mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Aliquot Sequences (https://www.mersenneforum.org/forumdisplay.php?f=90)
-   -   Subproject #3: 700k-800k sequences to 100 digits (https://www.mersenneforum.org/showthread.php?t=12537)

frmky 2010-01-13 04:35

766688, 766776, 766848 is done.
I'll reserve 765380, 765384, 765456, 765504, 765600.

unconnected 2010-01-13 06:53

765018: i2300->i2461, size 117, 2^2*7
769224: i1474->i1695, size 100, 2*3
768360: i1577->i1727, size 107, 2^8 <- i'll keep it for more

schickel 2010-01-13 07:21

[QUOTE=Joshua2;201681]whats it mean when u get a lot of WARNING: couldn't find factor in factor.log?[/QUOTE]I'm going to assume here that it happened while the composite was small? If so, there are some numbers in the 20-25 digit range that msieve has problems with. Do you also have yafu installed? If so, aliqueit will switch over to yafu, which can usually factor the composite when msieve can't.

And welcome back!

10metreh 2010-01-13 07:24

[QUOTE=Joshua2;201681]whats it mean when u get a lot of WARNING: couldn't find factor in factor.log?[/QUOTE]

That means that yafu has failed to factor a number. This generally happens with small composites like c25 which were split off by P-1. When this happens, aliqueit calls msieve instead. If msieve fails too, you will get WARNING: couldn't find factor in msieve.log and aliqueit will run into an endless loop. In this case, find the composite in aliqueit.log and split it separately with ECM. Then restart aliqueit with the extra switch -f <one_of_the_factors>.

Also, you don't need to post your .elf files. Just submit them to [url]factordb.com[/url] by running aliqueit <the_sequence> -s 0.

frmky 2010-01-13 08:55

766880, 765380, 765456 are done.

I'm going to try a truly automated run overnight, so I'll take all of
765624, 765732, 765864,
767296, 767520, 767550, 767640, 767802,
768042

Greebley 2010-01-13 16:49

I made a download that includes the elf or alq files for all open aliquot sequences. The elf files had a program run over it that removes the 'shorthand' values (ones with +,*,^,#,!, etc on the left side of the equation). Also the final partial line is gone as this causes issues in windows.

You can find it posted below under 'All the open aliquot sequences'. It is current as of Jan 11, 2009. It also includes at least part of the two 'broken' sequences in the database - 417336 and 749320 from a previous download.

Note that it is post #14 that has the up to date ones.

10metreh 2010-01-13 18:00

[QUOTE=Greebley;201771]Also the final partial line is gone as this causes issues in windows.[/QUOTE]

I thought that bug had been fixed - I only get a "WARNING: @index ***: partial last line. Resuming..." message and then things proceed as normal.

unconnected 2010-01-13 18:04

taking 780066, 780330, 780576, 780930

10metreh 2010-01-13 18:10

[QUOTE=unconnected;201782]taking 780066, 780330, 780576, 780930[/QUOTE]

BTW, how is 770k-780k getting on?

Greebley 2010-01-13 18:53

[quote=10metreh;201781]I thought that bug had been fixed - I only get a "WARNING: @index ***: partial last line. Resuming..." message and then things proceed as normal.[/quote]

For windows, it doesn't correctly put in a new line so you get two indices on the same line. So it does continue, but you have to go back and add the newline in or the db is missing one index. I am pretty sure this wasn't ever fixed.

It is my understanding that everything works for unix.

frmky 2010-01-13 19:02

765384, 765504, 765600, 765624, 765732, 767296 are done.


All times are UTC. The time now is 22:39.

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