the whole development process plowed along at a pace I didn’t expect. During the 48 hours, I slept for maybe ten of them, and apart from a couple of showers and meals, I didn’t stop working. I had a to do list that was divided into two sections – essentials, and ‘nice to haves’. In the end, I managed to implement everything on the list with the exception of two items from the ‘nice to haves’ side. They were animated legs for the player and enemies, and more levels (I had set a minimum of ten levels, with a ‘nice to have’ number of fifteen, but in the end I just stuck to ten).

Here’s a .gif of one of the more interesting bugs I encountered…

One of the most challenging parts of the jam was coming up with a name, and looking at the other submissions, I reckon lots of other participants will agree with me. About half of the games are variations on ‘Shapeshifter’ – you’ve got ‘ShapeShift’, ‘ShiftShape’, ‘ShiftShaper’, etc. My own name isn’t much better – I settled on ‘Shifty Ninja’ after spending several minutes stringing words related to the game or the theme together – as soon as I said ‘Shifty Ninja’, I thought it had a nice ring to it.


I finished the game with a couple of hours to spare, which were spent playtesting it as much as possible. I’m sure more than a few bugs escaped my attention and made it into the finished product, but it’s playable, and (in my opinion) fun, so that’s the important thing!

A few thoughts on what I did right for this jam…

  1. I stuck with my idea. Within a few hours of starting work on my stealth game, I had three or four other ideas and had to stop myself from abandoning what I had and starting again. I’m glad I stuck with my original idea rather than wasting hours flip-flopping between a bunch of half-baked concepts.
  2. I stuck with what I know. Top down movement? Easy to code. I only had 48 hours, so there was no way I was going to tackle a genre I have no experience with, as half of the time would have likely been spent just getting the core mechanics up and running.
  3. I kept the art simple. Art is not my strong point, so I kept it at one tileset, one wall sprite, and enemies are just different coloured versions of the player. Trying to do any more than that and I still wouldn’t be finished.

And what I did wrong…

  1. I spent almost no time on design. I had my idea, and immediately started coding and building. Maybe I work better like that, but I think spending a couple of hours on an initial spec would have saved me some headaches later in the jam.
  2. I tried to power through tiredness rather than taking a break a couple of times. At about the halfway mark, I was really struggling with a bug, and wasted maybe two or three hours in a very lethargic state trying to fix it. After a four hour nap, I fixed it in five minutes.

Overall, it was an incredible experience, albeit a very draining one. I’ll definitely be participating in future jams now that I know I can actually make a game from scratch in 48 hours.

If you’ve got a few minutes to spare, I’d really appreciate it if you gave my entry, Shifty Ninja, a go!

Here’s the link to the Ludum Dare page.

And here’s a link to the page.

Finally, here are a few screenshots:




Leave a Reply

Your email address will not be published. Required fields are marked *