* If you\'re running Windows, run it as a service (console client) so no pesky console or crummy screensaver (unless you like that sort of thing -- Seti@home has a far more captivating display but remember folks, \"People who don\'t fold eat puppies\"). Run \"FAH504-Console -configonly -local\" to configure and follow instructions (especially, say \"yes\" when asked to install service).
* You can set your username when you configure the client. Just make sure it isn\'t taken so you don\'t contribute towards someone else\'s :)
* Our team number is 49268
* Say \"yes\" to \"Allow receipt of work assignments and return of work results greater than 5MB in size (such work units may have large memory demands) (no/yes)?\". The peeps say you can seriously bring down your machine with this option but I\'ve never seen my cores use more than a couple of megs of RAM. If you have spates of RAM you should be fine.
* Under advanced options, you can similarly say \"yes\" to \"Set -advmethods flag always, requesting new advanced scientific cores and/or work units if available (no/yes)?\", although the same caveat as above holds. Supposedly, if the client ever crashes, it will unset the \"-advmethods\" flag, potentially robbing you AMD users of awsome SSE madness.
* Again, if you\'re running Windows and you used the configonly flag, the service probably isn\'t running yet. If you used the -config flag instead, the client is running in console mode but not as a service. In either case, you can manually start the service by launching services.msc from the Run... menu option under Start. If you\'re running a *nix box you\'ll have to use some other sort of magic.
* If you\'re running more than one client at a time, you\'ll need the -local flag, which tells the client to use local configuration options, else the clients get confused. You\'ll also need to assign each client a different machine id for the same reason. If you have Hyperthreading but only one core, you\'ll get more work units done with HT enabled, but according to the folks at F@H your work units take longer overall so that in the end scientific progress is delayed. This is not exactly something I agree with but that\'s my point of view. If you have two cores/CPUs, go ahead and run two clients. If you have a farm of quad opterons in your basement then go mad and post some pictures here ;]
* System load: The clients are designed to give higher priority to other tasks, so you shouldn\'t notice a hit when using other programs (why do you need other programs anyway? Folding comes first! Remember, \"People who don\'t fold eat puppies\", heh heh)
* Monitoring results: If you want to keep tabs on your folding you can either watch the logs being created by the client, or use something like FahMon: http://fahmon.silent-blade.org/ (other altneratives exist of course). I have another machine\'s work folder mapped as a network drive so I can monitor it semi-remotely. Also good for headless setups.
Too bad they\'re not using BOINC. I\'m already running SETI@home and Einstein@home, and I sure wanted to add Folding@home just to join the team. But I don\'t want to install another number crunching app/service to do it ...
Yeah I know, you\'ll just say me to dump SETI and Einstein and go for Folding instead =)
