mersenneforum.org Factoring for a publication
 Register FAQ Search Today's Posts Mark Forums Read

2021-06-10, 06:01   #188
bur

Aug 2020

12C16 Posts

Quote:
 Originally Posted by VBCurtis I'm curious to hear how many relations / hours this job needs at the smaller parameters you chose; this thread may not be the place for it, you can PM me details if you decide not to post about them. I don't run jobs that have initial yield below 3.5 (35k relations per 10kQ), but that doesn't make it right.
Yes, that's also part of the reason why I continued with it. ETA increased overnight to a total of 60 hours.

Sieving is fast per q though, approximately 180 s / 10,000 q (usually 600 s). Due to the low rels/q yield that was initially 25 rels / s, now it's down to about 10 rels / s.

Max, if this is too off-topic, let me know, then I'll take that elsewhere. But maybe it's interesting for other SNFS jobs of your numbers?

 2021-06-10, 09:17 #189 fivemack (loop (#_fork))     Feb 2006 Cambridge, England 72×131 Posts Line 56 (6,-7) complete by SNFS. Required rather more relations than I was expecting (29.7m raw worked, 29.3m raw didn't) and about 1100 thread-hours, so three or four days on a not-very-idle i9-7940X. 14e siever, average 1.6 to 1.9 million relations per million-Q Not quite sure what to work on next - I can probably do a 180-digit GNFS in two weeks, I have done quartic SNFS of difficulty 250 for the Fibonacci numbers but that took 150k thread-hours which is getting quite substantial, especially when 120 of my threads are still waiting for an electrician.
2021-06-10, 11:38   #190
Max0526

"Max"
Jun 2016
Toronto

89310 Posts

Quote:
 Originally Posted by bur Yes, that's also part of the reason why I continued with it. ETA increased overnight to a total of 60 hours. Sieving is fast per q though, approximately 180 s / 10,000 q (usually 600 s). Due to the low rels/q yield that was initially 25 rels / s, now it's down to about 10 rels / s. Max, if this is too off-topic, let me know, then I'll take that elsewhere. But maybe it's interesting for other SNFS jobs of your numbers?
Right on topic as for me, bur! Keep it here please for all others (composites and people). :-)

2021-06-10, 12:35   #191
Max0526

"Max"
Jun 2016
Toronto

15758 Posts

Quote:
 Originally Posted by fivemack Line 56 (6,-7) complete by SNFS. Required rather more relations than I was expecting (29.7m raw worked, 29.3m raw didn't) and about 1100 thread-hours, so three or four days on a not-very-idle i9-7940X. 14e siever, average 1.6 to 1.9 million relations per million-Q Not quite sure what to work on next - I can probably do a 180-digit GNFS in two weeks, I have done quartic SNFS of difficulty 250 for the Fibonacci numbers but that took 150k thread-hours which is getting quite substantial, especially when 120 of my threads are still waiting for an electrician.
That was a tough corner point and the 90th line done in the project!
Thank you for finishing it, fivemack!

> Not quite sure what to work on next
Line 99 (4, -9) -- two SNFS jobs got released by thyrex. If you pick one/them up, do you need me to generate 4 SNFS polys for each?

If you are looking for ECM / SNFS / GNFS challenge in general, we have unfinished leftovers after the last GCD run.

survived 5120 curves at B1=3M (B2=300M)
Code:
c153 (10, -8) GCD = 278106774146897761889088065378881257457214004932010734645371580984833113728381371759350143981110354929176767086199946165773758751608762941001961811169829
same c 153 (9,  -8) GCD = 278106774146897761889088065378881257457214004932010734645371580984833113728381371759350143981110354929176767086199946165773758751608762941001961811169829
no official results / ECM level unknown
Code:
c133 (5,  -10) GCD = 9545342495904618816640292834610815871220823986206246542435770890095240888878169242902309265848865468979768283180199480693110167603219
same c133 (6,  -10) GCD = 9545342495904618816640292834610815871220823986206246542435770890095240888878169242902309265848865468979768283180199480693110167603219
c150 (9, -9) GCD = 904224293268543843801569160544524954043974152721386519658926836632420221588920131493335396941480653249031357624804216365494500922135668590974674993099
same c150 (10, -9) GCD = 904224293268543843801569160544524954043974152721386519658926836632420221588920131493335396941480653249031357624804216365494500922135668590974674993099
c159 (10,  -10) GCD = 130715471304737731942896783128177037171192857024887336805411473265434714847285168042987524345122789642978895577107728740264664543351159368054787653395381786139
same c159 (9,  -10) GCD = 130715471304737731942896783128177037171192857024887336805411473265434714847285168042987524345122789642978895577107728740264664543351159368054787653395381786139
c173 (7,  -10) GCD = 12864712255988255321120686908357785941428688500920022805855938815039530086069901747360196720971339311554550717720578514148182515159971529413269696244808399728892293014244381
same c173 (8,  -10) GCD = 12864712255988255321120686908357785941428688500920022805855938815039530086069901747360196720971339311554550717720578514148182515159971529413269696244808399728892293014244381
c183 (8,  -10) GCD = 260619676534535520006065925766601144379916756293638823393521545289079042453073513762919243226752208436523359004606829311659266977101397987864247642777246238394102711162002636768376591
same c183 (9,  -10) GCD = 260619676534535520006065925766601144379916756293638823393521545289079042453073513762919243226752208436523359004606829311659266977101397987864247642777246238394102711162002636768376591

2021-06-10, 12:38   #192
EdH

"Ed Hall"
Dec 2009

3,823 Posts

Quote:
 Originally Posted by swishzzz I think generally tn means that the collection of curves run has about a 1-1/e (about 63%) chance of finding a factor of n digits if one were to exist, so that would mean the point on the graph where success probability is closest to 63%.
Thanks! That will be where I count from now on.

2021-06-10, 13:08   #193
bsquared

"Ben"
Feb 2007

22·3·293 Posts

Quote:
 Originally Posted by bsquared Thanks for the new polys... if the methods to generate a5 thru a8 apply in general then I'm interested to see the formulas. I've started sieving line 105 already, should be done when I wake up tomorrow.
Line 105 is done, c158 split as p77 * p85 by SNFS.

Should I continue with SNFS on line 103 or something else a higher target?

2021-06-10, 13:15   #194
swishzzz

Jan 2012

3·29 Posts

Quote:
 Originally Posted by fivemack Line 56 (6,-7) complete by SNFS. Required rather more relations than I was expecting (29.7m raw worked, 29.3m raw didn't) and about 1100 thread-hours, so three or four days on a not-very-idle i9-7940X. 14e siever, average 1.6 to 1.9 million relations per million-Q Not quite sure what to work on next - I can probably do a 180-digit GNFS in two weeks, I have done quartic SNFS of difficulty 250 for the Fibonacci numbers but that took 150k thread-hours which is getting quite substantial, especially when 120 of my threads are still waiting for an electrician.
Seems like 3000 digits (for the size of the full discriminant) is around the upper limit of the numbers we can reasonably completely factor unless we get quite lucky with ECM. I'll focus on those after my GNFS runs, maybe finish half a t55 on the c179 on line 87.

2021-06-10, 13:37   #195
bsquared

"Ben"
Feb 2007

66748 Posts

Quote:
 Originally Posted by Max0526 no official results / ECM level unknown Code: c133 (5, -10) GCD = 9545342495904618816640292834610815871220823986206246542435770890095240888878169242902309265848865468979768283180199480693110167603219 same c133 (6, -10) GCD = 9545342495904618816640292834610815871220823986206246542435770890095240888878169242902309265848865468979768283180199480693110167603219 c150 (9, -9) GCD = 904224293268543843801569160544524954043974152721386519658926836632420221588920131493335396941480653249031357624804216365494500922135668590974674993099 same c150 (10, -9) GCD = 904224293268543843801569160544524954043974152721386519658926836632420221588920131493335396941480653249031357624804216365494500922135668590974674993099 c159 (10, -10) GCD = 130715471304737731942896783128177037171192857024887336805411473265434714847285168042987524345122789642978895577107728740264664543351159368054787653395381786139 same c159 (9, -10) GCD = 130715471304737731942896783128177037171192857024887336805411473265434714847285168042987524345122789642978895577107728740264664543351159368054787653395381786139 c173 (7, -10) GCD = 12864712255988255321120686908357785941428688500920022805855938815039530086069901747360196720971339311554550717720578514148182515159971529413269696244808399728892293014244381 same c173 (8, -10) GCD = 12864712255988255321120686908357785941428688500920022805855938815039530086069901747360196720971339311554550717720578514148182515159971529413269696244808399728892293014244381 c183 (8, -10) GCD = 260619676534535520006065925766601144379916756293638823393521545289079042453073513762919243226752208436523359004606829311659266977101397987864247642777246238394102711162002636768376591 same c183 (9, -10) GCD = 260619676534535520006065925766601144379916756293638823393521545289079042453073513762919243226752208436523359004606829311659266977101397987864247642777246238394102711162002636768376591

(9, -10) GCD c159 factors as
Code:
PRP30 = 474691209469028714385947310001
PRP30 = 343529014609848460590154775291
PRP56 = 13581444414842401684004799943789251748189207442075102209
PRP44 = 59020980441390309457852487922385042008431281
by ecm and siqs. factors entered into factordb.
(The others all have at least 5120 curves @ 3M/300M)

Last fiddled with by bsquared on 2021-06-10 at 13:54

2021-06-10, 14:28   #196
Max0526

"Max"
Jun 2016
Toronto

37D16 Posts

Quote:
 Originally Posted by bsquared Line 105 is done, c158 split as p77 * p85 by SNFS. Should I continue with SNFS on line 103 or something else a higher target?
Thank you for the line 105 split, bsquared!

Line 103 SNFS sounds great for me, if you are up for the complexity of it.
It is a target as it opens a new point on the plot, last composite. Do you need polys generated for it?
I am marking the sheet for you.

2021-06-10, 14:54   #197
bsquared

"Ben"
Feb 2007

22·3·293 Posts

Quote:
 Originally Posted by Max0526 Thank you for the line 105 split, bsquared! Line 103 SNFS sounds great for me, if you are up for the complexity of it. It is a target as it opens a new point on the plot, last composite. Do you need polys generated for it? I am marking the sheet for you.
a2 was the best poly, sieving now. Getting a little over 2 rels/q average so far with a poly very highly skewed toward rational and the 14e siever.

2021-06-10, 14:56   #198
EdH

"Ed Hall"
Dec 2009

3,823 Posts

Quote:
 Originally Posted by EdH . . . It will not be until tomorrow, but I WILL factor the c124!!
Done:
Code:
Thu Jun 10 10:29:57 2021  p57 factor: 131602702254471142369840998682715878213808538243164717269
Thu Jun 10 10:29:57 2021  p68 factor: 14058479411029392739744800321961901720197429606455508685678591243029
I'll work on line 140 (7, 10) unless I see opposition. . .

 Similar Threads Thread Thread Starter Forum Replies Last Post baih Miscellaneous Math 9 2020-09-21 07:11 xx005fs GPU Computing 3 2018-10-27 14:49

All times are UTC. The time now is 08:21.

Sat Jul 24 08:21:56 UTC 2021 up 1 day, 2:50, 1 user, load averages: 1.85, 1.78, 1.64