mersenneforum.org 3,748+ c204 Smaller-but-Needed
 Register FAQ Search Today's Posts Mark Forums Read

 2021-09-28, 02:32 #89 Max0526   "Max" Jun 2016 Toronto 38A16 Posts All are CADO spin-offs of Plutie's best with CADO and msieve/cownoise skews: Code: Y0: -5364063500450744582974088002674833158482 Y1: 38191391760434248123003 c0: -323053497049455517568711972226437285633256011200 c1: -32881533164508662100579880137642753307503 c2: -262894880327249064803188214098306 c3: 1404314030873951443478061 c4: 1920699805618928 c5: 240240 # CADO # skew: 271337941.305 # lognorm 62.75, E 54.98, alpha -7.77 (proj -2.37), 5 real roots # MurphyE(Bf=1.000e+07,Bg=5.000e+06,area=1.000e+16)=3.391e-15 # msieve skew: 481007898.57511 # size 7.424e-20, alpha -7.768, combined = 3.438e-15 rroots = 5 Code: Y0: -5364063508666891384227654733092861839050 Y1: 38191391760434248123003 c0: 22871723956720938259362411318932924652447651674724 c1: 104738178237916821258210599650730734608771 c2: -659793559025584614995447176531252 c3: -137306982760992449325011 c4: 1662284621391728 c5: 240240 # CADO # skew: 535088083.918 # lognorm 62.93, E 55.06, alpha -7.87 (proj -2.37), 5 real roots # MurphyE(Bf=1.000e+07,Bg=5.000e+06,area=1.000e+16)=3.374e-15 # msieve skew: 695929313.35178 # size 7.114e-20, alpha -7.871, combined = 3.380e-15 rroots = 5 Code: Y0: -5364063500462896472769990013479602745034 Y1: 38191391760434248123003 c0: -713990161291777371623202145207818984374249797200 c1: -32713808997208599257281929078750325616355 c2: -264234204449356184148657122001370 c3: 1401869730307664624785453 c4: 1920317602998128 c5: 240240 # CADO # skew: 282002112.573 # lognorm 62.79, E 55.19, alpha -7.60 (proj -2.37), 5 real roots # MurphyE(Bf=1.000e+07,Bg=5.000e+06,area=1.000e+16)=3.308e-15 # msieve skew: 462084249.65 # size 7.153e-20, alpha -7.597, combined = 3.335e-15 rroots = 5 Code: Y0: -5364063512426790517621366607496639457642 Y1: 38191391760434248123003 c0: -65610093342843097238682465977264696732382560535060 c1: -741105475056053815340626246194061025254449 c2: -524865775533559254998534167181272 c3: -768622622839623204122579 c4: 1544027845954928 c5: 240240 # CADO # skew: 925068831.430 # lognorm 64.01, E 55.18, alpha -8.83 (proj -2.37), 3 real roots # MurphyE(Bf=1.000e+07,Bg=5.000e+06,area=1.000e+16)=3.280e-15 # msieve skew: 1289209761.56 # size 6.818e-20, alpha -8.830, combined = 3.300e-15 rroots = 3 Last fiddled with by Max0526 on 2021-09-28 at 03:01
 2021-09-28, 04:30 #90 Max0526   "Max" Jun 2016 Toronto 2×3×151 Posts Similar msieve output: Code: Y0: -5364063502976178739336880662454761924711 Y1: 38191391760434248123003 c0: 332097428849770708325237168273205952335337119516 c1: 18109818606056862400890163879130278018596 c2: -491782570120299055422600002647243 c3: 906787991570777130541645 c4: 1841269563127328 c5: 240240 skew: 523672984.47 # size 7.424e-20, alpha -7.768, combined = 3.441e-15 rroots = 5 Code: Y0: -5364063506562591718854798984784384757665 Y1: 38191391760434248123003 c0: 26632093892502554253880366841686054358036459616744 c1: 31903270149328699502605338559388617840266 c2: -651809097836850031097043886121287 c3: 236345926992259238606029 c4: 1728469293945728 c5: 240240 skew: 709204499.17 # size 7.114e-20, alpha -7.871, combined = 3.382e-15 rroots = 5 Code: Y0: -5364063502803158291292213845602219037655 Y1: 38191391760434248123003 c0: 2760601683535597371997408984508696229590191424 c1: 13710440622589173824726865489581401028400 c2: -479231398083326591691658358969851 c3: 940203695634188752149069 c4: 1846711421949728 c5: 240240 skew: 503598307.18 # size 7.153e-20, alpha -7.597, combined = 3.338e-15 rroots = 5 Happy test-sieving!
 2021-09-28, 17:06 #91 Plutie   "Evan" Dec 2020 Montreal 23·32 Posts Very late update: Found the original polynomial pre-cownoise. (inverted signs, but it's the same E score.) Code: ### root-optimized polynomial 2 ### n: 533439167600904850230361756102700151678687933392166847323827307497363839257031077774321424872955045754669625577486179222154434651598903112919949771321416511589029559325246084363632977829645558547714072241 Y0: -5364063517160195101439064344720425686774 Y1: 38191391760434248123003 c0: 515349887751874687116560925109512114626453230780 c1: -17974344725264404895365666536735938178777 c2: 101346622782727207541871449878174 c3: 1497180969353527609911507 c4: -1395152266302128 c5: -240240 skew: 214513572.301 # lognorm 62.56, E 54.79, alpha -7.77 (proj -2.37), 3 real roots # MurphyE(Bf=1.000e+07,Bg=5.000e+06,area=1.000e+16)=3.435e-15 ### Best MurphyE so far is 3.435e-15, av. exp_E 56.25, av. E 55.95
 2021-09-29, 15:56 #92 chris2be8     Sep 2009 2·7·157 Posts msieve's square root stage, -nc3, can't handle negative leading algebraic polynomial coefficients. Which is probably why cownoise reversed all the signs. I hit this once and managed to get -nc3 to work by reversing the signs in the .poly etc, with the same relations, output from -nc2, etc. But I'd not want to rely on that working, so I've been careful to avoid it. With regard to skew's it would be interesting to test sieve a few .polys varying the skew but keeping everything else the same. Obvious skew's to try are the default given by msieve and the one cownoise suggests.
2021-09-29, 19:07   #93
frmky

Jul 2003
So Cal

32·13·19 Posts

Quote:
 Originally Posted by chris2be8 msieve's square root stage, -nc3, can't handle negative leading algebraic polynomial coefficients. Which is probably why cownoise reversed all the signs.
That's a simple fix:
https://github.com/gchilders/msieve_...bcb3dd3ab22175

 2021-09-30, 18:03 #94 swellman     Jun 2012 62548 Posts I ran some test sieving on Plutie's record polynomial using the following parameters Code: n: 533439167600904850230361756102700151678687933392166847323827307497363839257031077774321424872955045754669625577486179222154434651598903112919949771321416511589029559325246084363632977829645558547714072241 skew: 214513572.301 c0: 515349887751874687116560925109512114626453230780 c1: -17974344725264404895365666536735938178777 c2: 101346622782727207541871449878174 c3: 1497180969353527609911507 c4: -1395152266302128 c5: -240240 Y0: -5364063517160195101439064344720425686774 Y1: 38191391760434248123003 lpbr: 33 lpba: 33 mfbr: 66 mfba: 96 alim: 225000000 rlim: 225000000 rlambda: 3.0 alambda: 3.7 And then again with the cownoise suggested skew value. Test sieving on the -a side in Q blocks of 10K using 4 threads on an otherwise fully tasked machine: Code:  skew e-score Q0(M) Yield #spec_Q Corrected_Yield Avg Speed (sec/rel) 214513572.301 3.435e-15 100 26896 539 27089 1.045 214513572.301 3.435e-15 200 28566 577 25902 1.137 428721961.354 3.496e-15 100 26898 539 27091 1.041 428721961.354 3.496e-15 200 28566 577 25902 1.142 Limited data but the skew seems to make little difference on e-score or sieving performance, at least for this job. If someone wants to investigate higher Q be my guest. Suggest negating the algebraic expression, for aesthetics if nothing else. (Greg has fixed the issue but still...)
 2021-10-07, 14:55 #95 Plutie   "Evan" Dec 2020 Montreal 1108 Posts Question - Assuming we're doing a forum team-sieve for this, would anyone be willing to help me test my setup for connection to my CADO instance? I'm willing to host and post-process if doable. (unless someone is already set to host it)
 2021-10-07, 15:18 #96 VBCurtis     "Curtis" Feb 2005 Riverside, CA 37×137 Posts Sure, just PM me connection details and I'll fire up a client.
2021-10-07, 21:12   #97
kruoli

"Oliver"
Sep 2017
Porta Westfalica, DE

2FF16 Posts

From the Aliquot number 3366 thread.

Quote:
 Originally Posted by kruoli [...] I'd like to host one of them; I will adopt your proposal which one.
(My emphasis.)
Quote:
 Originally Posted by VBCurtis [...] Barring any new voices objecting, we will put this aliquot number on f-small and the C204 with you hosting a team sieve. Future team sieve discussion can be on the 3,748+ thread.
I still would like to run it. If you think we are ready, we can proceed!

 2021-10-07, 21:52 #98 VBCurtis     "Curtis" Feb 2005 Riverside, CA 37×137 Posts Kruoli- We're ready, I think- I didn't realize you were waiting on us! I suppose a little test-sieving might be in order, but I think I have a good idea of parameters. I'll post them here this weekend once I have some time to consult my notes, and likely Charybdis will comment on the ones that might have better options available. A=30, 33LP will need around 950M raw relations to yield a reasonable matrix. I'll still test Plutie's system tonight, as he wishes to know if he can host a future job. Speaking of hosting- Kruoli, do you have a static IP address? Or, how are you going to set up incoming connections from the internet? You don't need to share your IP address presently or publicly, but most American internet connections don't allow incoming connections by default so this is a bit of a problem for hosting CADO team-sieves. I hope your location does not have that problem. Last fiddled with by VBCurtis on 2021-10-07 at 21:52
2021-10-07, 22:02   #99
kruoli

"Oliver"
Sep 2017
Porta Westfalica, DE

13×59 Posts

Quote:
 Originally Posted by VBCurtis I didn't realize you were waiting on us!
Nah, I was not actively waiting; I was following the thread and looking for "keywords". This is a voluntary project and I do not have a reason to rush anybody.

Quote:
 Originally Posted by VBCurtis I suppose a little test-sieving might be in order, but I think I have a good idea of parameters. I'll post them here this weekend once I have some time to consult my notes, and likely Charybdis will comment on the ones that might have better options available.
Thanks for test sieving!

Quote:
 Originally Posted by VBCurtis I'll still test Plutie's system tonight, as he wishes to know if he can host a future job.
I did not want to hinder you doing this. Sorry if I "came the wrong way" (do you say that?).

Quote:
 Originally Posted by VBCurtis Speaking of hosting- Kruoli, do you have a static IP address? Or, how are you going to set up incoming connections from the internet? [...]
Unfortunately, I do not have a static IP address, but I have a domain with dynamically updated DNS records of my IP. Inside my home network, I usually use host names instead of IP addresses, so I assume CADO will be fine using a domain, but I'm not sure how it would handle a change of IP while a work unit is processed. The IP gets renewed daily, the DNS record is updated immediately afterward.

I already have some services accessible through my router via port forwarding, so I assume this should not be a problem for us.

 Similar Threads Thread Thread Starter Forum Replies Last Post fivemack Factoring 3 2017-09-19 08:52 skan YAFU 6 2013-02-26 13:57 akruppa Factoring 114 2012-08-20 14:01 fortega Data 2 2005-06-16 22:48 marc Factoring 6 2004-10-09 14:17

All times are UTC. The time now is 04:14.

Thu Dec 2 04:14:08 UTC 2021 up 131 days, 22:43, 0 users, load averages: 1.56, 1.23, 1.20