mersenneforum.org > Math P-1 factor probability distribution given n, B1 and B2 bounds
 Register FAQ Search Today's Posts Mark Forums Read

 2020-10-07, 12:49 #1 gLauss   Nov 2014 2510 Posts P-1 factor probability distribution given n, B1 and B2 bounds I want to calculate how likely a potential factors in a given bitrange will be missed by P-1, given B1, B2 and n. For ECM, I can get this information from here. It can't be the same as ECM for P-1, because P-1 will use the fact that the factor is of form 2kn+1. Please read about the https://en.wikipedia.org/wiki/Dickman_function first, I will use the same notations. $\Psi(x,B1)$ should be the number of B1-smooth numbers below x, similiar for $\Psi(x,B1,B2)$ . Assume n is prime and large, too. So I need to calculate $p(n,B1,B2,k_{max}) = \frac{\left\{k\in\mathbb{N}|k. The denominator should be (if Chebyshev bias is not considered) $\pi(2nk_{max}+1)$ divided by $\varphi(2n)=n-1$ or around $\frac{(2nk_{max}+1)}{(n-1)log(2nk_{max}+1)}$ or around $\frac{(2k_{max}+1)}{log(2nk_{max}+1)}$. However, I can't come up with a method to count the elements in the nominator, not even if I consider stage1 only. By definition there are $\Psi(k_{max},B1,B2)$ potential values for k where k will be power-smooth with B1/B2 bound, but I don't know how to count for how many of those $2kp+1$ will be prime, too. Any ideas? And if I consider Brent-Suyama extension, then I am completly lost... PS: Reading the source code of Prime95, it seems that it the P-1 factor probability for a mersenne number is the same as the P-1 factor probability for a random number with n bits less. However, I do not yet see why this is. Last fiddled with by gLauss on 2020-10-07 at 12:57
 2020-10-07, 20:53 #2 preda     "Mihai Preda" Apr 2015 5×271 Posts Another P-1 calculator is here: https://github.com/preda/gpuowl/blob...m1/pm1prob.cpp Oh, I think the two are related or similar, so no new insights. Last fiddled with by preda on 2020-10-07 at 20:55

 Similar Threads Thread Thread Starter Forum Replies Last Post Unregistered Information & Answers 13 2011-07-15 15:12 nuggetprime Math 2 2011-03-19 22:14 JuanTutors Software 20 2004-09-26 09:47 dsouza123 Math 2 2004-06-02 02:16

All times are UTC. The time now is 09:53.

Tue Apr 20 09:53:00 UTC 2021 up 12 days, 4:33, 0 users, load averages: 1.95, 2.00, 1.90