gack! hide the women and children!

[FoBoT]

we are about to get trampled by 3 teams! :mad:

now that things have stabalized at df i will turn all my online p3's to SoB. after i get my other 15 AMD reinforcements on DF in a week or two, i will put the celeryrons onto SoB as well :D

phear the celery!

http://www.happiestguys.com/images/CeleryMan010550shrunk2.jpg

[Beyond]

:scared:

[rsbriggs]

I've got a little bit going on it. P4 boxen seem to really like that client...

[Supp]

Yup, turned my P4-2.0 over to SoB when I found out that in comparsion to AMD XP2000+ it really sucks in DF... :gone:

[wirthi]

I tried to run this project some weeks ago, but the two of my clients deleted their cache after a days work each. See http://free-dc.org/forum/showthread.php?s=&threadid=3515. Auto-Deletion of the current work is not very good for this project :)

[rsbriggs]

Yes, there can be some problems with the client. You really need to install and start the client, then reboot and restart it. And I'll be crunching along for days then after doing like block 400 of 512, the client will just decide to get a new block when it connects. (Can't link to it, but look at my "lost my block for no reason" threads.)

And the Linux client just throws a SIG_SEGV on all my boxes, so I can only run on Windows Boxen....

But the stats and charts are great, and I still like to run that project.....

[rsbriggs]

I tried to run this project some weeks ago, but the two of my clients deleted their cache after a days work each.But the good news is, you still get credit for the work you DID crunch via the intermediate stats (if you have "send intermediate results" turned on, and you really should for this client, so as to not lose a week or mores work if something goes wrong.). And after 10 days the numbers that were abandonded just go back in the pool of things to be crunched, so there isn't any real loss to the project, either.

[FoBoT]

Originally posted by wirthi
I tried to run this project some weeks ago, but the two of my clients deleted their cache after a days work each. See http://free-dc.org/forum/showthread.php?s=&threadid=3515. Auto-Deletion of the current work is not very good for this project :)

the times i have lost my work were only:

1- i tried to use the "on completion, exit w/o new test" feature, twice when i turned this option on, in the middle of a block, it erased my block and got me a new one. once when useing this feature, it restarted me for no apparent reason

2- trying to run it on off line boxen. i only run SoB on connected, always on PC's now

YMMV - good luck!

other notes - as rsbriggs says, when i tried the linux client, many months ago, i had the same issue. i am not linux guru enough to deal with that

and i only run the normal client, the one that puts the little "17" in the systray, i haven't messed with the service junk

[FoBoT]

Originally posted by rsbriggs
But the good news is, you still get credit for the work you DID crunch via the intermediate stats (if you have "send intermediate results" turned on, and you really should for this client, so as to not lose a week or mores work if something goes wrong.). And after 10 days the numbers that were abandonded just go back in the pool of things to be crunched, so there isn't any real loss to the project, either.

actually, i am quite sure on this, but someone can correct me if they have changed this recently

with intermediate blocks turned on, you will get credit for those on an intermediate basis only. if that block is never finished, you actually have that intermediate credit pulled back off the stats. you only actually get permanent credit for a block once the final block info is returned and credited, sorry :(

[PCZ]

we are about to get trampled by 3 teams!

Can't have that now, can we

How do you set this this project up ?

[rsbriggs]

Ummm - I believe the Linux client is broken....

I don't think we are in much danger. But then you never know about those Dutch Power Cows...
http://www.dutchpowercows.org/

[PCZ]

OK I will just run it on a couple of P4's

[wirthi]

Hi,

yes, you get credit for partitially calculated workunits (for every "block").

But, honestly, I'm trying to actually "do" something with my participation. Just to improve in stats is great, but that's not why I run DC projects and spend a lot of time on money on it ... I want to help science, at least a bit.

[rsbriggs]

I'd consider mathematics as "legitimate science", and going down in the history books isn't bad, either !! At this point, I believe the next prime the project encounters on the way to attempting to prove the Serpinski conjecture will be one of the largest prime numbers every found, if not THE largest found to date.

This project is actually attempting to accomplish more than the GIMPS "lets find a really big prime number" project.

I want to start on the Riemann hypothesis next :D :crazy:

http://match.stanford.edu/rh/

WARNING: Don't go to the link unless you think you can follow some of the little sub-notes presented on the page. Like the following, for example:
Abstract: We reduce the Riemann hypothesis for L-functions on a global field k to the validity (not rigorously justified) of a trace formula for the action of the idele class group on the noncommutative space quotient of the adeles of k by the multiplicative group of k. :whistle: :dunno:

[FoBoT]

Originally posted by rsbriggs
Ummm - I believe the Linux client is broken....

I don't think we are in much danger.

linux client - some people can get it to work and some can't, AFAIK, NEEDED: people who can make the current linux client crash (http://www.free-dc.org/forum/showthread.php?s=&threadid=3224)

danger-
i guess i should have supplied a link, according to these Ars provided stats (http://marcc.no-ip.org/sob/Team.php?Team_ID=60) it says


When do they get you?
Team_Ninja_-_The_Counts_of_Crunch 27 days
Dutch_Power_Cows 19 days
:dunno:

[PCZ]

Lets see what the stats say 24 hrs from now.

[FoBoT]

the 24 hour stats graph shows

good job everyone, we are already back into the top ten on daily rate, #9 for the day. we should creep up a bit more tomorrow, since a lot of that production was started yesterday

:drums:

[wirthi]

Originally posted by rsbriggs
I'd consider mathematics as "legitimate science"
Guess you misunderstood me on that. I also think that this is legitimate science, but nobody benefits from calculating half of the workunit and then restarting it. You only do science if you complete a packet - an I was not able to do so.

In fact I think it's one of the most useful projects. I don't know what the benefit of solving the Sierpinski problem is, but at least this project WILL solve it. IF (seti@home, FAH, DF, FAD, DPAD, ...) are successfull (like finding the aliens, learning how proteins fold exactly, finding a cure for cancer, ...) the benefit will be much bigger, but nobody (not even the project coordinators) do know when and if a solution is found at all.