This week has seen a lot of progress and some very busy engineers who have so much to do that performance isn’t at the top of the list. is. There has been a ton of progress on the game in the form of getting lots of art assets done and put in the game. I was finally able to ask our main engineer about our performance problems that arose last week when we were trying to get the game onto the iPad. She said that in order to increase performance, we add a line of code that basically tells the device to render everything in low quality. This only effects the appearance of the game by a little bit given that everything is 2D, but the fact that was her answer for solving a performance issue made me frustrated.
Something in the code, or engine is just eating up low end computers and mobile devices, when it shouldn’t be. Our game is a roughly 40mb file, but is eating up 100 percent of a CPU and at times using 1.5gb of ram. This should not be happening. Hailin, and I are trying to figure out what is causing all the problems, but we have very little flash, and actionscript knowledge or experience. We really need some help to get it performing up to snuff.
When taking on the ability to directly manipulate thoughts of other people there are many problems that present themselves. Trying to write down all the thoughts of lots of different individuals becomes problematic because of the number of thoughts a single person can have. The next problem is how to plan for all the possible combinations of thoughts that a player could make. If the player is able to take any thought from any NPC and place that thought in any other NPC in any order then it creates an almost endless number of permutations.
How then to try and bring those permutations to a more reasonable level and still let the player feel like they are making meaningful choices? We have decided to try and limit the players choices in a logical way. We started with a room with 4 people who each have 3 thoughts. Just this limited number of possible thoughts presented thousands of possible permutations. Moving forward we decided that the player would only be able to swap one thought for another, and this took our permutations down about 150. But this still seemed like way to many possibilities for the player to choose from.
We really wanted the solution to be obvious to the player so we wrote our the NPC thoughts in such a way that the solution was obvious to us. We brought in a few people to play test our thoughts and see what solutions they could see and would want to try. What we thought had been obvious to us was not what other people saw. This really clued us in that everyone interprets thoughts differently and our design had some flaws. We could not have a single solution to a problem and expect everyone to manipulate the same thoughts to get there.
So knowing this we decided our game really needed to have many possible solutions to every problem. We decided to start way small for our prototype. We have one room with one NPC who has 3 thoughts. We start the player with 2 thoughts in their inventory and give them a short tutorial on how to place a thoughts. With only being able to switch one thought, this limits our possible outcome to about 9. This is a much more reasonable scope for the prototype than dealing with hundreds, thousands, or millions of permutations.