mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Aliquot Sequences (https://www.mersenneforum.org/forumdisplay.php?f=90)
-   -   Unilateral sieve by fivemack: c160 from 4788:i2765 (https://www.mersenneforum.org/showthread.php?t=16323)

jrk 2011-12-09 15:12

Unilateral sieve by fivemack: c160 from 4788:i2765
 
line 2765 poly:

[code][color=blue]# sieve with ggnfs siever 14e on alg side from Q=20M to 85M[/color]
# aq4788:2765
n: 2814575633606932762559115069533933368677889149925801155928197648179619287546158555273702253059022542550239875837596670776891953100123181330796798182098206571983
# norm 1.894764e-15 alpha -7.376608 e 1.626e-12 rroots 5
skew: 11068740.95
c0: 1161978843716318456646952812120907588608
c1: 2689494311189461863540738853548620
c2: 287918016978138708257077050
c3: -37048723565902501500
c4: -2009829632263
c5: 148680
Y0: -7168562641279545574915385511745
Y1: 63337248332342941
rlim: 40000000
alim: 40000000
lpbr: 29
lpba: 30
mfbr: 58
mfba: 60
rlambda: 2.5
alambda: 2.5
[/code]

debrouxl 2011-12-10 13:56

It's 14e, so RSALS could participate in the team sieving, should ECM fail to split the composite. However, the number of WUs in progress has been cut by more than half over the past few days, even though the WUs are slightly easier today than they were several days ago...

frmky, fivemack, other takers ?

schickel 2011-12-10 14:42

[QUOTE=debrouxl;281765]It's 14e, so RSALS could participate in the team sieving, should ECM fail to split the composite. However, the number of WUs in progress has been cut by more than half over the past few days, even though the WUs are slightly easier today than they were several days ago...

frmky, fivemack, other takers ?[/QUOTE]The situation isn't really clear from your comment....does that mean you have fewer machines participating in sieving?

I've still got ~2 weeks or so to go on my current job, so that lets me out for the start of this one....

unconnected 2011-12-10 14:55

I can do the postproccessing if nobody else wants.

debrouxl 2011-12-10 16:39

Yes, seems like fewer machines are participating in sieving. It's not the first time the power of the grid experiences variations such as the one we're currently noticing, over the course of several days: 9K pending WUs on Tuesday, and 3K-4K today.

schickel 2011-12-10 16:47

[QUOTE=debrouxl;281781]Yes, seems like fewer machines are participating in sieving. It's not the first time the power of the grid experiences variations such as the one we're currently noticing, over the course of several days: 9K pending WUs on Tuesday, and 3K-4K today.[/QUOTE]Maybe variations in the number of machines at schools participating?

Anyway, anything you can help us out with is certainly appreciated....not quite there yet on the ECM, so not to worry.

Andi47 2011-12-10 16:50

[QUOTE=debrouxl;281781]Yes, seems like fewer machines are participating in sieving. It's not the first time the power of the grid experiences variations such as the one we're currently noticing, over the course of several days: 9K pending WUs on Tuesday, and 3K-4K today.[/QUOTE]

Hmmm... The 8[SUP]th[/SUP] of December is a [URL="http://en.wikipedia.org/wiki/December_8#Holidays_and_observances"]holiday[/URL] in quite a few countries. Maybe several people have made a journey and switched off their computers over the long weekend?

fivemack 2011-12-10 17:18

My monster is working on sieving 2340:693 for at least another week, and my smaller machine on 1920:2385; if you expect your current resources to factor the number by 19th December, go ahead. If not, tell me; I'll throw 48 cores at it starting 19th December and it'll be done by Christmas.

pinhodecarlos 2011-12-11 19:41

[QUOTE=Andi47;281783]Hmmm... The 8[SUP]th[/SUP] of December is a holiday in quite a few countries. Maybe several people have made a journey and switched off their computers over the long weekend?[/QUOTE]

That was not the problem. debrouxl, you need to check more often RSALS stats...lol What happened was this: ripli from SETIKAH@KOREA team moved a bunch of cores to RSALS and run it until he/she reached first place in overall stats. Meanwhile BOINCstats team reacted to SETIKAH@KOREA team first place approach but then ripli reduced his/her power by two thirds when BOINCstats team matched SETIKAH@KOREA points ouput. Now both are reducing power. You can check all stats progress in here:

[URL]http://stats.free-dc.org/stats.php?page=proj&proj=rsa[/URL]

Don't forget that having lots of downtime from server side is probably also an issue here.

Carlos

schickel 2011-12-12 00:09

[QUOTE=pinhodecarlos;281867]That was not the problem. debrouxl, you need to check more often RSALS stats...lol What happened was this: ripli from SETIKAH@KOREA team moved a bunch of cores to RSALS and run it until he/she reached first place in overall stats. Meanwhile BOINCstats team reacted to SETIKAH@KOREA team first place approach but then ripli reduced his/her power by two thirds when BOINCstats team matched SETIKAH@KOREA points ouput. Now both are reducing power. You can check all stats progress in here:

[URL]http://stats.free-dc.org/stats.php?page=proj&proj=rsa[/URL]

Don't forget that having lots of downtime from server side is probably also an issue here.

Carlos[/QUOTE]In other words, we need someone to move up in the rankings so as to draw the stats junkies into a pissing contest, huh?

fivemack 2011-12-12 09:25

Executive decision
 
Hello all.

I'll run this one, starting now - I've paused 2340:693 and set 48 cores to work starting at 20M. Should be done by the end of the week.


All times are UTC. The time now is 17:28.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.