PostmortemThis time I managed to submit something playable and it is a great victory for me, nevertheless the voting results. These are some comments about my experience.
Figuring out what to do
That was difficult. At first I couldn't find any idea that I could implement (both programming and art), and after that I couldn't find any idea at all. I lost the first day, and in the second day I spent some time watching some youtube videos of Spectrum, Amstrand and Amiga games.
I found some videos of games I used to play about 20 years ago (Saboteur, Trantor, Xecutor, Xevious, etc), and I started to feel nostalgic and sad because I was watching videos instead of writing code :)
Finally, the third day I started writing some base classes and implemented a basic scroll. OK, go for it: create a game.
I'm not an artist. Repeat: I'm not an artist. I googled a tutorial on how to make a starfield with Gimp (yes, a tutorial... I had to repeat it 4 times to get something decent). Oh, it doesn't look that bad after all. I started drawing some spaceships with mtPaint. I knew the application, but it's the first time I've used it... it reminds me to my old Deluxe Paint II :) It's perfect for pixel art (I can even use the keyboard!).
A shot'em up sounds OK for the theme: one way trip, a suicide mission to destroy... etc. So carry on.
At this point I made my first great mistake: it's an horizontal shooter, meaning that the graphics are more complicated. After sweating bullets for two hours I had a player ship and one enemy. Then my partner noticed I wasn't using the same perspective for the player and the enemies... so I present you the spinning enemy:
It's far from perfect, but I can tell you is the best animation I've ever done under pressure! (and in any other way, to be honest).
I had a great time drawing the explosions, although it was pretty clear I wouldn't have enough time to make a lot of different enemies. I'm too slow :(
With the music I was lucky. I used SoundTracker about 10 years ago and it hasn't changed at all (I don't know if this is good or bad). I started with a bass line and it was easy to get a shot'em up feel. I'm happy I used some time for the music, it adds a lot to the final result.
I made some effects with Audacity and the laptop microphone, and finally I generated some extra sounds with sfxr (although my first try with mouth-generated explosions was quite nice too).
Evidently I didn't have any plan whatsoever, so I had to face a lot of problems when I was tired and after lots of hours programming (after my work day, of course) because I didn't know how how to implement what I had in my head.
One of the critical moments was when I was programming the circular hordes: according to the sprite it was a good idea that they shoot to the player, but I was getting the maths wrong because the aliens seem to have sight problems (or even worse, a drinking problem).
That was Saturday, 8 hours left, and trying to understand why your Python code for the formula of the line between two points is wrong... it's quite stressing!
I think I wasted a lot of time with that and refactoring code because I didn't know PyGame well enough. Besides I had to test the game script to see if it was playable just hours before the deadline.
It has been a nice experience, although I think I can improve the results and (probably) make the whole process more pleasant and less stressing. So count on me for the next PyWeek (hopefully another game).