PDA

View Full Version : Folding@Home eMac lab install




stoid
Jul 15, 2003, 10:26 AM
I am going to be installing an brand new mid-level eMac 24 computer typing/yearbook lab at a high school either later today or tomorrow (depending on when the shipment arrives, and I want to put F@H for the MacRumors team to run in the background. I recently started the app version at my house (user name: stoid), but have not been able to run it much on my G4 tower because it is in the guest room, and we have had relatives in town since I started. I'm guessing the console version would be the best (does it still show up in the dock, because if so, it won't work.) Any hints/tips/things-I-should-know-(i.e. computer won't sleep when running, won't run while computer is sleeping, has to be initiated each time at log-in, etc.) Thanks guys, and I look forward to giving the MacRumors team a boost.



Mr. Anderson
Jul 15, 2003, 10:29 AM
Excellent!!! I'd go with the console version and use the scripts from mc68k - pm him or Rower to get the info!

And thanks in advance for all the WUs!!!!
We need the boost. :D

D

britboy
Jul 15, 2003, 10:45 AM
Originally posted by Mr. Anderson
Excellent!!! I'd go with the console version and use the scripts from mc68k - pm him or Rower to get the info!

D


Actually, I've been meaning to ask about that script. I just recently installed a new hard-drive in my tibook (went from a 20 to a 40gig), and thought that it would be a good time to use mc68k's script. However, it was only folding at a rate of about 2% a day for the major WU's. I thought that was a bit slow, so changed to the graphical client, and that now runs at circa 13% a day for 70 point units. Was I doing something wrong, or is that normal?

MrMacMan
Jul 15, 2003, 11:43 AM
Originally posted by britboy
Actually, I've been meaning to ask about that script. I just recently installed a new hard-drive in my tibook (went from a 20 to a 40gig), and thought that it would be a good time to use mc68k's script. However, it was only folding at a rate of about 2% a day for the major WU's. I thought that was a bit slow, so changed to the graphical client, and that now runs at circa 13% a day for 70 point units. Was I doing something wrong, or is that normal?

Thats not normal.

bertinman
Jul 15, 2003, 12:45 PM
not sure, but when I tried the script for the first time it used the .24 core. .25 is out now... might be the big difference.

-bert

yzedf
Jul 15, 2003, 01:07 PM
With SETI I do the following; Command Line version... open a terminal and type the following:

$ ./setiathome > /dev/null &

./setiathome = run seti

> /dev/null = transfer output to /dev/null (nowhere land)

& = run in the background so that when you close terminal, it still runs.

Set up a script that runs this for you on startup, and you are set.

mc68k
Jul 15, 2003, 01:21 PM
Originally posted by bertinman
not sure, but when I tried the script for the first time it used the .24 core. .25 is out now... might be the big difference.

-bert what do u mean by .24 and .25? the scripts do not copy the cores, that's done from stanford dynamically.

crenz
Jul 15, 2003, 01:49 PM
Originally posted by stoid
I am going to be installing an brand new mid-level eMac 24 computer typing/yearbook lab at a high school either later today or tomorrow (depending on when the shipment arrives, and I want to put F@H for the MacRumors team to run in the background.

Whew. :eek: Make sure the school knows about it, or else you might be in trouble.

mc68k
Jul 15, 2003, 02:00 PM
Originally posted by crenz
Whew. :eek: Make sure the school knows about it, or else you might be in trouble. it's that kind thinking that keeps us from getting more points :(

our lab here put a new disclaimer on the library comps stating that the FBI could be monitoring them under the Patriot Act...it's all scare tactics

bertinman
Jul 15, 2003, 02:26 PM
Originally posted by mc68k
what do u mean by .24 and .25? the scripts do not copy the cores, that's done from stanford dynamically.

haven't done it recently. But I could have sworn there was an executable that needed to be downloaded... the script had a version on 3.24 or something like that and there was a 3.25 on the web site.

OH, I don't meen "core" as in gromacs of tinker... I mean the aplication itself.

I don't fold on my mac anymore... I've been in denile for a month but it is now clear that my wife has stolen it for her own use :P :(.

anyway, sorry for any confusion and if I don't know what I'm talking about.

- bert

crenz
Jul 15, 2003, 02:39 PM
Originally posted by mc68k
it's that kind thinking that keeps us from getting more points :(

Well, I'm definitely not out to spoil your fun! However, the folding craze is nothing new. It was the same thing with distributed.net a few years ago, until the craze somewhat cooled down. We had problems in our university's labs with people installing RC5 crackers everywhere, trying to get credit for cracking with machines that didn't belong to them.

Admittedly, folding makes much more sense than cracking RC5, but I'd still be wary to use machines that belong to somebody else without their permission.

mc68k
Jul 15, 2003, 03:02 PM
Originally posted by crenz
Well, I'm definitely not out to spoil your fun! However, the folding craze is nothing new. It was the same thing with distributed.net a few years ago, until the craze somewhat cooled down. We had problems in our university's labs with people installing RC5 crackers everywhere, trying to get credit for cracking with machines that didn't belong to them.that's the challenge :D

Admittedly, folding makes much more sense than cracking RC5, but I'd still be wary to use machines that belong to somebody else without their permission. in theory, yes. i think folding has a noble goal, but i don't think it really has an impact on disease research. distributed.net is much more mature in every aspect of the distributed arena then stanford is (or will ever be) IMO

i agree being wary is essential

pEZ
Jul 15, 2003, 06:34 PM
Definitely - I don't fold because I think the protein I'm folding right at this moment is the cure for alzheimers - I fold because the more we know about the human body, the better.

MrMacMan
Jul 15, 2003, 07:54 PM
Originally posted by pEZ
Definitely - I don't fold because I think the protein I'm folding right at this moment is the cure for alzheimers - I fold because the more we know about the human body, the better.

At the first part of the sentence I was about to write

'wtf pEZ, you fold' but then I read the whole thing. ;)

Keep those labs coming!

meeble
Jul 16, 2003, 04:53 AM
in theory, yes. i think folding has a noble goal, but i don't think it really has an impact on disease research.

really?

what is the point if this data can't be used for disease research?

MrMacMan
Jul 16, 2003, 07:25 AM
Originally posted by meeble
really?

what is the point if this data can't be used for disease research?

I'm sure it is used for research.

mc68k is ccrraaazzy sometimes.

--pwned.

bertinman
Jul 16, 2003, 07:06 PM
Originally posted by MrMacman
I'm sure it is used for research.

mc68k is ccrraaazzy sometimes.

--pwned.

they need something to print out so they have enough "scratch" paper. Millions of uses for scratch paper: doodling during class, lining the hamster cage, making paper airplanes and even for printing on the other side as to not waste the use of "new" paper on first drafts! :p

Stanford has a lot of students... those students need to be provided for--they need a steady supply of scratch paper, and they need it at all times!

mmmdreg
Jul 19, 2003, 10:47 PM
heh go install rc5-72. It's commandline and you can "install" it and make it run in "quiet" mode so that it starts up on boot and doesn't show itself. Of course, if people were looking at the processes, dnetc would show, but will anyone really care?

MrMacMan
Jul 19, 2003, 10:50 PM
Originally posted by mmmdreg
[B]etc/B]

I'm pretty sure he is installing folding on these because it is in the title of the thread, rc5-72 will soon probably reach its goal anyway.

mc68k
Jul 21, 2003, 02:21 AM
Originally posted by MrMacman
rc5-72 will soon probably reach its goal anyway. maybe by some freak stroke of luck--

time working 230 days

.033% keyspace searched

http://stats.distributed.net/projects.php?project_id=8

mmmdreg
Jul 21, 2003, 02:52 AM
Originally posted by mc68k
maybe by some freak stroke of luck--

time working 230 days

.033% keyspace searched

http://stats.distributed.net/projects.php?project_id=8

lol hence we need more macs on the case. Altivec helps G4's completely kill PC's at this. G5's maybe more so. With the percentage of Mac users on it nearly 7%, it obviously shows we're proportionally greater in number there than in real life. A few more G4/5's cracking for xlr8 and we may move into the top 5 very soon. Go us!. However, there *still* are random teams waay below us who are killing us on a day-to-day basis. So one day they may overtake. We have to watch ourselves. Day-by-day, we're generally in the top 5 although a handful above us are below us in the overall rankings. Obviously, there is little scientific benefit to crunching raw numbers like this, but if you're bored of SET and folding, why not give this a go, over all the other alternatives.