mpz_probab_prime_p(n, reps) : how long?

[jojo 29]

My god! many years?! I get a similar result using the bench test of the 
GIMPS project (http://www.mersenne.org/bench.htm)...

Is it possible to get probab_prime_p() faster?


Jojo.

>From: Torbjorn Granlund <tege at swox.com>
>To: "jojo 29" <jojo29118 at hotmail.com>
>CC: gmp-discuss at swox.com
>Subject: Re: mpz_probab_prime_p(n, reps) : how long?
>Date: 10 Mar 2005 11:18:27 +0100
>MIME-Version: 1.0
>
>"jojo 29" <jojo29118 at hotmail.com> writes:
>
>   First, excuse my poor english (It's not my native language, I'm french).
>   Then, do you have an idea of how long an Athlon XP 2200+ (1.8 GHz) needs 
>to
>   execute the function mpz_probab_prime_p(n, reps) with n, a 33 million 
>digits
>   prime number and reps = 10?
>
>I am afraid that a single iteration will take many, many years.
>
>(Literally, I do not intend to exaggerate.)
>
>--
>Torbjörn