mersenneforum.org  

Go Back   mersenneforum.org > Factoring Projects > NFS@Home

Reply
 
Thread Tools
Old 2017-10-24, 17:19   #2212
swellman
 
swellman's Avatar
 
Jun 2012

2·1,429 Posts
Default

Reserving C224_122_119 for postprocessing.
swellman is online now   Reply With Quote
Old 2017-10-25, 15:46   #2213
richs
 
richs's Avatar
 
"Rich"
Aug 2002
Benicia, California

3×379 Posts
Default

Reserving C192_1577431459_23 please.
richs is offline   Reply With Quote
Old 2017-10-25, 19:17   #2214
frmky
 
frmky's Avatar
 
Jul 2003
So Cal

2,039 Posts
Default

C226_141_59 is done.

Code:
p84 factor: 362717407573062759649737199143733658562292627766699815889194128031404204438544404439
p142 factor: 5584640606072561151802930019243122240833798331945858851893157042850138505313340336371766706494184014505008264594637551042137666886143607864541
frmky is offline   Reply With Quote
Old 2017-10-26, 13:21   #2215
swellman
 
swellman's Avatar
 
Jun 2012

1011001010102 Posts
Default

Reserving 141^62+64^141 on 15e.
swellman is online now   Reply With Quote
Old 2017-10-27, 16:03   #2216
VBCurtis
 
VBCurtis's Avatar
 
"Curtis"
Feb 2005
Riverside, CA

52×173 Posts
Default c226_143_57 results

Log at https://pastebin.com/Dw1swigP
Code:
prp96 factor: 897094747754482045770531544477700088352397640594837413588145052113208015535658806042142533635811
prp130 factor: 5402116304783573647710016304889406303241500724198159102155379461705248829675304254503761226405413342784878208865325882006231555611
400 hours on 6 threads of a busy i7-5820 to solve a 23.5M matrix that filtering marks as density 84, and LA notes sparse part has density 71. In hindsight, I should have reduced the relations available to filtering until I got a matrix the matched the target-density setting. "top" listed msieve as using 9GB for this matrix.
This was my first 14e/33 job, and 590M unique relations (of roughly 720M raw) were far too many. Future 14/33 jobs should target 650-680M raw relations.
VBCurtis is online now   Reply With Quote
Old 2017-10-27, 16:46   #2217
RichD
 
RichD's Avatar
 
Sep 2008
Kansas

22×3×7×37 Posts
Default

Taking C169_2158316902961_17 next.
RichD is offline   Reply With Quote
Old 2017-10-27, 17:04   #2218
fivemack
(loop (#_fork))
 
fivemack's Avatar
 
Feb 2006
Cambridge, England

6,361 Posts
Default

Quote:
Originally Posted by VBCurtis View Post
In hindsight, I should have reduced the relations available to filtering until I got a matrix the matched the target-density setting.
I'm not quite sure why you suggest that, rather than increasing the target density (you were running with the default target density 70)
fivemack is offline   Reply With Quote
Old 2017-10-31, 04:26   #2219
VBCurtis
 
VBCurtis's Avatar
 
"Curtis"
Feb 2005
Riverside, CA

432510 Posts
Default

Quote:
Originally Posted by fivemack View Post
I'm not quite sure why you suggest that, rather than increasing the target density (you were running with the default target density 70)
Two possible explanations:
1. After experimenting with a few filtering runs with varying relation counts (see http://mersenneforum.org/showpost.ph...postcount=1186), I accidentally forgot to set TD on the full set, saw the matrix was the same size I got from the first filtering run with TD set to 140, and let it run without realizing I hadn't set TD
2. I edit/cut the log to delete the filtering runs irrelevant to the actual job, and accidentally deleted the target-density line.

Based on the post linked above, I'm quite sure I had target_density set during my trials.

However, on my current LA job I tried what I suggested two posts up, and the ETA (after 1% of the job, about 4 hrs) for the smallest matrix I could build with density matching my target density (I used 134) was about 5% longer than the ETA of the matrix I got from the full relation set, even though that matrix is density 104 when much higher was requested.

More specifically, 360M unique 32LP relations built a 19.0M matrix with density 104, while 339M unique relations produced a 17.0M matrix with density 134. The former matrix has a shorter ETA, after letting both run about 4 hrs so the ETA estimate settled down.

Interesting tidbit: 340M unique rels and TD=134 built a 17.9M matrix at density 132, while 338M unique rels and TD = 132 built a 17.1M matrix.
Tidbit #2: 333M unique rels and TD = 140 failed, as did 337M rels and TD = 136.
On this dataset, TD over 134 was nearly impossible to build, and matrices were MUCH more sensitive to input # of rels than I expected. I'll be using TD = 132 as my default for a while, and also running similar experiments on future 14/32 jobs.

Last fiddled with by VBCurtis on 2017-10-31 at 04:28
VBCurtis is online now   Reply With Quote
Old 2017-10-31, 04:29   #2220
VBCurtis
 
VBCurtis's Avatar
 
"Curtis"
Feb 2005
Riverside, CA

52·173 Posts
Default

Reserving 3408:1671 from 15e.
VBCurtis is online now   Reply With Quote
Old 2017-10-31, 15:42   #2221
RichD
 
RichD's Avatar
 
Sep 2008
Kansas

310810 Posts
Default C169_2158316902961_17 factored

84 hours to solve a 11.3M matrix using -t 4, TD=120. (TD=124 failed)

Code:
p58 factor: 5021316583534880940885444734079370333361932652310779490947
p112 factor: 1214281023169276509053011690942441596344967221458926915648042712870516387083258213891473300070212158822642727987
https://pastebin.com/68hDDQnp
RichD is offline   Reply With Quote
Old 2017-10-31, 17:02   #2222
swellman
 
swellman's Avatar
 
Jun 2012

2×1,429 Posts
Default

Quote:
Originally Posted by swellman View Post
Reserving 141^62+64^141 on 15e.
Someone pointed out to me the typo here - this should be listed as
141^62+62^141
. It’s listed as the above on the 15e status page. Don’t think it matters, just clarifying which number is being factored.
swellman is online now   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
restarting nfs linear algebra cubaq YAFU 2 2017-04-02 11:35
Linear algebra at 600% CRGreathouse Msieve 8 2009-08-05 07:25
Linear algebra crashes 10metreh Msieve 3 2009-02-02 08:34
Linear algebra proof Damian Math 8 2007-02-12 22:25
Linear algebra in MPQS R1zZ1 Factoring 2 2007-02-02 06:45

All times are UTC. The time now is 19:58.

Wed Sep 30 19:58:29 UTC 2020 up 20 days, 17:09, 0 users, load averages: 2.01, 1.98, 1.88

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.