mersenneforum.org  

Go Back   mersenneforum.org > Great Internet Mersenne Prime Search > Data > Marin's Mersenne-aries

Reply
 
Thread Tools
Old 2004-03-24, 11:04   #1
nucleon
 
nucleon's Avatar
 
Mar 2003
Melbourne

5×103 Posts
Unhappy Automated P-1 thoughts.

Okay I've only done the automated P-1 method two days running, and I don't like it.

Today as an example, I'm running the P-1 checks for 8hrs/day. At the start of the day the client grabs 10 double checks. From the logs, I gather that only one hadn't been previously P-1'd, so the sole P-1 check was completed 90mins into the 8hr cycle, so 6.5hrs is wasted (as I'll terminate the double check LL prematurely)

So if I want to do P-1 checks, I think it's best I'll claim them manually from this forum. It's a pity 'cause I wanted the automatic method to work as I wanted prime credit (Rank 542 and counting) :)

I think I'll leave the automated method for P-1 factors until it's integrated in the prime95 client. (i.e it's just a tick box to select)

-- Craig (account S90106)

Last fiddled with by nucleon on 2004-03-24 at 11:05
nucleon is offline   Reply With Quote
Old 2004-03-24, 12:08   #2
dave_0273
 
dave_0273's Avatar
 
Oct 2003
Australia, Brisbane

1110101102 Posts
Default

I agree with you. When GP2 first left the forums, i tried the semi-automated method. But like you i just couldn't get enough work to do. I don't really care about primenet credit and thats why I now continue to do it manually. I don't mind posting sets for other people to do. Now that I have written a program to do it all, it only takes me a couple of minutes to find exponents that haven't been p-1 tested and post them on the forum.
dave_0273 is offline   Reply With Quote
Old 2004-03-24, 12:23   #3
markr
 
markr's Avatar
 
"Mark"
Feb 2003
Sydney

23D16 Posts
Default

Good points, Nucleon...

I agree, exponents without P-1 are too thinly spread in the current DC ranges for the fully automated method, unless a line of normal work is left at the top of the worktodo to be done after the P-1 tests. Maybe the earlier P-1 work by the Mersenne-aries has already filled in most of the gaps?

One way to get P-1 work through PrimeNet might be to find exponents that need P-1 just ahead of PrimeNet's assignments but still in its ranges, add them to the worktodo, & connect manually. PrimeNet will remove any that are assigned to others.

P-1 testing adds to factoring credit, not LL credit, so I guess it depends whether you want to move ahead in one or both rankings. Of course a single P-1 test only gives 0.001 credit! Happy crunching!
markr is offline   Reply With Quote
Old 2004-03-25, 02:45   #4
Complex33
 
Complex33's Avatar
 
Aug 2002
Texas

5×31 Posts
Default

The way I handle it on a friends computer with 4gigs of RAM is pickup a bunch of DC's until I get about 200 or so that are needing P-1 factoring. Dump the rest and there is enough work for a while :)
Complex33 is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Automated LLR testing with LLRnet mdettweiler No Prime Left Behind 24 2011-11-04 19:20
Automated PRP using LLRNet axn Sierpinski/Riesel Base 5 73 2008-11-26 03:46
automated factor submission ltd Sierpinski/Riesel Base 5 22 2007-07-18 22:22
Automated PRP discussions ltd Sierpinski/Riesel Base 5 20 2006-09-02 22:19
Semi-automated P-1 testing GP2 Marin's Mersenne-aries 2 2003-09-29 19:01

All times are UTC. The time now is 02:30.

Sat Jun 6 02:30:56 UTC 2020 up 73 days, 4 mins, 0 users, load averages: 1.30, 1.30, 1.26

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.