When the time came, and the theme was set, I was initially dismayed. It was … not one of my favourites. Oh well.
Fortunately, I knew I needed some things working in XNA, regardless of what I did, so I could just start coding, while watching Ludum Dare screencasts at Twitch. so, I wrote an engine which rendered textured quads, and set up the camera so that I could set them on screen with pixel precision.
Some 9 hours in, I finally started getting an idea I liked.
The initial concept was, that you’d play a hero, who’d have to stop a disaster, every 10 seconds, and rush to the next one. I imagined a caped character, saving planes and trains, stopping robberies and alien invasions.
Well, that wasn’t going to get done in 39 hours. So, time to simplify.
The times I’d played Arkham Horror, or the last time I spent over-analyzing the setting in Sailor Moon, probably cued something.
What is there was just one disaster type, that of interdimensional invasion? You’d fight the creatures rushing in from portals, and… stuff. I was kind of fuzzy on the details, but closing the portals or taking the fight to the enemy should be an option. Initially I had civilians whom the monsters would attack, and locations the player could visit, but time pressures -> snip.
Onwards, with the post-mortemizing!
What Went Right:
Quite a bit. I was about 90% satisfied with my experience and results, although this was my first LD, and my second build-a-game-over-the-weekend participation.
Starting with the Engine.
Certainly not always a good idea, but in this case, I had a pretty clear idea what I wanted to do – I didn’t want to do any modelling, but I wanted to have access to rotations and whatnot. This was also my XNA learning project, and I made good use of the first nine hours, despite having no idea of what I’d make.
That being said, I had to rewrite the core of the thing, as my initial version was badwrong. Of course, I figured the correct way to do it immediately after a good night’s sleep, when the competition had finished.
Simple Design
Not having an overambitious design cut down on time needed, and I stil barely squeezed the minimum in, within the time limit.
Focusing on Finishing and Staying on Target
To get a game, not a game demo, I needed win and lose conditions. I sort of got there – if I’d just slapped a “you win” screen in front of the player, when they reached the final battle portal stage, it would have been complete. Now, while there’s a distinct lose condition, there wasn’t clear win condition. It has a screen showing controls, and even some sound effects.
It’s not just a tech demo, but an actual, if tiny and not very entertaining, game.
Coder Art
I like drawing, and I’m not horrible at it. I intended to leverage that, but I slapped in some quick placeholder art — which ended up being pretty much final. It’s aesthetics grew on me, like an alien fungus. I didn’t end up spending much time on it, just adding tiny little animations, which made things quite lively enough
Tiled
The Tiled editor, while very simple, is great at quickly putting levels together, and I know some simple tricks with it, so that was fun.
What Went Wrong
Not that much, really.
Using 3d.
Not in and as itself, but learning and building took a huge chunk of time, which I could have avoided, if I’d just stuck to blitting. As it turned out, I didn’t really need all those 3d features at the time.
Of course, now that I’ve rewritten the engine and so forth, I’m starting to see some benefits from the 3d engine in the background, but for the purposes of the competition, it was redundant.
Getting Zero Peachy Keen Features Implemented
I had a plan to give players power-ups depending on how they performed during each 10-second segment. Various attacks, heals, and other neato things. As I got none of them in, the player was left with two basic attacks, both which are kind of meh.
Using CamStudio for Timelapse
Not a big thing, but I lost first three hours of video due to hitting esc at the same time as CamStudio offered a save dialog, and that was that. ’twas a pain, mostly.
Afterwards, I familiarized myself with Chronolapse, which is much better regarding timelapse recording. I also lost a few minutes fighting with CamStudio, which may or may not count.
(I linked the timelapse videos in previous posts, if you’re interested)
Not Paying Enough Attention To Rules
I didn’t know there was a hour’s grace time to upload the product, after the actual competition time was done. I could have put in a bit more work on the game had I known that. Ah well.
Level Design, or Lack Thereof
My test level end up being pretty much the final one. I could have made the gameplay much more pleasant by simply redoing the level’s layout. Few things are as permanent as those meant to be temporary.
Still, I’m not very good at level design, in general. Need to improve there.
Coder Art
On the other hand, this was another thing, where an investment of, oh, 3-4 hours would have given me much more stylish graphics. Not that I had the time, due to spending most of the first day building the engine.
My LD27 Entry
Tags: postmortem
For me it’s C = 164 ^^