mersenneforum.org  

Go Back   mersenneforum.org > Factoring Projects > Aliquot Sequences

Reply
 
Thread Tools
Old 2011-10-06, 21:10   #45
science_man_88
 
science_man_88's Avatar
 
"Forget I exist"
Jul 2009
Dumbassville

8,369 Posts
Default

Quote:
Originally Posted by EdH View Post
The warnings are just because aliqueit.log has two sets of the minor factors now and the second time through, aliqueit is trying to use the second set on the cofactor that was left after using the first set. I hope that made sense.
in other words it treats the file it made the first time almost like the original elf file and tries going off that but it's a reading of the original and the new so it's a mix of both.
science_man_88 is offline   Reply With Quote
Old 2011-10-06, 22:35   #46
schickel
 
schickel's Avatar
 
"Frank <^>"
Dec 2004
CDP Janesville

2·1,049 Posts
Default

Quote:
Originally Posted by science_man_88 View Post
in other words it treats the file it made the first time almost like the original elf file and tries going off that but it's a reading of the original and the new so it's a mix of both.
Right. When Aliqueit starts up it reads the log file to see if there are any factors in there from a previous run; if there are, it tries to divide them out, but since it has already done some trial division, the small factors are already divided out, so they don't divide the remaining composite.

Aftter a re-start, the factor does not divide warnings can be safely ignored....
schickel is offline   Reply With Quote
Old 2011-10-06, 22:38   #47
science_man_88
 
science_man_88's Avatar
 
"Forget I exist"
Jul 2009
Dumbassville

20B116 Posts
Default

Quote:
Originally Posted by schickel View Post
Right. When Aliqueit starts up it reads the log file to see if there are any factors in there from a previous run; if there are, it tries to divide them out, but since it has already done some trial division, the small factors are already divided out, so they don't divide the remaining composite.

Aftter a re-start, the factor does not divide warnings can be safely ignored....
sorry if I'm a major annoyance.btw way it's got to doing 864 curves at B1=1e6

Last fiddled with by science_man_88 on 2011-10-06 at 22:39
science_man_88 is offline   Reply With Quote
Old 2011-10-06, 23:46   #48
schickel
 
schickel's Avatar
 
"Frank <^>"
Dec 2004
CDP Janesville

2·1,049 Posts
Default

Quote:
Originally Posted by science_man_88 View Post
sorry if I'm a major annoyance.
Always happy to help. It's always nice when everything suddenly clicks into place and things start running like they're supposed to.
Quote:
btw way it's got to doing 864 curves at B1=1e6
Aliqueit tries to balance the run time for ECM against the expected run time for GNFS. How long does it look to be taking per curve? (You'll have to look at the aliqueit_ecm_temp.log file in the work folder.) That will tell you how long it will take to do 800+ curves at 1M.

If you want to jump right to GNFS, start Aliqueit with the -e switch. That tells it to skip the ECM portion and jump right to the next stage. Remember, though, that if you do that without "enough" ECM work, you have a chance of being disappointed when an 8(16) hour GNFS job nets you a "small" factor....

I also posted a reservation for 563064 in the reservation thread so bchaffin's workers will not factor the c108. Let us know how you're getting on....good luck!

Last fiddled with by schickel on 2011-10-06 at 23:47 Reason: Darn tyops; dern grammar
schickel is offline   Reply With Quote
Old 2011-10-07, 00:05   #49
science_man_88
 
science_man_88's Avatar
 
"Forget I exist"
Jul 2009
Dumbassville

8,369 Posts
Default

Quote:
Originally Posted by schickel View Post
Always happy to help. It's always nice when everything suddenly clicks into place and things start running like they're supposed to.Aliqueit tries to balance the run time for ECM against the expected run time for GNFS. How long does it look to be taking per curve? (You'll have to look at the aliqueit_ecm_temp.log file in the work folder.) That will tell you how long it will take to do 800+ curves at 1M.

If you want to jump right to GNFS, start Aliqueit with the -e switch. That tells it to skip the ECM portion and jump right to the next stage. Remember, though, that if you do that without "enough" ECM work, you have a chance of being disappointed when an 8(16) hour GNFS job nets you a "small" factor....

I also posted a reservation for 563064 in the reservation thread so bchaffin's workers will not factor the c108. Let us know how you're getting on....good luck!
my mom wants me to turn off my computer at night but I haven't in a long while ( in fact I think it's been at least since we moved it last) looks to be about 29 s/curve from what few i read.so according to that 6.96 hours is what i get for 864 curves if that's correct it should finish that by about midnight-1am tonight (my time) unless it's slowed down majorly. if I remember correct that should mean about 4:00am-5:00 pm in the UK, if that holds true I'm likely not going to be usually looking at my PC at that time ( usually turn the screen and mouse off by 10pm or later when I try to goto sleep).

Last fiddled with by science_man_88 on 2011-10-07 at 00:14
science_man_88 is offline   Reply With Quote
Old 2011-10-07, 11:23   #50
science_man_88
 
science_man_88's Avatar
 
"Forget I exist"
Jul 2009
Dumbassville

8,369 Posts
Default

here's the news I woke up looked at it and it was saying msieve149 was not valid as a command ( I stupidly hit close and see I've basically accomplished nothing).
science_man_88 is offline   Reply With Quote
Old 2011-10-07, 11:48   #51
schickel
 
schickel's Avatar
 
"Frank <^>"
Dec 2004
CDP Janesville

2·1,049 Posts
Default

Quote:
Originally Posted by science_man_88 View Post
here's the news I woke up looked at it and it was saying msieve149 was not valid as a command ( I stupidly hit close and see I've basically accomplished nothing).
Don't panic!

Yes, you actually did acomplish something: you know that Aliqueit wasn't able to crack the composite with ECM, so it was going to move on to the bigger guns. Try starting it this way:
Code:
aliqueit 563064 -e
That tells aliqueit to skip the ECM attempt and proceed straight away to the next step. Let us know what happens when you try that....
schickel is offline   Reply With Quote
Old 2011-10-07, 11:56   #52
science_man_88
 
science_man_88's Avatar
 
"Forget I exist"
Jul 2009
Dumbassville

8,369 Posts
Default

more warnings including the one about msieve149:
Attached Thumbnails
Click image for larger version

Name:	newwarnings.jpg
Views:	86
Size:	46.0 KB
ID:	7111  
science_man_88 is offline   Reply With Quote
Old 2011-10-07, 12:10   #53
schickel
 
schickel's Avatar
 
"Frank <^>"
Dec 2004
CDP Janesville

1000001100102 Posts
Default

Quote:
Originally Posted by science_man_88 View Post
more warnings including the one about msieve149:
OK, don't worry about the factor doesn't divide warnings, we already covered those. The msieve149 not recognized means that it can't find the msieve executable.

The fact that a directory was created for the job is good, that means we're making progress. Since I use the Perl setup to run NFS jobs, I can't be certain of this, but try changing this line in the aliquiet.ini file:
Code:
//Name (and path, if needed) of your msieve executable.
msieve_cmd = \aliquot\auto\msieve149
If it's doing what I think it's doing, it's basically trying to do the equivalent of this:
Code:
md <gnfs_really_long_number>
cd <gnfs_really_long_number>
msieve149
Which doesn't work becuase msieve is in the parent directory. The .ini change should remedy this.
schickel is offline   Reply With Quote
Old 2011-10-07, 12:15   #54
science_man_88
 
science_man_88's Avatar
 
"Forget I exist"
Jul 2009
Dumbassville

8,369 Posts
Default

Quote:
Originally Posted by schickel View Post
OK, don't worry about the factor doesn't divide warnings, we already covered those. The msieve149 not recognized means that it can't find the msieve executable.

The fact that a directory was created for the job is good, that means we're making progress. Since I use the Perl setup to run NFS jobs, I can't be certain of this, but try changing this line in the aliquiet.ini file:
Code:
//Name (and path, if needed) of your msieve executable.
msieve_cmd = \aliquot\auto\msieve149
If it's doing what I think it's doing, it's basically trying to do the equivalent of this:
Code:
md <gnfs_really_long_number>
cd <gnfs_really_long_number>
msieve149
Which doesn't work becuase msieve is in the parent directory. The .ini change should remedy this.
I can tell you've found every error in this book lol.
science_man_88 is offline   Reply With Quote
Old 2011-10-07, 12:23   #55
schickel
 
schickel's Avatar
 
"Frank <^>"
Dec 2004
CDP Janesville

2×1,049 Posts
Default

Quote:
Originally Posted by science_man_88 View Post
I can tell you've made every error in this book lol.
FTFY.....it's always easier to troubleshoot errors that I've made myself.
schickel is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Aliqueit.exe discussion bsquared Aliquot Sequences 591 2020-09-09 16:51
Factoring projects algorithms - What is Aliqueit ? Romuald Factoring 75 2016-08-27 09:46
Resuming aliqueit johnadam74 Aliquot Sequences 4 2016-03-28 12:32
Apparent aliqueit issue with specifying factors pakaran Aliquot Sequences 2 2015-09-12 23:10
Tried out aliqueit.exe: ggnfs failing Greebley Aliquot Sequences 35 2010-02-13 15:23

All times are UTC. The time now is 08:34.

Fri Oct 30 08:34:09 UTC 2020 up 50 days, 5:45, 1 user, load averages: 1.69, 1.68, 1.71

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.