Six Month Recap

Posted June 23, 2014, 7:51 p.m.

After cranking out six games in as many months, it seems like a good time to review what kinds of games I ended up making, how they could have been better, and focus on what kinds of games I'd like to make in the months to come. It feels like way longer than 6 months since I started doing game jams, probably because ever since I got a little more involved in game development it's what I end up thinking about most.


In December 2013 I made Molten Progeny, a WASD/Mouse score attack shooter. It was originally going to be a tower defense game, but we scrapped the idea as the deadline got closer. It was pretty scary making a game in 48 hours for the first time ever, but I managed to get the hang of implementing a pretty common control scheme, score system, ability cool down, and a surprisingly complicated piecewise animation system. The game was surprisingly polished, especially for a first attempt!


The next month I joined a dozen people to make Ensphere, a puzzle platformer in Unity that involved transforming between two states to push blocks around. I can't claim to have done much on the project, but got my feet wet with Unity and C#. The final game didn't feel very coherent either... I haven't really revisited Unity since this project, but will certainly consider it - especially if I'm trying to make a more fully-featured and stable game in the future rather than the quick Flash games I've made so far.


In February I made one of my favorite projects to date - Terrarium, which is an exploration platformer. I learned a lot about what Flash can and can't do without hampering performance on this project as I tried to put in some atmospheric lighting that was constantly changing alpha values on 4 different layers - bad idea. I also got to tweak the conveyance more with this project. Since the player isn't directly told anything I had to learn how to draw the player's attention and communicate how cause and effect work in the game's world through sound and visuals alone.


In March I made Crashdown, a Titanfall de-make that used WASD/Mouse controls and platforming. There were a lot of crashing (ha) issues with this game as something about the collision boxes tended to make things explode. I was able to get a pretty reliable jumping system implemented that would allow for wall running, wall jumping, and a midair jump just like in Titanfall. I was pretty happy with that part!


In April I made yet another WASD/Mouse shooter, Strider Submerge. Of the games I've made thus far, this one ties with Terrarium for most polished. There's a ship upgrade system, a novel submersion mechanic, and decent enemy A.I. too. On the other hand, it's fairly short and could use alternate weapons to change up how the player deals with enemies.


For May I made a tile-based puzzle game called Pixel Garden. Flowers seemed like the thing for May, so I rolled with it. The game was a simpler version of Triple Town with an irrigation component thrown in that acts as a land/upgrade trade-off. I ended up making a fairly robust tile system for this game in order to retrieve spatial info for the combination logic.

Games I'd like to make

So far I've put together a couple of different takes on the shooter genre and mixed that in with some platformers. There's still plenty of space to explore within those genres, but to broaden my repertoire a bit I'd like to try my hand at some less twitchy, more strategic experiences more akin to Pixel Garden. Lucky for me, the latest version of Stencyl (3.1) has a tile extension built right in. I had already written my own, so using this one feels less like cheating. Making a rogue-like or turn-based strategy game will be much more straight forward with coordinate lookups built right into the system.

Here are some of what I would consider deeper game components I'd like to incorporate into my next few projects:

  • Procedural map generation

  • Branching game structure/story

  • Item collection and inventory

  • Equipment slots

  • Experience systems, level ups, skills

  • Item crafting

  • Resource management

  • Turn-based combat

I also feel like I need to put together a game that makes more use of a physics engine for general competency. Rhythm Grocer has sort of fallen by the wayside as well. I'd like to get that project far enough along that I'd feel comfortable posting it as a complete game. It still needs more content and some bug fixes as it stands.

Tile Swapping

Back to the Stencyl 3.1 update, I've been playing around with the tile extension to see if I could work my way toward procedural map generation. One thing that needs to be done in map generation after roughing out general shapes is to replace tiles based on their surroundings to make them look coherent. I spent a morning banging out a crude behavior to do this and was pleased with the results. For any arbitrary collection of tiles (in this case, dirt mounds), my script will choose the appropriate fence tiles to build a consistent shape - albeit without gates.



This makes me feel good about the possibility of using Stencyl for tile manipulations and potentially randomized dungeons.

Dialogue Trees

As for branching game progression, a third party put together a pretty decent dialogue extension for Stencyl 3.1. It handles player choices and text scrolling just like you'd expect. Compared to Shadowrun Returns' dialogue branches, it's also way easier to use. With Shadowrun Returns, copying dialogue was difficult and setting up variables and conditional statements was disastrous. If you were only supposed to have a particular dialogue option when 3 variables were true, you'd have to nest a few "and" statements and select your variables from a list of all variables in the game, including character stats. Finding 3 needles in a haystack is no fun.

Shadowrun Dialogue Tree

Third-Party Stencyl Dialogue Tree

For that reason it's seeming like a good idea to try fleshing out my ideas for the Secrets and Promises campaign for Shadowrun in Stencyl. That, and the fact that any content I make for Shadowrun Returns would have an audience limited to those who already own the game. Admittedly I haven't figured out if this crazy extension can do everything I want it to yet. If it can't, I suppose Twine would be the simplest way to prototype the sort of relationship systems I'm trying to experiment with on that project.

Anyhow, I'm excited to experiment more with strategic elements in my games and to have some tools that will help me with that. Still trying to get the courage to try Unity again, but for now I'm doubling down on Stencyl.