mersenneforum.org Poly select and planning for 2,2210M
 User Name Remember Me? Password
 Register FAQ Search Today's Posts Mark Forums Read

2019-09-06, 03:34   #23
EdH

"Ed Hall"
Dec 2009

335010 Posts

Quote:
 Originally Posted by VBCurtis While I await more RAM, I've started a CADO server for polyselect for this number. Point your client to the usual address as used on the 2330L job, with port number changed to 44400. The job should take about 8M thread-seconds, so just 2-3 days. EDIT: reserving degree 6, admin 1.05M, admax 20M. For those curious: nq=6^6, incr 4620, P=8 million.
Code:
ERROR:root:Download failed, URL error: <urlopen error [Errno 111] Connection refused>
ERROR:root:Waiting 10.0 seconds before retrying

 2019-09-06, 07:10 #24 VBCurtis     "Curtis" Feb 2005 Riverside, CA 10000111001012 Posts Bummer! I copied the network settings from 2330L params file, so I figured it would work without issue. Maybe I did something dumb like mistype the port. If I have time Saturday I'll see if I can figure out the problem; won't have time to check before then. Then again, the job will finish late Saturday on its own. I'll be running a deg 5 search with CADO later (unless Vebis beats me to it), so we'll have another chance.
 2019-09-06, 11:58 #25 EdH     "Ed Hall" Dec 2009 Adirondack Mtns 335010 Posts Is the new port available from outside your local network?
 2019-09-06, 14:27 #26 VBCurtis     "Curtis" Feb 2005 Riverside, CA 103458 Posts There is no local network; it's assigned an IP address via DHCP from my university.
2019-09-07, 19:50   #27
VBCurtis

"Curtis"
Feb 2005
Riverside, CA

52×173 Posts

Quote:
 Originally Posted by EdH Code: ERROR:root:Download failed, URL error: ERROR:root:Waiting 10.0 seconds before retrying
I found the problem; I ran CADO from home, my connecton to the campus machine died, so CADO stopped responding.
Unfortunately, I learned Friday that my office is scheduled for destruction Tuesday, so I went in last night and unplugged all the silicon. My department is as surprised by this turn of events as I am, so I and my machines are homeless for a week or two until they find some office space for me.

On the bright side, RAM arrived last night.

 2019-09-13, 01:32 #28 swellman     Jun 2012 23×3×7×17 Posts Code: # norm 8.388914e-21 alpha -9.216995 e 7.639e-16 rroots 5 skew: 20979898852.98 c0: 27649131585183391812916066541481716806536590015577581424 c1: 27330385748632515681889738973504668871302682226 c2: -2155465644405759903230110401342981943 c3: -213599886258400208384412088 c4: 4879065781035618 c5: 121716 Y0: -96818882190388663272841084874018768984485 Y1: 74413548803525738089
 2019-09-13, 02:14 #29 VBCurtis     "Curtis" Feb 2005 Riverside, CA 52×173 Posts That may be a record for highest skew on a usable poly in history. 20 billion! Recall that GGNFS does not run with skew above Q; if you change skew to 400M or 500M, I bet the score is pretty lousy. Then again, Max might be able to spin it such that skew drops by a factor of 4 or 8 while score remains similar, so there's some hope. Skews like this are why tiny c5s aren't usually searched for poly select above c200.
2019-09-14, 01:29   #30
swellman

Jun 2012

B2816 Posts

Quote:
 Originally Posted by VBCurtis That may be a record for highest skew on a usable poly in history. 20 billion! Recall that GGNFS does not run with skew above Q; if you change skew to 400M or 500M, I bet the score is pretty lousy. Then again, Max might be able to spin it such that skew drops by a factor of 4 or 8 while score remains similar, so there's some hope. Skews like this are why tiny c5s aren't usually searched for poly select above c200.
I can’t even get cownoise to spit out an answer for optimal skew! Perhaps Max can spin it down. But let’s face it, the best quintic e-score is likely to have at least a 9-handle.

The search goes on.

 2019-09-14, 03:51 #31 VBCurtis     "Curtis" Feb 2005 Riverside, CA 52×173 Posts I finally finished my deg 6 CADO run: Code: skew: 402972.111 c0: 2118696065289163977962187507742615436240832 c1: -45195693963724945835426077433010009156 c2: -298923418394166123716233260770564 c3: 328843002406215196716601281 c4: 2641231812968768558129 c5: -1065578287083828 c6: -276312960 Y0: -8247746467927728998793444647326297 Y1: 69974668288047521820527 # MurphyE (Bf=1.718e+10,Bg=1.718e+10,area=1.288e+16) = 1.17e-08 Cownoise says skew 586407.73298 score 1.21942919e-15 Prior records for deg 6: 210 digits 1.49e-15, 212 digits 9.42e-16 This composite has first three digits 103, so the poly score should be a bit closer to 210 digits than 212. I'd like a 1.30 in deg 6 to call the search complete, or 1.10 in deg 5. I'm running CADO on 0-1.05M in deg 6 next; by then I'll be back to 2330L matrix-solving. Edit: this was ~7M thread-seconds of search. Last fiddled with by VBCurtis on 2019-09-14 at 03:52
2019-09-14, 15:30   #32
fivemack
(loop (#_fork))

Feb 2006
Cambridge, England

23·3·5·53 Posts

Quote:
 Originally Posted by VBCurtis That may be a record for highest skew on a usable poly in history. 20 billion! Recall that GGNFS does not run with skew above Q
I think that's just not true; the yields you get are definitely lower for large skew, but it will run.

(sieving the demonstrated polynomial from 400M to 400M+100k with 16e and 33-bit large primes gives a yield of 40 relations)

2019-09-14, 16:24   #33
swellman

Jun 2012

1011001010002 Posts

Quote:
 Originally Posted by fivemack I think that's just not true; the yields you get are definitely lower for large skew, but it will run. (sieving the demonstrated polynomial from 400M to 400M+100k with 16e and 33-bit large primes gives a yield of 40 relations)
Then we should start sieving as soon as possible!

 Similar Threads Thread Thread Starter Forum Replies Last Post VBCurtis Cunningham Tables 69 2020-05-08 06:39 VBCurtis Operation Kibibit 25 2020-01-07 01:57 amphoria YAFU 22 2016-09-17 09:47 VBCurtis Msieve 0 2016-04-11 21:33 jux YAFU 5 2016-01-02 01:01

All times are UTC. The time now is 14:43.

Tue Sep 29 14:43:45 UTC 2020 up 19 days, 11:54, 0 users, load averages: 1.91, 2.19, 1.92