I don't understand the specifics of the whats, hows, and whys of the program, but I know a bit about Proteins and stuff, Qaz. I'll try and explain it to you as best I can.
Proteins are really complicated "machines". They are the the things that make your body, allow the processes that work inside you to work, and they can be both very simple, and extremely complex.
Proteins have to build themselves before they can work. Your DNA is like a recipe book - it tells all the ingredients (Amino acids, which are made of groups of moelcules, and can be as small as only three molecules in size) how and where to join together, so that they can make Proteins.
The chemical bonds that form between these Amino acids force the Protein to bend and twist into very, very, very specific shapes. This bending is known as "folding".
This folding is done at an incredible speed - some fold at speeds of millionths of a second!
This alone makes it impossible to phyically watch the Proteins fold, let alone the infinitesimally small size of them. Electron microscopes (The most powerful microscopes in the world) can only just see them, but they aren't anywhere near fast enough to allow us to watch the folding take place.
This is why the Folding@home program exists. It folds Proteins by estimating what physical forces are pulling, pushing, twisting, or bending the Protein.
It does this by creating "snapshots" of the Protein whilst it folds. Each snapshot is a still picture and is simply a single moment in time - like a photo.
When you put all the snapshots together, you end up with a moving image. These snapshots are known as frames in the program.
The complexity of the Protein affects how quickly a computer can model how this is done, but obviously, the more complex the Protein, the more bonds between the Amino acids there will be, so the harder it is to do, and the longer it takes to fold.
Each workunit is not the entire folding sequence of a single Protein - far from it, in fact.
Each work unit is merely a few moments in a fold, since it has been estimated that it would take 30 YEARS for a single CPU to fold an entire Protein! This is obviously useless to the scientists, especially since we know of literally thousands of Proteins. It would take Millennia for the results they need to come through!
The scientists have overcome this problem by breaking the folds into small chunks that are more easily completed by the average computer.
As the results of each work unit is sent in, they build up a picture of how (And why) many Proteins fold, sort of like a jigsaw puzzle. This information can then be used to develop cures or preventions to Genetic diseases.
This is why it is so important that Folding@home gets as many people as possible to fold for them - the more computers running the program, the more results that come in, and the faster they get the information they need.
...
Sorry this turned into an Uberpost, but i've answered several questions in one there. Please remember that I don't know the specifics of the program or of the folding process, so I may be wrong.
So then if you want more info, it would be best to check out the FAQ's on the folding@home site:
Obviously these will be better at providing info for you than me, since the people that wrote them will understand the process far more than me.
There's also an FAQ on how the points are calculated, if anyone is still reading this...
