We're off and running even more!

[rsbriggs]

w00t - we just cracked the top 3,000 !! We're up over 27,000 places in eleven days, and have moved up to 408th place in factoring, with 5.277 CPU years, as well....

FreeDC Rank: 2997., 26 LL Tests completed, 7 factors found, P90 CPU: 21.091 LL + 5.277 fact. = 26.368 Years total

408. FreeDC 21.091 26 5.277 7 133.98

Hopefully we can nibble another .01 off the ARS Technica LWP on the next repository update, too ( evil grin :D )

[rsbriggs]

Well, we nibbled on the ARSians DC Repository GIMPS LWP today:

Their LWP on the project dropped from 99.75 to 99.68 and we climbed from 0.08 to 0.15 !

At this rate in another week we'll have.... Hmmmm.... Well, never mind...

Just know that we're "doing good"...

[magnav0x]

Let's knock ARS's down below 99% ! :cheers:

[willy1]

Still moving up!


FreeDC Rank: 1935., 43 LL Tests completed, 8 factors found, P90 CPU: 35.512 LL + 6.041 fact. = 41.553 Years total

[magnav0x]

Wow top 2,000 :D

[rsbriggs]

And moving them down. ARS Technica LWP: 99.61 - we're nibbling on them....

And, to everyone - a job well done. It is absolutely amazing that a team that doesn't even turn two weeks old until tomorrow has cracked the top 2000 in GIMPS. Some people crunch for years and never break into the top 10,000....

:cheers: :notworthy :elephant:

[magnav0x]

I'm very proud of all that have came to support the independent Free-DC GIMPS effort. Every time I see us climb in the ranks I think of this originaly posted in the 'Um?' thread:


A Free-DC Gimps team starting out now would rank low in the Repository, and barely register at all on the overall Gimps rankings. We don't have the firepower available to make that much headway on our own

Look at us now, ranked 1911, after two weeks :smoking:

[ECL]

It's a little early to be getting cocky, isn't it?

I just checked the Primenet server:

Rank Account ID LL P90* Exponents Fact.P90 Exponents P90 CPU
CPU yrs LL Tested CPU yrs* w/ Factor hrs/day
------ -------------- ------- --------- -------- --------- --------
1. Team_Prime_Rib 14474.9 18786 1576.22 3283 61941.78
2. curtisc 13885.6 14337 10.777 184 62108.25
3. TempleU-CAS 3646.39 10342 20.622 148 16818.35
4. SW 3471.09 9169 5.655 106 16174.36
...
1874. FreeDC 36.869 45 6.042 8 217.66

1874 is better than 30,000 or wherever we started, but our total production so far isn't even a blip to the real producers. The teams we've passed are basically the roadkill you find at the bottom of the list on all projects - the teams with a couple of machines, the guys who gave up years ago. Turning in that first LL P90 year jumped us two-thirds of the way up the list. The road ahead is all uphill and gets progressively steeper. We'll make significant progress in the next few weeks, but it'll start getting harder to make the kinds of gains we've seen in our two weeks in the project. Realistically, we'll soon stop passing 200 teams every time one of us turns in an exponent.

Those who are playing for points in the DC Repository will observe that we currently have 0.22 points on Gimps as opposed to Ars' 99.61 points. It will take some time to wear that down enough to make a difference.

We get points for suiting up and playing the game, but we're still not in Ars' league on this project, at least on LL. We're a more credible threat on factoring, largely because factoring is less popular project-wide (look at the #4 team's factoring score above). But Ars has still done 392.6 times the work we have on LL and 260 times on factoring. If you trust the P90 CPU Hours/Day metric, they've got 284.5 times our processing power. We have a lot of machines which have not yet returned their first work, so we'll start seeing our numbers rise quite a bit in the next couple of weeks. We'll climb past some more of the quitters and the one-man teams, but I hope nobody's holding too closely to the illusion that we're a major Gimps force. At our present rate, it'll take us about 800 weeks to reach where Ars is now. The current #100 team is about 22 weeks away.

Unless somebody like Beth133 joins us and brings over beaucoup P90 hours, this is going to be a long, slow fight. Let's see where we are in, say, six months. We'll have a better handle on our true production rate and things may be more clear.

[magnav0x]

No one here expects to get into the top 100 ;)

[rshepard]

No one here expects to get into the top 100

That's not entirely true-- I think it's a viable target, just not one we can reach quickly. How far we get depends entirely on how much long-term commitment we can get.

[rsbriggs]

The fact of the matter is, we should be able to reach the top 100 in a matter of months, (under a year, anyway) given a long term commitment by the people crunching for the team right now.

One thing deceptive about the P90 CPU hours / day number is that the team is only two weeks old (14 days old today!), but when FoBot merged his stuff in, the team had 21 months added to its age. So, our hours/day production isn't being divided by 14 days to calculate the number, it's being divided by about 650 days. Watch the number and see how it moves up from day to day, even with the extra 600+ days included in the division. There is more production going on here than meets the eye - we are producing at least 2 p90 CPU years per day.

It takes about 430 LL CPU years for a top 100 place currently. My best calculation is that we could reach the top 100 in under 6 months in LL (especially with NineBall producing the way he is), and a top 10 in Factoring isn't impossible (even if it doesn't count in the official stats or the repository).

That said, I think that people realize that it is an uphill battle from 1,000th place to 100th place, and no one here is expecting to bite off big 1% chunks of ARS's LWP - they've had hundreds of people and thousands of machines crunching for years to get where they are.

In the mean time, we're having great fun nibbling away at their LWP by a few hundredths each day, and enjoying the rocket ride past what you call the "roadkill" along the way. At 1874 this morning, with a couple LL's due in each day for the next week... w00t !! :cheers: :|party|:

[Xyzzy]

Originally posted by rsbriggs
That said, I think that people realize that it is an uphill battle from 1,000th place to 100th place, and no one here is expecting to bite off big 1% chunks of ARS's LWP - they've had hundreds of people and thousands of machines crunching for years to get where they are.More like a few hundred machines running about 2.5 years... :)

[rsbriggs]

Ahh - well that makes it even less scary. :)

( But I still look with awe at the pictures of some of the Prime-Monsters people run..... )

And hasn't team Prime rib been crunching primes since 1997 or 1998 (or something like that) ?

[Xyzzy]

We started in May 2001 but several people have merged in results from before that time...

[rsbriggs]

Kind of what happened to our team. We started on Novermber 3, 2003, but after a couple people merged in old results, our start date shows as February, 1999.

[robreid]

:cheers: Nice going guys, I'm sure it will work the same for you as it did when we started the ars gimps team, once you start rising you get noticed and more lone gimpsters start joining you.

Anyway, just swung by to say that I had noticed that the The Mersenne Database has been updated to move all of Angus's results from TPR to FreeDC.
This is unlikely to show up on primenet but will be reflected on the Gimps Top Producers (http://www.mersenne.org/top.htm).

As a result I have removed Angus from the TPR stats but I did a dump of all his data (http://www.teamprimerib.com/rr1/dev/Angus.csv) I had so its available when you guys get your own stats up and running.

[outlnderFDC]

Two years ago when I started running Gimps, no one here wanted a team. So, I joined TPR. Now you form a team.:bang: :bang:

[willy1]

Well, come on over - the water's fine.

:hifi:

[outlnderFDC]

I'm not sure that would be appreciated by either team.

If I moved, Free-DC would move into the Top 100 in LL years and the Top 25 in Factoring years.

It would also knock TPR to the #2 position overall.

[willy1]

Well, I can certainly see why Ars would not want that...


That's a LOT of crunching!! :shocked:

[rsbriggs]

I'm not sure that would be appreciated by either team.

If I moved, Free-DC would move into the Top 100 in LL years and the Top 25 in Factoring years.

It would also knock TPR to the #2 position overall.

And the downside to any of that is what exactly????? ( evil grin :D )

[magnav0x]

I think I could live with that......

[rsbriggs]

yeah. I'd only feel bad for about 10 seconds.... Maybe just 5... :Pokes:

[FoBoT]

Originally posted by outlnderFDC
Two years ago when I started running Gimps, no one here wanted a team. So, I joined TPR. Now you form a team.:bang: :bang:

yeah, i know, check the "um?" thread, it was over a year old, then these dudes decided to resurrect it ;)

:drums:

[magnav0x]

yeah. I'd only feel bad for about 10 seconds.... Maybe just 5...

:rotfl: :rotfl: :rotfl: :rotfl: :rotfl:

....ooooh that's funny :D .......


:rotfl: :rotfl: :rotfl: :rotfl: :rotfl: