So basically Folding@Home is a distributed computing program used by Stanford University that will use your processing cycles that are not being used by other programs from either your CPU and/or graphics card (NVIDIA or ATI based) to work on complex protein folding sequences’ trying to find cures for diseases, such as Alzheimer's, ALS, Huntington's, Parkinson's disease, and many cancers. If you would like to learn more about this you can find more information here http://folding.stanford.edu/ .

Now there are several different programs and ways to go about helping with the cause. I will list the current programs available and try to explain a little about each one.

1: Basic easy mode programs include the system tray CPU client and the GPU client

    A: The system tray CPU client will utilize approximately 100% of your single core CPU’s processing power depending on the Work Unit (WU). This client is extremely easy to setup and is pretty much a set it and forget it program for the basic computer users.
        This client can be downloaded from Stanford here http://folding.stanford.edu/English/Download

    B: The GPU2 client which will utilize your single chip graphics card (NVIDIA or ATI based) at 100% of unused cycles (this can be changed by a slider bar in the advanced configuration tab). This client is just as easy to setup and use as the system tray client. If you have a multi-chip graphics card or more than one graphics card that you want to use please refer to this guide for setup https://www.evga.com/forums/tm.aspx?high=&m=9849&mpage=1#9849 .
        This client can be downloaded here http://folding.stanford.edu/English/DownloadWinOther

    C: There is also a console client for the CPU that is basically the same as the system tray client except it uses a command line input system similar to DOS. This client can be found on the same page as the system tray client.

2: For a slightly more advanced user there is thenew SMP2 client. The SMP2 client is a command line input system similar to DOS that can utilize multi-core processors  to 100% of unused processing cycles. While being slightly more difficult to setup than the system tray client's above it is still fairly easy to setup by following the respective guide for your OS The new SMP2 client also utilizes a bonus point system that will give more points for completing the work units faster in order to obtain bonus points you will need a passkey wich is talked about further down this thread.
Guide for 32-bit window's OS's http://folding.stanford.edu/English/WinSMPGuideDEINO
Guide for 32 or 64-bit windows OS's  http://folding.stanford.edu/English/WinSMPGuideMPICH

These two clients can be downloaded here http://folding.stanford.edu/English/DownloadWinOther found under Windows V6 Beta Smp2/CPU clients

3: For the power user's there are a few different options that will run on virtual machines (requires a CPU with virtualization technology) with2-8 core proccesors

    A: NOTFRED'S is a SMP client that will run on a Virtual Machine and is capable of utilizing up to 100% of your CPU's unused cycles.
        For a guide on how to install this client you can go to http://forums.evga.com/tm.aspx?high=&m=9778&mpage=1#9778


    B: Linuxrouter's BigAdvanced Image is a Linux based SMP client that is capable of folding -BigAdv projects on windows with the help of the virtual machine. This client now has a web interface for setup and allows for auto starting the client.
        For a the guide on how to install you can go to http://forums.evga.com/tm.aspx?m=4464

4: There are also clients for MAC, Linux, and Playstation3 available.

Now for the fun part, for each work unit you complete you will be awarded a certain number of points depending on the work unit and how long it took to complete the work unit. For certain work units you will need a passkey from Stanford in order to qualify for bonus points. This passkey can be obtained here http://fah-web.stanford.edu/cgi-bin/getpasskey.py . With the points you are awarded you will be ranked against all others that are folding. You can also join a tem of people that are folding and be ranked separately within that team (I'm kind of partial to team EVGA Team #111065).

I would like to give credit to and thank Horvat for the guide to setup multi-GPUs and NOTFRED's guide. Also I would like to give creditt to and thank LinuxRouter for the excellent work he has done porting the Linux SMP image to VMware so that we can be all that we can be in the folding army.

If anyone has any suggestions on how to better this guide please let me know.

Blue ribbon!  Now we can point noobs here instead of explaining stuff.  

nice!
BTW, the win smp 6.29 with the A3s can utilize more than 4 cores

swapped links and removed 4 core limit from smp info thanks for the feedback and changed smp client to smp2 client with updated info

Did you put your folding name in here?
http://www.evga.com/community/myProfile.asp

Then after you completed your first wu, Stanford needs to run an update (hourly) and then EVGA needs to run an update as well (once every 24h, don´t know the exact time)

Barbarossa

Then after you completed your first wu, Stanford needs to run an update (hourly) and then EVGA needs to run an update as well (once every 24h, don´t know the exact time)

I think EVGA runs the update at midnight CST (GMT -6)

I am running GPU folding however it is using 100% of one of my cpu cores. Is this normal?

Thanks for the guide!

Amazing guide, has helped me on my installs so far. Just having problems with a laptop's ATI Radeon Mobility HD 3670 (cpu client works but gpu doesn't even after newest drivers) but it's not a home computer. Just another machine to help the good cause meh.

Thanks for the nice, simple guide!  I've passed it along to some friends.

Running the right client on your 930, you should outproduce the 260.

One thing is true though, the cpu wus take longer to complete.
Have a look in here for the cpu
http://forums.evga.com/tm.aspx?m=539797
Whether you can run the bigadv depends on clock speed and computer usage.
But you def can run the regulare smp wus. They only take a few hours, up to one day

demongimp

what is the team #for team fold with evga?

 
111065

next update - 3 hrs 
at EOC
http://folding.extremeoverclocking.com/search.php
EVGA -on left hand side and MY folding - every 24hrs
 

raxavier69

how long before the forums show folding= for me?

 
If you mean the folding points on the forum, you can set that up via Member Profile -> My Profile. There is a field called EVGA Folding Name. Once you enter that, your points will show up on the forum.

you have to have folded points - have completed WU's
you don't yet
 
check here
http://folding.extremeoverclocking.com/search.php
 
and to get a folding sig
http://piro.pirocast.net/badges/badgemaker.html
 
just post in your EOC number to the badgemaker
 
http://forums.evga.com/tm.aspx?m=17992

Zibri

In my opinion the folding client for GPU is really badly coded.

It's very easy to support more cards (like many gpu password crackers do).

Another reason: when folding, one of my card is al 100% and 60 degrees...
When password cracking, the same card goest up to 72-74 degrees.
That means that the gpu folding program could be at least 10-20% faster if written correctly and optimized.

just my 2c

Why don't you suggest some improvements to the GPU client coding via Pande group at Stanford and make all our cards fold much better? That way we can find the cures to the diseases we fold for all that much faster?  
Not sure what you mean by "support more cards". You can fold as many GPU clients as your board will support cards.
Also, how does an increase in heat directly correlate to improved efficiency and production?
How is password cracking in any way related to folding btw? Cite direct coding correlaries please.