F@H problem

lee87

lee87

lee87

Permabanned
Joined
31 Jul 2005
Posts
2,592
Hi, i've been running seti for a while now and have decided to switch over to folding@home.

I've set it up on 2 of my computers but not sure if I have done it correctly. I think I have done it wrong as each system is doing the same project. Is it meant to be doing this or should they be different.

Thanks
 
It's fine if each of your computers is doing the same project. If you look in the FAHLog.txt file in your Folding@Home folder, the client will print information like this as it starts work on the unit:

Code: 
Folding@Home Gromacs Core
[08:22:14] Version 1.86 (August 28, 2005)
[08:22:14] 
[08:22:14] Preparing to commence simulation
[08:22:14] - Assembly optimizations manually forced on.
[08:22:14] - Not checking prior termination.
[08:22:26] - Expanded 3973119 -> 20311029 (decompressed 511.2 percent)
[b][08:22:27] Project: 1481 (Run 1, Clone 119, Gen 47)[/b]

That last line is the important one. Each project actually consists of thousands of different work units. The same project has to be run several times with different starting parameters, etc. (runs/clones) and then when the unit is completed the results are used to make a new work unit, which is the same project but has a different generation number.

Simple answer: As long as at least one of the four numbers given is different, the work units aren't the same. All four numbers have to match for two units to be identical.

Normally the only situation in which the servers may give the exact same unit twice, is if you try to download a new unit before you've finished the last one on the same PC.
 
What we usually refer to as the "project" number is really just one area of study, one protien in a particular situation. Under that main project there are loads of generations and runs. Each WU sent out is entirely unique. The server knows who downloaded it and is waiting patiently for it come back in. The first WUs sent out in a project must come back in before the second can go out becasue they are generational; each WU sent out is built on the work performed by the previous cruncher. It happens that most WUs under a particular "project" will behave similarly so we tend to refer to each WU of a project as being project xxxx.

For instance, I'm working on
Code: 
[20:50:17] Project: 2107 (Run 30, Clone 53, Gen 1)


If anybody asked, I'd say I had a p2107. However, in reality the WU is fully described by all four numbers, the project, run, clone, and generation. If you look at your FAHlog.txt files you'll find that each computer has a unique set of numbers.

EDIT: pwnt by mattus! :D
 
My main system is a Athlon X2 3800+ overclocked to 2.65ghz with 1GB Ram.
The other system is quite old, it's only a AthlonXP 3000+ with 1GB Ram.

I keep them on for about 15 hours a day, so it should get quite a bit work done, nothing compared to the amount some people do though.
 
Sounds hot. Be sure to parp your victims as you head up the charts becasue with a rig like that you'll be delivering mass stompage at a tremendous rate.

BTW, do you have a total of 3 clients running across those two machines? Is there a FAH core for each CPU in the X2? I'd hate to see CPU cycles going to waste...
 
I've got 2 clients for the X2 running, 3 in total. Went through the guide you made to find out how to set it up, excellent guide btw.
 
You must log in or register to reply here.
Back
Top Bottom