How Berry came to be ~Some kind of a post-mortem

its-over-now-berry

Hi! I’m Fabio, developer of “Berry | The last straw”! :)

This is long, still didn’t sleep much, I might have written a load of crap, got confused a little and put some ramblings inside, you’re warned! 😛

For those who find it too long, the abridged TL;DR:

What went wrong:

  • Bad project management and planning;
  • Redoing everything from scratch two times to “improve” code;
  • Not thinking carefully instead of implement the basics faster;
  • “Thinking” too much looking at the screen/paper;
  • Bad sleeping pattern and posture;

What went right:

  • Albeit rushed, and with flaws, I found the game fun to play;
  • I really liked how the circular bars came out;
  • I liked how the character turned out;
  • Haxe skills improved (I think);
  • Liked how the soundtrack came out (melody got stuck in my head);

There are probably more points in it, but I didn’t put it because, although I said abridged version, I did it without writing the full text first :P. And, now, the main text.


Ludum Dare started, for me, on April, 17th, at 22h local time (I’m from São Paulo, Brasil, BTW) and ended Monday, 20th, also at 22h. It is my first time on LD or any other jam and the first time I make something for release.

As soon as the theme was announced I got a little bit lost because it wasn’t the theme I ranked first in my wanted list, but still I wanted to give it my best and do something.

At the time I was on the LD IRC channel and somebody posted an online spreadsheet for theme brainstorming. There were some pretty nice themes, some were a little bit too unconventional (I put things like “Gorgonzola Cheese Past Expiration Date“, “Nostrils” and “Hello, World!“), but still I didn’t get an idea.

I had an idea after looking into some food photos on my PC and seeing photos of berries and grapes, and I thought about this seed-shooting strawberry thing. I had made some comics about living fruits before (in 2006, I think, and it was called Frooty), with some zombified fermented grapes. Decided to mix this with the strawberry thing in the game, for a survival arena shooter. I was really excited about my idea.

While I didn’t have much experience/training with Haxe and OpenFl, though I made some prototypes on subjects I find interesting, the idea looked simple enough to be done and I thought I could complete it fast enough (code-wise) and do a good work with graphics and audio.

I did a few things and found some progress on character movement and other things, like the health bars, rudimentary AI and other basic stuff. But I tend (like a lot of people) to not be satisfied with the work I’ve done.

I put almost all the code into an “old/scratch” folder and did everything from scratch, to make the code “cleaner” and to try to improve on some calculations. This was the start of everything that made the project go slower.

Instead of redoing it (and I did it TWO times) I should have made it carefully from the very beginning. Instead, I wanted to have the basics working as fast as I could and forgot to properly organize code and rename things. I also wasted a lot of time doing other things and “thinking” while looking at the screen/paper.

The finishing of all screens was rushed and done with a lot of copy/paste from within the project itself, and I packed submitted in a rush. There were lots of things I wanted to implement that I couldn’t because I spent so much time thinking and working the basics.

Some of them are here:

  • Instead of a square arena, I wanted it to resemble a fruit bowl/have an ellipse shape;
  • More characters for mobs, a secondady playable character;
  • I wanted to make it more top-down 2.5d;
  • Obstacles placed randomly;
  • Better collision system for mobs/player;
  • Better management for full screen mode;
  • More refined UI, scoring system and a high score table;
  • Better controls on the Android port;
  • Improve stability and performance (Windows native is the most stable :/);
  • Better sound effects and music;
  • Work more on art;
  • HTML5 support;

But, while it has (lots of) flaws, I find the game quite fun to play and, also, it’s my first “finished” game, the first step on a journey I’ve been wanting to take for the last few years but couldn’t, so I’m quite proud of it.

I also liked the work I did on the main character, on the soundtrack (I get myself humming it a lot and put it on repeat a little bit :P) and some of the code I did, specially on the Health/Ammo bars.

Taking part on this LD helped me improve my Haxe/programming skills in general (I think) and made me realize the important of a lot of things during game/app development, how important and nice all the feedback is, be it good or harsh and how important speaking with people is (even if just on chat or social networks), because I tend to avoid discussions and conversations.

It also helped me give even more value about my real work (I’m a web developer, working with PHP) and work of other developers, since a lot of people I work/worked with tend to think that just by working from home, in front of a computer, my work is easy peasy, when it’s not.

Resuming everything: I did a lot of errors, the end product is flawed, but I’m proud of it because I made it! Taking part in LD has been a really nice experience and I totally want to take part in it again and also on the Mini-LD and other game jams out there. 😛

And, that’s it! Thanks for reading and sorry if I talked too much, talked a lot of crap or if my english isn’t that good.

I hope to have a lot of fun playing the games made for LD and have even more fun building my future games.

P.s.: I plan to expand on the “Berry” idea. I will try to make a post-compo version, but I’ll make a separate game for this improved idea and I hope it’ll be fun to do, without the time constraint, and fun for everyone to play! :)

P.s.2: If you’d like to give the current game a try, the page it’s here: http://ludumdare.com/compo/ludum-dare-32/?action=preview&uid=50909

P.s.3: I did 2 timelapse videos and will post’em here when I upload them. I had some problems recording the timelapse and also editing things (I edited out most of the personal stuff, like Facebook and E-mail, because of personal life and work-related things :P)

Tags: post-mortem