Rc5-72

[rsbriggs]

I'm fixing to start cranking it up a little in RC5-72....
Just so you know....

[black_civic55]

thats great, the more the merrier.

By the way.... is there a tweak or anything to make this a little faster?? My 2.2GHz P4 is only doin about 1.5 million keys/second

[rsbriggs]

That seems a bit slow - this 2.4 Ghz is doing 4.2 - 4.4 mil / sec while I'm on surfing around... That's the best you get when you benchmark everything?

Here's my current core Bench Mark, for a vanilla 512 MB/2.4 Ghz P4 Dell box:

[Mar 31 20:04:55 UTC] RC5-72: Benchmark for core #7 (SGP 3-pipe)
0.00:00:16.51 [4,670,172 keys/sec]

[black_civic55]

my best when benchmarking is

[Mar 31 20:49:56 UTC] RC5-72: Benchmark for core #1 (SES 2-pipe)
0.00:00:16.39 [2,950,446 keys/sec]

[n7vxj]

Damn thats slow!! You Guys really don't wanna know what my XP1700 @2.3 gig is doing!!

[jasong]

1.75GHz Sempron(2500+):
about 6.1 million/sec when not taxed(There's a ChessBrain game going on right now, so...)

[jasong]

Guys, I forgot to join the team. :o That's been remedied.

Anyway...Since someone said they wanted to reach top 100 before moving on:

Team Free-DC has 1,853,682 blocks and turned in 6,163 blocks yesterday.

Team-Total blocks, blocks yesterday, days to pass at yesterday's rates

University of Florida 2,064,708 3,835 90
The Hardware Group 2,127,575 2,073 66
Team Beyond Reality 2,126,761 3,889 120

My contribution wasn't included in the calculations(see above), but I figure I can shave 1-2 days off the total time.

[rsbriggs]

I'll be adding 1,000 blocks/day or so by the first of next week - just as soon as I can transition some comps out of Fad...

These darn qm3 blocks in FaD just won't finish :bang: The one running on this comp is at 86% with close to 300,000 seconds used so far (CPU rating 149)....

[black_civic55]

well since the other thread is 3 pages long we'll continue with this one. Not sure if anyone has stated or asked yet but whats after this?? I happened to notice in EON we only have 3 people working. Not a huge problem yet but will be soon.

[em99010pepe]

Originally posted by black_civic55
well since the other thread is 3 pages long we'll continue with this one. Not sure if anyone has stated or asked yet but whats after this?? I happened to notice in EON we only have 3 people working. Not a huge problem yet but will be soon.

Do you guys want to try OGR?

Carlos

[black_civic55]

id be down for it, i dont really care! lets just get us in the top 100 first for this project! (wishing i had a NOS button)

[jasong]

I just want to go where I can make a difference and be heard above the noise. If you read the post in this forum entitled "idea for next gauntlet," you'll see where I WANT us to go, for 2-4 weeks or until a Gauntleteer finds a Riesel prime, whichever comes first. Or I could also go for EON, although I think we should warn them first in case they decide to optimize anything to take advantage of our not-yet-existing EON Gauntlet.

One more thing, there's a project in the forum called ChessBrain. Basically, it makes your computer into a networked Chess computer. It only uses(and maxes out) cpu cycles when a game is playing AND it's ChessBrain's turn. My experience has been about a half hour's worth of cpu time each day. I don't know how this time amount changes for other computers, but if there's enough interest I can ask. I'm just thinking it's a fun project; even if you don't know anything about chess, like me. It doesn't use up a lot of resources, most of the time it sits idle. One thing that confuses me about the client is that sometimes the client will minimize to the system tray(via a command by a sister program that ONLY sits in the system tray), but sometimes it stubbornly sticks around. I have a theory that if a game is in session, the "minimize to system tray" option temporarily breaks. I dunno.:confused:

[CaptainMooseInc]

RieselSieve is too painful to get a bunch of people running it unless the project makes a more user friendly client such as SoB has.

Also the "sieving" part of RieselSieve I hear is bad right now because the program becomes inefficient around 150T. Most people have been asked to switch to the PRPing aspect instead of Sieving until a fix can be found.

-Jeff

[black_civic55]

yea i agree i tried RieselSieve and i was not a fan. plus also over heated my comp. Buutttttt if others wanted to i would join along.

never liked chessbrain. rather use spare time for something that is needed.

[jasong]

Originally posted by black_civic55
never liked chessbrain. rather use spare time for something that is needed. Can I assume that you're running rc5-72 at the moment and believe IT'S useful?

Not that I'm telling you what to do. We wouldn't be human if we didn't participate in seemingly senseless pursuits, generally referred to as "hobbies." Myself, I'll never understand stamp collecting.

Crunch on.

[black_civic55]

The "Bovine" RC5 effort was formed to take the responsibilities of coordinating and maintaining the RC5 servers that are needed to distribute key blocks to work on to all of the participating client programs.

Yea its not curing cancer and such but its more important then a chess game.

[black_civic55]

were now 110 guys!! keep it up!

[b2riesel]

RieselSieve is too painful to get a bunch of people running it unless the project makes a more user friendly client such as SoB has.

We now have a installer for the client. We have registration PHP pages. The LLRNET client now talks to a MySQL server for work. We have proxy servers that you can run as a team server. We have many new options available over what we had a few weeks ago. We also are open to any suggestion that people have.

Our last prime was ranked #23 and all our new primes will be even bigger. We will be hitting top 20 primes next month. With 79 primes left to find..and them getting bigger every day we check....we will start finding some of the largest primes ever found by years end.

While many argue whether a project has scientific merit or not...every large prime we find....every effort we put into tweaking our clients...every new angle we attack the problem from...is a furtherment of number theory and our understanding of the nature and behavior of the constructs we use to create gigantic primes.

Lee Stephens
B2
www.rieselsieve.com

[rsbriggs]

Originally posted by black_civic55
were now 110 guys!! keep it up!

Hang in there. I'll be back over on RC5 big time once we make a little more progress in SOB. I have a goal I want to hit in SOB (40T), then I'll put the whole pharm on RC5 until we're at least in the top 100....

[black_civic55]

were almost 109, probly 1 or 2 more days. Hurry up Briggs! kidding.. take your time

[rsbriggs]

I put an AMD box back on it... :)

[black_civic55]

we're now 109!! probably be 108 tomorrow!

[rsbriggs]

I've got a few boxen transitioning over - it may take a while to flush out the current stuff they are doing... Onward and upward towards top 100 !!!

[black_civic55]

yesterday 108! come on everyone, few more weeks till 100! no slowing down :cheers:

[em99010pepe]

Originally posted by black_civic55
yesterday 108! come on everyone, few more weeks till 100! no slowing down :cheers:

I still have one machine running RC5-72.

Carlos

[jasong]

If this stuff is already supplied somewhere, please tell me:

Going by the last stats update

Team Days to overtake

WinBiff Users 2
alt-hacker.org 4
Everything2.com 17(I'm thinking about checking out their site. Intriguing name)
Iron Cow 36
RC5 for Jesus 34(Crazy. I've never known Jesus to use a brute-force attack for ANYTHING, lol)

[black_civic55]

hmm i haven't seen those dates, but ill jump back on to help out a lil more, i was testing out some BOINC

[black_civic55]

105 now!

[em99010pepe]

102.

I'm still running RC5-72.

Carlos

[MrTRS]

I still have one machine on RC5-72. I turned the other 5 off untill it cools off
around here.

Ray

[em99010pepe]

Soon I will be :stomp: by ECL.

Carlos