Sunday, December 13, 2009


I participated in the LudumDare contest this weekend. I knew I couldn't make an entire game but I hoped I could at least make a little part of it. boy was I wrong... Everything went to hell because I rushed things, entity management system was horrible, everything was full of hacks that didn't surpass the pressure of hard work, the engine was a disaster. But I was okay with that, I knew at the start I probably couldn't make it and I was cool. When I finally decided to drop off I was actually a bit happy as I learned a lot of important lessons, made a lot of mistakes and all in all this will help me in the future. However one thing happened I didn't account for. My game was going to be called "The world is beautiful out there...", I wanted to have a small guy jump through levels at a half fast pace, when I gave up I made a post on LudumDare blog saying that I can't finish the game. In the last few words I wrote That next Time I will finish the game and that I will win. I felt this was a pretty cool thing to say and decided to add an image to make it sound even cooler. In the same posy I put an image I composed in photoshop from the assets I've drawn as a preview level. Just a house on a hill going to water.
I showed the blog post to a friend and he noticed noticed that it could go great if it was the last the cool image was the continuation from that level. An incredibly short game, just 3 screens, and all you do is walk from the house to the cliff. It could be done by ANYONE in just those 5 hours left until the end of LD. Anyone except me. I grabbed my code and tried to forcibly put it together. My engine was so bad, so utterly putrid I couldn't add 3 simple classes to it and make 1 level. It would be a great "Well, I didn't manage to do my actual game but this makes me cool" farawell game.
God I feel bad right now. I failed so hard it hurts.

Sunday, December 6, 2009

El Poho going good

btw El Poho is the name of my game engine until I find a better name.
Ok, I finally added polygon collisions into it. It's a smaller version than the previous one but it's very easy to add features in it. For now the main features of it are:
-great scalability (should be easier to make a game in it now)
-Polygon-based collision system
-compact (most unnecessary features like GUI and a great number of other stuff was not implemented)
and as a bonus:
-simple yet powerful particle engine!

However I'm still planning on adding a simple GUI system for things like buttons and text at the very least, I should probably add an event system as well.
Another feature I'm working on is quite an important one: Scheduling.
So far it runs at 30 frames per second and it just waits some time between every frame to mantain that, but I need to add a feature that skips the drawing step completely if it runs too slow. So far every object in the game has a few functions like draw() and think(). With the scheduler I'll be able to set every object to think at diferent time periods, like every 1 second or every 1/20 seconds, which will make it a lot faster when I'll have to add AI support to it. With scheduling I can have AI's in Idle states that think every 5 seconds or so, or think every 1/20th second if on full alert, this'll be an enormous performance boost.

For testing the engine out I'm making a small game where you control a dust cloud. Not much to do for now as I'm having some trouble in porting things to the new collision system (what worked back then doesn't work now).

Ok, that's all for now.
maybe I'll update thig blog later this week.