mersenneforum.org  

Go Back   mersenneforum.org > Prime Search Projects > Riesel Prime Search

Reply
 
Thread Tools
Old 2011-07-22, 19:18   #265
kar_bon
 
kar_bon's Avatar
 
Mar 2006
Germany

53178 Posts
Default

Quote:
Originally Posted by SaneMur View Post
I tried running the NewPGen output for k = 11235813213455 into LLR, but apparently LLR can't parse the header NewPGen created:

Code:
ABC 5*389*5776767719*2^$b-1 //NewPGen:3598503246:M:0:2:322
5*389*5776767719 = 11235813213455, and NewPGen required entering it into the app in that format.

Does this header need to be tweaked in any way to have it read by LLR?

Any ideas what might be wrong?
NewPGen will create such ABC-file because (as suggested) the k-value is too big and should be typed as factorization instead.
So the result file also contains such lines.

Edit the resultfile with the 'normal' header for LLR like:
1:M:1:2:258

and replace the wrong k-value with "11235813213455".

Easier way: use srsieve.

Call srsieve as follows:
srsieve -f -G -N 1000000 -P 100000000 "11235813213455*2^n-1"

-f write found factors to srfactors.txt (not needed really)
-G create a prp-resultfile called "t17_b2.prp"
-N max n-value (-n 0 omitted)
-P max sieve depth
and the sequence to search for.

You'll get something like
Code:
47345069:M:1:2:258
11235813213455 4
11235813213455 28
11235813213455 60
11235813213455 64
11235813213455 108
11235813213455 124
11235813213455 160
in t17_b2.prp.

Please read more about the options of srsieve and the tool srfile to convert result-files.

Srsieve is much faster and will find small primes (n-values < 50, NewPGen won't!) for such a sequence.
kar_bon is offline   Reply With Quote
Old 2011-07-22, 21:47   #266
SaneMur
 
Jul 2011

32×13 Posts
Default

Quote:
Originally Posted by kar_bon View Post

Easier way: use srsieve.

Call srsieve as follows:
srsieve -f -G -N 1000000 -P 100000000 "11235813213455*2^n-1"

-f write found factors to srfactors.txt (not needed really)
-G create a prp-resultfile called "t17_b2.prp"
-N max n-value (-n 0 omitted)
-P max sieve depth
and the sequence to search for.

{...snip...}

Srsieve is much faster and will find small primes (n-values < 50, NewPGen won't!) for such a sequence.
Wow!

You are not kidding, thanks! I sieved the exponent up to 1,000,000 and the prime max to 10,000,000,000 and it didn't even take half an hour!

That is what I call really fast!

I left off the -f option but I did pipe the entire console window output to a text file with:

srsieve -G -N 1000000 -P 10000000000 "11235813213455*2^n-1" > my_results.txt

I couldn't read anything as it flew by!

This program is really awesome, thanks again.
SaneMur is offline   Reply With Quote
Old 2011-07-22, 21:52   #267
SaneMur
 
Jul 2011

11101012 Posts
Default

Riesel primes with k = 11235813213455 are prime with n =

4,28,108,124,2524,2540,7388,11584,14196,16948,
20796,38184,46592,64464

Still searching up to n = 100K.
SaneMur is offline   Reply With Quote
Old 2011-07-22, 22:32   #268
SaneMur
 
Jul 2011

32·13 Posts
Default

Riesel primes with k = 1123581321345589 are prime with n =

17, 19, 125, 449, 565, 1339, 2767, 8237, 9665, 17849

still searching up to 100K
SaneMur is offline   Reply With Quote
Old 2011-07-23, 00:19   #269
SaneMur
 
Jul 2011

1658 Posts
Default

So kar_bon doesn't go crazy



PRIME REPORT (updates in bold)

Riesel primes with k = 11235813
exist for n =

2, 23, 26, 28, 80, 83, 98, 127, 152, 182,
347, 388, 392, 400, 416, 542, 830, 839, 1292, 1436,
2572, 4280, 9724, 13843, 15992, 17084, 34076, 44483, 45692, 52036,
85864, 97640, 113716, 161927

Searched up to: 270K

--------------------

Riesel primes with k = 1123581321
exist for n =

33, 42, 117, 157, 177, 373, 870, 2077, 2977, 6874,
7065, 10663, 11410, 16705, 17973, 35698, 52353, 64050, 64702, 64857

Searched up to: 100K

--------------------

Riesel primes with k = 11235813213455
exist for n =

4, 28, 108, 124, 2524, 2540, 7388, 11584, 14196, 16948,
20796, 38184, 46592, 64464, 77260

Searched up to: 100K

--------------------

Riesel primes with k = 1123581321345589
exist for n =


17, 19, 125, 449, 565, 1339, 2767, 8237, 9665, 17849, 67987

Searched up to: 68K
SaneMur is offline   Reply With Quote
Old 2011-07-23, 00:40   #270
kar_bon
 
kar_bon's Avatar
 
Mar 2006
Germany

53178 Posts
Default

Quote:
Originally Posted by SaneMur View Post
So kar_bon doesn't go crazy
Don't worry, all is online!

Another hint and a help for me:

If an odd k-value is divisible by 3, there could exist twin primes for the n-values found (i.e. k*2^n-1 and k*2^n+1 are prime).

You can easily test this:

edit the header of your LLR-input file from

xxxx:M:1:2:258

into

xxxx::P:1:2:257

and run this 'new' file as input for LLR (with a new output file for the primes).

Posting both side primes will save me some time here.

Thanks.

Last fiddled with by kar_bon on 2011-07-23 at 00:59
kar_bon is offline   Reply With Quote
Old 2011-07-23, 03:04   #271
VBCurtis
 
VBCurtis's Avatar
 
"Curtis"
Feb 2005
Riverside, CA

2×2,003 Posts
Default

SaneMur:
You don't need to post updates on your progress on a daily basis. I suggest posting found primes when you reach 500k, or whatever the cutoff is for top-5000 list (currently 667k, I believe), or perhaps both. Any results that can be found in a day or two amount to clutter in the forum- but a monthly update is most welcome, even if you don't reach 500k in that time.
If you read through this thread you are posting in, you'll see that updates are not terribly frequent, yet are still somewhat regular either by Big Round Number or by time.
-Curtis
Edit: Even better, use the "post small primes here and tell us about your progress" thread for these updates below 667k- that thread is designed exactly for folks like you who are just starting out and playing with small primes.

Last fiddled with by VBCurtis on 2011-07-23 at 03:08
VBCurtis is offline   Reply With Quote
Old 2011-07-23, 16:05   #272
SaneMur
 
Jul 2011

11710 Posts
Default

Quote:
Originally Posted by kar_bon View Post
Posting both side primes will save me some time here.

Thanks.
Sure I can do that.

Also:

Reserving k = 10000000001
SaneMur is offline   Reply With Quote
Old 2011-07-23, 16:33   #273
SaneMur
 
Jul 2011

32×13 Posts
Default

Quote:
Originally Posted by VBCurtis View Post
SaneMur:
I suggest posting found primes when you reach 500k, or whatever the cutoff is for top-5000 list (currently 667k, I believe), or perhaps both. Any results that can be found in a day or two amount to clutter in the forum- but a monthly update is most welcome, even if you don't reach 500k in that time.

{snip}

Edit: Even better, use the "post small primes here and tell us about your progress" thread {snip}
Sure thing!
SaneMur is offline   Reply With Quote
Old 2011-07-23, 16:44   #274
Thomas11
 
Thomas11's Avatar
 
Feb 2003

111011011012 Posts
Default

Quote:
Originally Posted by kar_bon View Post
edit the header of your LLR-input file from

xxxx:M:1:2:258

into

xxxx::P:1:2:257
Shouldn't it be the LLR output file (containing the primes), which he shall use as input for the "plus" side?
Otherwise he would rerun the whole file, which is obviously not properly sieved for the k*2^n+1 side...
Thomas11 is offline   Reply With Quote
Old 2011-07-23, 21:16   #275
kar_bon
 
kar_bon's Avatar
 
Mar 2006
Germany

2,767 Posts
Default

Quote:
Originally Posted by Thomas11 View Post
Shouldn't it be the LLR output file (containing the primes), which he shall use as input for the "plus" side?
Otherwise he would rerun the whole file, which is obviously not properly sieved for the k*2^n+1 side...
Sure, misleading used phrase of mine. Thanks for the remark.
kar_bon is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Best Work for Finding Primes Unregistered Information & Answers 9 2012-06-24 13:50
Which settings to choose for best work? jmb1982 Software 2 2009-04-07 09:33
Help test 210885 - Find a new top 5000 prime! SlashDude Riesel Prime Search 121 2008-01-03 08:47
Help test 2995125705 - Find a new top 5000 prime! SlashDude Riesel Prime Search 538 2007-05-08 01:42
The fastest way to a top-5000 prime? lsoule 15k Search 13 2005-09-19 20:24

All times are UTC. The time now is 01:58.

Tue Apr 7 01:58:05 UTC 2020 up 12 days, 23:31, 2 users, load averages: 2.79, 2.39, 2.31

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.