mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Msieve (https://www.mersenneforum.org/forumdisplay.php?f=83)
-   -   Polynomial Request Thread (https://www.mersenneforum.org/showthread.php?t=18368)

Max0526 2019-07-18 15:17

C162 poly
 
My best CADO poly so far:[code]Y0: -19842869335459075015251022616866
Y1: 93388358975407157387
c0: 1663678916559999938652381253611265374
c1: 59807525913835586945325216936586
c2: -20679164131397310880416811
c3: -33095437817726307969
c4: -1879830200070
c5: 1690200
skew: 2361877.22856
# lognorm 49.49, E 43.08, alpha -6.41 (proj -2.57), 5 real roots
# MurphyE = 1.13275429e-12[/code]

VBCurtis 2019-07-18 15:59

[QUOTE=swellman;521862]No worries. Keep in mind that’s likely a 15e job.[/QUOTE]

C180 is far easier on 15e, for sure; I think a test-sieve that shows 14e is 20% slower is acceptable in exchange for the shorter queue and faster result on 14e.

The test-sieving might be a bit complicated, as 15e may benefit from lim's half the size of the ones you choose for 14e (or, perhaps, 70% size). e.g. if you try 268M for 14e, try 134M and 200M and 268M for 15e.
EDIT: I posted before adding the poly:
[code]n: 5569405272001321328542088675673286615239616651676919352988365836178676833273524154355871452912145865149105882$
skew: 21872667.352
c0: 12107295535342871823236436212001956583790830
c1: 1920658626946783575842554229662615341
c2: -251726862055821673188246578096
c3: 3503018700597041947423
c4: 417824834197750
c5: 1167600
Y0: -86131594648870461324526977473657588
Y1: 1502101774554069680002831
# MurphyE (Bf=4.295e+09,Bg=2.147e+09,area=2.416e+16) = 1.48e-08[/code]
cownoise says skew 33532655.49939 E-score 8.70334954e-14. So, our best so far.
I'll do a slightly longer run today, should have another candidate tonight. This was a search from 4k-300k with nq set to 15625, incr=210, and P=3M.

Edit #2: This was about 480k (not 1.3M, re-edited) thread-seconds of search. Not much effort for the results!

Max0526 2019-07-18 18:35

C180 poly
 
[QUOTE=VBCurtis;521864]C180 is far easier on 15e, for sure; I think a test-sieve that shows 14e is 20% slower is acceptable in exchange for the shorter queue and faster result on 14e.

The test-sieving might be a bit complicated, as 15e may benefit from lim's half the size of the ones you choose for 14e (or, perhaps, 70% size). e.g. if you try 268M for 14e, try 134M and 200M and 268M for 15e.
EDIT: I posted before adding the poly:
[code]n: 5569405272001321328542088675673286615239616651676919352988365836178676833273524154355871452912145865149105882$
skew: 21872667.352
c0: 12107295535342871823236436212001956583790830
c1: 1920658626946783575842554229662615341
c2: -251726862055821673188246578096
c3: 3503018700597041947423
c4: 417824834197750
c5: 1167600
Y0: -86131594648870461324526977473657588
Y1: 1502101774554069680002831
# MurphyE (Bf=4.295e+09,Bg=2.147e+09,area=2.416e+16) = 1.48e-08[/code]
cownoise says skew 33532655.49939 E-score 8.70334954e-14. So, our best so far.
I'll do a slightly longer run today, should have another candidate tonight. This was a search from 4k-300k with nq set to 15625, incr=210, and P=3M.

Edit #2: This was about 480k (not 1.3M, re-edited) thread-seconds of search. Not much effort for the results![/QUOTE]
This one doesn't spin up, sorry.

Max0526 2019-07-18 18:43

C162 poly
 
Better CADO poly (and a new record for C162):[code]Y0: -14746203136816823944613660192562
Y1: 21823399221738572899
c0: 119595929340090554827356446639608491975
c1: 158467115924738283758548791610975
c2: -111894390619333624786737429
c3: -99793395630116531767
c4: 16117210329462
c5: 2237040
skew: 2768078.21821
# lognorm 50.62, E 42.81, alpha -7.81 (proj -2.30), 5 real roots
# MurphyE = 1.25115306e-12[/code]

swellman 2019-07-18 20:12

[QUOTE=Max0526;521878]Better CADO poly (and a new record for C162):[code]Y0: -14746203136816823944613660192562
Y1: 21823399221738572899
c0: 119595929340090554827356446639608491975
c1: 158467115924738283758548791610975
c2: -111894390619333624786737429
c3: -99793395630116531767
c4: 16117210329462
c5: 2237040
skew: 2768078.21821
# lognorm 50.62, E 42.81, alpha -7.81 (proj -2.30), 5 real roots
# MurphyE = 1.25115306e-12[/code][/QUOTE]

We have a winner! I’ll test sieve and queue in 14e.

Thanks Max!

VBCurtis 2019-07-19 01:30

[QUOTE=VBCurtis;521864]I'll do a slightly longer run today, should have another candidate tonight. This was a search from 4k-300k with nq set to 15625, incr=210, and P=3M.[/QUOTE]

Run #2 scored a bit better:
[code]n: 5569405272001321328542088675673286615239616651676919352988365836178676833273524154355871452912145865149105882$
skew: 2843064.177
c0: -10324367348401364499966384538905410388196
c1: 9851107221741126908874566563861144
c2: 8123815724328353410492353289
c3: -1327065403846903596526
c4: -885691407890250
c5: 19656000
Y0: -58893861824040225665437104882200001
Y1: 290320644085824192450599
# MurphyE (Bf=4.295e+09,Bg=2.147e+09,area=2.416e+16) = 1.5e-08[/code]
cownoise: skew 3567519.16518 E-score 8.80489346e-14

This run covered admin 3e5, admax 1e6, P=3.5M, incr 420, nq 15625.
Run 3 will cover 1e6-3e7 with incr 4620. I should have a last poly candidate within 24hr.

Max0526 2019-07-19 15:59

C180 poly
 
[QUOTE=VBCurtis;521899]Run #2 scored a bit better:
[code]n: 5569405272001321328542088675673286615239616651676919352988365836178676833273524154355871452912145865149105882$
skew: 2843064.177
c0: -10324367348401364499966384538905410388196
c1: 9851107221741126908874566563861144
c2: 8123815724328353410492353289
c3: -1327065403846903596526
c4: -885691407890250
c5: 19656000
Y0: -58893861824040225665437104882200001
Y1: 290320644085824192450599
# MurphyE (Bf=4.295e+09,Bg=2.147e+09,area=2.416e+16) = 1.5e-08[/code]
cownoise: skew 3567519.16518 E-score 8.80489346e-14

This run covered admin 3e5, admax 1e6, P=3.5M, incr 420, nq 15625.
Run 3 will cover 1e6-3e7 with incr 4620. I should have a last poly candidate within 24hr.[/QUOTE]
Made it a tiny bit better:[code]Y0: -58894023406930142580350424534733634
Y1: 290320644085824192450599
c0: -13147922596530357029389073843158947477435
c1: 195193748172087378144925660332776
c2: 8659583139549371264172638515
c3: 705608802632225330474
c4: -940390812650250
c5: 19656000
skew: 3570529.44
# size 1.400e-17, alpha -6.289, combined = 8.818e-14 rroots = 5[/code]

VBCurtis 2019-07-20 04:53

The last poly run completed, with a score of 1.56e-08. That's 4% better than the last one I posted.
Alas, the run was performed on a machine in my campus office, where the power is cut for 12 hrs this evening. I ran a standard CADO invocation without specifying a target folder, meaning the work was saved to /tmp and thus lost in the power outage.
So, I'll do the run again Saturday and have a poly Sunday. This will be my last effort on this C180.

VBCurtis 2019-07-21 18:19

C180 final run from CADO:
[code]skew: 1008681.119
c0: -466854068483609003256414775158074518182
c1: 468888524961519328283072109202337
c2: 2156022386419175817647117226
c3: -4001303278228806795581
c4: -1197497048095548
c5: -140595840
Y0: -34305459183446976423691552471637917
Y1: 1868038466414918025342517
# MurphyE (Bf=4.295e+09,Bg=2.147e+09,area=2.416e+16) = 1.56e-08[/code]

EDIT: cownoise: skew 1657241.61733 E-score 9.39539185e-14
Happy factoring.

Max0526 2019-07-21 23:01

C180 poly
 
[QUOTE=VBCurtis;522033]C180 final run from CADO:
[code]skew: 1008681.119
c0: -466854068483609003256414775158074518182
c1: 468888524961519328283072109202337
c2: 2156022386419175817647117226
c3: -4001303278228806795581
c4: -1197497048095548
c5: -140595840
Y0: -34305459183446976423691552471637917
Y1: 1868038466414918025342517
# MurphyE (Bf=4.295e+09,Bg=2.147e+09,area=2.416e+16) = 1.56e-08[/code]

EDIT: cownoise: skew 1657241.61733 E-score 9.39539185e-14
Happy factoring.[/QUOTE]
Spun it up a bit:[code]Y0: -34305743965911181377944515938354567
Y1: 1868038466414918025342517
c0: 474686265203189101896493769508628103832
c1: 450873946025418521924569636862563
c2: -3824013687701916272964867576
c3: 3303745460989437625181
c4: 1090327869055548
c5: 140595840
skew: 1673765.73
# size 1.504e-17, alpha -6.589, combined = 9.401e-14 rroots = 1[/code]

Max0526 2019-07-21 23:17

C180 poly
 
Next one down is:[code]Y0: -68611484010809621750976096682765951
Y1: 1868038466414918025342517
c0: 3801241994152566273755280343048396075185
c1: 1771433052284523964831685644645603
c2: -7627209275682637426571678151
c3: 3308324205976963985285
c4: 545532822862974
c5: 35148960
skew: 3503114.11
# size 1.289e-17, alpha -6.820, combined = 8.574e-14 rroots = 1[/code]


All times are UTC. The time now is 22:02.

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