mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   NFS@Home (https://www.mersenneforum.org/forumdisplay.php?f=98)
-   -   2021 “small” 15e post processing reservations and results (https://www.mersenneforum.org/showthread.php?t=26367)

jyb 2021-03-19 17:47

[QUOTE=swellman;574123]Yes, I believe I’ve located the problem. In your original [url= https://www.mersenneforum.org/showpost.php?p=573028&postcount=104]post[/url], you had a line “lss:0”, but it should read “lss: 0”. Msieve is apparently ignoring this line in its current erroneous form and is sieving on the rational side.

I should have caught this typo but did not. My apologies.

Moving forward, I can

A. Fix the typo and see what happens.
B. Ignore the typo and extend Q further
C. Let this job finish and start another job called C172_434040_3112a with the correct syntax and sieve on the algebraic side albeit over a smaller range. You can download both data files, combine them and proceed with LA.

A is the fastest option, but there’s a chance Greg will have to intervene if something breaks.

Any suggestions? Greg?

-Sean[/QUOTE]

It's actually the ggnfs sievers that are ignoring the line, not msieve. The real longer-term solution of course is that the sievers shouldn't be so fragile that they require a space after the colon on every line of the .poly file. But they do.

So as far as immediate solutions for finishing the job go, I suspect that C would actually be faster than B. A is a bit of an unknown quantity, because I don't know exactly how NFS@Home will handle a change like that. My guess is it wouldn't have any effect, since it thinks that it's already handed out all the work. But it's up to you and Rich, of course.

richs 2021-03-19 19:02

[QUOTE=jyb;574136]It's actually the ggnfs sievers that are ignoring the line, not msieve. The real longer-term solution of course is that the sievers shouldn't be so fragile that they require a space after the colon on every line of the .poly file. But they do.

So as far as immediate solutions for finishing the job go, I suspect that C would actually be faster than B. A is a bit of an unknown quantity, because I don't know exactly how NFS@Home will handle a change like that. My guess is it wouldn't have any effect, since it thinks that it's already handed out all the work. But it's up to you and Rich, of course.[/QUOTE]

Sean, I would prefer to work with just one file, so I'll do some test sieving to see how many more Q we need on the rational side to get to 260M relations. And it was my fault for missing the space.

swellman 2021-03-19 21:31

Ok. I’ll add 20M to the range immediately, just keep the job moving.

frmky 2021-03-19 23:21

At this point I recommend to just keep going. You'll probably need to sieve to about 200M. Not a big problem, really. :smile:

richs 2021-03-20 13:55

I estimate we will have to go to 230M based on test sieving.

[B]Done.[/B]

chris2be8 2021-03-20 16:47

f46_151p1 is done:
[code]
p56 factor: 56716017332497737007943447942089613541504316740174903547
p194 factor: 44733100374449220218692374489860972569801642657534698241621213444520906020421560170403326414709087432631539868532341825200179500622109712189571637615656380557297750931043982055164242850413603133
[/code]

Log at [url]https://pastebin.com/ZmCu35Xs[/url]

Chris

swellman 2021-03-21 19:15

C184_151_55
 
[CODE]prp66 factor: 120601796245117744432926416623034946129337699526364837410877535259
prp119 factor: 13466088200721119318183155736098896116231311182612024480255570253028228916190646795791989742536269092187784439544106919[/CODE]

Log at [url]https://pastebin.com/yAA9MC3V[/url]

richs 2021-03-23 17:08

C172_434040_3112 in LA. ETC 28 March.

jyb 2021-03-24 00:18

Taking 5m3_367.

jyb 2021-03-25 19:48

5m3_367 factored
 
[code]
prp93 factor: 985816600114738579524735301598577943592370169052614234608050594023899023287471667084838782083
prp107 factor: 95690678285681363850612928673524921279571093383785293918713592311111122033116069694421977759302200277929799
[/code]

[url]https://pastebin.com/EzfytDMk[/url]

jyb 2021-03-26 04:28

Taking 5m3_407.


All times are UTC. The time now is 07:43.

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