Another Ludum Dare done, another dollar! Except, no dollar I suppose. But lots of fulfillment and fun! I very much enjoyed playing everyone's games this jam, and I'm pretty satisfied with how my entry turned out as well. A great jam overall!
I made a post-mortem for my last solo project and it was received pretty well, so I've decided to do it again. I found it to be pretty cathardic for me as well. I thought about doing one before the end of the jam, and that may have been smart to get a little bit more exposure for my entry, but I really liked having the results to show in my last post-mortem so I ended up waiting. Also, laziness happened. So here we are!
Brainstorming
I'll be honest, I wasn't too thrilled about the theme. Ironically it felt claustrophobic-- I didn't have any ideas off the top of my head, and I was worried about being able to come up with a unique take. But in a way it was a good thing, as it really got me pumped for the challenge and got me brainstorming hard.
I usually bounce ideas off my wife at dinner Friday night for Ludum Dare, and this was no exception. We ended up coming up with a handful of decent ideas, and in the end I was deliberating between three candidates. I had my IDE open and was ready to get started on a puzzle game about cramming material into your brain the night before a test when I realized I had never made a puzzle game before... so I chickened out and went with something a bit more familiar, a half-baked idea for a quirky action-strategy-whatever that became High Castle.
Some players very quickly made the connection to the arcade and SNES game Rampart. I loved Rampart as a kid, and I distinctly remember the frustration of not being able to fit your wall pieces together late into a battle because of all the junk and mangled walls laying around taking up space. High Castle was directly inspired by that memory of having nowhere to fit your walls. I decided to add in the crumbling island in an attempt to increase the "running out of space" theme strength, which obviously was going to be due to the fact that the island was flying, right? A fantasy setting it is then!
Not Math Again
Like a good LD jammer, I only had a loose idea of what I wanted to do when I started. I knew it was going to be an island floating in space (or somewhere dark and sparkly), and I knew it would involve defending a keep by placing walls. The first thing I worked on was the code behind wall placement. I quickly realized that getting arbitrary shapes that could be rotated to be placed on a grid and set corresponding grid cell properties would not be a trivial task. And so my first panic began.
Each shape contained its own mini-grid, and depending on its rotation that smaller grid had to be mapped to the larger grid and set cell properties accordingly. Coordinate transformations? That sounded like some linear algebra crap, and it had been a long time since I had taken linear algebra (not to mention how I barely passed... the second time).
But it's amazing how something that was so difficult to wrap my head around in a school setting could be so thoroughly and completely dominated in the course of a few hours when there's a fire lit under my butt, and a passion for the application of the concept. After a few scribbles, google searches, and intense braining, I got the math down and the code working beautifully.
Making It Up As You Go
As stated earlier, I only had a rough concept to start with. As I built the game and test played it, I came up with new mechanics or changes that I thought would be fun to try. This is a risky sort of thing to do, but I think that it is also a very game-jammy thing to do when you've got a very short period of time to work on things. You're driving on a bridge while construction is happening just a few feet ahead, barely keeping up. Hey! That's not a bad idea for a game...
My original (very loose) plans were to mimic Rampart more closely and have "rounds" where you'd fire your weapons and then rebuild in relative peace, then rinse and repeat. Your wizard would conjure random wall shapes during the build period that you'd have to rotate and place.
Then as I was going to sleep Friday night I thought about how cool it would be if debris was floating around in space and you had to grab it and slam it down onto the island. You'd still be handed random pieces, but I felt it would feel a lot more interactive. After prototyping and testing that concept, I thought, hey, wouldn't it be simpler and more hectic if everything was just happening at once? Yes, old me, yes it would be.
It felt haphazard and dangerous, but for a 48-hour code jam the quick [build > idea > prototype > integration] cycle seemed to work very well.
Pesky Theme
Although I took steps to align High Castle more with the theme as development went on, I got some comments about how, while fulfilled, the theme wasn't as center-stage as it should have been. I have to agree with these comments, even though the game focuses on winning 'space' and losing it due to breaches or the island falling apart, it felt more like background noise. I blame the 'making it up as you go along' strategy for this.
If I could go back and try to fix this, I might make the island fall apart more methodically, perhaps in sections rather than lots of small pieces at a time. I'd also try to add 'land debris' pieces that you could pick up and slap on to the edge of your island to build it back up. These changes might help make the theme more central to the gameplay. But that's just a theory.
Difficulty Is Difficult
In my last solo project, I got a lot of comments that the game was too easy. I also got a few that the game was too hard. I even got a lot saying that the game was right where it should be. One of the issues with that project was that most everything was procedurally generated, including aspects of the game that directly affected difficulty. This time I decided to maintain more control and manually build the map and difficulty arc.
While the results were much more consistent and objectively better than my last project, I still don't think I was able to nail down the difficulty curve. For me this is one beast that I just can't seem to slay. There isn't a lot of time during a code jam to manually test your game, and you don't have access to a lot of play-testers (is it even kosher for a compo game to get play-tested by anyone but the creator?), so I understand why it's hard. Maybe next time I should try to be more mathematical about it or something.
In the post-jam edition of High Castle I implemented difficulty modes to try to accomodate more types of players. I thought about doing this for the official entry, but I didn't think there was enough time left to implement it.
Complexity Kills
As I was wrapping the game up, I realized I had no tutorial. As I was thinking about how to throw together a quick tutorial or controls screen, I began to realize I had slightly drifted into "too complex" territory. Shoot enemies with mouse click from your tower? Cool. Grab floating debris and build walls? Alright. Gain power and capture structures by building full enclosures? Ok, getting weird but not too bad to follow. Sometimes your tower and other buildings produce other towers that you can place by right clicking on them then left clicking to place? I think that's where it crossed the line. All the other mechanics were more or less part of a larger system, but the turret building system was just weird and out-of-place, a relic from the game's conception as inspiration from Rampart that didn't belong anymore.
And of course this made it hard to build a controls screen or a tutorial in a short amount of time. And so I was highly unsatisfied with the player's educational experience of the game mechanics ended up. 48 hours isn't enough time to cram in mechanics that need extra explanation, it's best that the mechanics remain simple and tightly cohesive to be as intuitive as possible. A lame attempt was made to shoehorn a sort of tutorial in-game through pop-up messages that would try to give quick and dirty explanations, but it became painfully apparent that people very quickly stopped paying attention to those while playing.
In the post-jam edition of High Castle I remedied that with a sleek visual in-game tutorial system and by removing the turret building system altogether, instead scattering unpowered turrets around the island that you are free to capture. I feel that it jives with the other mechanics a whole lot better, and just makes more sense to a new player.
Story Time
In my last solo entry, I threw together a simple last-minute story line before I finalized the game. The story actually turned out mildly compelling, but didn't really deliver at the end and was honestly just filler for presentation. Players care about gameplay above all, and could care less about the story after all. Right?
Wrong. I got several complaints that the story seemed to be building towards something and then delivered a wholly unsatisfying conclusion. I learned something important from that project: players like stories. So, for this project, I tried to put in a bit more effort with the story. It was an incredibly stupid story, I won't lie about that, but I tried to present it more dynamically with small cinematics and funny quips. I think it paid off, people seemed to enjoy it. It was fun to do as well, so I'm looking forward to continuing to improve my story telling methods in the future.
Bit Tones And Bit Tunes
Making the audio is strangely always a stressful task for me, not because it's that difficult but because I feel like it will take a lot more time than it actually ends up taking. As I'm building music I feel that hours are passing by but it's only been minutes. In contrast, when I'm coding I feel like minutes have passed by when it's been hours.
The sound effects so far have been very rudimentary using BFXR, which produces very 8-bit sounds. They go well with my rudimentary graphics style which is about the best I can do on the art front. I try to make the sounds as painless as possible (it's easy to make VERY annoying sounds with BFXR) but I still got some small complaints, especially from players who got swarmed by enemy infantry throwing hundreds of spears that all made an annoying 'tch' sound when they hit something.
I haven't been making game music very long (only since I've been doing compos) and I finally began to notice a "style" emerge. If you listen to the menu music for my last solo entry Sanguinite and then listen to the menu music for High Castle back-to-back, it's painfully obvious. I'll have to try to mix it up more next time.
Results
With the difficulty issues, the overly-complex mechanics, lack of a decent tutorial, and apparently annoying sounds, I was content with my entry but not expecting a whole lot from the results. While I was very satisfied with the level of polish I was able to achieve and how the idea panned out in general, the game was highly flawed even for a game jam entry.
And so the results have completely blown me away. Unexpectedly almost every category was a personal best. I managed to land in the top 50 in four categories, notably beat my previous placements in 'Overall' and 'Fun', and even land a top 10 spot in audio. Audio? That one was a real surprise; while I didn't think my audio was bad, I didn't think it was excellent, especially with the few complaints I got about annoying noises. Who cares, I'll take it! Suffice to say High Castle did a lot better than I expected it to.
Conclusion
Well, this turned out to be somewhat of a novel, but that's alright. The Ludum Dare experience is intense and there's always a lot to say about it. It was a bit of a wild ride and I re-learned a bunch of scary math, I made a lot of good snap-decisions and a lot of neglectful snap-decisions, and I ended up with a game that I'm very content with. I was also floored by the unexpected results, and that was a really nice treat. Thanks to everyone who voted!
Here's the game page if you haven't played High Castle yet!
Here's the post-jam edition if you want to try it out!
I played a ton of awesome entries too! Playing new games for a few weeks after the jam is over is always so cathartic to me and is the real reward for participating in Ludum Dare. I wasn't able to play as many games as I would have liked, and I'm sure I'm forgetting some that I really liked, but here's a list of some of my favorites!
We are very happy with our results, thank you so much for voting and playing our entry. We got a lot of valuable feedback and have created a Post Jam version that contains both content, bug fixes and balance changes. If you like RTS games, give it a try :smiley: 
