Premise of the game itself is quite simple:
Go down the dungeon, floor by floor, beat up enemies using four different abilities, steal the demon treasure from the bottom floor.
As some people who read "What are you working on?" know, I took part in .
Ludum Dare TL;DR: You make a game in 48 hours, all by yourself, using engines/libraries/tools of your choice. All gameplay related code, art, sound, music, you have to make yourself.
There's also a theme revealed at the start of the event which all participants should follow, this time it was An Unconventional Weapon, a massive step up from previous one...
I was no stranger to the LD concept and how to work with it, I took part in the LD31 last December with my FTEQW based entry Mobster Massacre, which taught me a lot about using FTEQW for more things than an average quake mod would require, it was afterall a total conversion using no quake assets at all. It also taught me a lot about time management in a time span like that, 48 hours is pretty damn tight, especially if you want to catch some sleep.
This time I felt much better prepared going into it, before starting I had an idea of a kinda diablo-like game that you play completely without a mouse. This is what I started with, just had to modify it a bit to fit the theme, obviously by making the weapons unconventional! I originally thought I'm gonna make a rudimentary inventory system, and a dozen different equippable unconventional weapons for your HJKL slots. Of course I didn't have time to implement that completely but more about that later.
As the very base I used an ultra stripped version of my own CleanQC4FTE template (haven't put this version up anywhere yet, sorry). I started actual development with basic camera, movement, very basic AI stuff, and also reading text files for map information, I knew I wanted tile based maps read from text files since that allows me to quickly design levels after levels.
Here's what things looked like early on, barebones code was done and a few models as well:
At that point I could move around, I had super basic enemy AI in, it actually loaded from a file, I had also started work on the 4 weapon slot system. I started by making it so that player has .entity slots[4]; defined for him, 4 different weapons he can use, but since there was not enough time to program interface for the player to actually switch weapons around between slots let alone actually creating those extra weapons, I ended up with the current system of just spawning the weapons and putting them into the slots at the start, essentially making the player have 4 different attacks they can always use rather than a variety of weapons they can switch. This was fine though, it still gave enough variety for the combat for the duration of the game.
Then there's the art, I created A TON of animations for the player and various monsters (in total ~300 frames, for a 48 hour game). This is something I haven't been too fond of previously, but I feel like I'm really getting the hang of it, and I don't feel like I wasted too much time on it. The 3D modeling itself was quite simple, flat shaded textureless vertex colored stuff mostly, a very wise choice for quick stuff, I also use it for my own prototype stuff outside of Ludum Dares. It can still look relatively good, not sure if I quite nailed it the way I'd like to but I'm happy enough with it.
The map itself uses only 1 model for the walls, just using .colormod to make it different color for various levels, and the floor itself is a large quad drawn with R_DrawPolygon stuff in CSQC, not the most elegant option. Lighting for the map was completely flat but I used clever tricks like drawing a vignette style .tga overlay that makes the edges of the screen darker, and also a very subtle additive blob on the center of the screen, kinda simulating player being a light source.
A screenshot of the final look:
Sound stuff came out quite naturally, all sounds are some recordings I did with my relatively good quality desk mic and edited afterwards with Audacity. The important thing is how well the sounds work in game, and I can quite safely say they work just like they should, providing satisfying feedback when using any of the weapons.
Music I had more trouble with, I still haven't gotten a good hang of actually composing. I still managed to make something with SunVox that serves the purpose of background music well enough. (I think I had like 2-3 hours to the deadline when I made the track...)
That was basically my semi-random ramblings about the development process itself on each field, but what matters most is the end result right?
I got a very playable game with no game breaker bugs in, only some very minor ones that don't affect much anything. There's 11 (0 to 10) different floors to play, the game is somewhat pleasing to the eyes, it doesn't kill your ears either. Overall I was really happy with what I could achieve in the 48 hours, and there aren't much things I would've done otherwise with the development. Another good learning experience.
(LordHavoc)