The weekend before christmas me and friends were participating in our first Ludum Dare Jam spending 72 hours making a little adventure game. It is called Neukoelln Nights and you may find more about it on my website. Since not everything turned out as we expected it, I reconsidered some aspects of the development process and here is what I came up with:
What went right?
1. fast decision
As for a short term competition like ludum dare, it is an advantage to get started very fast. Since the theme “alone” didn’t quite fit to the universal gameplay we had in mind before, we had to reconsider a new one. After brainstorming gameplay ideas and scenarios with the group, we came up with an old school adventure game idea very fast (2h). The idea of a one-room-adventure in which you need to get rid of all party members except the girl of your dreams, was crisp and seemed manageable and funny enough to entertain the players as well as ourselves during the development process.
2. skill mix
With four members beeing a relatively big group we found the configuration turning out pretty well for an adventure game. Our two artists David and Merlin dealt with the high amout of environmental pixel graphics and character Sprites.Robet, responsible for code started from scratch and managed to set up the core scene at day two so we could put in art on the fly whenever necesary. Being primarily in charge for design, I planed the level scale, quest tree, character motivation and sound.
Using browser based java script has a lot of advantages (and some disadvantages mentioned later on):
- easy testing on a local host
- simple deployment and access for the players
- no downloading
- playable on mobile devices
What went wrong?
It turned out that canvas has problems running smoothly on firefox and chrome as far as we can tell. The bug makes the game freeze in different situations in the game and sometimes even causes the browser to crash entirely.Along with the advantages of web deployment comes the pain of providing a flawless experiences on every browser. Not everyone has the latest updates for his browser and not every browser supports every feature yet.
2. complexity & time
We didn’t manage to put everything in the game we wanted (of course). The first version of the game featured 9 different characters with animations for walking, idle and talking, interactive environment, an item system and a dialogue menu, an intro screen and a whole bunch of audio. Everything a good adventure needs. At the evening of day two, when we realized our plans where to ambitious for a three day project, I had to trash my allready finished work on the quest tree and start thinking about a more simple version using only the features we allready implemented. That also meant trashing two days of valuable manpower but we had to go on. For the purpose of documentation take a look at these drafts:
So far our hero would walk and talk to other people triggering events. Walk and talk. Thats not much but we came up with a simplyfied dialogue structure which featured a trial and error way of solving the puzzle. This means in every game stage there was a person you could talk to, whos actions would have forced a party guest to leave (progress) and a person who made the player fail. With every correctly triggered person you would advance through the phases and in every phase the situation was slightly different so the player would understand his influence on the guests. Take a closer look to this discarded structure sheet:
In the end we didn’t even had the time to let the player fail correctly, resulting in some kind of interactive story in which the players only challenge is to talk to every character randomly, advancing automatically from phase to phase.
To finish the game we would have needed two to three more days setting up the engine, making it a title people could spend 10-20 minutes on and have fun with. Otherwise we could have cut content and freedom in order to achieve more gameplay but the result would have been a rather poor p&c adventure I guess.