Hey DRingdahl
From what I read, the THINK portion of UD (which was run originally for the cancer project) was developed by the people now running FAD. I think FAD started after UD stopped using THINK (instead they now use LIGANDFIT, which is apparently much better than THINK). The FAD team improved THINK after UD stopped using it, which is probably how they can claim 40x faster than UD, as UD used the THINK software years before FAD deployed it. Also, they say that the 40x improvement was made with help from Intel, so what platform are they benchmarking to claim their 40x improvement? Is it only on P4 Prescotts that happen to have SSE3? What kind of boost do AMD processors get? Again from what I've read, THINK was used by UD to filter out the background noise... to get a list of candidates that could possibly be used for a drug... either a yes or no. LIGANDFIT actually scores the molecules and gives a likeliness that the drug would work. This will definitely take a lot more processing time than the THINK client, so there's another way they could claim 40x faster.
It is nice that FAD finally developed a Linux version of the THINK client, but I remember back to the problems UD was having with them when they asked for a Linux client. UD has developed a linux version of the wrapper application, but FAD refused to license a linux version of the THINK client to them, so noone running linux could run THINK.
And yeah, their website looks like something I threw together in a few minutes, and does not give a professional air to their project. Their mystery meat navigation bar looks like it's from 1995. This statement makes me unsure of their motives: "income from partnership agreements to exploit the results will allow Find-a-Drug to become self-funding." Although directly after they state "results from public interest projects such as the Bioterrorism Antidote project will be made available to approved academic and government research laboratories without charge," which projects are deemed public interest projects? Can they be swayed by someone offering them enough money to make the cancer drug they find "non-public interest?"
UD makes their money licensing the wrapper application, and the grid.org cancer project is the best way to demonstrate the client working on a very large scale.
Anyway, I'm sticking with UD, but everyone is free to their own opinion.