In my previous post I talked about the process of tearing the game down, so that I could build it up again. Today I created a spreadsheet containing information about each scene in the first few minutes of the game. It proved to be far more time intensive than initially anticipated.
Afterward, I exported each file as a .avi using the SCAT tool I mentioned in the previous post. The HTML5 video player can’t read .avi, so I had to convert them to .mp4 using Handbrake. Again, a time consuming process. Surpsingly, the size of each file shrunk considerably. One scene went from 9mb to 500k! That’s an 18x reduction. needless to say, this is fantastic becuase it makes for smaller files for me to have to stream, but I can also switch between scenes more quickly.
The footage from the SCAT tool also provides a debugging overlay for information, so I had to dive into the source code for SCAT and remove that overlay. The tool was written in Visual Basic, and I hadn’t used VB before, so that was a decent lesson this morning. When all was said and done, we had clips that looked like this, below.
Organizing the data
Now that I had about 30 clips chopped it, it was time to work on the spread sheet and make sense of this mess.
My notes are pretty meticulous, and I did that for a reason. I wanted to make sure that all of my clips would be able to align with one another. I also wanted to maek sure that all of the scenarios played out correctly. For example, if you fail to catch an auger in the bathroom, they may make their way out the door and into hall-2. If you do catch the auger in the bathroom though, you’d never see the clip in hall-2, so I needed to plan for every scenario.
Moreover, this allowed me to plan out how to architect this application. I’ll obviously break this up into several functions, but now I’m considering things like passing in callbacks as parameters, so that the clip will know whther or not it has to return to another clip, and if so, at which point in time.
Certainly this is no easy task, but at least having a general outline of information at my finger tips helps me to visualize how this will begin to look.
Timing is everything
To verify that I had the correct time stamps for each of these events (Ex: when a clip starts or stops, when a trap can be triggered), I watched a few people play through the game on YouTube before finally deciding to play the 32X version through an emulatior so that I could save my progress at 30 second intervals, then sit in each room and make note of any activity. This is waht allowed me to have such precious.
Tedious, yes, but in a game of this nature your timing needs to be exact.
There’s still lots to go, but I figured I’d post today’s work for those of you who were interested. If you’d like to help, please contribute to this excel spreadsheet. At this moment I am looking for time stamps for when a scene starts / ends, whether or not character can be captured, and if so, at which time.-----------------------