Yeah I got the deduce bit. Friggin easy trig. It seems to have absolutely nothing to do with z4+1 at all. Just expand RHS and it simplifies down to cos2(Theta)
BTW I got the first part before you replied, but thank you anyway!
How I did the first part was I found the roots of z4+1 (using z = cis[(360k+180)/2],
put them in the form (z-z1)(z-z2)(z-z3)(z-z4),
expanded and simplified down to z4+1 = (z2-zsqrt2+1)(z2+zsqrt2+1)
I like the completing the square method though, thanks for that