Factoring database - Page 35

Syd · 2009-06-08, 20:05

Something that shouldn't happen, I thought I fixed this one. Well...

Quote:

Originally Posted by Mini-Geek

By the way, I just noticed that the graphs are now smoothed by a log scale instead of blocky to the digit. Nice update, Syd!

Thanks

Mini-Geek · 2009-06-09, 14:42

That just happened again on 233760:
http://factorization.ath.cx/search.p...n=last&fr=&to=
(I noticed factors from it were being processed and then saw that it stopped there, I didn't run it, mataje did. I repaired the line that was squared.)

Mini-Geek · 2009-06-10, 18:02

Might be considered too soon to bump, but according to his profile Syd has visited the forum since I posted this and it's still not fixed, so...

Quote:

Originally Posted by Mini-Geek

That just happened again on 233760:
http://factorization.ath.cx/search.p...n=last&fr=&to=
(I noticed factors from it were being processed and then saw that it stopped there, I didn't run it, mataje did. I repaired the line that was squared.)

Syd · 2009-06-10, 19:04

Quote:

Originally Posted by Mini-Geek

Might be considered too soon to bump, but according to his profile Syd has visited the forum since I posted this and it's still not fixed, so...

Please give me some time, I want to use this one to track down the bug.

Mini-Geek · 2009-06-10, 20:13

Quote:

Originally Posted by Syd

Please give me some time, I want to use this one to track down the bug.

Ok, no problem, just wanted to make sure you didn't miss it.

axn · 2009-06-11, 05:56

Syd, could you make it so that, the "Show whole sequence (text)" uses <PRE> </PRE> tag instead of the current option of listing lines with <br>.

This would vastly simplify the parsing of that page by software.

Syd · 2009-06-11, 09:02

Quote:

Originally Posted by axn

Syd, could you make it so that, the "Show whole sequence (text)" uses <PRE> </PRE> tag instead of the current option of listing lines with <br>.

This would vastly simplify the parsing of that page by software.

Now it does. Maybe adding "&raw=1" to the query is also useful for you

raw

@Mini-Geek

Fixed your sequence, but could not track the bug down yet

axn · 2009-06-11, 10:37

Quote:

Originally Posted by Syd

Now it does. Maybe adding "&raw=1" to the query is also useful for you

raw

Awesome!! That's what I am talkin about!

henryzz · 2009-06-11, 17:53

it seems to me that for a little while the server has been stopping at t30 ecm for 95-99 digit numbers
the minimum limit for t35 is listed on workerstatus as 95 digits

henryzz · 2009-06-12, 09:15

for some reason my workers have duplicated
i restarted the client and it hasn't gone away
the workers that haven't returned work for quite a while under my name can be removed

gd_barnes · 2009-06-13, 10:57

Syd,

The database is a remarkable piece of work! I just discovered it yesterday and am having a lot of fun with it.

There are a couple of factorizations that I am "stuck" on that there is no way to have your database automatically queue them up for factoring. They are:

2293*2^317-1
2293*2^325-1

I've done all of the trial factoring on them and then done the low, medium, high, and very high ECM's as well as the P-1 and P+1 for good measure. Both have been factored down to remaining composites that are 91 and 90 digits respectively.

What happens is that the only buttons that are now provided are: "ECM to low limits", "ECM to medium limits", "ECM to high limits", "P-1", and "P+1".

There is no "ECM to very high limits" or "sieve" button so that I can queue them up to find the final factors when workers are available. To manually keep doing the ECM to high limits would be very time consuming and quite ineffeicient for your workers. I tried that a few times and quickly found it to be futile.

This problem seems to occur only for approximately 87-95 digit factors but it doesn't always happen. For factors below that, there always seems to be a "sieve" button available. For factors above that, there seems to always be an "ECM to very high limits" button available; both which allow work to be queued up for processing.

These 2 factorizations are now the only incompleted ones for 2293*2^n-1 up to n=325 in the database so I'd really like to get them completed. :-) This sequence is interesting to me because 2293 is the smallest k-value where k*2^n-1 has no known prime as of yet up to n=4M!

If anyone else can check this and just tell me if I'm not understanding something, that would be great too. Otherwise, it seems like a "bug" or "feature" of the system at the current time.

Thank you,
Gary

2009-06-11, 05:56	#380
axn Jun 2003 3²×5×113 Posts	Feature Request Syd, could you make it so that, the "Show whole sequence (text)" uses <PRE> </PRE> tag instead of the current option of listing lines with <br>. This would vastly simplify the parsing of that page by software.

2009-06-13, 10:57	#385
gd_barnes May 2007 Kansas; USA 3²×13×89 Posts	Syd, The database is a remarkable piece of work! I just discovered it yesterday and am having a lot of fun with it. There are a couple of factorizations that I am "stuck" on that there is no way to have your database automatically queue them up for factoring. They are: 22932^317-1 22932^325-1 I've done all of the trial factoring on them and then done the low, medium, high, and very high ECM's as well as the P-1 and P+1 for good measure. Both have been factored down to remaining composites that are 91 and 90 digits respectively. What happens is that the only buttons that are now provided are: "ECM to low limits", "ECM to medium limits", "ECM to high limits", "P-1", and "P+1". There is no "ECM to very high limits" or "sieve" button so that I can queue them up to find the final factors when workers are available. To manually keep doing the ECM to high limits would be very time consuming and quite ineffeicient for your workers. I tried that a few times and quickly found it to be futile. This problem seems to occur only for approximately 87-95 digit factors but it doesn't always happen. For factors below that, there always seems to be a "sieve" button available. For factors above that, there seems to always be an "ECM to very high limits" button available; both which allow work to be queued up for processing. These 2 factorizations are now the only incompleted ones for 22932^n-1 up to n=325 in the database so I'd really like to get them completed. :-) This sequence is interesting to me because 2293 is the smallest k-value where k2^n-1 has no known prime as of yet up to n=4M! If anyone else can check this and just tell me if I'm not understanding something, that would be great too. Otherwise, it seems like a "bug" or "feature" of the system at the current time. Thank you, Gary Last fiddled with by gd_barnes on 2009-06-13 at 11:00

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

2009-06-09, 14:42	#376
Mini-Geek Account Deleted "Tim Sorbera" Aug 2006 San Antonio, TX USA 17·251 Posts	That just happened again on 233760: http://factorization.ath.cx/search.p...n=last&fr=&to= (I noticed factors from it were being processed and then saw that it stopped there, I didn't run it, mataje did. I repaired the line that was squared.)

2009-06-11, 17:53	#383
henryzz Just call me Henry "David" Sep 2007 Cambridge (GMT/BST) 1011011111110₂ Posts	it seems to me that for a little while the server has been stopping at t30 ecm for 95-99 digit numbers the minimum limit for t35 is listed on workerstatus as 95 digits

2009-06-12, 09:15	#384
henryzz Just call me Henry "David" Sep 2007 Cambridge (GMT/BST) 2×3³×109 Posts	for some reason my workers have duplicated i restarted the client and it hasn't gone away the workers that haven't returned work for quite a while under my name can be removed