
Here is the postmortem for our LD22 jam entry : Gemclusion. This was our second Ludum Dare (this first being LD21), and learning from our experiences from the last one we decided from the beginning to make sure to keep the game simple (K.I.S.S.).
Warmup
This time around we benefited greatly from an accidental warmup contest: Game Prototype Challenge. It’s another for fun game dev contest, but it’s on a smaller scale than LD (29 entries in the last one). During the GPC contest, we made a point to keep it simple and have fun with the contest. This is an important thing to remember when doing these contests. Too often we get caught up in the malaise of trying to make the ultimate game and forget about having fun. It was after the GPC contest that we happened to remember that the next LD was starting soon. Once we realized that LD was coming, we started getting ourselves ready. The next week or so was spent relaxing and waiting for the Theme.
Theme
During the voting rounds there were main themes that we were hoping for. In particular, I was hoping that we would get more of an unconventional theme (like midgets) … sadly we will never know that greatness that would have been mighty midget magnets … but I digress. In the end Alone was chosen as the theme. Once it was announced, we immediately sprang into action and started talking out some game ideas that would fit an interpretation of Alone. Since we wanted something unconventional, we decided to stay away from any ideas involving a person being or becoming alone. It was at this time that we came up with a rough puzzle game idea where the goal was to get a single puzzle piece alone.

A small doodle of our LD22 Prototype
Prototype
Since we liked the idea we decided to do a quick and dirty prototype. It would have no effort placed in sounds, art or guis; instead it would focus just on the Gameplay. So I went to work with the quickness that is afforded to me by my bearded-ness. [PIC] Within moments I had re-made my doodle in Unity [PIC]. Once the level was in place, I was able to quickly get all everything working for the prototype. It included moving the pieces with the goal of destroying all the pieces but one. To destroy a piece, you had to move it through an opening and into an invisible wall. It took about 2 hours to get the prototype completed.

LD22 Entry - Alpha 0.5
[PLAY THE PROTOTYPE]
Gameplay
Over the span of the weekend, the gameplay was expanded upon a bit. Most of Saturday morning was spent adding new gameplay elements (warp points, breakable walls, pass-through walls, merging pieces and gate walls).
- The warp points would teleport a piece from one spot on a level to another.
- The breakable walls were special wall blocks that would ‘break’ when a piece ran into them. The walls could be set to take 1 or more impacts before breaking.
- The pass-through walls looked just like the other wall blocks, but a piece could slide right through them.
- The merging pieces would allow for pieces to be combined and form a new piece.
- Then we added gate walls that would only allow certain pieces to pass-through them.

A small doodle of the invisible walls that would destroy pieces

A list of the various gameplay elements
The rest of Saturday was spent testing the gameplay and trying to come up with level designs.
Sunday morning we realized that we were shooting for too much. Here we had several different gameplay elements that we were trying to fit into 20 levels. It was a much bigger task then we were anticipating. So we changed directions and decided to cut out gameplay elements that a) were not much fun or b) would take too much effort to get the 20 levels done in time. During this process we also changed the game pretty significantly. Originally the idea was to get the other pieces to disappear by having them leave the game-board. After cutting through the game elements, we ended up where you had to merge all the pieces together. We removed the warp points, breakable walls, pass-through walls and decided to make the levels closed in. This turned into a really good direction and helped spark a 8+ hr session where were able to create all 20 levels.
Graphics
Almost all of the graphical work was texture based. The game only used primitives for the in game items. We didn’t start any work on the GUI / HUD images or the in game textures until Sunday. We wanted to wait to make sure we had all the gameplay done … after all it’s a game. While the art makes up 90% of it, that other 10% is very important and often can take more time then the art.
Audio
We originally had planned to make a few different tracks for the game using FL Studio. Sadly, we haven’t really used it much and we were not familiar enough to get that done. In then end we used sounds that we already had from a friend and then added him to the credits.
Sleep
We have done several game jams of various lengths (24+ hrs) and several times we made the awesome decision to stay up the whole time. One time I was awake for 44 1/2 hrs straight … and that includes the 5 – 6 hr drive to and from Pittsburgh (so 10 – 11 hrs driving total) and the time for the competition (it was a 24hr game jam + the time for setting up / packing up + the time for the awards etc). Nothing I can say will ever describe how bad it sucks to be up for 44 1/2 hrs with all that activity.
So this time around we made sure to go to sleep at a reasonable hour and not try to stay up for the whole weekend. I can say that it was one of the best decisions that we made this time around. Some of the better ideas we had with the gameplay came after getting a good nights sleep. We set clear goals for the day and made sure to get as close as we could before going to bed.
Final Thoughts
So there was our Postmortem for Gemclusion. Hopefully it was a good read and provided some useful advice to someone for something for some reason … yeah that would be nice :). Please be sure to play and rate our game. We are looking for any feedback that we can get that will help improve the game.
We are currently planning on expanding the game and releasing it on PC and/or Android for $1 – $2. We are not sure how many levels it will have (20 – 50 is probably a good bet) and we are still deciding the gameplay elements (probably some of the originals with a twist + some new ones). We might also change the movement a bit to coincide with a phone for the Android version. Additionally we plan to update the art (both in-game and GUI / HUD items). It should be a lot of fun.
I will post again with and update when we get things going after the holidays. Until then have a safe and happy holidays!!
Good luck with the upgrade system, those are pretty addictive