Judging will be over in just over 24 hours. I had a blast with this competition, and I thought I’d sum up what I learned, what went well, and what didn’t.

Hide-n-Stab is a multiplayer stealth game – player characters and NPCs are identical cloaked figures, and your goal is to stab other players in the back. Attacking will reveal you to everyone around as a human, so you have to be careful and wait for the right chance to strike.

This was my first Ludum Dare and only my second game jam ever – I previously entered the Game Boy Jam, which gave 10 days. 48 hours was a very different experience. I wasn’t even planning on entering this time, and was out at dinner when the theme was announced. I checked Twitter, saw the theme, and immediately came up with the idea for Hide-n-Stab, and I knew that I had to give it a shot.

Anyway, here’s what happened:

What went well

• I used a fantastic technology stack. The game was written in Haxe, with a Flash game client and a Neko server. I used OpenFL with the HaxePunk game engine, which I’m a contributor to. I’ve been using this stack for a long time so I was able to get right to work and build something quickly, and the server and client were able to share a large amount of code.
• I had my asset production pipeline down solid as well. Graphics were drawn in Inkscape with individual body parts as separate layers; I used the Inkscape command line tool to split each layer into a separate image and used Spine to do skeletal animation. I had the fully animated character finished, in its more-or-less final form, the on first night, only a few hours after hearing the theme.
• I serendipitously came up an idea that fit the theme well from the start. This let me tailor gameplay to fit the theme as closely as possible throughout the entire process.
• I had just moved to a new apartment and was starting a new job on Monday after the competition ended, so I had focus and free time to devote to the competition. Also, my very patient wife kept our 3-year-old occupied so I could keep working.

What didn’t

• Developing an online game in 48 hours is insane. I spent probably half of my time debugging esoteric networking bugs. Some of the more elusive had to be solved after the competition was over.
• I posted Hide-n-Stab to Hacker News and did not anticipate the outcome – it was on the front page for an entire Saturday. The server was instantly filled with players and remained so for several days. I was completely unprepared – the server had never been load tested, it wasn’t running on the most robust VM, and it didn’t even have a limit on the number of concurrent players! I spent the weekend running around with my hair on fire, migrating to a better server and trying to tweak things to accomodate the huge influx of players, eventually opening a second overflow server. To be clear – I’m very glad to all the people who played from Hacker News, and it was a great chance to test the game, fix bugs, and get feedback. The reason this is a negative is that I wasn’t completely ready for it.
• I didn’t have enough time allotted to fix bugs. As a result, the final version was playable, but had some intermittent breaking bugs.

Next time

• I’ll definitely try to use HaxePunk again. Great experience, can’t recommend enough.
• I will almost certainly not attempt another MMO-style game in such a short time period.
• I’ll plan to enter the competition from the beginning and brainstorm theme ideas instead of waiting around and deciding to enter after it’s already underway.
• Allocate enough time to thoroughly test the game before release, and aim to put out a fully functional version.

Thanks to everyone who played and/or voted! I played some fantastic games this round so the competition is fierce. Good luck to everyone!

Also, if you enjoyed the game, tune in to www.hide-n-stab.com – I’m working on a post-competition version of the game, with browser, desktop, and mobile versions. It’ll have new features and unlockable cosmetic upgrades. You can also follow me on Twitter (@monsterfacegame) to hear about progress.