~~Judging ends today (i think), so~~ I just learned that there are other wednesdays than the current one...anyway it's time for a little post mortem of our very first LD entry Membrum. Technological base for the game is libGDX.
A little summary of how it went
I immensly enjoyed playing the games you guys pushed out in the past LD's, so i've always been a fan. While working on another (game)[tank-tycoon.com], my girlfriend Anne and i decided, that we want to give it a try, finish something while struggling for a year with our other project in our spare time. I made sure i wouldn't have to work on monday of the LD and also asked my sisters boyfriend Christoph if he wants to make some music for the LD game, which he did.
Saturday
I woke up 10:00 and immediately checked for the theme while laying in bed. I thought about it for some minutes. An idea i had before came into my mind: the management of a body. Since it should be a small world, i thought of a platformer where the level is a body that you are maintaining. The walls are flesh, there are different rooms, for example with a beating heart as a central object in it.
I left my bed and my already Anne what ideas she got for the game. She had three broad setups, which i all liked:
* being an animal in a cage, in a zoo
* managing ants
* being a small parasite, a flea for example, living on a body
I progressed to the morning toilet and thought about all 4 concepts. Considering the scope, i mentally cut the whole platformer part and made it all directly clickable - after all, the platforming was not really necessary for the core concept (though i wonder what game this would be). After some brainstorming the idea for membrum was fleshed out best, as can be seen here:
You can see the resemblence, the alcohol habit on the top right, the grid layout on the bottom right. There, i still had spine and ribs visible, which was quickly cut. Also, please not the penis. * Should we add genitalia? Let me know in the comments * :D
Anyway, after i convinced Anne that membrum will work best (the scope of the others seemed to big for me), we started. She did graphics and i programmed and designed. I described an ideal future: Saturday we should finish the core mechanics. Sunday is for making the game fun and finishing all mechanics. Monday is for polishing, testing and delivery. Guess when i finished all core mechanics! I think you had a hunch, it was Monday around 18:00. Anyway, around 22:00 on saturday we had made a lot of progress - there was a little guy on the screen, different stages of body parts were shown and all organs were moving around. The idea was already quite clear, so the design work was really minimal at this point. We quit for the day at this point and went playing Heroes of the Storm, which we also did for 2 hours during the afternoon.
Sunday
Sunday started very productive. Until Lunch we added sound and metabolism (which was never touched again, still working like this). I wanted organs to decay visibly - we probably took 3 hours until we found a solution. Originally i wanted to take white organs, color tint them healthy and then change the tint color to something sick. This did not work for Anne, since drawing white organs sucks. We agreed to have multiple stages and just fade them into each other.
In the afternoon i added the bars for energy, poison and age. Around 22:00 we had decaying and repairable organs as well as a functioning body. I let Anne play the first round of the game. It was ok, but not really a game yet. I had a bad feeling, because Sunday evening should have been the "game is basically done, now bake it better". Events still were missing. I was wondering if libGDX was the right choice, if i should switch to Unity to be more productive.
We then played HotS again for 2 hours, which we did also in the afternoon for 1 hour.
Monday
Monday Anne had to work, so i was alone for the day. Christoph had delivered small parts of the music, which i did not add yet - the core was still missing. My sister joined on monday and helped out with some pixel art for habits. I didn't know what she would deliver, to my surprise it looked really good and i scolded myself for not giving her enough work on the day before, where she asked if she could do something for the game.
I started the day by doing some light refactoring for some systems to work better. I wrote everything as quick and dirty as possible, which worked out excellent. Honestly, by the time it was running i knew the context of the programming problem very good and could easily come up with a good design. Not that i had time to do it unless it was a blocker. The consequences of my next big mistake now came back to bite me: i had not prepared any build and deployment, nor had i done it before. I lost 1 hour before i had a desktop build running, 2 more hours before my html build was running mostly. I lost all energy and drive in this period. I like stuff to simply work, not solve problem after problem. This drained me, so at 18:00 after implementing illnesses, habits, the event system and the first 2 events, i wanted to deliver a game that received 0 minutes of polish and balancing. I was feature complete, which i wanted to be 2 days earlier and out of energy. I asked Anne to play a round...and she had fun! She laughed about the events and found joy in growing the organs. The life she had was kind of hilarous. I did know that she was no objective source, but seeing this revitalized me. I pushed hard, added more events, deaht and wind conditions at 21:00 and Christophs music at 22:30.
Anne went to bed and i wanted to work another hour. In reality, i got into the flow and worked until around 2:00 on tuesday, where i made everything ready and deployed it to itch. I wrote a small blog entry and went to bed after 17 hours in front of the PC programming, still full of energy and happiness. I got 4 hours of sleep before going to work. It was interesting to see how much this does to my working quality, namely a lot.
Points
What went right:
- we finished and released a game, yay! My first finished game since i was a kid, where i quickly made a lot of shitty games with * blitz basic (anyone else?) *
- the game overall turned out better than expected
- the idea worked out and all areas were taken (some) care of
- Pizza! We ordered Pizza on Saturday evening and it lasted exactly until monday evening. Cold Pizza is best Pizza.
- ludum dare rocks
What went wrong:
- there is no introduction - there is too much to learn in the first minute, while the game already shoots events up to you as if you knew it all
- we had no time to polish and balance, i did the broadest of balancing monday midnight
- the events are disruptive and can get annoying. I haven't found a solution i really like (should not be more click work). * Do you have an idea? *
- i kind of had to fill every role except for drawing and sound/music. I "produced" not well enough, making collaboration less efficient and created some frustration.
- no preparation - i set up the project and git on saturday, took care of the builds on sunday. All stuff i could have done before, easily 6-8 hours of work (including coosts for collaboration).
- 3 days non stop in front of the PC is probably not ideal. I think my productivity went down. Maybe i'll interchange playing heroes of the storm with taking a walk :D
Next?
We are not done with the game yet. We want to
* polish & balance
* add some events (* do you have ideas? * Please comment! :)
* Make the android build properly playable (there is no hover...)
* i'm also interested in tracking what the player does, for balancing and game design reasons. * Does anyone do this? *
Thank you all, especially you who played membrum and gave some feedback. Also thanks at @pov for making this event possible. We'll be back next time, fully committed and better prepared. See you then!