I hate exams. Especially long exams. Especially long exams which you were previously told would be an hour shorter.
And especially long exams which you were told would be shorter that test you on stuff like IEEE 754 floating-point numbers. Yep, I’ve had to use that in real life… not.
taking over the world... 
I have no idea what the exam was on, but my guess would be that if you ever do any serious scientific computational work you’re going to run into this, especially if you need to know whether or not your model will need to use single or double precision floating point.
From a hardware point of view it is also useful so you can understand why, for example, the current Cell processor isn’t interesting for a lot of scientific codes (because it does single precision fast, but not double)..
If you’re subscribed to linux-aus, just look at a thread a few weeks back on free computing textbooks. Yes, it’s *that* exam.
Unfortunately I’m not on that list any more, too little time to read it! 😦