Project 1: update: Milestone 2 Requirements

As mentioned on the project announcement, Project 1 Milestone 2 has two parts, a demonstration, and some writing:

1. The demonstration requires you to demonstrate a flocking “toy” – a program that demonstrates flocking behaviors, and allows some fiddling with it (so that it might be fun to play with – we did not require it to be a good toy). It’s OK if the “interaction” is to tweak internal parameters in the code, but if this is the case, you’ll need to demonstrate it in person.
2. The writing requires you to think about the performance of your system, and maybe to experiment with it a bit.

For the demonstration part, you have two choices:

1. We will have a lab session during the class time slot on Friday, September 23rd. You can show your program to the course staff (Mike, David, Gautam) – maybe sure one of us crosses you off the list. I think we’re all interested in seeing what you’re doing, so preferably you will show it to all of us.
2. You can send email with a web link where we can see the program in action. You might need to provide some instructions on how to play with the program. Please do not send the program by email – just send a web link in the email that we can visit.

The lab sessions seemed to work pretty well, so we encourage people to take advantage of them. Its an opportunity to see what others are doing, and share what you’ve learned about implementing Javascript graphics and whatnot. By now, you are all more expert at it than the course staff!

For the questions, please place a file in your handin directory named “M2.*” (where * is an appropriate extension – be it html, txt, doc, pdf, …). Please make sure the file name begins with M2, or we might not find it.

Please discuss the following questions:

1. How does your flocking system perform? Is the performance good enough for your game ideas?
2. How does this performance change as the size of the flock (number of objects)?
3. At what point does your program break down as you increase the flock size (how many “boids” can you handle before the system feels too slow)?
4. When things start to get slow, where is the bottleneck in your program? How did you figure that out? (We appreciate that we haven’t given you tools for doing this scientifically)
5. What might you do if you felt like you needed to get your system to perform with even larger flocks? (note: you may not need to do this. you can probably handle a big enough flock for your game needs – or it’s probably easier to change your game needs).