Thursday, December 27, 2012

LD#25 postmortem

As is becoming a regular feature with Ludum Dare challenges, shortly after the challenge I like to write a postmortem of the game I created. The theme for Ludum Dare #25 was "You are the Villain." The game that I created for the challenge was "Evil Gnome's Revenge." This game has you playing the Evil Gnome from one of my earlier Christmas games (Santa's Snowball). As the gnome is not able to steal toys, he is instead going to release Santa's reindeer from the corral that he keeps them in.

What went right

The first thing that I did right was choosing a simple game to develop.  Even simple games take longer than you think they will to create. As polishing games is one of my weaknesses, I figured that the sooner the game was functional the  sooner I could start polishing the game and perhaps end up with a really great game. While I was able to get the basic chase mechanics working right away, I still spent a lot of time coding so didn't get as much polishing done as I would have liked. Missing features include fully animated reindeer, idle actions for the reindeer, and an evil version of a Christmas song for the theme.

The second thing that went right was getting proper rest. As Todd was having his 24 hour podcast while the challenge was going on, I had plans on listening to the whole podcast while I was fine-tuning the game. When I realized that I simply was not far enough along. There is a big difference between an hour of work done when exhausted and a productive hour of work. By spending a few hours sleeping I was able to regain my focus and get the game done. I honestly believe that the extra productivity gained from getting a few hours of sleep more than made up for the hours lost.

What went wrong

While I  liked the theme when voting, shortly before the announcement there was a major high school shooting which really  put a damper on it.  When I heard the theme I knew that I wanted a non-violent game.  It took me about an hour to come up with an idea that would fit the theme yet would not be violent. Had the shooting not happened I would have gone with one of my other ideas which probably would have resulted in a more entertaining game.

The other big problem that I had was the realization that I was writing far too much code. A lot of the code I was writing was fairly straight-forward and should be handled by my GameJam library. The library is in a early state so this is understandable, but I am going to have to find ways of creating classes that incorporate a lot of the generic code that I was writing into them.  Basic movement logic and collision detection are common to most games so having that inheritable in a generic game screen class would be a huge time-saver.


While I ended up with a fairly solid game, it simply is not quite as polished as it should have been. I am going to have to find spare time  before the next game jam and update my library to incorporate a lot of the more generic game elements so rapid game development will be easier.

Thursday, December 13, 2012

GameJam library release 0.01

For 2013 I plan on focusing more on my larger scale potentially commercial projects with Blazing Games content coming from game-jam competitions such as the Ludum Dare challenge that is taking place this weekend. As these challenges allow for use of your own libraries as long as they are shared before hand, I am planning on creating a GameJam library to make it easy to create games quickly. The first release of this library is located at but I don't recommend anybody use it yet. This is still a very primitive release. For those who are curious, I am going to outline my plans for this library.

First, I want to get basic 2D game elements, such as scrolling maps and multiple layers of sprites fully working. This is mostly there but the code needs a lot of cleaning yet before I would be happy with other people using it. I suspect that seeing where things need improvement while participating in a couple of game-jams will get me to the ready for consumption point.

Once this is done, I will add a couple of kits to the game focused on the major elements of common Genres of games. The idea here is that there would be core mechanics for a number of types of games that can be mixed and matched. That way if I go into a game-jam and think that a RPG would be ideal, the core elements such as character sheets, inventory systems, conversations, dice rolls would all be ready to go so I could get the game skeleton done in a few hours and spend the rest of the competition focusing on the content of the game. Essentially, I want the ease of a game-making tool yet with the flexibility of total control of the code. More important, at least to me, I want the library to be mine. I am making a slight exception to my-own-code ideal by using the starling library for 3D acceleration of the 2D graphics but when I get to my final step in this project, that will not be the case.

Once I have the ActionScript library in a stable state, I will port the library to HTML5. The decision to develop the library in AS3 initially is due to the fact that I can get something useful much quicker as I am far more comfortable with ActionScript. As the world is shifting to JavaScript, it only makes sense to shift development to that platform. I will probably have both Canvas and WebGL versions of the sprite/map renderers to take advantage of hardware on browsers that support WebGL while still allowing IE users to play the games.

I have no timeline for the library (and am actually not putting in as much time towards the library as I hoped) but plan on updating the library before every game-jam that I plan on entering.