No Prime Left Behind Math Project

[vaughan]

Seconded.

As AMDave said above there is an AMDUsers team. Hopefully ototero can get some team stats; maybe a request to Carlos might help to get some link to the team stats data instead of having to screen-scrape?

[Brucifer]

http://nplb.rieselprime.org/stats/stats.php?a=reset

The user names have their team in front of their username. Looks at present that I'm the only amd user crunching. Not many crunching so there isn't much of a daily stats thing here, not much different than the way Ototero was/is doing the MoneyBee stats. I don't even know if there is any discussion regarding the dc-vault or not. Not really worried about it either for that matter. It's a small project in total number of people participating, but is a continuation of the overall prime hunting effort of a number of prime related projects.

[Ototero]

Ok, teams stats will be up tommorrow, individuals, a bit later.

[spikey_richie]

Cracked it. Not sure what I did wrong the first time, but I just re-tried it and it's working.

Having to manually set affinity though

[Brucifer]

@vaghan -- thank you for seconding the project.

@Ototero -- thank you for taking on the added hassle of another stats effort that isn't an easy one, I really appreciate it!!

@spikey_richie -- thank you for continuing to work at getting the project to run on your system!! Rather hard to be a team with only one!

@anyone else that is in the least interested. This is a good beginner's project to work at getting started on an old p4 system with linux. One of the gutted down small distro linux version works great. No graphics needed as it just slows the system down for nothing. Old amd athlon's like 22/2400's work too, just not as fast as the p4's will. Those q6600's really fly through the stuff. Best is non-overclocked stuff as reliability is preferred. On running the llrnet application, it's pretty much hands off until you find something. In looking at the stats on the project, it appears that by returning around 10,000 results there is a pretty good chance of finding a prime number. The stuff we are working on should go in the top 5,000 listing of known primes, sort of like the Guinness book for large primes. My first one fell in the middle and entered at 2,037. While not a record like what the GIMPS people are working at, it's still a kick to get one. I intend to make myself a t-shirt with mine listed on it, just for geeky bragging/show_off rights for a little family humor.


And while it isn't boinc, there's still cool things to play with in the DC world outside of the boinc envelope for a little something different to do when you are sitting around bored looking at that boinc screen.

edit: found my 2nd one late this afternoon 773*2^426630-1, 128432 digits ranked 2,017 at entry into the database, and has been verified. This is a kick!

[CarlosEduardo]

Hey Bruce...catch me on the rankings...

[Brucifer]

Hey, give me at least a week or so Carlos. How long you been working on the LLR side of these projects now? And how many quads? A nice list of primes you have indeed!!

Until this past Saturday, all my previous time on this prime number searching stuff (a couple years at least) has gone into some heavy sieving for PSP/RS. That doesn't find any primes as you know, just greases the skids for others to find them. Sort of the curse of the sievers. So to me this has been a fun excursion into the llr side of things. And though the primes on this specific project aren't the really huge ones, there is a good chance here for people not set up with heavy assets in the Intel chip hardware, or having lots of corporate/institutional assets behind them to join in and find a prime.

[spikey_richie]

When is this battle due to finish? My MM stats are taking a hit...

[Brucifer]

The "battle" that Carlos was originally commenting on was this weekend past. Sort of expired like noon thirty Sunday east cost time or something like that.

[CarlosEduardo]

Hey, give me at least a week or so Carlos. How long you been working on the LLR side of these projects now? And how many quads? A nice list of primes you have indeed!!

I've been running for almost an year, I mean, dedicated crunching but I really had a push since October last year when I bought my quad-core.

Until this past Saturday, all my previous time on this prime number searching stuff (a couple years at least) has gone into some heavy sieving for PSP/RS. That doesn't find any primes as you know, just greases the skids for others to find them. Sort of the curse of the sievers. So to me this has been a fun excursion into the llr side of things. And though the primes on this specific project aren't the really huge ones, there is a good chance here for people not set up with heavy assets in the Intel chip hardware, or having lots of corporate/institutional assets behind them to join in and find a prime.

I know about your sieve past...don't forget that you invited me to help RS (sieve) a few years ago (2004).