April 28th, 2010
Roly-Poly samurai earned an average of 2.3/5 for fun, 2.6/5 for production values, and 3.1/5 for innovation. Someone also gave it a Bronze Award for Ninja Concentration, which probably means there were only two other entries with ninjas. The winning entry got 3.7/4.2/3.9 and no ninja-based awards, so I'm content with that. Reading the comments, I was surprised how hard people seemed to find it, but maybe I just got too much practice to be objective. A lot of people said the controls seemed backwards, which I can't really argue with. Neither option seemed particularly intuitive to me at first, and I ended up just going with the direction that had been used in the game which inspired Alan. I suspect people would have complained whichever way around it was, but ideally there would have been an option to reverse it.
Also, the winning game used pyglet and doesn't run horribly slowly, despite having a lot more images at a time. It frequently slows down briefly, but runs at what appears to be full speed otherwise. Probably I should look and see how it works.
Someone eventually replied to my complaints about the coursework specification. It turns out we don't need to write a shell at all, just the parser for it. Poor reading comprehension on my part, there. What's annoying me now is that according to John, the comp sci department only has JavaCC 3.2 (the latest is 5.0). I don't notice any differences between the versions in JavaCC itself, but 3.2 predates Java 1.5, so the bit which parses your Java code to find when it ends throws an error when you use syntax that was introduced in 1.5. Which means no generics or for-in loops. After replacing the generics, javac started to give warnings about type-safety or something ("unchecked call to addElement(E) as a member of the raw type java.util.Vector"), which afaik I can only avoid by using generics.
Term started on Monday. I had two exams that day, two on Tuesday, and my essay was also due in on Tuesday. Now I have nothing to worry about except the automata coursework due in two weeks, and then five more exams in the latter half of June.
Of the exams, Combinatorics didn't go very well. There was a 14-mark question that I (quite easily) got a recursive formula for the answer, but decided to come back and derive the closed form if I had time, which I didn't. There was a 12-mark question that I wrote down some observations for, but didn't know where to actually begin answering. And there was a 13-mark question where I got a summation formula, noted that it looked like the closed form would be 3^n, then tried to prove that and got 3^n + 2^n, which contradicted values that I actually worked out.
There was also one 9-mark question where I didn't understand the notation: A ⊗ B, where A and B are sets. It didn't make sense as cartesian product (A × B) in context, so I wondered if it was a typo for intersection, but no correction was given. So I asked the lecturer (I think the person who writes the exam is always present as an invigilator in case of these situations), thinking that if he said it was correct I'd have to do a different question. But then he told me what it was, which I didn't expect: symmetric difference. (Actually, at first he said it was cartesian product. I ummed a bit and then he corrected himself.) I don't remember ever actually using that beyond simple exercises, but I at least remembered what it was, and it made the question even easier than intersection would have been.
So, I think I'll be very lucky to get 70% raw on that; but with the curve, who knows? I think it was a pretty hard exam in general, rather than me just being unprepared. But it's hard to tell with combinatorics, and I always got quite low marks on the assignments (usually 60-70%, IIRC).
For the rest, Vector Analysis was good: I even finished with a few minutes to spare. Analysis III I had a little bit left to clean up when time ran out, but it was otherwise pretty good. Algebra I was okay, despite being a horribly confusing module.
My essay ended up at 4026 words, after I cut a lot of bits that I decided were unnecessary. It ended rather abruptly, but I had no idea what I could write for a conclusion. There also wasn't any particularly hard maths in it, which I suspect will end up costing me marks, but I'm not too worried.