Register FAQ Search Today's Posts Mark Forums Read

 2021-07-19, 02:25 #518 LaurV Romulan Interpreter     "name field" Jun 2011 Thailand 24×617 Posts Nice. I have never seen one of my own. In fact, I didn't see a 3-way split for quite a long time either. But I don't work so large composites as you guys.
 2021-07-21, 22:40 #519 charybdis     Apr 2020 593 Posts A rather lucky ECM factor from aliquot sequence 2360: Code: Using B1=11000000, B2=35133391030, polynomial Dickson(12), sigma=1:2026970909 Step 1 took 12373ms Step 2 took 5994ms ********** Factor found in step 2: 8640369870909863595929431692681480218474037077400129286983763 Found prime factor of 61 digits: 8640369870909863595929431692681480218474037077400129286983763 Prime cofactor 129269788173717510352741587971842512872817913474983535163294531334491305593503434770571399 has 90 digits
 2021-07-22, 01:29 #520 Batalov     "Serge" Mar 2008 Phi(4,2^7658614+1)/2 5·7·277 Posts Hey, cool, must be a good,lucky day for factors. I also just got an uncharacteristically large hit (for seq 2081190): Code: > cat EC.44 GMP-ECM 7.0.3 [configured with GMP 6.1.1, --enable-asm-redc] [ECM] Input number is 26659319056687683461054149321917986804957358719605281966548936444886896624104353412491288953593160598080321001018543288026127232842327373237 (140 digits) Using B1=43000000, B2=240490660426, polynomial Dickson(12), sigma=1:3707545622 Step 1 took 182540ms Step 2 took 55670ms Run 2 out of 200: ... Run 5 out of 200: Using B1=43000000, B2=240490660426, polynomial Dickson(12), sigma=1:3798568687 Step 1 took 182177ms ********** Factor found in step 1: 32756224828930142289475723150116220166815023466048767025627 Found prime factor of 59 digits: 32756224828930142289475723150116220166815023466048767025627 Prime cofactor 813870316128197392584899168196680294640704883649117820738362327353816069049664431 has 81 digits
 2021-07-22, 09:16 #521 Brownfox     Dec 2017 23·32 Posts Yup, today is officially Large Factor day: Code: GMP-ECM 7.0-dev [configured with MPIR 2.6.0, --enable-openmp] [ECM] Input number is 19665584029300424610206656998744724640118849009115415651634333787794657657987385775417086739192039189614555026416812029751027390818218736070170758482564486193541449884496154967983274557938323133 (194 digits) ... Run 89 out of 1033: Using B1=43000000, B2=240490660426, polynomial Dickson(12), sigma=1:3631092842 Step 1 took 758234ms Step 2 took 198812ms ********** Factor found in step 2: 14206140704026661221619235881597509147461207215069425801 Found probable prime factor of 56 digits: 14206140704026661221619235881597509147461207215069425801 Probable prime cofactor 1384301650885825087848187760503761406631430275388925914731703001560119263184362769750089653715830933780189618374942258401954949617045411733 has 139 digits
2021-07-22, 09:22   #522
kruoli

"Oliver"
Sep 2017
Porta Westfalica, DE

23×3×5×7 Posts

Quote:
 Originally Posted by Batalov I also just got an uncharacteristically large hit (for seq 2081190): [...]
A group order of $$2^3 \cdot 269 \cdot 8{,}807 \cdot 32{,}887 \cdot 48{,}017 \cdot 50{,}341 \cdot 59{,}791 \cdot 137{,}251 \cdot 1{,}251{,}461 \cdot 4{,}406{,}351 \cdot 18{,}528{,}707 \cdot 25{,}929{,}229$$, wow!

 2021-08-04, 20:35 #523 sean     Aug 2004 New Zealand 32·52 Posts New personal best ... Code: GMP-ECM 6.4 [configured with GMP 6.0.0, --enable-asm-redc] [ECM] Input number is (177!+1)/1891548004136643904823/271 (299 digits) Using B1=110000000, B2=776278396540, polynomial Dickson(30), sigma=1450894984 Step 1 took 1259302ms Step 2 took 246545ms ********** Factor found in step 2: 107357044729767130172699954387081510561518701754865074363224611 Found probable prime factor of 63 digits: 107357044729767130172699954387081510561518701754865074363224611 Composite cofactor ((177!+1)/1891548004136643904823/271)/107357044729767130172699954387081510561518701754865074363224611 has 237 digits sadly still a long way off the pace for the current top 10.
 2021-08-19, 06:50 #524 SethTro     "Seth" Apr 2019 2·5·41 Posts happy me Code: Resuming ECM residue Input number is 0x33eec89e881f1de8ef38dfa9da48d6d6cdf00aca90a788c587f719c01f68b3fc01f2a859fb804df05fa379 (103 digits) Using B1=1000000-1000000, B2=500000000, polynomial Dickson(3), sigma=3:46621 Step 1 took 0ms Step 2 took 277ms ********** Factor found in step 2: 50936553852286051757980931161033 Found prime factor of 32 digits: 50936553852286051757980931161033 Prime cofactor (0x33eec89e881f1de8ef38dfa9da48d6d6cdf00aca90a788c587f719c01f68b3fc01f2a859fb804df05fa379)/50936553852286051757980931161033 has 72 digits Number is a random C103 from factordb, but to me it the first "real" proof that my new fast stage1 ecm code works.
2021-08-19, 13:45   #525
bsquared

"Ben"
Feb 2007

E0216 Posts

Quote:
 Originally Posted by SethTro Code: Resuming ECM residue Input number is 0x33eec89e881f1de8ef38dfa9da48d6d6cdf00aca90a788c587f719c01f68b3fc01f2a859fb804df05fa379 (103 digits) Using B1=1000000-1000000, B2=500000000, polynomial Dickson(3), sigma=3:46621 Step 1 took 0ms Step 2 took 277ms ********** Factor found in step 2: 50936553852286051757980931161033 Found prime factor of 32 digits: 50936553852286051757980931161033 Prime cofactor (0x33eec89e881f1de8ef38dfa9da48d6d6cdf00aca90a788c587f719c01f68b3fc01f2a859fb804df05fa379)/50936553852286051757980931161033 has 72 digits Number is a random C103 from factordb, but to me it the first "real" proof that my new fast stage1 ecm code works.
Congrats! If it's an implementation you want to share it can be added to the list here. I'm always interested to know more about new programs. It looks like it uses or is compatible with gmp-ecm's param3 approach?

2021-08-20, 00:56   #526
SethTro

"Seth"
Apr 2019

6328 Posts

Quote:
 Originally Posted by bsquared Congrats! If it's an implementation you want to share it can be added to the list here. I'm always interested to know more about new programs. It looks like it uses or is compatible with gmp-ecm's param3 approach?
Yes I'm using gmp-ecm's param3 approach. It produces bit for bit identical files to gmp-ecm which can be used with ecm -resume

You can find (and if you are very brave play with) it at [1].

I'm working on merging it into ecm[2] but that's a non-trivial effort (I'm on the third 10+ hour day of work) to get the code working and will be another large effort to get it upstreamed

[1] https://github.com/sethtroisi/CGBN/t.../sample_05_ecm
[2] https://github.com/sethtroisi/gmp-ec...pu_integration

Last fiddled with by SethTro on 2021-08-20 at 00:56

 2021-11-20, 19:36 #527 fivemack (loop (#_fork))     Feb 2006 Cambridge, England 22×1,613 Posts 55274240166617786177517698361930925108950901219054105198393711339667425579 divides L1643 This is a 207-digit GNFS, I started the ECM on 21 August 2020, polynomial selection the second half of November 2020, sieving 6 March - 21 October 2021 (with a big gap 25 April - 12 August while we moved house - it took a while to get electricity and Ethernet to the garage, and the rack-mount machines are not acceptable housemates), and the linear algebra just finished. 26 threads of Skylake Xeon made fairly short work of a reasonably large matrix (with one interruption since the house power had to go off to wire the new heat-pump in): Code: Mon Oct 25 03:39:44 2021 matrix is 62544722 x 62544898 (30647.8 MB) with weight 9315321859 (148.94/col) Mon Oct 25 03:39:44 2021 sparse part has weight 7283597063 (116.45/col) Mon Oct 25 03:51:01 2021 filtering completed in 1 passes Mon Oct 25 03:51:14 2021 matrix is 62544722 x 62544898 (30647.8 MB) with weight 9315321859 (148.94/col) Mon Oct 25 03:51:14 2021 sparse part has weight 7283597063 (116.45/col) Mon Oct 25 03:53:52 2021 matrix starts at (0, 0) Mon Oct 25 03:54:04 2021 matrix is 62544722 x 62544898 (30647.8 MB) with weight 9315321859 (148.94/col) Mon Oct 25 03:54:04 2021 sparse part has weight 7283597063 (116.45/col) Mon Oct 25 03:54:04 2021 saving the first 240 matrix rows for later Mon Oct 25 03:54:18 2021 matrix includes 256 packed rows Mon Oct 25 03:54:44 2021 matrix is 62544482 x 62544898 (28165.6 MB) with weight 6756100917 (108.02/col) Mon Oct 25 03:54:44 2021 sparse part has weight 6382715051 (102.05/col) Mon Oct 25 03:54:44 2021 using block size 8192 and superblock size 878592 for processor cache size 36608 kB Mon Oct 25 04:00:29 2021 commencing Lanczos iteration (26 threads) Mon Oct 25 04:00:30 2021 memory use: 39043.6 MB Mon Oct 25 04:01:47 2021 linear algebra at 0.0%, ETA 658h35m ... Mon Nov 15 21:16:12 2021 matrix is 62544482 x 62544898 (28165.6 MB) with weight 6756100917 (108.02/col) Mon Nov 15 21:16:12 2021 sparse part has weight 6382715051 (102.05/col) Mon Nov 15 21:16:12 2021 using block size 8192 and superblock size 878592 for processor cache size 36608 kB Mon Nov 15 21:22:24 2021 commencing Lanczos iteration (26 threads) Mon Nov 15 21:22:25 2021 memory use: 39043.6 MB Mon Nov 15 21:22:49 2021 restarting at iteration 211179 (dim = 53900111) Mon Nov 15 21:24:19 2021 linear algebra at 86.2%, ETA 141h24m Mon Nov 15 21:24:47 2021 checkpointing every 70000 dimensions Sat Nov 20 03:05:47 2021 lanczos halted after 245048 iterations (dim = 62544482) Sat Nov 20 03:10:49 2021 recovered 33 nontrivial dependencies Sat Nov 20 03:11:17 2021 BLanczosTime: 367021 33-bit large primes both sides, 2 on the rational and 3 on the algebraic; large-prime bound 400M both sides. The yield-per-ideal has several points where a linear fit stops fitting well and you need to fit again with similar slope but lower intercept, and peaks around 143M. Code: n: 222842705654835428750310157418180649369425196738437743023292433223227387705941603491512689351466942188943979579791406241779390337921282895215018272769713788019782823508427649965898109154797136335087701532469 Y0: -4363620109222009261304576630621125251928 Y1: 2979763403111336301386407 c0: -966293160631088157475504471774620629860730881530 c1: -24199932911880159036837638477570208148921 c2: 368106549361642705326074034908420 c3: -1407927385984843714287529 c4: -182638606556036380 c5: 1408522500 skew: 60223320.205 # lognorm 65.19, E 57.32, alpha -7.87 (proj -2.18), 3 real roots # MurphyE(Bf=1.00e+07,Bg=5.00e+06,area=1.00e+16)=1.73e-15 477,194 thread-hours on my cluster, which got some upgrades during the sieving and is now around a hundred hyper-threaded cores. Attached Thumbnails   Last fiddled with by fivemack on 2021-11-20 at 19:51
 2021-12-31, 21:00 #528 nivek000   Dec 2021 2×5 Posts 43^113-42^113 is a 185-digit SNFS for OEIS A289985 (n=42, k=113) or A289629 (n=113, k=42). Factors are Code: P51 = 104372443967980461443712852210951709032363177730327 P135 = 340268945841017188896399386143641494373451035998499887856217897604638740935342990980415482060468236293905432473046135908690706699506253 NFS ran for about 3 days on 5 threads, but with an older, slower GGNFS. Perhaps would be 1.5-2 days on a newer version. - Kevin Moderator note - please use code tags for easier reading. Last fiddled with by swellman on 2021-12-31 at 21:52

 Similar Threads Thread Thread Starter Forum Replies Last Post wblipp Factoring 463 2019-05-30 07:19 FactorEyes Factoring 2 2011-04-09 05:45 MatWur-S530113 PrimeNet 11 2009-01-21 19:08 jasong Programming 16 2006-11-07 01:03 GP2 Hardware 7 2003-11-24 06:13

All times are UTC. The time now is 23:48.

Tue Jan 25 23:48:19 UTC 2022 up 186 days, 18:17, 0 users, load averages: 1.25, 1.27, 1.31