Quote:
Originally posted by MadHacker
does that mean everybody is doing the same work?
how is duplicate calculations on a molicule avoided?
|
Yes and No. We are all using the same protein, but the work is defined by how the client processes it. If you have noticed Generation 0 takes a while (about 2-3hrs on my comps) and that is the generation the rest of the 250 generation are based on.
And it is a good thing if there are duplicate results as that is what they are trying to achieve if I understand it correctly. The more similar the results the more progress they are making. They are trying to "match" an already completed and mapped protein with this "simulator" client. Once the client can match this protein properly then it will be able to start on non pre-mapped proteins and then that is where the real work starts to happen.
In other words the current client is basically a "beta" client. What we are doing now is stream lining it to perfection before any real work on other proteins is completed. Which we will all being doing once the client is acceptable to do this.
This is an extreamly important thing we are doing which some are incapable of understanding. A lot of people quit the project a while ago because they discovered we are not doing any "real" work on proteins (yet) like Folding@home is doing. What they fail to understand is it takes a lot of resources and power to make a client work as good and accurate as it can. Once this client is streamlined it will kick the hell out of Folding@home for quality of results. Without the effort we are putting in now none of this would be possible.