LLRnet/PRPnet rally January 3rd-10th
 

Hi everyone,

To kick off the new year, we are planning a 7-day rally starting Monday, January 3 at 7:00 PM GMT and finish 168 hours later on Monday, January 10 at 7:00 PM GMT. For those of you who can't convert GMT to local time :razz:, that's 2:00 PM EST and 1:00 PM CST.

Since we finished the 5th Drive not long after last rally (thus completing our year-end goal quite nicely ahead of schedule), this one will be on our next priority, the 6th Drive. Our goal on this drive is to complete it (at n=1M) before the end of Q2 2011. (At this point, that goal is starting to look a little easier than we thought, so we may up it to Q1 depending on how things look after this rally.)

Though we're crunching a different range this time, the servers are the same: LLRnet port 3000 and PRPnet port 9000. (To make a long story short, when the 5th Drive finished, we moved the work from our former 6th Drive server on another port into port 3000 so that people wouldn't have to move their machines. As for port 9000, it's our only PRPnet port at the time and as such is always loaded with our current priority.) Both servers should be processing work in the vicinity of n=870K-900K during the rally, making for medium-sized top-5000 primes that will stay on the list for a good while.

LLRnet config info.:
server = "www.noprimeleftbehind.net"
port = 3000

PRPnet config info.:
server = "www.noprimeleftbehind.net"
port = 9000
Or, in terms of a prpclient.ini config line:
server=G9000:100:1:noprimeleftbehind.net:9000

Be sure and change your userID to whatever ID you prefer. Most people use their ID here at mersenneforum. It is also best to set the cache to something in the vicinity of 2 to 5 pairs.

Info. for downloading the newest LLRnet client can be found at [URL="http://www.mersenneforum.org/showthread.php?t=13165"]LLRnet supports LLR V3.8! (LLRnet2010 V0.73L)[/URL].

Info. for downloading the newest PRPnet client can be found at [URL="http://www.mersenneforum.org/showthread.php?t=12223"]PRPnet[/URL].

As a point of interest for those who might be new to NPLB and for others who haven't done any testing with us for a while, we have some milestones that we would like to shoot for in the near future. Please refer to our [URL="http://www.noprimeleftbehind.net/stats/index.php?content=drive_progress"]primes graph and drive progress[/URL] as well as to our [URL="http://www.mersenneforum.org/showthread.php?t=13333"]"Countdown meter"[/URL]. NPLB's goal from the start has been to fill in holes in the prime search ranges for Riesel base 2. We'd like to get the dots in that [URL="http://www.rieselprime.de/Data/00300.htm"]k=300-1001[/URL] range fully extended to n=1M by year end 2011 so that they are closer to where k<300 (worked by RPS) is at. Right now, we're focusing on the k=600-800 range; once that is up to n=1M, we'll repeat the process for k=800-1001.

Last rally, we had a surprise upset in the teams arena, with ROLP taking the championship title by a 20,000-test lead over PrimeSearchTeam (the victor in the previous two rallies). Most of the lead was due to one of our big contributors (Ian a.k.a. MyDogBuster) temporarily departing from his usual work at our sister project, Conjectures 'R Us, and making an appearance at the rally; whether he'll be here this time is known only to him, but PrimeSearchTeam may want to plan to bring some extra firepower to cover that contingency. :wink: (FYI: ROLP's output from last rally, including Ian, is just about the max that we can crank out. If you can top that, then you have a rally victory for sure, unless some other enterprising team can do even better. :smile:) Free-DC made a very good showing, primarily driven by Brian a.k.a. PCZ, almost giving PrimeSearchTeam a real run for their money. If they can scrounge together a few more heavy hitters this time, they stand a decent chance at taking the title.

In the individual stats, Brian and Lennart (a.k.a. sm5ymt) had a very close race, with Lennart winning by just 1110 tests. Since both of them were cranking out much more than that per day (and near the end, PCZ was outputting just about that much more than Lennart each day), it was sufficiently close that had the rally gone on for another day, we may well have seen a different result.

If this rally is as competitive as the last one, we should see a number of very interesting races. Whereas in some past rallies, the outcome could have been roughly predicted from the start, this time I have to say it is totally up in the air. Even one computer could make the key difference! :smile:

Hope to see you there--let's see if we can make some serious headway on the 6th Drive!

Max :smile:

Primes found during the rally:
[code]
prime who port 24 new (plus 1 confirmed)
651*2^893622-1 MiHost 3000
759*2^894625-1 PCZ 3000
759*2^902553-1 Beyond 3000
741*2^903981-1 gd_barnes 3000
737*2^906168-1 gd_barnes 3000 (confirmed)
707*2^907156-1 kar_bon 3000
719*2^915504-1 vaughan 3000
735*2^916709-1 vaughan 3000
611*2^916918-1 vaughan 3000
651*2^919214-1 Beyond 3000
629*2^920824-1 PCZ 3000
633*2^920967-1 Beyond 3000
669*2^926649-1 PCZ 3000
735*2^931109-1 Lennart 9000
687*2^931420-1 Steve 9000
625*2^933969-1 Sloth 9000
683*2^937166-1 Lennart 9000
767*2^937280-1 Steve 9000
721*2^946433-1 PCZ 3000
677*2^951134-1 gd_barnes 3000
793*2^951963-1 PCZ 3000
705*2^963099-1 AMDave 9000
665*2^963554-1 Bibi 9000
657*2^970025-1 PCZ 3000
693*2^971735-1 Beyond 3000
[/code]

stats page is up and ready :smile:

Max.

CST is GMT-6
[url]http://wwp.greenwichmeantime.com/time-zone/usa/illinois/chicago/time.htm[/url]
so the CST start and finish time is 13:00 hrs CST

EST is GMT-5
[url]http://wwp.greenwichmeantime.com/time-zone/usa/dc/washington-dc/time/[/url]
so the EST start and finish time is 14:00 hrs EST

Can you post the corrections please.

[QUOTE=AMDave;244378]Max.

CST is GMT-6
[url]http://wwp.greenwichmeantime.com/time-zone/usa/illinois/chicago/time.htm[/url]
so the CST start and finish time is 13:00 hrs CST

EST is GMT-5
[url]http://wwp.greenwichmeantime.com/time-zone/usa/dc/washington-dc/time/[/url]
so the EST start and finish time is 14:00 hrs EST

Can you post the corrections please.[/QUOTE]
Thanks for the heads-up--I'm not sure how I made that mistake, I usually do so well with those kinds of conversions...*bangs head*
(Edit: Ah, I see how I did it now. I changed the abbreviations, thinking that I had done the numbers themselves earlier, when in reality I had not.)

Also, thanks for getting the stats page up! :smile:

The rally starts in 45 minutes! :smile:

To all:

There're now many rejected pairs on port 3000!

Please, be sure to delete old reservations (for the LLRnet client delete the workfile.txt and the z*****-file) and update the server/port information in llr-clientconfig.txt as given in the first post.

In the download-zip for the WIN-version of LLRnet2010 there's a batch called '_new.bat': execute it and all old/temp file will be deleted.

Just now moved the brunt of my cores to the rally...I can't believe that I (of all people) missed the start. Bah. :rolleyes:

This time I have a total of 10 cores participating: a Q6600 quad and E4500 dualcore (as before), plus now an AMD Phenom X4 9750 that I now have access to. This should be my best rally yet...hopefully I will (finally!) find a prime on port 9000. :smile:

It's bitter/sweet for me. I have my Q6700 on @ 2.66GHz and my new Core i5 laptop running at about 4 x 1.8GHz equivalent. But there are 4 more fast cores here that people want to actually use for computing. The cheek of it!

[QUOTE=mdettweiler;244467]Just now moved the brunt of my cores to the rally...I can't believe that I (of all people) missed the start. Bah. :rolleyes:

This time I have a total of 10 cores participating: a Q6600 quad and E4500 dualcore (as before), plus now an AMD Phenom X4 9750 that I now have access to. This should be my best rally yet...hopefully I will (finally!) find a prime on port 9000. :smile:[/QUOTE]

There are no primes on port 9000, you need to switch to a different port, may I suggest 3500.... :whistle:

[QUOTE=Flatlander;244473]It's bitter/sweet for me. I have my Q6700 on @ 2.66GHz and my new Core i5 laptop running at about 4 x 1.8GHz equivalent. But there are 4 more fast cores here that people want to actually use for computing. The cheek of it![/QUOTE]
Is it a laptop or desktop? If a laptop, then I can understand why people wouldn't necessarily want the CPU running full tilt if they're planning to use it, but for a desktop there isn't really any perceptible difference. (Not that it's always easy to explain that to a non-techie...)
[QUOTE=Beyond;244474]There are no primes on port 9000, you need to switch to a different port, may I suggest 3500.... :whistle:[/QUOTE]
Ah, but primes come in bunches--9000 is just in between bunches right now. When they come, it will be a windfall (like the one that hit 3000 earlier today). :wink:

[QUOTE=mdettweiler;244477]Is it a laptop or desktop? If a laptop, then I can understand why people wouldn't necessarily want the CPU running full tilt if they're planning to use it, but for a desktop there isn't really any perceptible difference. (Not that it's always easy to explain that to a non-techie...)
[/QUOTE]
A dual core laptop and a dual core desktop. I might try running it on the desktop tomorrow, see how noisy it is.