Before this class, I thought about game design as a relative solitude process, much like more traditional arts, where the artist build, reflect and refine the artifact on their own so that the artwork is largely a product of individual expression and creativity. I was also assuming a perfectionist mindset, as traditional artists usually have high standard judging their own creation and often spent enormous time and effort just for tiny improvement. It turns out that those presumptions are all very wrong through this class!
The first thing that struck me was the importance of rapid play testing. As I was more familiar with traditional arts, the concept was largely foreign to me — usually artists don’t mind much about audience’s opinions when creating arts. However, the case is completely different for games, because they are pieces of art that by definition require interaction with the audience. Moreover, such interaction is an integral and important part of the art itself. In our projects, we started play testing from the very beginning, as soon as we have a very crude prototype made from scratch papers. This turns out to be a very effective methodology and allow us to spot problem faster and come up with new ideas when watching the game happening in real time.
Another thing that struck me was that it’s ok to forget about perfectionism. While traditional art focuses a lot on artist expression, games also emphasize the experience it bring to the audience — and a fun game doesn’t need to be perfect. I’ve learnt to hear more about what players think about the game, on top of what I think, and this helps me a lot to figure out the key problems and improvements and postpone the minor ones.
During the final project, I took the major responsibility of designing one of the puzzles. The game we were working on was an alien-themed escape room, and I decided to go for a deciphering puzzle — but with “real” linguistic element, rather than artificial letter/word substitution. I was very passionate about this idea because it was novel and I can leverage my linguistic skills (I was using a complete constructed language). However, we spent much more effort on getting the puzzle workable that I’d ever anticipated. The major challenge was to 1) adjust the difficulty, and 2) account for variations in player groups. First, something that seems relatively easy to someone with background in linguistics (which I am), or someone that know the answer (which my team members are, because of lack of confidentiality protocol within our group!), can be hard or near impossible for an average player with probably no linguistic background, and it’s not at all obvious whether this is the case. Moreover, to match other parts of the game I sometimes need to make adjustment to an old play tested version of the puzzle, and it’s not obvious whether a seemingly minor adjustment have major impact on difficulty. This happened when I need to update the challenge sentence from a single sentence to a short passage of three sentences — it unexpected makes a previously working version near impossible, because a longer challenges have more glyphs to be matched, and it does not work like 1+1+1=3. Fortunately, a single play test exposes the problem and we came up with adjustments (let players fill in blanks in a template rather than figuring out the whole passage) to lower difficulty on the spot. Secondly, it turns out that different groups perform very differently on the same puzzle. For example, groups with multi-lingual background tend to perform better, and the similarity between languages the group know and the constructed language has an impact. The communication pattern of the groups also have a great impact. More talkative groups make much faster progress even if lots of time they seem to be talking about insubstantial details from a bystander viewpoint, while groups that don’t talk that much can just stare at each other and stuck until time runs out. There are some remaining variations that are probably just caused by different level of linguistic talent. To compensate for such variation, we found a progressive hint system really helpful. We give out small pieces of hints at relatively short interval, so groups never get stuck for too long. All the hints, when put together, pretty much tells players the answer right away so that we make sure a group that doesn’t perform so well can still proceed to next part of the game, while stronger group may choose not to see some hints to earn bonus points.
Overall, I’ve learnt a lot about the game designing process and team collaboration skills. I will certainly apply the methodologies and skills I’ve learnt in this class in my future game projects.
