BOINC NFS sieving - NFS@Home - Page 108

RichD · 2014-03-01, 04:27

GC_11_227 splits as:

Code:

prp71 factor: 24013573097415000462985655593434665537916678988110599464127913117714211
prp135 factor: 439286544732716591887311221013349373538132387268354397485024871768431029824624530196541345827372316603652416192358915641775905314464951

45 hours (in BL) built a 8.25M matrix using "-nc target_density=114 -t 4" on i5/2500

pinhodecarlos · 2014-03-01, 14:57

I would like to try GC_3_496 but I need some info regarding the latest msieve. Been out of post-processing almost a year and things changed since then. Don't have unlimited bandwidth anymore so I am limited to post-process 2-3 numbers/month, msieve version changed, etc.
What version should I use for 64-bit windows machine? And, do I need pthreadGC2.dll and pthreadVC2.dll files on the msieve folder?

Carlos

swellman · 2014-03-01, 15:59

I've been using the executable and associated file found here.

pinhodecarlos · 2014-03-01, 18:11

I have to give up GC_3_496 due to heat problems. Anyway, LA has an ETA of 76 hours for 9.8M^2 matrix. Sorry.

swellman · 2014-03-01, 21:22

Quote:

Originally Posted by swellman

I'll take W_2_781 if it's still available.

Code:

prp60 factor: 854643608564173566055954493461963414327770125674674003206399
prp111 factor: 829470699876789341315231711633049803730963085776901862130515192148745174538740550793880454144054148383946858223

RichD · 2014-03-01, 23:34

I'll take GC_3_496 next.

fivemack · 2014-03-03, 10:23

GC_6_303 ETA is now Tuesday afternoon; a runaway Firefox meant it made little progress over the weekend.

xilman · 2014-03-03, 13:35

Quote:

Originally Posted by fivemack

GC_6_303 ETA is now Tuesday afternoon; a runaway Firefox meant it made little progress over the weekend.

I ditched FF last year for that very reason. Now running Midori which does pretty much everything I ask of it and is much lighter weight than FF and Chrome.

Paul

fivemack · 2014-03-04, 13:37

Code:

Tue Mar  4 11:48:46 2014  prp78 factor: 724901011535214037845874972824565483006757602317236033423370083435767781234797
Tue Mar  4 11:48:46 2014  prp111 factor: 891413068611899193736220698858807819904586081225032674362266394055114505379699337711097101194312538214688008561

About 90 hours for 13.2M matrix (density 112) on six threads of i7/4930

I suppose the very best automation protocol would not have required me to copy-type the 78-digit factor from my phone, but I deliberately don't have efficient access to my compute cluster at home from my desk at work.

Reserving F1881, will set up this evening and edit to insert ETA then

fivemack · 2014-03-04, 13:41

Code:

Tue Mar  4 12:22:44 2014  prp72 factor: 215290914147000790812689007424442844957319292819239150613345580364711241
Tue Mar  4 12:22:44 2014  prp99 factor: 151825369037900078063922325167583391673084915398865559605969446693935136477201824284072243950242799

123 hours for 9.7M matrix density=96 on sort of three-to-four-ish cores of i7/2600 - would probably have run in 84 hours on three cores.

ETA for 10C236 is Tuesday morning; 12.8M matrix density=96, 151 hours on three cores anticipated.

RichD · 2014-03-05, 01:37

GC_3_496 splits as:

Code:

prp68 factor: 34558630114648526421540494777207045843755870156121410528411719897833
prp135 factor: 890198443460579047649250096723450598716680985003533287186615497352057375239425980651670627348360194020701310297797549559585881599490339

74.5 hours with 9.9M matrix (-t 4 -nc target_density=122) on Core-i5

2014-03-01, 14:57	#1179
pinhodecarlos "Carlos Pinho" Oct 2011 Milton Keynes, UK 3×17×97 Posts	I would like to try GC_3_496 but I need some info regarding the latest msieve. Been out of post-processing almost a year and things changed since then. Don't have unlimited bandwidth anymore so I am limited to post-process 2-3 numbers/month, msieve version changed, etc. What version should I use for 64-bit windows machine? And, do I need pthreadGC2.dll and pthreadVC2.dll files on the msieve folder? Carlos Last fiddled with by pinhodecarlos on 2014-03-01 at 14:58

2014-03-01, 18:11	#1181
pinhodecarlos "Carlos Pinho" Oct 2011 Milton Keynes, UK 4947₁₀ Posts	I have to give up GC_3_496 due to heat problems. Anyway, LA has an ETA of 76 hours for 9.8M^2 matrix. Sorry. Last fiddled with by pinhodecarlos on 2014-03-01 at 18:11

2014-03-04, 13:37	#1186
fivemack (loop (#_fork)) Feb 2006 Cambridge, England 2³·11·73 Posts	F1463 done Code: Tue Mar 4 11:48:46 2014 prp78 factor: 724901011535214037845874972824565483006757602317236033423370083435767781234797 Tue Mar 4 11:48:46 2014 prp111 factor: 891413068611899193736220698858807819904586081225032674362266394055114505379699337711097101194312538214688008561 About 90 hours for 13.2M matrix (density 112) on six threads of i7/4930 I suppose the very best automation protocol would not have required me to copy-type the 78-digit factor from my phone, but I deliberately don't have efficient access to my compute cluster at home from my desk at work. Reserving F1881, will set up this evening and edit to insert ETA then Last fiddled with by fivemack on 2014-03-04 at 13:45

2014-03-04, 13:41	#1187
fivemack (loop (#_fork)) Feb 2006 Cambridge, England 6424₁₀ Posts	GC_6_303 done Code: Tue Mar 4 12:22:44 2014 prp72 factor: 215290914147000790812689007424442844957319292819239150613345580364711241 Tue Mar 4 12:22:44 2014 prp99 factor: 151825369037900078063922325167583391673084915398865559605969446693935136477201824284072243950242799 123 hours for 9.7M matrix density=96 on sort of three-to-four-ish cores of i7/2600 - would probably have run in 84 hours on three cores. ETA for 10C236 is Tuesday morning; 12.8M matrix density=96, 151 hours on three cores anticipated.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Boinc Statistics for NFS@Home borked ?	thomasn	NFS@Home	1	2013-10-02 15:31
BOINC NFS sieving - RSALS	debrouxl	NFS@Home	621	2012-12-14 23:44
BOINC?	masser	Sierpinski/Riesel Base 5	1	2009-02-09 01:10
BOINC?	KEP	Twin Prime Search	212	2007-04-25 10:29
BOINC	bebarce	Software	3	2005-12-15 18:35

2014-03-01, 15:59	#1180
swellman Jun 2012 6024₈ Posts	I've been using the executable and associated file found here.

2014-03-01, 23:34	#1183
RichD Sep 2008 Kansas 6500₈ Posts	I'll take GC_3_496 next.

2014-03-03, 10:23	#1184
fivemack (loop (#_fork)) Feb 2006 Cambridge, England 1918₁₆ Posts	GC_6_303 ETA is now Tuesday afternoon; a runaway Firefox meant it made little progress over the weekend.