Seventeen or Bust: Just a couple more handful of computers needed

[SinfulWeeper]

For top rate that is.

When TPR jumped to over 35M I got very discouraged. With the recent addition of IJump's great fleet we are now just shy of 30M, a decently OC'ed Barton 2500+ or good P4 will put our TeAm in the 30M range depending on time of day.

Then it occured to me. A non-OC'ed 2GHz P4 turns out about 280000-320000 cEMs (can be slower if other programs are running)
If we can get just a few more people to join in or have more members who have not produced in a while. Only about 20 good computers will put us back in top production. That is a rather insignificant number and should be easy to attain I would think.

Seventeen or Bust FAQ's can be found here if you are unfamiliar with the project and are interested. Also I am sure any member would be willing to help out if you have any questions.

If you had problems with the old client and did not like 'babysitting' it. The new version 1.1.0 solves all those problems. I have not babysat any of the computers in my fleet since installing. It simply works great
Version 1.1.0 can be downloaded here.

Feel free to post your questions/thoughts/rants

 

[winr]

I will try to add a 900 Duron tomorrow.

I have went from around 400 to around 650/700.
I will try to get up to 800/850.

 

[LANMAN]

I added my home AMD 1400+

--LANMAN

 

[NicColt]

does this project have a conclusion like in all keys searched and if so is there like a % complete number ?

 

[Allio]

I'm going to try to convince the techie at school to let me run the client on a few computers. So many P4-2GHzs going to waste

 

[Baldy18]

Originally posted by: NicColt
does this project have a conclusion like in all keys searched and if so is there like a % complete number ?

As far as I know the project does have a completion goal (when the 17 thingies are found). However, there is no % to speak of since no one really knows when another one will be found. At least that is the way I understand it.

 

[Confused]

Originally posted by: NicColt
does this project have a conclusion like in all keys searched and if so is there like a % complete number ?

As Baldy18 said, there is a goal, which is to find primes for all 17 remaining values below 78557, and solve the Sierpinski problem.

Due to the fact that we don't know when these values will be found (if we did, the there would be no point running the project!!) then it's only a matter of "when". Hence, there isn't a % value you can give, like RC5. In theory, this project could go on forever, but only time will tell how long it will last, really. But one thing is certain, the more computers are running it, the higher the chance that it will complete sooner


Garry

 

[SinfulWeeper]

Glad to see the additions peeps

does this project have a conclusion like in all keys searched and if so is there like a % complete number ?

Baldy18 and Confused pretty much nailed it on the spot.
So far 5 of the 17 primes were found. Which leaves 12 left to go. One of the primes was even found by a member on our team, sean2002 (SeanD).

How long it will take to find the other twelve primes, no one knows. They could all be found tommorrow, or it may take many years, know one will know until it's all said and done.
I hope we all answered your question

 

[IJump]

I am probably going to add a few more too.... it kills me to be so close to #1 in the daily rate......

Go TeAm!!

 

[SinfulWeeper]

Originally posted by: IJump
I am probably going to add a few more too.... it kills me to be so close to #1 in the daily rate......

Go TeAm!!

:Q
You have more available...

Have at 'em

:beer:

 

[LANMAN]

SPIKE !!

:Q



--LANMAN

 

[MereMortal]

They haven't fixed the linux version yet, have they? (I won't run it with the open sockets problem.) And can this project be run by a proxy? We just received our new "toy" and I'm trying to decide what project to benchmark with. I think I estimated at one point it should a number of McEMs/sec, but I may have remembered incorrectly. What's the going rate for Xeon 2.4 GHz these days?

 

[Ken g6]

Can I cache work units yet? And are Athlon XPs any good at this? If not, maybe some cycle trading could be set up?

 

[SinfulWeeper]

WU's as far as I can tell can not be cached yet , that is a big problem for those on dial-up :|
Athlon XP's run SoB very nicely though

MereMortal, according to Mathguy, the linux version should be out soon. I do not know if it will address your problem though
Shoot him a PM or email, he might be able to better fill you in than I . Also short of dnetmhz who has a 1.8GHz, I do not know of anybody else using Xeon's for the project. Being he just started, I want to ask him the rate on his machine.

 

[LANMAN]

I have dual Xeon's running tonight... I'll let you know what the results are tomorrow.
(Dual 2.4Ghz)

--LANMAN

 

[Allio]

Originally posted by: SinfulWeeper
WU's as far as I can tell can not be cached yet , that is a big problem for those on dial-up :|

I'm on dialup and having no problems... it can work from the cache until the end of the test, which is something like 150 blocks. Or more? I can't remember. I've done 80 or so without connecting with no problems.

Can't wait to see those Xeon results

 

[SinfulWeeper]

Originally posted by: Allio3

Originally posted by: SinfulWeeper
WU's as far as I can tell can not be cached yet , that is a big problem for those on dial-up :|

I'm on dialup and having no problems... it can work from the cache until the end of the test, which is something like 150 blocks. Or more? I can't remember. I've done 80 or so without connecting with no problems.

Can't wait to see those Xeon results

Mind telling me how you get them? I might be able to assimilate some dial-up computers if thats the case .

 

[IJump]

Originally posted by: SinfulWeeper

Originally posted by: Allio3

Originally posted by: SinfulWeeper
WU's as far as I can tell can not be cached yet , that is a big problem for those on dial-up :|

I'm on dialup and having no problems... it can work from the cache until the end of the test, which is something like 150 blocks. Or more? I can't remember. I've done 80 or so without connecting with no problems.

Can't wait to see those Xeon results

Mind telling me how you get them? I might be able to assimilate some dial-up computers if thats the case .

When work is downloaded, an entire test is downloaded (I have seen tests with as many as 225 blocks). For my P4 2.53, 225 blocks would still take about 2 or 3 days to complete, so you would only need to connect every couple of days. The only problem that I see, is that you don;t get a new test until the previous one is finished, so the computer is not processing after the test is finished until it connects again. It may be working for 3 dyas, then down for 1 if it doesn't get connected. That is still better than not working at all.

 

[Confused]

**Cough**

 

[LANMAN]

The dual Xeon stats don't look so good.

2.4Ghz was slower then my P4 1.7 laptop!!

196.8

Anyone else have stats on Xeon's??

I have another running on a 1.8Ghz Xeon.. I'll go check the rate on that beast after lunch...

--LANMAN

 

[amcdonald]

Lanman, I have a dual 1.7 xeon with SOB.
the 2 rates right now are ~230 & ~250
As with other pentiums, I've noticed that it takes a few hours for them to climb to their limit.
They start out around ~190 a few minutes after restarting the client. When I check back later they are ~240.

 

[MereMortal]

Hmmm. Well even at 200 kcEMs/sec/CPU, this beast looks like it should do close to 10 McEMs/sec. I'll keep that in mind. :Q

 

[LANMAN]

Originally posted by: amcdonald
Lanman, I have a dual 1.7 xeon with SOB.
the 2 rates right now are ~230 & ~250
As with other pentiums, I've noticed that it takes a few hours for them to climb to their limit.
They start out around ~190 a few minutes after restarting the client. When I check back later they are ~240.

I have some problems then. Double checking the system (it was just out of the box a few days ago) it's actually a dual 2.6 Xeon.
After running over 31 hours it's still sitting at 196,072.9 and 195,997.1 on the second CPU.

Again checking my laptop running a P4 1.7, it's sitting around 261,512.9

Sitting near double the clock speed, shouldn't I be near 500 on this 2.6 Xeon??

HELP!!!

--LANMAN

BTW: Fresh install of Win2k SP3, with bare minimum services running.

 

[amcdonald]

did you run the service handler portion of the client?

 

[Baldy18]

Are you running the client with the service handler. It has instructions with it for maximizing speed for cpu with heyperthreading which I would figure your XEONs have. Confused puts a link to it in his daily stats threads.