mersenneforum.org Pascal's OPN roadblock files
 User Name Remember Me? Password
 Register FAQ Search Today's Posts Mark Forums Read

2019-06-02, 03:07   #727
RichD

Sep 2008
Kansas

2×1,663 Posts

Quote:
 Originally Posted by ThomRuley (52379047267^23-1)/(52379047267-1)
This number has been fully ECMed with 20K curves at 260e6. I have no plans doing anything more with it at this time.

2019-06-02, 12:34   #728
ThomRuley

May 2003

111110002 Posts

Quote:
 Originally Posted by RichD This number has been fully ECMed with 20K curves at 260e6. I have no plans doing anything more with it at this time.
Have you done anything with
(7489^67-1)/(7489-1)?

All I can do at the moment is ECM, so if you haven't done anything with the above, I'll take it instead.

2019-06-02, 12:51   #729
RichD

Sep 2008
Kansas

2·1,663 Posts

Quote:
 Originally Posted by ThomRuley Have you done anything with (7489^67-1)/(7489-1)? All I can do at the moment is ECM, so if you haven't done anything with the above, I'll take it instead.
No work by me...

 2019-09-17, 09:20 #730 lavalamp     Oct 2007 Manchester, UK 17·79 Posts I'll take 41442572371^19-1 from t800. It seems to be going a little slower than I would have expected, but perhaps I'm out of practice. Unless I'm missing something obvious, this is a simple deg 6 poly right? p*(p^3)^6-1 and skew 1? ETA is in the 8 - 10 day range.
 2019-09-17, 15:41 #731 chris2be8     Sep 2009 37728 Posts I'd recommend a skew of about 0.017 since c6 is much larger than c0, That should increase yield a bit (but test it). You should be OK to switch skew part way through a run. Chris
2019-09-17, 16:08   #732
axn

Jun 2003

2×3×827 Posts

Quote:
 Originally Posted by lavalamp I'll take 41442572371^19-1 from t800. It seems to be going a little slower than I would have expected, but perhaps I'm out of practice. Unless I'm missing something obvious, this is a simple deg 6 poly right? p*(p^3)^6-1 and skew 1? ETA is in the 8 - 10 day range.
Have you tried the degree 5 & 4 ? (p^4)^5-p and (p^5)^4-p ?

Also, are you sieving on the algebraic side of the degree-6?

2019-09-17, 17:12   #733
Brownfox

Dec 2017

32×7 Posts

Quote:
 Originally Posted by axn Have you tried the degree 5 & 4 ? (p^4)^5-p and (p^5)^4-p ? Also, are you sieving on the algebraic side of the degree-6?

Degree 4 will be much slower at this size of exponent, in my experience. Degree 5 may be worth a look but I think the original polynomial from lavalamp will probably be best.

2019-09-17, 19:52   #734
lavalamp

Oct 2007
Manchester, UK

24778 Posts

Quote:
 Originally Posted by chris2be8 I'd recommend a skew of about 0.017 since c6 is much larger than c0, That should increase yield a bit (but test it).
Oh yes, that nearly doubled the speed. I had previously thought that skew should be a minimum of 1, but it seems not.

I tried the degree 5, it's somewhat slower and needs more relations.

I'm sieving the algebraic side as it is a little faster.

 2019-09-21, 19:20 #735 lavalamp     Oct 2007 Manchester, UK 134310 Posts I've had an issue with SNFS and wondering if anyone can shed some light on it. It's happened on two different jobs so far, and the symptoms are exactly the same. This is the log file: Code: Thu Sep 19 20:26:04 2019 Found 11742819 relations, 48.3% of the estimated minimum (24306044). Thu Sep 19 20:26:04 2019 LatSieveTime: 1600.9 Thu Sep 19 20:26:04 2019 -> making sieve job for q = 4000000 in 4000000 .. 4025000 as file test.job.T0 Thu Sep 19 20:26:04 2019 -> making sieve job for q = 4025000 in 4025000 .. 4050000 as file test.job.T1 Thu Sep 19 20:26:04 2019 -> making sieve job for q = 4050000 in 4050000 .. 4075000 as file test.job.T2 Thu Sep 19 20:26:04 2019 -> making sieve job for q = 4075000 in 4075000 .. 4100000 as file test.job.T3 Thu Sep 19 20:26:04 2019 -> Lattice sieving algebraic q from 4000000 to 4100000. Thu Sep 19 20:26:04 2019 -> gnfs-lasieve4I14e -k -o spairs.out.T0 -v -n 0 -a test.job.T0 Thu Sep 19 20:26:04 2019 -> gnfs-lasieve4I14e -k -o spairs.out.T1 -v -n 1 -a test.job.T1 Thu Sep 19 20:26:04 2019 -> gnfs-lasieve4I14e -k -o spairs.out.T2 -v -n 2 -a test.job.T2 Thu Sep 19 20:26:04 2019 -> gnfs-lasieve4I14e -k -o spairs.out.T3 -v -n 3 -a test.job.T3 Thu Sep 19 20:26:39 2019 -> Return value 3221225477. Updating job file and terminating... Thu Sep 19 20:26:39 2019 LatSieveTime: 35.0657 Code: Sat Sep 21 09:28:17 2019 Found 10760363 relations, 42.8% of the estimated minimum (25118864). Sat Sep 21 09:28:17 2019 LatSieveTime: 1605.97 Sat Sep 21 09:28:17 2019 -> making sieve job for q = 4000000 in 4000000 .. 4025000 as file test.job.T0 Sat Sep 21 09:28:17 2019 -> making sieve job for q = 4025000 in 4025000 .. 4050000 as file test.job.T1 Sat Sep 21 09:28:17 2019 -> making sieve job for q = 4050000 in 4050000 .. 4075000 as file test.job.T2 Sat Sep 21 09:28:17 2019 -> making sieve job for q = 4075000 in 4075000 .. 4100000 as file test.job.T3 Sat Sep 21 09:28:17 2019 -> Lattice sieving algebraic q from 4000000 to 4100000. Sat Sep 21 09:28:17 2019 -> gnfs-lasieve4I14e -k -o spairs.out.T0 -v -n 0 -a test.job.T0 Sat Sep 21 09:28:17 2019 -> gnfs-lasieve4I14e -k -o spairs.out.T1 -v -n 1 -a test.job.T1 Sat Sep 21 09:28:17 2019 -> gnfs-lasieve4I14e -k -o spairs.out.T2 -v -n 2 -a test.job.T2 Sat Sep 21 09:28:17 2019 -> gnfs-lasieve4I14e -k -o spairs.out.T3 -v -n 3 -a test.job.T3 Sat Sep 21 09:28:20 2019 -> Return value 3221225477. Updating job file and terminating... Sat Sep 21 09:28:20 2019 LatSieveTime: 3.03377 At this point the script terminates but I don't notice for several hours and later get annoyed that my machine has been idle all this time. If I manually edit q to skip over the 4M to 4.1M range, the script progresses fine as before. Both jobs failed in exactly the same 4M to 4.1M range, both are 6th degree polynomials for p^7-1.
 2019-09-21, 22:24 #736 RichD     Sep 2008 Kansas 2·1,663 Posts The only tidbit I can offer, if it is p^7-1, then it should be sieved on the rational side.
 2019-09-22, 15:51 #737 chris2be8     Sep 2009 111111110102 Posts What screen output did you get? That should tell us if the sievers said anything interesting. What did test.job.T* contain? How much was written to spairs.out.T* (if one is empty but the rest contain relations that would tell us which thread had the problem)? What happens if you try to restart sieving the 4M to 4.1M range? That would say if it's a transient error or not. And you would be ready to catch screen output if it fails. What script are you using? And what OS version, software version, type of CPU etc? Chris

 Thread Tools

 Similar Threads Thread Thread Starter Forum Replies Last Post Xyzzy GPU Computing 1 2017-05-17 20:22 Mark Rose GPU Computing 52 2016-07-02 12:11 firejuggler GPU Computing 12 2016-02-23 06:55 Elhueno Homework Help 5 2008-06-12 16:37 jchein1 Factoring 30 2005-05-30 14:43

All times are UTC. The time now is 07:23.

Thu May 13 07:23:47 UTC 2021 up 35 days, 2:04, 1 user, load averages: 1.75, 1.74, 1.76

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