mersenneforum.org Manual assignment
 Register FAQ Search Today's Posts Mark Forums Read

 2009-10-28, 00:08 #1 Unregistered   230716 Posts Manual assignment Hi guys, I am novice in PimeNet so sorry for stupid questions but I cant realize whats wrong... I'm trying to get Manual Assignment from http://mersenne.org/manual_assignment/ with custom exponent range for double-check test (actually it doesn't work for any work type to me). When I set an exponent number (say 54321123) it gives me an error: Error code: 40 Error text: No assignment available meeting CPU, program code and work preference requirements, cpu_id: 182578, cpu # = 0, user_id = 30876 Somebody can explain this please?
2009-10-28, 02:29   #2
lfm

Jul 2006
Calgary

52×17 Posts

Quote:
 Originally Posted by Unregistered Hi guys, I am novice in PimeNet so sorry for stupid questions but I cant realize whats wrong... I'm trying to get Manual Assignment from http://mersenne.org/manual_assignment/ with custom exponent range for double-check test (actually it doesn't work for any work type to me). When I set an exponent number (say 54321123) it gives me an error: Error code: 40 Error text: No assignment available meeting CPU, program code and work preference requirements, cpu_id: 182578, cpu # = 0, user_id = 30876 Somebody can explain this please?
It means there is no double check exponents ready for you to test. Try test factoring or first time checks or try other ranges.

2009-10-28, 07:42   #3
markr

"Mark"
Feb 2003
Sydney

3·191 Posts

Quote:
 Originally Posted by Unregistered Hi guys, I am novice in PimeNet so sorry for stupid questions but I cant realize whats wrong... I'm trying to get Manual Assignment from http://mersenne.org/manual_assignment/ with custom exponent range for double-check test (actually it doesn't work for any work type to me). When I set an exponent number (say 54321123) it gives me an error: Error code: 40 Error text: No assignment available meeting CPU, program code and work preference requirements, cpu_id: 182578, cpu # = 0, user_id = 30876 Somebody can explain this please?
Welcome to GIMPS!

To expand a bit on what lfm said... There's very few numbers of that size available for double-checking, because very few LL tests have been done that far ahead of the main wave. Have a look at http://www.mersenne.org/primenet/. It shows how many candidates have had LL tests and how many are assigned & available for different work types. In the whole range from 54000000 to 55000000, there's only one number available for double-checking. If you really, really wanted to double-check it, it would be possible to find out what it is and grab it with the manual assignment page. (Personally, I prefer smaller numbers. That LL test would take way too long for me.)

If you were interested in a particular range, maybe explore it with the PrimeNet reports, like the Exponent Status page http://www.mersenne.org/report_exponent/. Remember, for Mersenne numbers 2^p-1, p is prime. So there's no details for your example 54321123 because it's not in the database. (Yeah, I know, that was just a random number you chose for illustration. I'm just being a trifle pedantic. )

 2009-10-30, 07:24 #4 Unregistered   23×5×163 Posts Thanks markr, you gave me good understanding
 2010-09-29, 15:08 #5 Commaster   Jun 2010 Kiev, Ukraine 3·19 Posts It seems that it's impossible to manually request an ECM on Fermat numbers. (And, probably, ECM on Mersenne numbers with known factors). Am I doing something wrong?
2010-09-29, 15:28   #6
rajula

"Tapio Rajala"
Feb 2010
Finland

32·5·7 Posts

Quote:
 Originally Posted by Commaster It seems that it's impossible to manually request an ECM on Fermat numbers. (And, probably, ECM on Mersenne numbers with known factors). Am I doing something wrong?
If nothing has changed since the last time I manually requested ECM on Fermat numbers, the easiest way is to simply add the work you want to do in the worktodo.txt. For example, to run 3 curves on F_20, you want to add the line

Code:
ECM2=1,2,1048576,1,3000000,300000000,3
When the client connects to the server, you will get an assignment key on the job.

(You can also get the assignments on the page http://mersenne.org/manual_assignment/. You will just have to change slightly the url you get from requesting ECM on a specific Fermat number. Namely, the manual assignment -page lets you request work type 5 (ECM) and you want 6, so just change in the url "pref=6". Remember to be logged in when you do the request.)

 2010-09-30, 14:02 #7 Commaster   Jun 2010 Kiev, Ukraine 3·19 Posts Thanks, rajula, that did the trick!
 2010-10-07, 04:10 #8 Commaster   Jun 2010 Kiev, Ukraine 3×19 Posts According to PrimeNet Stats, 40M-49M have a bunch of available P-1 assignments. Nevertheless, requesting P-1 in that range, results, as usual, in DC-assignments. Server is greedy? My hardware is too weak(I doubt)?
2010-10-07, 05:26   #9
Uncwilly
6809 > 6502

"""""""""""""""""""
Aug 2003
101×103 Posts

19·499 Posts

Quote:
 Originally Posted by Commaster My hardware is too weak(I doubt)?
What are the memory settings in Prime95? You need to set it to at least 300MB.

 2010-10-07, 05:53 #10 ckdo     Dec 2007 Cleves, Germany 21216 Posts You can't reserve "behind the lines" P-1 assignments via the manual assignment form. You will have to assemble a worktodo file manually. I've been doing that for exponents below 40M for months now, and unless you have some profound experience in determining what exactly to put in your worktodo file, and how to react to the various errors that may occur, I'd recommend you don't even try it and stick to whatever work PrimeNet gives you.
 2010-10-07, 09:25 #11 Commaster   Jun 2010 Kiev, Ukraine 3×19 Posts Uncwilly, 2GB ^_^ ckdo, I managed to request some P-1 in 50M-51M range... 40M-49M range just seem to be way more delicious :) (They aren't behind the lines, PrimeNet shows them as "available")

 Similar Threads Thread Thread Starter Forum Replies Last Post 3mg4 PrimeNet 9 2016-12-13 18:24 ig1 GPU Computing 10 2012-11-04 14:44 tichy PrimeNet 4 2010-12-17 09:57 xode PrimeNet 14 2009-05-25 20:48 JuanTutors Software 2 2008-12-02 01:57

All times are UTC. The time now is 21:09.

Wed Apr 14 21:09:21 UTC 2021 up 6 days, 15:50, 0 users, load averages: 2.87, 2.56, 2.48