[evanb]

Mathematica has symbolic and infinite-precision addition, so you can't automatically take advantage of obvious compiled code.

[sfpotter]

What? Arbitrary precision arithmetic implemented in a compiled language will be faster than the alternative. This is no great mystery. The same is true of essentially all low-level symbolic or numerical math algorithms. You need to get to a fairly high level before this stops being true.

[creato]

Of course. The point is whether you interpret a call to arbitrary_precision_add or compile the call doesn't matter much.