Welcome. You'll find that folding will stress your systems much more than LinX overnight. Don't be surprised if you start losing bidadv wus after two days of processing. You may want to downclock to 4.0 if this starts to happen. In fact, don't use bigadv until you've finished your first 10 smp wus with the passkey. This will definitely tell you how stable your system is.
Come to think of it, I don't remember the last time I shut down and blew the computer. (I am serious... and don't call me Shirley.)
---
rklapp's links for the uninitiated:
---
To start with, try jedi95's FAH Tracker V2 explained here:
http://forums.evga.com/tm.aspx?m=219556Use this to set up your cpu and gpu clients. I've decided to remove the individual client links that I used to have here because the FAH Tracker is easier to use. Keep checking for updates because jedi is the man!
---
Although pretty, don't open the Viewer. It slows down your clients and could cause your system to crash. If you want to play with pretty molecules, check out
http://fold.it from my alma mater, University of Washington. Go Huskies!
---
Most i7 with 8 cores use bigadv after 10 WU are completed. Use 7 cores if running a fermi or mulitple gpus. When I'm pissed off at the 670x wus, then I shut off the smp client and run the uni-core client. Please do not delete wus because it messes with the server and makes more crappy work for the rest of us.
---
* FaH passkey (needed for smp bonus points)
http://fah-web.stanford.edu/cgi-bin/getpasskey.py---
* HFM.net (use to monitor each of your clients and smp bonus points)
http://hfm-net.googlecode.com/files/HFM%20Release%200.5.1.198.msi---
Highlites from Pande's 2010 lecture and his quest for exaflops:
http://forums.evga.com/tm.aspx?high=&m=632909---
If you're wondering how points are assigned, read the FAQ:
http://folding.stanford.edu/English/FAQ-Points---
Problems uploading your wu to Stanford, check out the server status:
http://fah-web.stanford.edu/serverstat.html---
Acronyms:
PPD = Points Per Day (How you measure the client's performance. The uni-core client will net you about 300 PPD while the bigadv client is over 25k PPD.)
WU = Work Unit (The number that Stanford assigns each project.)
TPF = Time Per Frame (How long it takes to process 1%.)
GROMACS = The programming core that Stanford uses.
www.gromacs.orgrklapp = PPD/watt for each client. This is how I measure the folding efficiency of each computer device.
http://forums.evga.com/fb.ashx?m=531159