Closed kevinywlui closed 8 years ago
Yes, definitely a great idea. I also have a conjectural formula for J1(p)..., see http://wstein.org/papers/j1p/ section 6.2.3. It's a low bound at least. You could make it available via proof=False with a reference to the paper.
We know the rational torsion order of J0(p) explicitly and the rational torsion order of elliptic curves. Should the multiple_of_order and divisor_of_order code reflect this?