Lonely Island – Post Mortem

Posted by (twitter: @Winterblood_Dev)
December 30th, 2011 6:36 am

Well, regardless of how the voting goes, I consider my effort a failure. I’ll explain why after breaking down the issues…

What went right

  • Theme – I drew a blank on interesting game mechanics for “Alone”, so decided to go for a mood piece with fairly conventional mechanics (especially since we had a new category just for that). I was happy with the general mood of the game.
  • Mechanics – I created a handful of simple scripts that I could re-use to quickly add objects and interactions. This really helped in the final few hours, even though I lost a lot of that work.
  • Art style – I was initially going for a desaturated, pastel look rather than completely grayscale. I abandoned this on day two because it was taking far too much time twiddling the colours to try to get the look I wanted. Grayscale was much simpler to implement and I felt it turned out more dramatic too.
  • What went wrong

  • Too much art – I spent the vast majority of my time building and lighting the environment. It’s pretty, I guess, but when I look at what other entries did using much simpler art, I feel that was wasted time.
  • Art first – I build most of the environment on Saturday, tweaked and polished on Sunday morning, and only started adding simple puzzles on Sunday afternoon. This was exceedingly stupid.
  • Not enough to do – the flip side of overdoing the art. There’s only three simple fetch puzzles to do, and no proper ending, which is pathetic. Because I wanted to make a story-driven experience and had already designed much of the enivronment, I struggled to come up with logical puzzles. At the last minute I realised that they could be emotional blocks rather than physical barriers to leaving, but didn’t have time to properly evoke that.
  • No backups – during submission hour, Unity crashed and corrupted the terrain data I’d spent the bulk of day 1 sculpting. I had to revert to a test build I’d done an hour earlier. Sadly, in that hour I’d added a lot of additional text and a proper ending. Next time I’ll make backups and bank test builds at regular intervals.
  • In this shot you can also see the debug plane – a quad with a mist texture on it that slices through the landscape (most obvious above the doorway) showing where the player can move. I used this to help sculpt the terrain and place objects on it, and intended to delete it before submission. It’s a minor niggle, but it annoys me!
    Lonely Island screenshot

    Self-analysis

    The main reason I consider Lonely Island a failure is that I made ALL THE SAME MISTAKES as in LD21. Read the post-mortem for that one here.

    You’d think because I’m a programmer by trade, I’d be able to knock up working game mechanics in a few hours. And I can, if I try. But that’s exactly what I do at work day-in, day-out. During Ludum Dare, I’d far rather play around with the art – it’s more fun, I learn new tricks and I get a pretty result a lot faster than with programming. And I guess that’s fine, if that’s what I wanted to do.

    But what I really want out of Ludum Dare is to prove to myself that I can design and build my own games – not just implement highly-polished cogs that go into the big machine. And I’m failing at that because I’m doing the fun stuff instead of the important stuff.

    So as far as I’m concerned, the theme of the next LD is “Self-Discipline”!

    Play (Unity) | Ratings Page


    5 Responses to “Lonely Island – Post Mortem”

    1. Ralkarin says:

      Thanks for sharing! I’m with you on the artsy stuff. I’m not an artist, but one of the things I love about making games is doing everything! I love playing with modeling, texturing, programming gameplay, etc… and I think one of the biggest challenges for things like LudumDare is finding the right balance.

      In a good game, I appreciate the gameplay more than the aesthetics, so I tend to whip out simple models (more complex than a cube, but it’s no mario), with simple animations and bang together some gameplay. If you check out my entry, you’ll notice the models are simple (possibly too simple, but I’m working on that), but it has a few various gameplay mechanics.

      By the way, I can’t play the whole game right now, but I love the aesthetics of your entry. As an extension (and heck, maybe your game does it already), it would be cool to see it with some cartoony color added, maybe a game where you have to paint the world or add color back or something.

    2. Ralkarin says:

      P.S. Your entry was not a failure. I think these competitions are a good feedback loop for learning, and if you learned something, you’ve succeeded in my mind.
      Edit: And you already played my game! Thanks for the feedback. :)

      • Winterblood says:

        Cheers! Like the idea of painting the world, wanted to do something similar since I played Okami (although I think the concept originated in Wizball on C64) – loved the transformation of the gameworld. Maybe LD23 will have suitable theme :)

    3. Mjiig says:

      For someone who enjoys doing the art in LD games, any game as pretty as that is not a failure. I can’t play it right now because I’m on linux, but the screenshot looks beautiful.

      I really hope they port unity to linux at some point, the games people make in it in LD are insanely impressive.

      • Winterblood says:

        Thanks!
        The spanking new Unity 3.5 dev preview exports to Flash and Google NaCl – hopefully one of those will work on linux. Should be rolled out properly by LD23. I certainly plan to provide as many options as possible for maximum accessibility.

    Leave a Reply

    You must be logged in to post a comment.

    [cache: storing page]