A week has passed, I’m more or less rested and it’s time for the customary wall of text some people call it postmortem.

During the week before the theme announcement I like to prepare for it by making a little brainstorming. Most of the time the final idea has absolutely nothing to do with any of the concepts, and this was no exception, but it’s still fun and you never know if THE BIG IDEA is just around the corner waiting to be found. This edition the concept of the game itself changed from the initial design to the final implementation as I was running out of time to implement the mechanic.

While discarding roots, submarines and caves as I feared everyone would use variations of this, I took a look at Trompin, my pet rabbit. Half joking with a friend it hit me that I could do a game dealing with rabbits, burrows and the likes. And mating.

title

You can play and rate Playbuck here

I then thought of a Metroidvania game borrowing the concept of generations from Rogue Legacy: you would get to an available partner (buck or doe, you wouldn’t be tied to a gender), mate and then pick one of your children as the new main character. It would carry around the parent’s skills plus one more, and replace it as the main character. To make this work well, I would have needed a lot longer to focus on level design, rabbit skills and craft a balanced experience.

With this in mind, at some point during Sunday the mating mechanic became a goal in itself. Since gameplay was pretty bland at the moment and the clock was ticking I thought it would add a somewhat original silly touch. It’s a bit too silly for my taste, but everybody enjoys a fart joke here and there, so hey…

 

What went well

  • Planning. Last time I almost screwed the submission for a number of reasons, one of them was poor planning. This time I managed to determine what should I focus on pretty quick and, most important, stick to that and handle changes with common sense.
  • Haxe/Flixel.  I’ve done quite a bit of AS3 coding, but Haxe was a first, and I’d never used Flixel either. I’ve been gladly surprised at the almost flat learning curve from switching to Haxe + OpenFL, and how many built-in functionality already comes with Flixel. I like to experiment a bit with technology from time to time and I think this time I’ve found something really interesting 😀
  • Music! This is the first edition where I can create some music (I used autotracker for LD28, but I think it doesn’t count. Also, I didn’t like the result at all), and although short and repetitive in the long term, it was a funny little tune to listen for 30s.
  • Rabbits! I’m not the one to judge, but still I believe the rabbit visuals turned out pretty well. Yes, there aren’t too many frames per animation (and some of them shamelessly reuse  frames from other animation states, in fact), and yes, the does are just reskins of the main character. My focus is on programming and art assets for me always take up quite a bit of time (and don’t look particularly good), but amazingly enough I managed to draw some cool sprites in a manageable time frame. I probably had a nice model :)
    IMG_20140504_185330


    Sketches for rabbit actions

    DSC_0011

    …and here’s the original

What went wrong

  • Change of objective. As I explained before, this is not per se a bad thing, but the game became a different thing, much simpler both in terms of gameplay and “concept” than what I had in mind.
  • Music! Yeah, even though this time I’ve managed to get music in the game, it tends to get a bit irritating, as the melody is pretty short and it gets repetitive soon. I guess it’s not too bad for a first try. At some point I’d like to take advantage of my keyboard and use something a bit more complex, but Famitracker seems like a reasonable program to start.
  • Lack of visual feedback and eye candy. Hearts, dust clouds…I ran out of time to add some visual effects that would have enhanced the appearance of the game and probably give some clues. Next time, I suppose.
  • Collisions/Sh*tty digging controls. They’re not terrible, at least, but as gameplay is pretty simple both should have been smoother, specially the digging controls. I wanted it took some time since the point when you start pressing the key and when a tile of dirt is actually removed, but I messed it when I thought about how it should behave.
  • A couple of issues with map loading. Even though Haxeflixel had an addon to support Tiled maps there were some things that I couldn’t manage to map well into the game. Next time I think I’ll probably code my own parser class (I can probably reuse some code from Conquer all the castles!), but despite this it saved me a lot of time.

 

Overall, this time I’m quite happy with the overall result, but as always there are lots of room for improvement.


Leave a Reply

You must be logged in to post a comment.

[cache: storing page]