Home | Rules and Guide | Sign In/Create Account | Write a Post | Reddit | #LD48 | #ludumdare on irc.afternet.org (Info)

Ludum Dare 31 — Coming December 5th-8th 2014! — Join the Mailing List!
Also check out the Ludum Dare 30 results!
  • Ludum Dare 31 begins: in 77 days, 15 hours, 8 minutes, 57 seconds
  • (Time might be off, we’ll have it right soon) | Real World Gatherings (Now Open!)


    Fry Fleshlings Fast – Post mortem

    Posted by (twitter: @zeh)
    December 21st, 2012 8:23 am

    It’s been nearly a week now, so I figured I’d post a post-mortem about Fry Fleshlings Fast, my Ludum Dare #25 entry for the theme “You Are The Villain”.

    The game is an ActionScript 3-game compiled in Flash. It runs online, or as Android AIR installs (I guess it could work as iOS installs too, but, you know, it’s a pain in the ass to create all certificates for that and only people with jailbroken phones would be able to install it since it’s not an App Store release, so why bother).

    The original idea I came up with was a game where the player acted as an invading Alien overlord. You’d have a top-down view of a map and your task was to kill everyone below. For this, I envisioned a mechanic where you would have actors – the fleshlings – roaming around a map, going from checkpoint to checkpoint. The player would have the ability to cut their paths short (by using lasers) with the objective of getting everyone to move to the same general area, where you could then drop a meteor attack on top of them, killing everyone. The challenge would be in using as few lasers as possible.

    Although the systems were different (using point- and line-based roaming paths, rather than a two-dimensional plane), the whole idea was heavily inspired by Jezzball, a very simple yet addicting game for Windows 95.

    Early screenshot showing the internal map/waypoint editor I had to build for the game

    Early screenshot showing the internal map/waypoint editor I had to build for the game

    What went right

    Reusing code: although I failed to do a post about it prior to starting the project, in this project I reused much of the code I developed for my previous Ludum Dare game attempt, Survival of the Tastiest. While much of the code is, of course, garbage – many shortcuts were taken – a lot of the support code and more abstract frameworks for handling objects and Starling entities came out very handy for this new project.

    I now feel like, for a Ludum Dare project, you can’t plan on working on a new engine and a new game at the same time. Unless it’s a very unique but simple mechanic, it’s just a lot of effort that means you can’t get both really right. It’s better to develop an engine with a simple game layer on top of it, or to work on a cool game idea based off a well established engine. Of course, this is not something I did right this time around – I was developing both again – but having somewhat of a starting point helped.

    Over time, it’s likely that my own entity engine – which doesn’t have a name – could shape up pretty well. There are many well known, polished engines for games in Flash out there, of course, but having one of your own is a great learning tool and, with time, allow you greater freedom since you can tweak and change it to fit your needs better.

    Programming a system I enjoyed: I never get to spend my (full-time) programming job on game-like systems anymore, and while creating this game I was dead set on creating a point-based pathfinding and routing system that allowed for routes to be created and cut dynamically during gameplay. I love the code I produced for that – one where I didn’t take any shortcut on – and it makes me happy knowing that no I have this additional library to carry with me whenever I need it. I believe challenging systems like that are what I do best – rather that repetitive tasks – and I was a great joy to get this working from the group up. This made this Ludum Dare effort really worth it.

    Not worrying too much about the idea: truth be told, until Saturday at noon, I didn’t even know if I was going to participate in Ludum Dare at all (because, like I said, I had other work to do). So I decided not to fret over it. I was thinking of ideas and discussing them with my girlfriend and friends, but not going crazy over it. Maybe because of that, the idea came naturally and I started implementing a few hours later.

    Screenshot showing a debug view of the game, including map waypoints as I tested the path cutting mechanic. Roaming Fleshlings are also visible.

    Screenshot showing a debug view of the game, including map waypoints as I tested the path cutting mechanic. Roaming Fleshlings are also visible

    What went so-so

    Scope: with my previous game attempt, I had a very audacious plan that failed to materialize in a game; it took so much effort to get something playable done, that by the time it was ready, I didn’t have enough time to make it easier, more approachable, or just fun. It was an engine, but not a real game. With this entry, I made sure to think smaller – creating something that could be accomplished without a lot of effort. This was also made more important because I actually had Emergency Real Work to do that same weekend, so I wanted to spend just a few hours with the Ludum Dare entry.

    In the end, while the scope of the game was much better aligned with a Ludum Dare project, it still failed to materialize into a fun game. I only spent 6 or so hours with it, and I feel the results were more or less the same: there’s a game engine in there, but it’s not that challenging, or fun. It’s a gimmick.

    Still, I think that, given enough time, it could be pivoted to something more fun.

    Testing the final assets and the laser animation and smoke particle system (heavily inspired by Faster Than Light, I must add)

    Testing the final assets and the laser animation and smoke particle system (heavily inspired by Faster Than Light, I must add)

    What went wrong

    Assets: Even if I decided to use simple assets this time around, I ended up spending too much time creating graphics, animation frames, and map textures for the game. Sometimes, it’s better to have extremely simple assets and concentrate on everything else.I’m convinced my next game will have circles, triangles, squares and lines first and foremost, and no animation frames. But then maybe it’ll finally have audio.

    Not giving enough time for testing: the gameplay concept was something kind of unclear, and my problem was assuming it would just work. It’s only later in the development process that I realized it wasn’t the case. The mechanic was there – you cut paths down, limiting fleshlings to a given area – but it wasn’t fun. For one thing, their behavior was too unpredictable (they would take random paths once they reached a path connection) so the idea of funneling them into a common area never worked out. It would be better if they had clear rules – for example, when reaching a connection point, if there’s a clear path ahead of them, take that instead of turning the corner randomly – or if the map had additional mechanics, like one-way routes (maybe by deploying signs of some sort).

    I feel like the need for those changes would have came out earlier in development if I was focused on getting a prototype working, rather than on getting a game working. I heard this is how World of Goo has come to exist: they developed a simple prototype over a week to see if the mechanic would work (as a part of a large, a-game-a-week effort), tweaked that as necessary, and only then decided to create a full game with its unique but fun mechanic.

    Much was dropped: once again, I had to drop a few features that made the game less cool than I thought it could be. The end-game meteor attack was removed – the new objective was just to kill everyone with lasers themselves – and potential new weapons were never seriously considered. In way, this made the game interface better – I didn’t need a HUD for weapon switching, so the map could take the whole screen – but it also meant it was a bit repetitive.

    Roaming Fleshlings and their bounding boxes (or their blood) when testing the killing mechanic

    Roaming Fleshlings and their bounding boxes (or their blood) when testing the killing mechanic

    Conclusion

    This game was great in that it taught me one thing: I normally plan a system and then try to develop a game out of it. Maybe, given enough time, this can be a solution; but it means you need to give time to work on the gameplay after the system is complete, because at first, it’ll only be an engine of possibilities.

    Many lessons learned for the next Ludum Dare. I like to believe I’m getting better. Code-wise, it has been a lot of fun, but maybe one day it’ll also be fun gameplay-wise!

    One Response to “Fry Fleshlings Fast – Post mortem”

    1. [...] have just published a post-mortem about its development on Ludum Dare’s official blog denoting what went right and what went wrong, and what [...]

    Leave a Reply

    You must be logged in to post a comment.


    All posts, images, and comments are owned by their creators.

    [cache: storing page]