mersenneforum.org Does NewPGen have a bug?
 User Name Remember Me? Password
 Register FAQ Search Today's Posts Mark Forums Read

 2008-12-11, 04:34 #12 Kosmaj     Nov 2003 362210 Posts And you have a level of arrogance approaching infinity Code: b=5684341886080801486968994140701; n=2481892977737648921240994084183; p=3433683820292512484657849089373; m=1;while(n>0,if(n%2>0,m=(m*b)%p);b=(b*b)%p;n=n>>1); m=1721733593148639177176135185028
 2008-12-11, 04:57 #13 CRGreathouse     Aug 2006 3·1,993 Posts Sure, though Code: lift(Mod(b,p)^n) would actually be faster for n large enough to matter.
2008-12-11, 05:23   #14
gd_barnes

May 2007
Kansas; USA

33×17×23 Posts

Quote:
 Originally Posted by henryzz what is the crossover point is it 3 ks or more or does it depend on the size of the k or something else

In my experience, use sr2sieve for 3 k's or more, regardless of the size or nature of the k-value. Use 1 or 2 instances of sr1sieve for 1 or 2 k's. I believe the k-value must be > 2^n where n is the lowest n-value being sieved, although I think that restriction also exists in NewPGen.

I haven't done any testing with exotic forms but I'm confident that this is the fastest way for the standard k*2^n-1 and +1 forms. In no case is NewPGen more efficient for these forms.

Keep in mind that you have to run srsieve first to get a "starting file" before using its output file as input to sr1sieve or sr2sieve. I generally use srsieve to sieve to P=250M, 500M, or 1G before changing over.

Once you get used to running them at the command prompt instead of using the GUI like NewPGen has, it's child's play to use them.

Gary

Last fiddled with by gd_barnes on 2008-12-11 at 05:26

2008-12-11, 05:32   #15
mdettweiler
A Sunny Moo

Aug 2007
USA (GMT-5)

3×2,083 Posts

Quote:
 Originally Posted by gd_barnes I believe the k-value must be > 2^n where n is the lowest n-value being sieved, although I think that restriction also exists in NewPGen.
Um...I presume you mean that the k-value must be *less than* 2^n, not greater?

2008-12-11, 07:43   #16
henryzz
Just call me Henry

"David"
Sep 2007
Cambridge (GMT/BST)

2·2,969 Posts

Quote:
 Originally Posted by gd_barnes In my experience, use sr2sieve for 3 k's or more, regardless of the size or nature of the k-value. Use 1 or 2 instances of sr1sieve for 1 or 2 k's. I believe the k-value must be > 2^n where n is the lowest n-value being sieved, although I think that restriction also exists in NewPGen. I haven't done any testing with exotic forms but I'm confident that this is the fastest way for the standard k*2^n-1 and +1 forms. In no case is NewPGen more efficient for these forms. Keep in mind that you have to run srsieve first to get a "starting file" before using its output file as input to sr1sieve or sr2sieve. I generally use srsieve to sieve to P=250M, 500M, or 1G before changing over. Once you get used to running them at the command prompt instead of using the GUI like NewPGen has, it's child's play to use them. Gary
thanks that was what i suspected
i have used srsieve and sr1sieve for a bit but i have never used sr2sieve although i should be able to work it out when i need to
i usually work on ks in pairs so i have been using sr1sieve

2008-12-11, 11:46   #17
Flatlander
I quite division it

"Chris"
Feb 2005
England

31·67 Posts

Quote:
 Originally Posted by henryzz thanks that was what i suspected i have used srsieve and sr1sieve for a bit but i have never used sr2sieve although i should be able to work it out when i need to i usually work on ks in pairs so i have been using sr1sieve
Well, I still find it confusing but here's a link I still refer to:
http://mersenneforum.org/showpost.ph...9&postcount=67

( I think you can ignore the cache sizes parameters and reference to Core(). )

 Thread Tools

 Similar Threads Thread Thread Starter Forum Replies Last Post Cybertronic Factoring 0 2014-03-22 10:07 pepi37 Software 12 2013-12-25 18:59 roger Information & Answers 0 2007-04-04 22:38 Cruelty Riesel Prime Search 3 2006-02-15 05:15 Zenmastur Software 4 2003-08-02 19:43

All times are UTC. The time now is 02:26.

Sat Dec 4 02:26:42 UTC 2021 up 133 days, 20:55, 0 users, load averages: 1.55, 1.37, 1.31

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, 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.