Author Topic: Meta Automata prototype :: 0.26  (Read 1092 times)

logFarm

  • Jr. Member
  • **
  • Posts: 89
  • Karma: +0/-0
    • View Profile
Meta Automata prototype :: 0.26
« on: Friday 2016-April-01 05:43:46 PM »
I've already made two 2D games using LWJGL:

- wobbegong

- wobbegong platform hopper

both are here and on indieDB

So I wanted to teach myself LWJGL in a 3D context and also implement
some multiplayer action using UDP.

Mission accomplished.

If I could create a decent game during this learning then all the
better but in this case a half decent game will have to do.


Planning

I thought I had done enough planning in order to make a decent game; I
thought I had reduced the scope of Meta Automata down to a set of game
mechanics that I could build into a computer game but that has proved
not to be the case, in my opinion.

The problem: my plan is too ambiguous.

I hoped I could go with a few basic fuzzy ideas and then make them
work.

It's frustrating, remembering how much time I wasted implementing a few
basic game mechanics only to find that they offered very little in
terms of fun.

I didn't do enough planning. I should have spent weeks on designing
the game on paper rather than months building the game with computer
code.

Even though this game is a prototype, an estimation of a finished
game, it's too ho-hum.

I created a ho-hum single player experience and then, retrospectively
added multiplayer functionality. This was a bad idea; the ho-hum and
the retrospective part.

The ho-hum-ness might be solved by tuning various aspects of game play
but the retrospective multiplayer functionality required me to change
beautiful coding in a way that feels rushed and haphazard.

I've learned:

1. Multiplayer functionality needs to be designed into the game from
   the very start.

2. Single player game play will not automatically translate into
   multiplayer game play.


Light Weight Java Game Library

http://wiki.lwjgl.org/wiki/Main_Page
https://www.lwjgl.org/license

I thoroughly enjoy programming with Java and so luckily for me LWJGL
is a pleasant way to learn OpenGL and associated technologies using
Java.

It is marvellous.

I develop using a Linux desktop but my Java game will need absolutely
no additional programming to make it run on Linux, Apple and Windows.

Having said that, I've never seen any of my Java games run on an Apple
computer.


What else have I learned from this prototype

Moments in the game:

- laying fences and watching them connect

- firing and destroying parasites

- picking things up; boulders and oil

- watching bird machines lay new machines

- sparking machines into life

- watching a repair robot at work

- Meta Automata's game objects create what appears to be complex
  behaviour because they move as if they are part of a flock but their
  logic is very simple. It's relaxing to watch.

- Testing multiplayer functionality is not easy when you are working on
  your own.


Conclusion

I will not be doing any more development on this iteration of Meta
Automata.

I have learned a lot by working on this project but it is time to move
on.