I wanted to wait some time before making this post. Mostly to get some feedback from players and also to take more time to look back at that week-end.

What I prepared
First, this wasn’t my first game jam but it was my first Ludum Dare et my first solo jam. I’m a programmer and I suck at pretty much everything else (including level design). I mostly make HTML5 games and I’m used to build my games from nothing. So I don’t use mapping tools nor I use canvas libraries.
So before the weekend, I wanted to make my own warmup and check out some tools. I took a look at Construct 2 following the two first tutorials. I found it nice and quite useful for prototyping. One reason I tried it was Rytlock’s Critter Rampage made by ArenaNet as a side retro game to one of their in game event in Guild Wars 2. It’s made with Construct 2 so I saw what could be made with it. Though, the free version seems to be way too much limited to make a complex game.
I also took a look at Tiled. I’ve tried this tool before but never used it for games I’ve made because i didn’t like working with XML. This time I asked myself if there was a way to make a JSON exporter until I found out that this feature was there out of the box. That was a good point.
Finally, I brainstormed a bit on what game to do. I wanted to make a stickman fight game, so either a beat’em all, a platformer or a vs fighting game. The main problem was I wanted a stickman engine to render it in real time and not use sprites. A bit too big for a 48h game with my experience. So i thought about a platformer with stickman sprites.
How went the weekend
For the weekend, I wanted to be up for most time. I know sleep is important so I planed good nights of sleep and even cleaned my appartement and prepared better food/drinks than usual. Being in CEST timezone, LD started at 4:00 am. I woke up at 4:30 after a good 9 hours sleep. At first I wasn’t inspired by the theme. But then I remembered my platformer idea and went for a simpler form : take a square. How to move it ? Make it roll. Not quite efficient, circles are better at that. Oh, why the square wouldn’t be jealous of the circles and make it beat them all ? That’s how went my brainstorming phase.
With that idea, I first wanted to see how it would look and if it was any good so I went for a prototype with Construct 2. I made the sprites with GraphicsGale (first time with it) and then imported them. The prototype was ready in 30 minutes and it was looking good. Great occasion to post it already as my entry. As there was a low number of entries, it had a guaranted visibility. Thanks to it I had some early feedback so I knew the idea was good enough.

Then went some long hours on reaching the prototype status with everything coded by myself. Took me most of saturday. I used Tiled to make a test level and see how i could use all the parameters present in the JSON object. That was quick and it took me no time to get the level displayed.

After a not so long night of sleep, sunday brought all the game mechanics implementations, some bug fixing (not much done finally) and all the polishing. The final version was uploaded nearly 30 minutes before the end of the 48 hours. By the way, I didn’t know about the 1 extra hour for submission so I really thought i had to be all set at the end of the 48 hours. Now I know for next Ludum Dare.
And there it is, ^2 is published, with it’s bunch of bugs but I quite enjoy what I’ve made.

The good points
Now, the good parts of the week-end.
First I learned well about tools. It made me want to use more Construct 2 or any other game building tool to prototype games. It’s really efficient with only basic knowledge of the tool. And now i think I’ll mostly use Tiled for mapping when the map type allows it. I might check out other mapping tools but I quite like this one. Also I made my first decent sprites and animations (though it was only b&w and basic shapes) and made me discover the tools to make then.
I had the healthiest weekend for a long time. I’ve been doing much more for me than I usually do so I didn’t have to think about it during the jam. It has also been a great time. I didn’t paticipate over IRC but it was nice to read the discusions there and how everyone was progressing.
As for code learning, I’ve found new ways to implement efficiently animations and game state management. Something profitable for my next games.
Finally, as for the game itself, I want to really finish it by fixing the collision bugs, add sfx and music and add some levels. After that, I even have ideas for at least two sequels. I think I’ll try to allocate myself some time to make them. It might even make me start learning about 3D.
The bad points
But not everything was perfect. Some things I could improve.
First, I should have taken more active pauses. I think I spent something like 35 hours in front of my computer during the LD. I’d say at least for 10 to 15 hour I wasn’t focused on making the game. In these 15 hours, some were dedicated to playing games, reading IRC or else. And I’d say at least 5 hours were dedicated to nothing. Really. Just not knowing what to do. I think in these times I should have just left for a walk or a nap or anything else not taking place in front of the computer.
The second bad point was bug fixing. I postponed it too much. In fact, I made a first rough version which was designed for a specific type of tiles (I didn’t have any roof at first for exemple, so I didn’t bother to make collisions for them). I kept telling myself to fix those bugs but I ended up doing this between midnight and 4:00 am in the night between sunday and monday. I finished by having something but not any better than what I already had so I chose to keep what I had. But now from user feedback, I see a lot of people are facing those bugs and end up in impossible scenarios as then don’t know what to do to get out of the bug. So from now on I’ll add “fix it early” as high priority on my todo-lists.
Last, I’m a bad level designer. One recurrent comment in player feedback is how the level is difficult (starting with the first jump and the first ennemy). I know I tend to be nasty with players when it comes to level design so I’ll try to make it smoother for next games.
And I think that’s all.
To conclude
Overall, it was a great week-end and I’m very happy with what I ended up doing. That’s one more game developped :). I’ve been wanting to participate in LD for some editions now, maybe 2 years. But every time I had something else scheduled on the LD week-ends. I’ll try to keep free spots for the next ones.
I’ve also made a walkthrough video of ^2. As many player can’t go through first ennemy or get stuck in the first trap, I thought it could help to see how the level can be achieved or at least how the game looks like.
TL;DR
First LD but not first game jam. I’ve been enjoying this week-end and I’m happy with what I did. I learned to use new tools (Construct 2, Tiled, GraphicsGale) and I now know the benefits of prototyping with a tool making it simple (Construct 2). I was organized enough but should have made more real pauses (not being in front of my computer while take them). I should fix bugs way earlier, postponing it ended up in a buggy game. And don’t be too nasty when level designing.
Entry page : http://www.ludumdare.com/compo/ludum-dare-26/?action=preview&uid=21322
Game link (web) : http://maxgun.fr/ld26/
Source on github : https://github.com/MaxguN/ld26-square
Tags: html5, post-mortem