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)

VBCurtis 2018-04-23 00:21

C205 poly:
[code]# norm 3.277324e-20 alpha -10.060107 e 1.785e-15 rroots 3
skew: 5437653703.58
c0: -902085823224242369884783360210562600475884684968972000
c1: 619114696957349641498752587505301367376415520
c2: 223020967372172737921370143866737514
c3: -184407634158824479270798031
c4: -9060082353529094
c5: 1147080
Y0: -5279616874146630312291050124793368106551
Y1: 22106657921291548003[/code]
1M may be too small a coeff for this size composite- note skew of 5e9. Once Rich has a similar-scoring poly from 4M range with its likely-smaller skew, I'll do a little test sieving to see if "too much skew" is a thing.

Interpolating from the record polys over 195 digits, I think 2.1 to 2.2x will be our final score, and polys better than 1.8 should be "polished" by Max in case of a +20% miracle. If we don't turn up 1.9's after a few GPU weeks, we'll have to hope a 1.7x polishes to 2.0+?

RichD 2018-04-23 02:07

C205 poly
 
Best so far at the 2/3 point through 4-5M. I'll continue on to 6M.
[CODE]N: 4705499686630466473042812716861336570115880660669072889905496053875945320601882581045264500921665604368264934151667308502968743487543692918887982034128889292133536046049610054928273805151278311296155020067
# expecting poly E from 2.44e-15 to > 2.80e-15
R0: -4039698644653792089233790873671872282468
R1: 628992977004761595913
A0: -3509676591085569433719077056369866148709592007490625
A1: 8547863309863428365739706586715652270298215
A2: 22098645276592814812963801523633769
A3: -29227035952296083163151503
A4: -24129478705425692
A5: 4373820
skew 1113014561.55, size 2.980e-20, alpha -8.438, combined = 1.898e-15 rroots = 5[/CODE]

[CODE]R0: -4095894508738186780047604677126675033348
R1: 51142235084952881977
A0: -203864180360429049803225186629129272035959287084033985
A1: 157854552003963279860605191097772126032185331
A2: 216425049444553148154467930662498615
A3: -96658374742712179889142043
A4: -55074355189386390
A5: 4081896
skew 2345726867.42, size 2.911e-20, alpha -9.789, combined = 1.848e-15 rroots = 5[/CODE]

And another probably not worth mentioning.
[CODE]R0: -4087060766188978876393254650649990075686
R1: 423055644592826487473
A0: 3108579649075688912904854953276812069711814563382809
A1: -13770309936635113341878614050905096223733067
A2: -26654612205147586352092138045345571
A3: -1470161736035081447132213
A4: -2019922505643758
A5: 4126200
skew 1358253988.77, size 2.494e-20, alpha -8.254, combined = 1.697e-15 rroots = 3[/CODE]

swellman 2018-04-23 20:34

C200
 
Still plugging along, approaching c5=1e5. No flares, in fact nothing above 3.700. Maybe my search range is too low for a C200? Unlucky? Who knows?

Speed increasing slightly, hoping it keeps ramping. I’ll keep going until advised to stop.

Max0526 2018-04-24 01:24

C205 poly
 
[QUOTE=RichD;485964]And another probably not worth mentioning.
[CODE]R0: -4087060766188978876393254650649990075686
R1: 423055644592826487473
A0: 3108579649075688912904854953276812069711814563382809
A1: -13770309936635113341878614050905096223733067
A2: -26654612205147586352092138045345571
A3: -1470161736035081447132213
A4: -2019922505643758
A5: 4126200
skew 1358253988.77, size 2.494e-20, alpha -8.254, combined = 1.697e-15 rroots = 3[/CODE][/QUOTE]
This one spins up a bit:
[code]
# norm 3.252800e-20 alpha -8.447565 e 1.783e-15 rroots 3
skew: 1312232174.16
c0: 2225847670065637875776930319888692431664405115720240
c1: -12461580131678746709321463720773050872950132
c2: -26554072360554529336998669364436360
c3: -1246457120217633606417101
c4: -2527392310914758
c5: 4126200
Y0: -4087060766199384962650843669198844432279
Y1: 423055644592826487473
[/code]

VBCurtis 2018-04-27 20:13

More C200 polys
 
Here are this week's greatest hits:
[code]# norm 1.225720e-19 alpha -7.022927 e 4.144e-15 rroots 5
skew: 435364456.32
c0: -4239426932400990647972761365453510040062785704960
c1: 68325256209029787624186971210366552287528
c2: 486905441969908713922404095934980
c3: -1344859521302865109432606
c4: -2774068872309403
c5: 1670340
Y0: -489207429140957914168476416467357529067
Y1: 2336147341067514029[/code]
[code]# norm 1.192045e-19 alpha -8.405494 e 3.990e-15 rroots 3
skew: 794051411.14
c0: 485742819063526923233512051613234369082980321127687
c1: 1049290820304017378679942952990959537623052
c2: -4031320415158215716824415340870013
c3: -5917161339405616756445738
c4: 9211724637102620
c5: 2260440
Y0: -460484931334971320265971423024624119430
Y1: 14730460983535622093[/code]
[code]# norm 1.191205e-19 alpha -8.912441 e 4.033e-15 rroots 3
skew: 887790779.63
c0: 104348040205722424665219955878237111923360337173967
c1: 1034607206215427415257987648764345294136469
c2: 6122545048910592575902288693875344
c3: -6056708183148752953776661
c4: -11669457258400663
c5: 2332440
Y0: -457606225078059226188808262699323431130
Y1: 3800844310833479243[/code]
I'm nearly halfway done with 2-3M; 1-2M is complete.

Max0526 2018-04-27 22:27

C200 polys
 
[QUOTE=VBCurtis;486402]Here are this week's greatest hits:
[code]# norm 1.225720e-19 alpha -7.022927 e 4.144e-15 rroots 5
skew: 435364456.32
c0: -4239426932400990647972761365453510040062785704960
c1: 68325256209029787624186971210366552287528
c2: 486905441969908713922404095934980
c3: -1344859521302865109432606
c4: -2774068872309403
c5: 1670340
Y0: -489207429140957914168476416467357529067
Y1: 2336147341067514029[/code][/QUOTE]
The above spins up. At the end it will be a battle of VBCurtis's 4.2s and 4.3s. It shows higher 4.1s too but I omit them.
[code]
# norm 1.300852e-19 alpha -6.390943 e 4.323e-15 rroots 5
skew: 191347710.12
c0: -23317238081768083886733557930987982989078886520
c1: 5434648917939758945859757093245249195236
c2: 104831526013350974422273328724150
c3: -757707626085281496548203
c4: -2642582630604403
c5: 3340680
Y0: -489207429140921134682390218900114863217
Y1: 4672294682135028058
[/code][code]
# norm 1.297069e-19 alpha -7.557417 e 4.321e-15 rroots 3
skew: 400828319.94
c0: 4275212665675748772941835174069059603709297784160
c1: -30082181216689000468642994290162318593256
c2: -42784600122729334767469873218640
c3: -1199686157927472342047507
c4: -1814378917201603
c5: 3340680
Y0: -489207429140689468566159009367746676581
Y1: 4672294682135028058
[/code][code]
# norm 1.264685e-19 alpha -7.194171 e 4.277e-15 rroots 1
skew: 337626375.01
c0: -2930648107022327049399992251437520332349254384036
c1: 15577994124214048127805389571629087241639
c2: 100813217371759697785073838016231
c3: -776067321070058277042391
c4: -2613409324037803
c5: 3340680
Y0: -489207429140912974290785430648995191375
Y1: 4672294682135028058
[/code]

Max0526 2018-04-29 00:30

C200 polys -- good news
 
[QUOTE=Max0526;486422]At the end it will be a battle of VBCurtis's 4.2s and 4.3s.
[/QUOTE]
Apparently, I was wrong, it'll be a battle of 4.5s and above. Coming soon.

swellman 2018-04-29 01:15

C200
 
My personal best on the C200.

[code]
R0: -852834520596787430067206034496828829587
R1: 1208407823911089329
A0: 137145669165954147951957178335300696926009862599232
A1: 1511357598918410395153445716820127867540792
A2: -1692153134917748760881216148265926
A3: -16511809439665950333096277
A4: 752683226054464
A5: 103740
skew 1661479156.81, size 1.086e-19, alpha -8.963, combined = 4.138e-15 rroots = 5
[/code]

Max0526 2018-04-29 01:53

C200 -- four records in a row from CADO
 
[QUOTE=Max0526;486532]Apparently, I was wrong, it'll be a battle of 4.5s and above. Coming soon.[/QUOTE]
[code]
Y0: -489207429141006631900117062794681583529
Y1: 4672294682135028058
c0: -109152923706857914577390790494597391944231527200
c1: 5795645017560106604536965122174413937870
c2: 140913043704203190308889704674718
c3: -553097550393985150383835
c4: -2948234205202003
c5: 3340680
skew: 208641447.59179
# lognorm 61.86, E 55.18, alpha -6.67 (proj -1.63), 5 real roots
# MurphyE = 4.53081598e-15
[/code][code]
Y0: -489207429140969628765301315470050865393
Y1: 4672294682135028058
c0: -1125122782775082452092567517746701723798804989176
c1: 3025680921490687412989206542361179128036
c2: 126679044891883601026836778141686
c3: -644398652469977989880939
c4: -2815948421849203
c5: 3340680
skew: 266291072.77407
# lognorm 62.21, E 55.18, alpha -7.04 (proj -1.63), 3 real roots
# MurphyE = 4.52526204e-15
[/code][code]
Y0: -489207429141502432553102304304026870805
Y1: 4672294682135028058
c0: -130350159351028219108446747738303628468541675776
c1: -2124977598538771670981420510096972305516
c2: 77881481007157843830645332376816
c3: 1074484060157966877663629
c4: -4720715863676803
c5: 3340680
skew: 165095306.96074
# lognorm 62.07, E 55.58, alpha -6.50 (proj -1.63), 5 real roots
# MurphyE = 4.48419936e-15
[/code][code]
Y0: -489207429140909266605357306443884849649
Y1: 4672294682135028058
c0: -493621402117203176187307992424387442500241583000
c1: 5952374268635619259616735221940151899940
c2: 98955822135076711437323239278198
c3: -784341736760936282050155
c4: -2600154391078003
c5: 3340680
skew: 206035416.09230
# lognorm 61.91, E 55.41, alpha -6.50 (proj -1.63), 5 real roots
# MurphyE = 4.39628853e-15
[/code]

VBCurtis 2018-04-29 02:23

Max-
You've never let me know what range and nq you're searching with CADO for this C200. I suspended my CADO work this week for other reasons, but await your information so I can decide how to best proceed.

So far, I've run 500k-900k with nq=3125. I'll finish to 1M soon, and then will decide based on your info.

4.5 is pretty nice for this number; the sort of flare I've been holding out for. I don't usually test-sieve multiple polys from the same original hit, but I'll test the top two against the best from a previous hit. If one of the 4.5s performs as well as the score difference from the 4.3s indicates it should, I think we can send this on to Kurt & co.

I doubt I'll get to test-sieving tomorrow, so let's plan on more search until the end of this week, and then decide after I get test-sieving info.

Max0526 2018-04-29 03:11

C200 poly search
 
@VBCurtis
I got to 600k with
[code]
tasks.polyselect.P = 10000000
tasks.polyselect.incr = 12
tasks.polyselect.nq = 3125
[/code]We probably intersected a bit on 500k-600k there, I'm sorry for not letting you know in time.
I am trying to run the last four polys through msieve during the weekend, it doesn't work at all, msieve doesn't see anything above 4.323 yet. After that I'll be out for the next four days, too much stuff at work. But I'll get back next Friday.
Thank you so much for the future test-sieving!


All times are UTC. The time now is 23:09.

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