mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   NFS@Home (https://www.mersenneforum.org/forumdisplay.php?f=98)
-   -   2022 Queue management of 15e (https://www.mersenneforum.org/showthread.php?t=27462)

swellman 2022-03-15 15:25

[B]QUEUED AS 9p7_313_1[/B]

9+7,313 is a c176 cofactor from the HCN project ready for factoring, best accomplished as a GNFS job on 15e.

[code]
n: 78843157761124969315587837918376120020673977495146866536503374331150581966912191403137620791580812931050325866821847754132831737430550505783726834338788688491040955957006925267
skew: 20342150.27
lss: 0
type: gnfs
c0: -5377962198161983727052124489126551577629600
c1: 1771437999969865121833151878911970092
c2: 187537198432561329143534184052
c3: -6936139847659631621153
c4: -566687942257278
c5: 8383320
Y0: -6232604848911270864558578513802319
Y1: 18033458842268629
# norm 3.859232e-17 alpha -8.283314 e 1.429e-13 rroots 3
lpbr: 31
lpba: 31
mfbr: 62
mfba: 91
rlim: 266000000
alim: 134000000
rlambda: 2.8
alambda: 3.5
[/code]

Results of test sieving on the algebraic side with Q in blocks of 1000:

[code]
MQ Norm_yield Speed (sec/rel)
30 3366 0.293
50 3327 0.346
70 3236 0.349
100 2911 0.417
150 2848 0.460
[/code]

Suggesting a sieving range for Q of 30-115M to generate 260M raw relations.

swellman 2022-03-18 11:21

[B]QUEUED AS 7m4_355[/B]

7-4,355 of the HCN project has sufficient ECM to start sieving. It is difficulty SNFS 240 and a quartic.

[code]
n: 193541927652261891086380584417085268168739735822576911271284692368571152349130168378344633077258918745214934219740467786292948271384720788264449198074077767364712964881138162112985926980085228773919652255281312503944394142998764070402661
skew: 1.0000
type: snfs
size: 240
c4: 1
c3: 1
c2: 1
c1: 1
c0: 1
Y1: -5575186299632655785383929568162090376495104
Y0: 1004525211269079039999221534496697502180541686174722466474743
rlim: 134000000
alim: 266000000
lpbr: 33
lpba: 31
mfbr: 96
mfba: 62
rlambda: 3.7
alambda: 2.7
[/code]

Results of test sieving on the rational side with Q in blocks of 1000:

[code]
MQ Norm_yield Speed (sec/rel)
40 2030 0.458
70 2419 0.416
100 2573 0.420
150 2550 0.434
200 2337 0.455
250 2116 0.513
300 2209 0.488
[/code]

Suggesting a sieving range of 40-315M to generate 640M raw relations.

swellman 2022-03-18 19:19

[B]QUEUED AS 10p7_284[/B]

10+7,284 is a c185 HCN now ready for GNFS. Many thanks to Gimarel and firejuggler for their poly search efforts.

[code]
n: 38983880699168921298316015054540271508889383127549270301358526648583703324178966253204238246585503306728645089126706429740754423705165969703974889766191659525379287808908446024743760793
skew: 17879343.00
type: gnfs
lss: 0
c0: 1341855282491731695564343525049920252082850
c1: 1870186313200834972489851377048141965
c2: 500320095246279481938411056682
c3: 10507905717386259288761
c4: -1305319932837906
c5: 15876000
Y0: -430189005887341704564111931811306996
Y1: 15738024048870026182457
# norm 6.282963e-18 alpha -7.400786 e 4.766e-14 rroots 5
rlim: 266000000
alim: 134000000
lpbr: 32
lpba: 32
mfbr: 64
mfba: 94
rlambda: 2.8
alambda: 3.5
[/code]

Results of test sieving on the algebraic side with Q in blocks of 1000:

[code]
MQ Norm_yield Speed (sec/rel)
30 2857 0.243
60 2938 0.272
100 2849 0.325
150 2477 0.366
200 2312 0.330
[/code]

Suggesting a sieving range for Q of 30-205M to generate 460M raw relations.

unconnected 2022-03-20 19:25

[B]QUEUED AS C186_11040_10202[/B]

C186 from 11040:i10202 for 15_e queue:

[CODE]n: 168918465384180643662716769709089718059972741101722173555698638045990352875245861820158511335351231353190493431651610430238093193234757922056654293175072760330553973006841109358197985261
# norm 5.339537e-18 alpha -8.220923 e 4.390e-14 rroots 5
skew: 13460888.89
c0: -600311797157436043820004087077790210111648
c1: 2723382055806526334785564861258639577
c2: 196110992253051405714289555116
c3: -101450361108133412492777
c4: -1040155032295308
c5: 63504000
Y0: -502124576662630967110746962406571804
Y1: 7146029031049943749427
rlim: 268000000
alim: 134000000
lpbr: 32
lpba: 32
mfbr: 64
mfba: 94
rlambda: 2.8
alambda: 3.5
type: gnfs
lss: 0
[/CODE]Thanks [B]Gimarel[/B] for the poly!
Suggesting sieve range is 30M-230M, this should produce ~470M raw relations.

VBCurtis 2022-03-20 19:38

[QUOTE=unconnected;602175]C186 from 11040:i10202 for 15_e queue:
[/QUOTE]
Let us know when you do the test-sieving, so we can queue the job with an appropriate Q-range.

unconnected 2022-03-20 20:24

[QUOTE=VBCurtis;602179]Let us know when you do the test-sieving, so we can queue the job with an appropriate Q-range.[/QUOTE]
Thanks for notice that, I've just forgot about it. Added to the post.

swellman 2022-03-20 21:27

[B]QUEUED AS 11m9_251[/B]

11-9,251 is a HCN just finished with ECM. It is a straightforward sextic of SNFS difficulty 261 but must be run on 15e. (A previous HCN job 11+9,251 of the same difficulty just fit into 15e_small.)

Also note that 3 LP on the rational side yielded a bit less than the algebraic side, FWIW.

[code]
n: 14801980590115457490677498287171010843576996063668718271456717546185407488618462563007513859056380011362634544087628866285932815343449391727971525107669020951444668481539038016056726822547973340280091613240116836715724138097524554331257330897
# 11^251-9^251, difficulty: 262.43, anorm: 2.01e+031, rnorm: 5.38e+049
# scaled difficulty: 265.50, suggest sieving rational side
# size = 4.732e-013, alpha = 0.000, combined = 6.645e-014, rroots = 2
type: snfs
size: 261
skew: 1.0340
c6: 9
c0: -11
Y1: -11972515182562019788602740026717047105681
Y0: 54763699237492901685126120802225273763666521
rlim: 134000000
alim: 134000000
lpbr: 31
lpba: 31
mfbr: 62
mfba: 91
rlambda: 2.7
alambda: 3.4
[/code]

Test sieving results on the rational side with Q in blocks of 1000:

[code]
MQ Norm_yield Speed (sec/rel)
40 1687 0.317
70 1479 0.371
100 1443 0.396
150 1326 0.462
200 1212 0.598
230 1254 0.463
[/code]

Suggesting a sieving range for Q of 40-230M to generate 260M raw relations.

swellman 2022-03-24 16:48

[B]QUEUED AS 11m3_759M[/B]

11-3,759M is a GNFS 191 from the HCN project with 31,500+ curves @B1=260M worth of ECM. It is ready for sieving with a decent polynomial I found, best run as a hybird 32/33 job:

[code]
n: 66762596638636820386929302649675220224468124723775690876489936370620746421733699748670159276884984614106259405051808107188938168897741416136338075611479877879276235168876966095751126963625667
skew: 147738603.183
type: gnfs
lss: 0
c0: 8470553493921159614495257467089552847885519231
c1: 669353813191452513944882623775763795741
c2: 27337552030309059697452037846779
c3: -120385332917097867954581
c4: -857569306262650
c5: 315000
Y0: -11621164813546903375498432694472690734
Y1: 251971934558916447297487
# MurphyE (Bf=8.590e+09,Bg=4.295e+09,area=5.469e+16) = 1.073e-08 = 1.852e-14
lpbr: 32
lpba: 33
mfbr: 64
mfba: 96
rlim: 266000000
alim: 134000000
rlambda: 2.8
alambda: 3.7
[/code]

Test sieving results on the algebraic side with Q in blocks of 1000:

[code]
MQ Norm_yield Speed (sec/rel)
30 2367 0.342
40 2586 0.316
70 2685 0.320
100 2694 0.349
150 2532 0.347
200 2363 0.349
250 2227 0.420
300 2108 0.428
[/code]

Suggesting a sieving range of 30-310 to produce 680M raw relations.

swellman 2022-03-28 11:58

[B]QUEUED AS 3p2_1674L[/B]

3+2,1674L is a GNFS 192 from the HCN project. It has completed ECM and is ready for sieving. Thanks to Gimarel for the record breaking poly.

[code]
n: 118774638495343044085764617595746668209262348394154072091504703448929588459098368911352625946762996320778680241182601579702534542835804685795107781414309880804992674911118346161218362744529909
skew: 10718472.39
type: gnfs
lss: 0
c0: 9376981406132107809876067048173307147544320
c1: 4127466627368161466860176065390233516
c2: -1618309897539727281336682707095
c3: -99550693665552197274611
c4: 15421737026411625
c5: 291060000
Y0: -4591369400118680824085671137781295259
Y1: 3834028250588570488691
# norm 1.320094e-18 alpha -7.316701 e 1.861e-14 rroots 5
lpbr: 33
lpba: 33
mfbr: 66
mfba: 96
rlim: 266000000
alim: 134000000
rlambda: 3.0
alambda: 3.7
[/code]

Results of test sieving on the algebraic side with Q in blocks of 1000:

[code]
MQ Norm_yield Speed (sec/rel)
50 3890 0.207
70 3700 0.325
100 3724 0.247
150 3138 0.386
200 2980 0.429
250 2809 0.435
300 2692 0.444
350 2520 0.468
400 2421 0.461
[/code]

Suggesting a sieving range for Q of 50-360M to generate 950M raw relations.

VBCurtis 2022-03-30 01:09

[B]QUEUED AS 13_2_923m1[/B]

13*2^923-1 is ready for 15e:
[code]n: 1959001988598696195169598875484173714164262864152755309295532999017308059740836144253975421850379001567296238667801325144547118498943949460556177156530470062305206052116825081469379755710398617447483621683888828244222028559070360588645063676920531985462193
type: snfs
skew: 0.74
deg: 6
c6: 13
c0: -2
Y1: -1
Y0: 22835963083295358096932575511191922182123945984
rlim: 134000000
alim: 266000000
lpbr: 34
lpba: 33
mfbr: 98
mfba: 66
rlambda: 3.6
alambda: 2.4[/code]

Test-sieving indicates a Q-range of 25M to 255M is sufficient for 1200M relations:
[code]#Q=30M 9025 rels, 0.115 sec/rel
#Q=60M 6436 rels, 0.126 sec/rel
#Q=90M 5461 rels, 0.145 sec/rel
#Q=120M 5079 rels, 0.157 sec/rel
#Q=150M 5162 rels, 0.161 sec/rel
#Q=180M 3400 rels, 0.188 sec/rel
#Q=210M 3890 rels, 0.182 sec/rel
#Q=240M 4273 rels, 0.198 sec/rel
#Q=270M 3668 rels, 0.209 sec/rel[/code]

If I understood Greg correctly, the BOINC 15e siever has been updated in exactly the same way as the f-small siever, meaning 34-bit large primes should be OK. If I am mistaken and the cap remains 33, then please change mfbr to 96, lpbr to 33, and Q-range to 25-290M for 890M relations:
[code]#Q=30M 4683 rels, 0.174 sec/rel
#Q=60M 4170 rels, 0.196 sec/rel
#Q=90M 3749 rels, 0.232 sec/rel
#Q=120M 3580 rels, 0.252 sec/rel
#Q=150M 3764 rels, 0.253 sec/rel
#Q=180M 2510 rels, 0.286 sec/rel
#Q=210M 2769 rels, 0.282 sec/rel
#Q=240M 3027 rels, 0.306 sec/rel
#Q=270M 2593 rels, 0.320 sec/rel
#Q=300M 2481 rels, 0.310 sec/rel[/code]

swellman 2022-03-30 03:22

Re: 13*2^923-1
 
Greg says go for it.

I’ll post it as a 34-bit job.


All times are UTC. The time now is 16:53.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2022, Jelsoft Enterprises Ltd.