Comparing GMP implementation on two different architectures?

[Yusuf Abdulghani (Local)]

Hi,

I am relatively new to GMP and currently evaluating GMP implementations 
on two architectures.

I am wondering what is the best way to compare GMP implementation on 
two different architectures (different CPUs).

Can I do it using the "speed" test?
Or should I be doing the gmpbench?
What does the output of the "speed" test indicate? If I do "./speed -C 
-s 100000 mpn_mul_1.10000", what does the output number next to 
"100000" indicate?

Thank you.

Yusuf