Factoring database - Page 65

fivemack · 2010-02-13, 00:49

Given the various 'complete this range to X digits' projects, it would be very nice to see the size of the last known term in an aliquot sequence (rather than the size of its unfactored cofactor) in the overview page. Is that something that could be implemented? Obviously I can kludge it by scraping the overview page and following links, but that's a lot more load on the site.

smh · 2010-02-13, 08:32

Quote:

Originally Posted by Mini-Geek

Looks like it's fine to me.
http://factordb.com/search.php?se=1&aq=947076
Maybe it was a problem fixable with the "Repair sequence" button, and someone did so without telling you?

I did try the repair option myself 2 or 3 times to no avail.

biwema · 2010-02-13, 20:49

Quote:

Originally Posted by frmky

Rather than fix them all manually, it might be easier to implement a "Retest Primality" button. There is a legion of volunteers here willing to hunt down the bad entries and click the button.

I also like the idea of this Retest primality button.
Just trial division up to some thousand and a fermat etst to the base 2, 3 and one more random number. That is very quick and all composites can be revealed.

In the very rare case of a strong pseudoprime number it is still possible to contact the administrator.

By the way: There is a simple way to find most of the broken aliquot sequences:
Go to the factor database, choose "squence overview", show only sequences ending with a prime, start with 2 and click "Ends with" (it sorts the primes descending). Large primes are very rare, so all P20 an larger are suspicious.

Now:
Start Length End Ends with
794280 830 Prime P107
794196 438 Prime P106
789090 597 Prime P101
789740 608 Prime P100
790248 1278 Prime P91
792756 703 Prime P90
790884 1193 Prime P89
794880 560 Prime P85
792864 887 Prime P85
792820 835 Prime P85
791196 1853 Prime P83
789816 1030 Prime P82

10metreh · 2010-02-13, 21:18

That works if a whole line has been set as prime, but it doesn't if a remaining composite (not the whole line) has been set prime.

biwema · 2010-02-13, 23:25

true.

I noticed that normally the remaining composite was et prime and then the whole even next exponent. So we have to repair 2 indexes of every sequence. (Actually the last "prime" index would change anyway)

Until now i did not see a sequence where only the remaining composite was set prime. I assume that aliqeit would detect that while veifying when starting factoring.

rekcahx · 2010-02-27, 23:26

Hi. I uploaded some results to the db. Here is snapshot from the worker status page.

10metreh · 2010-02-28, 17:17

Sequence 552150, the longest single sequence, seems to have disappeared from the database. If you try to show the whole sequence, only line 8197 appears. Is there hacking going on?

Andi_HB · 2010-03-02, 23:30

Is the DB down? I can not Report Factors

Mini-Geek · 2010-03-03, 22:50

http://factordb.com/search.php?query=159
http://factordb.com/search.php?se=1&aq=824768

The DB has 159 listed as "C" (composite with no known factors) even though it already knows it's 3*53. This is holding up Aliquot sequence 824768 at line 918. I can't find a way to fix this. Syd, can you please get 159 set straight?

richs · 2010-03-04, 05:57

The same problem with the factorization of 159 exists with Home Prime base 5 of 675, index 41.

fivemack · 2010-03-04, 13:55

Is there a nicer way of doing

(((x + (1-(0-1)^x)/2))/2)!+(0-1)^x

to get x! \pm 1 ?

2010-02-13, 00:49	#705
fivemack (loop (#_fork)) Feb 2006 Cambridge, England 2³·11·73 Posts	Small request Given the various 'complete this range to X digits' projects, it would be very nice to see the size of the last known term in an aliquot sequence (rather than the size of its unfactored cofactor) in the overview page. Is that something that could be implemented? Obviously I can kludge it by scraping the overview page and following links, but that's a lot more load on the site.

2010-02-27, 23:26	#710
rekcahx Oct 2009 Oulu, Finland 2·3·5 Posts	Results uploading Hi. I uploaded some results to the db. Here is snapshot from the worker status page.

2010-02-28, 17:17	#711
10metreh Nov 2008 4422₈ Posts	Sequence 552150, the longest single sequence, seems to have disappeared from the database. If you try to show the whole sequence, only line 8197 appears. Is there hacking going on? Last fiddled with by 10metreh on 2010-02-28 at 17:17

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Database for k-b-b's:	3.14159	Miscellaneous Math	325	2016-04-09 17:45
Factoring database issues	Mini-Geek	Factoring	5	2009-07-01 11:51
database.zip	HiddenWarrior	Data	1	2004-03-29 03:53
Database layout	Prime95	PrimeNet	1	2003-01-18 00:49
Is there a performance database?	Joe O	Lounge	35	2002-09-06 20:19

2010-02-13, 21:18	#708
10metreh Nov 2008 2×3³×43 Posts	That works if a whole line has been set as prime, but it doesn't if a remaining composite (not the whole line) has been set prime.

2010-02-13, 23:25	#709
biwema Mar 2004 3×127 Posts	true. I noticed that normally the remaining composite was et prime and then the whole even next exponent. So we have to repair 2 indexes of every sequence. (Actually the last "prime" index would change anyway) Until now i did not see a sequence where only the remaining composite was set prime. I assume that aliqeit would detect that while veifying when starting factoring.

2010-03-02, 23:30	#712
Andi_HB Mar 2007 Germany 410₈ Posts	Is the DB down? I can not Report Factors

2010-03-03, 22:50	#713
Mini-Geek Account Deleted "Tim Sorbera" Aug 2006 San Antonio, TX USA 10AB₁₆ Posts	http://factordb.com/search.php?query=159 http://factordb.com/search.php?se=1&aq=824768 The DB has 159 listed as "C" (composite with no known factors) even though it already knows it's 3*53. This is holding up Aliquot sequence 824768 at line 918. I can't find a way to fix this. Syd, can you please get 159 set straight?

2010-03-04, 05:57	#714
richs "Rich" Aug 2002 Benicia, California 528₁₆ Posts	The same problem with the factorization of 159 exists with Home Prime base 5 of 675, index 41.

2010-03-04, 13:55	#715
fivemack (loop (#_fork)) Feb 2006 Cambridge, England 2³·11·73 Posts	Is there a nicer way of doing (((x + (1-(0-1)^x)/2))/2)!+(0-1)^x to get x! \pm 1 ?