20030423, 13:01  #1 
Feb 2003
2^{5} Posts 
Testing an expression for primality
Can anyone recommend a software (windowsbased) for testing the primality of a given mathematical expression? Most that I know of are for specific types of primes. (e.g. fermat, mersenne).
 Ray 
20030423, 18:40  #2 
∂^{2}ω=0
Sep 2002
República de California
2·3^{2}·653 Posts 
A google search for "general purpose primality proving software" turns up the following page:
http://directory.google.com/Top/Science/Math/Number_Theory/Prime_Numbers/Primality_Tests/Primality_Proving/Software/ which has links to all the relevant such programs. 
20030423, 21:37  #3 
"Mark"
Apr 2003
Between here and the
6838_{10} Posts 
Try WinPFGW (there is also a linux version available). It is based upon George's FFT code used in the GIMPS project. It can do both PRP and primality testing.
Although the project is open source, nobody has been able to write a version of PFGW that can run on other nonx86 CUPs. 
20030424, 10:37  #4 
Feb 2003
2^{5} Posts 
Thanks.

20030425, 05:05  #5 
Aug 2002
3·83 Posts 
WinPFGW is somewhat limited in proving primality of numbers which don't have special forms... for that you might want to try one of the other linked programs.

20030501, 06:37  #6 
Feb 2003
40_{8} Posts 
Yes, I'm now using WinPFGW, and it has the option to evaluate an expression for primality, aside from those of special forms. It's not that limited after all.[/i]

20150823, 18:09  #7 
Mar 2009
Indiana, United Stat
2^{4}·3 Posts 
Best primality proving program?
I have a question about which software is the most appropriate or fastest for proving. I thought that this thread is the most appropriate. But this thread may be forgotten or outdated after 12 years?
Go to http://primes.utm.edu/primes/search.php . Where it says digits, put in 411101 for the minimum and 411101 for the maximum. Click on “Search database for matches.” The result shown would be somebody’s discovery of 1874512^65536+1. I spent months sieving in NewPGen for k*1874512^65536+1 where k is a number between 2 and 400,000,001. It’s now sieved past 140.7 trillion. My understanding or misunderstanding was that Generalized Fermats must be easier to discover if there are more primes of this type than any others in Chris Caldwell’s Top 5000 database. I also wanted to see if I could discover an arithmetic progression this way. I’m running Windows Vista and since July 7, I’m only in line 89 in OpenPFGW. Is OpenPFGW, Prime95, LLR, or something else the fastest prover for NewPGen results for Windows Vista for Chris Caldwell’s Top 5000 list? That is what I please would like to know. EDIT: There’s another reason why I sieved above 400,000 digits and chose something based off of 1874512^65536+1. I wanted to find a prime with more than “401k” (401*1024 or 410,624) digits. 
20150823, 19:11  #8  
"Bob Silverman"
Nov 2003
North of Boston
2^{3}·937 Posts 
Quote:
You say Generalized Fermat's must be easier to discover". Easier than WHAT? You also say "if there are more primes of this type than any other... in the datbase". Why don't you just count them? Then you will know rather than asking "if". There are either more primes of this type of there aren't. There is no "if" Finally one must ask why having more primes of a given type in a given database means that they must be easier to find. How does one reach this conclusion? It could also mean "the tools people have are amenable to primes of this type but not other types". In CS, "easier" means "has lower time complexity". Is this what you mean? Quote:
Your terminology is not well defined. 

20150824, 18:45  #9  
Mar 2009
Indiana, United Stat
2^{4}×3 Posts 
Responding to R.D. Silverman
Quote:
I meant that they would take less time to discover. If you want to call it lower time complexity, then fine. I know discovering an arithmetic progression is trivial or superfluous. If I discover one then it’s a happy accident. Somebody discovered that 1874512^65536+1 is prime, so if I discover that k*1874512^65536+1 is prime then I’ll see if there’s a progression. I doubt if this progression will be discovered, but I’m searching for this as an afterthought. This seems to be the first time we’ve heard of each other and I had forgotten to mention something. Go to http://www.primenumbers.net/prptop/prptop.php , scroll down, where it says find by discoverer, select Matt Stath, and I have found 49 probable primes in the past. The top 2 that I discovered with over 40,000 digits were with NewPGen and OpenPFGW. After being on Henri & Renaud Lifchitz’ List, I’m trying to be on Chris Caldwell’s List. My question is still about whether OpenPFGW, Prime95, LLR, or something else would be the fastest prover. However, Jean Penne’s LLR has discovered 4,854 primes at http://primes.utm.edu/bios/top20.php...&by=PrimesRank . I’m going to test for a least a day to find out if LLR or Prime95 are faster than OpenPFGW. If anybody else is reading this and cares to respond about a program that is faster, then please do. EDIT: I had forgotten to mention that according to http://primes.utm.edu/top20/page.php?id=14 , I'm aiming too high with too many digits for an arithmetic progression. Last fiddled with by stathmk on 20150824 at 18:56 

20150824, 18:48  #10  
Sep 2002
Database er0rr
29×151 Posts 
Quote:
Last fiddled with by paulunderwood on 20150824 at 18:49 

20150824, 19:22  #11  
"Mark"
Apr 2003
Between here and the
2×13×263 Posts 
Quote:
If you wan to get a number in the Top 5000 ASAP, then it will require a bit of luck as there are a number of factors including: how many people are searching (if a distributed project), how large a prime do you want to find, what sieving software is available, what PRPing software is available, can primality be proven, how much horsepower you can dedicate to the search, do you have a GPU. I'm certain there are many other factors that I haven't even listed. 

Thread Tools  
Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Primality testing nonMersennes  lukerichards  Software  8  20180124 22:30 
Testing Mersenne cofactors for primality?  CRGreathouse  Computer Science & Computational Number Theory  18  20130608 19:12 
a new Deterministic primality testing  wsc812  Computer Science & Computational Number Theory  36  20130304 06:25 
Testing if expression over the reals is negative  CRGreathouse  Math  3  20090405 19:12 
a new primality testing method  jasong  Math  1  20071106 21:46 