mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Aliquot Sequences (https://www.mersenneforum.org/forumdisplay.php?f=90)
-   -   Reserved for MF - Sequence 3366 (https://www.mersenneforum.org/showthread.php?t=18449)

EdH 2020-05-13 15:49

[QUOTE=charybdis;545252]Could do, but I don't want to leave you without any jobs to run!

4000 curves at 8e7 done, now running 1500 at 15e7. Memory use in stage 2 is peaking just below 1.2G per process.[/QUOTE]
No problem at all. I'm playing in lots of areas. I think your data would be better for Curtis' work. (Notice how I sneaked out from under some responsibility?)

Thanks for the memory note. This may drive which machines I add to my ECM setup. My script(s) simply count up until I stop them, so let me know when I should stop. I don't even know how many curves I can do per how much time.

VBCurtis 2020-05-13 15:56

Thanks for all the ECM help! Nice to have friends with firepower. I'll not schedule those 1000@15e7 myself. :)

I choose non-standard GMP-ECM bounds after experimentation showed that, for example, a t50 is just as fast (or faster) using B1=6e7 as it is using B1=43e6, but the 6e7 curves have a higher chance to find larger factors.

Same for 15e7 vs 11e7 for t55 curves.

charybdis 2020-05-13 16:50

1500@15e7 done, starting GNFS now.

EdH 2020-05-13 17:43

I've stopped my local ECM (344 @ 15e7), but I'm leaving Colab running as a timed experiment.

charybdis 2020-05-17 11:46

c170 is done:
[code]p80 factor: 22518523926942820703763817501041320587763580874359805852200157171767426294316763
p91 factor: 3847115059798621443407081712134100568163094800824460322922394663279671623565737306421686491[/code]

The c189 at index 2212 has had 4000 curves at 15e6 which is about a t45.

charybdis 2020-05-17 22:24

If no-one else is running ECM I'll throw some more curves at the c189, but everyone's welcome to help of course!

VBCurtis 2020-05-18 01:12

I'll have a t50 done overnight, specifically 3000@6e7 and 2000@8e7; I've got 50 threads on it now.
If you run curves, head for 11e7 or 15e7?

charybdis 2020-05-18 01:51

OK I'll run curves at 15e7 overnight and see where that gets us.

charybdis 2020-05-18 12:29

Found a p54 factor. Next line has a c143 which will soon be ready for GNFS; I'll do it if no-one else wants it.

VBCurtis 2020-05-18 14:15

You found it, you do it! That's just a few hours for you, anyway.

charybdis 2020-05-18 17:38

The server and clients are having some connection problems so I'm unable to do the c143 at the moment. If anyone else wants it then go ahead and take it.


All times are UTC. The time now is 09:55.

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