Mass-X Post-Mortum!

Posted by (twitter: @charlottegore)
April 22nd, 2015 6:11 am

For LD32 I knew I wanted to tackle the ultimate Classic Ludum Dare game: The 2d pixel art platformer with a cute and hopefully distinctive mechanic inspired by the theme. My aim, this time, was to try to improve on my ‘fun’ score and if that’s what the final verdict is I will be so happy.

There was no need to kill this robot.

The result was Mass-X, where the mechanic is being able to switch positions with robots. I didn’t include any puzzle elements – I wanted it to be all about trying to find creative ways to kill as many robots, as possible or get through levels as fast as possible, or basically just bounce around for fun.

What went badly:

Day one was murder. I wanted to have the basic mechanics of the game coded by lunch. It actually took me the entire day. My custom engine itself didn’t have all the features I needed, was full of bugs (probably still is!) and the collision detection stuff was a total shambles, so it was all incredibly slow and difficult.

I also found working with programmer art incredibly depressing. The game looked beyond awful and I was rapidly losing faith in my ability to make this something worth playing. By the end of the first day it looked like this…

… and I wanted to give the whole thing up and die. 3 months of preparation for this rubbish?

It was actually about 6pm (6 hours before the deadline, my time) by the time I’d recorded the music and sound effects and done all the tile art for the sprites (and got the animations into the game) and was finally ready to begin designing levels… at which point I started discovering even more bugs with the collision detection and found out that my level editing tool was completely broken. I had to constantly save, export then restart the software every time I need to create a new level or make a change to a level. Nightmare.

The result is that the game is far shorter than I intended. Far, far, far shorter. I really wanted to prove that my tools were good enough to make a lot of content quickly and I have learnt that, in fact, no. They’re not. Not even close.

Final “What went badly” was the fact that I completely forgot that this was a browser game and I completely screwed up the mouse controls. If the cursor goes outside of the browser window, your cursor freezes. There’s a bug in the cursor-to-world translation, too. Basically the controls are a mess which is fine until the last two levels, at which point many people find it impossible. Quick workaround is to reduce the width of your browser window, creating some buffer space at the bottom. I apparently developed the game like this which is why I didn’t find the problem until it was too late.

What went well

Going Pixel Art was terrifying. Something about sharing stuff you’ve drawn… it’s so much more personal than 3d models. I feel like Mass-X is far more of me than any other game I’ve made.  I stuck with 10×10 pixel tiles which worked out nicely. They’re a lot less work to draw than 16×16 (156 fewer pixels to colour in per tile!) but more expressive and detailed than 8×8. I know I have a long way to go with pixel art (MOAR DEPTH!!!) but I accidentally achieved a consistent style which I think works pretty well.

Swanky Paint – it’s a quirky clone of Deluxe Paint 2 with one absolutely killer feature for me: It supports bloom. It meant that I could draw the art ‘bloomed’ then save them without. It meant that when rendered in game the art looked like it did in Swanky rather than all blown-out and too bright. Swanky Paint saved my bacon.

Guitar Midi Synth – I have this thing on my guitar which makes it output midi, which meant I was able to compose music on an instrument I actually understand and can improvise with. I’m quite pleased with the music even if it’s pretty rough and filthy. The sound effects were also a huge amount of fun to make – lots of layered sounds from my own voice and synths, all put through fancy effects. I’m especially thrilled with the bumper sound and the swapping noise.. a dirty buzz and a whoosh and a bang. :)

I also think the basic idea of the game works as well. I hope it’s fun for other people too. It has a lot of potential as a puzzle platformer but I kept it more action/adventure approach than I could have – I’ve seen some fantastic LD games with the same basic mechanic but done as a puzzle platformer instead. It turns out this particular mechanic has a lot more potential than I realised.

The engine is a little trooper as well. It turns tilemaps into single meshes which are incredibly quick for GPUs to draw. The post-processing effects are a bit of a trial for really bad GPUs, but you can get 60fps on an Intel HD 3000 as long as you don’t try to run it in 1080p.

Finally… deploying. This time I was able to deploy for HTML5 and for Windows, OSX and Linux (32 bit) with a single command. I’d done a lot of work preparing this side of things and it really paid off.

Lessons For Next Time

Assuming I want to make another 2d pixel art platformer I’m going to want to fix my level editor’s bugs, add more features to make drawing levels quicker – cut and paste would help. I also need some sort of game boilerplate that means i can start by writing object logic rather than having to set up switching between titles and levels and all that.

My personal challenge for next time? I want to see what the verdict on Mass-X is first. Then I’ll know what I need to focus on next. Fun is, I think, the most important category. It’s the truest test of a would-be game designer’s skill, I think.

Thank you for reading and for playing Mass-X. It’s a short, broken little game but I think I actually love this one.

Play Mass-X here


Leave a Reply

You must be logged in to post a comment.

[cache: storing page]