CSAT For Computer Science Undergrad Admissions 2016 Entry

How are things for others? I haven't touched CSAT since actually.. Thought I'd try and kick things off again... This is an attempt to Q15 (might be completely misguided), any thoughts? What are others doing for prep?

I don't think Q15 is answerable. Ignore the curvature of the earth for the moment. If you start n + n/(2 k \pi) miles north of the pole (where k is a positive integer), your journey will have you circling the pole k times, and returning to your starting point. Thus we're after the minimum value of n + n/(2 k \pi) for k a positive integer -- but that set has no minimum element! If you do consider the curvature of the earth, you get the same problem.

Gavin

How can you be so sure that the answer would be a polynomial of degree 2? Why can't it be a polynomial of degree 3, or 4, etc?

Hi just saw this thread! I found Section A pretty easy and 16 and 25 from Section B but I'm still struggling with the other two from Section B. I am preparing for it by looking at MAT and STEP papers at the moment.

Did you have to refer to Fermat's Little Theorem for 25? I understand how to do it once I've referred to Fermat's Little Theorem... But I wonder how you would do it if you didn't know about the theorem? Or perhaps you just pick a different question... I actually couldn't do 25 until I randomly watched a video about the theorem online and I was like wait I've seen that somewhere!

Shots fired by a CompSci Prof from The Other Place!

Sure it is. You may want to read the question carefully - it states "without visiting any point more than twice". Your assumption thus contradicts the question.

Who's assumption? I think you meant to reply to @gavinlowe not me... BTW he *IS* an Oxford CS Professor.

Yes, i meant to reply to him, got mixed up. I don't see what him being a cs prof at oxford has to do with the question (profs can be wrong too, which he is in this case).

Sure it is. You may want to reread the question - it states "without visiting any point more than twice". Your assumption thus contradicts the question.

I actually tried it with Fermat's Little Theorem as well at first but then I also found another way. Not 100% sure it makes sense but I think it does.

I took a similar approach too, although I didn't use induction for the last part. (Hint: note that when n can be written in one of the forms 5k, 5k + 1, 5k + 4 where k is a non-negative integer, it is trivial to show that the expression is divisible by 5. Now consider values of n in one of the forms 5k + 2, 5k + 3.)

So terrified for this exam. Especially since I'm struggling on Q1 xD Can't get anything other than b = a(sqrt(2)) - c(sqrt(3)) and since sqrt(2) and sqrt(3) are irrational, integer multiples of them are still irrational, so the difference between integer multiples of them cannot be a positive integer. (Like you can't make npi = a(sqrt(2)), etc.). So it's a,b,c=0? But, doesn't feel right xD