Wolf3d Haven Forum

Please log in or register. Smile

Join the forum, it's quick and easy

Wolf3d Haven Forum

Please log in or register. Smile

Wolf3d Haven Forum

Would you like to react to this message? Create an account in a few clicks or log in to continue.
Wolf3d Haven Forum

A friendly Wolfenstein 3D community, about Wolfenstein 3D, the game that gave birth to first person shooters...


    1,111st post and fun looking for probable primes where every digit is 1.

    stathmk
    stathmk
    Veteran
    Veteran


    Male
    Number of posts : 1740
    Age : 42
    Location : Indiana, United States
    Job : fast food worker & wolfensteingoodies.com webmaster
    Hobbie : old games & young dames
    Registration date : 2008-04-08

    1,111st post and fun looking for probable primes where every digit is 1. Empty 1,111st post and fun looking for probable primes where every digit is 1.

    Post by stathmk Thu Sep 18, 2014 12:15 pm

    This is my 1,111st post on https://wolf3d.darkbb.com. 1,111 isn't prime. It's 11x101. If you have a number with a prime number of digits, every digit is 1, and it also happens to be a probable prime, then it's a repunit probable prime.

    Repunit curiosities: http://www.elektrosoft.it/matematica/repunit/curiosity.htm

    - Go to http://www.alpertron.com.ar/ECM.HTM.
    - Set the number of digits in a group to 999.
    - Uncheck Automatic Switch to SIQS
    - Check Verbose mode
    - Check Use Cunningham tables on server
    - Click change

    - Type a completely random number. It's not very likely to be prime.
    - Type (10^1-1)/9. It's 1 and wouldn't normally be called prime.
    - Type (10^2-1)/9. It's 11 and prime.
    - Type (10^19-1)/9. It's 1,111,111,111,111,111,111 and prime.
    - Type (10^23-1)/9. It's 11,111,111,111,111,111,111,111 and prime.
    - Type (10^317-1)/9. It takes a few minutes to prove as prime.
    - Type (10^1031-1)/9. It takes less than an hour to prove as prime.
    - (10^49081-1)/9 is a probable prime.
    - (10^86453-1)/9 is a probable prime.
    - (10^109297-1)/9 is a probable prime.
    - (10^270343-1)/9 is a probable prime.
    - The next repunit probable prime would have more than 3 million digits and would take hours to 3 days to prove as a probable prime with Windows XP or faster.

    I volunteered for the Repunit Project on about 11-11-2011. I’ve now tested over 1,111 repunits. Now that I’ve had a 1,111st post about repunits, I'm going to finish my current testing list, tie up the loose ends of what I'm doing, and quit the project this year. I need to move on and work on http://www.wolfensteingoodies.com and other aspects of my personal life.

    Just for fun, you might want to a test few repunit candidates for the Repunit Project. Go to http://www.elektrosoft.it/matematica/repunit/repunit.htm to contact Maksym Voznyy, Danilo, and Giovanni, they will give you a list of numbers, and tell you how to set up. No previous computer code experience is necessary. Maybe you could test 11 or 12 numbers.  It takes me a little longer than 2 days on 1 of my Windows Vista processors to test one number with over 3 million digits. It's 6 times faster with NVidia Cuda, which I don't have. Contact Danilo about Nvidia Cuda. You can have your computer test it in the background while you sleep, are at work, do homework, check email, or play Windows games because it’s a low priority program.

    If you discover a Repunit probable prime, then it will be very near the top at http://www.primenumbers.net/prptop/prptop.php

    This is just for fun.  I don't want to obligate anybody to do any of this if they don't want to.

    EDIT: I've asked 3 of the mathematicians and dark_wizzie for permission to post this as my 1,111st post on the forum and nobody objected. I have no plans to post another thread about Mathematics unless we discover the bigger repunit probable prime with over 3 million digits! I joked to dark_wizzie that at the rate that he's posting, he should reach 11,111 posts in the year 2022.

    EDIT 2:  Danilo has contacted me back about the Nvidia Cuda GPU:

    Hi Matt,

    I am fine with your post. Just a small detail you might want to clarify. The GPU program is just for sieving the candidates, it cannot test for primality. Also I would estimate that a GPU is about 100 times faster than sieving on a CPU.

    Hope you are fine.

    Cheers,
    Danilo


    Last edited by stathmk on Sat Jan 21, 2023 5:49 am; edited 1 time in total
    stathmk
    stathmk
    Veteran
    Veteran


    Male
    Number of posts : 1740
    Age : 42
    Location : Indiana, United States
    Job : fast food worker & wolfensteingoodies.com webmaster
    Hobbie : old games & young dames
    Registration date : 2008-04-08

    1,111st post and fun looking for probable primes where every digit is 1. Empty Repunit Project

    Post by stathmk Wed Oct 05, 2016 5:22 pm

    I don't know if http://www.elektrosoft.it/matematica/repunit/repunit.htm is going to be continued or updated again.  It says that it hasn't been updated since Oct 26, 2013.
    stathmk
    stathmk
    Veteran
    Veteran


    Male
    Number of posts : 1740
    Age : 42
    Location : Indiana, United States
    Job : fast food worker & wolfensteingoodies.com webmaster
    Hobbie : old games & young dames
    Registration date : 2008-04-08

    1,111st post and fun looking for probable primes where every digit is 1. Empty 5,794,777 and 8,177,207-digit updates!

    Post by stathmk Sat Jan 21, 2023 5:47 am

    For some technical reason that I don’t know if I want to learn and explain, (10^49081-1)/9 wasn’t proven as a prime until March 2022 at https://primes.utm.edu/top20/page.php?id=57 .  Then it went onto Hacker News at https://mersenneforum.org/showthread.php?t=28252 .
     
    I had forgotten to mention in 2021 that at http://www.primenumbers.net/prptop/prptop.php and http://www.primenumbers.net/prptop/searchform.php?form=%2810%5En-1%29%2F9&action=Search that (10^5794777-1)/9 and (10^8177207-1)/9 were discovered to be probable primes in 2021.
     
    I think that by default that Microsoft Word 2007 can fit 3,486 digits of the 11 point Calibri font on 1 page.  That’s if you don’t change the font, margins, spacing, or anything else.  This means that it would take 2,345 and three-quarters pages to print out the 8,177,207-digit number!  It means that it would take almost 7,132 pages to print out the 24,862,048-digit proven prime at https://primes.utm.edu/top20/page.php?id=3 !
     
    The Repunit Project is still at http://www.elektrosoft.it/matematica/repunit/repunit.htm if you’re interested and want to use some of your computer time for fun.  I’m listed as one of the members who contributed at https://www.kurtbeschorner.de/ .

    Sponsored content


    1,111st post and fun looking for probable primes where every digit is 1. Empty Re: 1,111st post and fun looking for probable primes where every digit is 1.

    Post by Sponsored content


      Current date/time is Thu Feb 02, 2023 11:53 pm