Welcome to Green Folding
Hi everyone, and welcome to my blog. I’m Chris, and if you haven’t guessed already, this is going to be blog about big fans :). Well, big fans of F@H that is. I’m assuming if you found your way here, you know what Folding@Home is. If not, well, I’m not going to tell you what it is. I’m too lazy for that. But, if things go as planned, you will eventually figure it out by reading these posts. Or, you could save yourself the mental anguish and just go read about this charitable venture at http://folding.stanford.edu/home/. If it leaves you with a warm fuzzy feeling, please sign up and help! The feeling just gets better, until after eight years of contributing you just might want to start a blog about what you have learned.
First post done! Yay + Yawn. And I haven’t even really told you anything yet. This is probably due to it being 4:53 in the morning, and I am positively loopy with excitement. I promise there will be a more sane post tomorrow at some point. Until then, FOLD ON!
Green Folding@Home 
I think being part of folding@home is cool but like home solar vs grid scale solar you are better to have centralized super computers – they are more energy efficient. NVidia titan z board delivers 2.7 tflops double precision using only 375 watts (https://en.wikipedia.org/wiki/List_of_Nvidia_graphics_processing_units). If you put two boards on a single motherboard (you may be able to put up to 4) and assume system consumption of about 200 watts, you are getting about 6 gflops/watt. Most computers deliver at best 0.5 gflop/watt and most low end gpu based machines (hd 5850, 6990 etc) deliver maybe 1 to 2 gflops/watt. If you scale to 5 peta flops (computing power of folding@home) the difference in electric consumption is 1 megawatt for the best system and 2.5 to 5 megawatts for a low end system. At 0.10/kwh you are looking at annual electricity costs of $ 700,000 for the best system versus 2 million for a low end GPU system. Folding could build a gpu based computing system for the energy savings realized by such a system in just 3 to 4 years. They should at least be telling people with CPU systems to send money for such a system instead of wasting electricity doing such energy inefficient calculations.
I agree with your statement (build a very efficient centralized folder); Only question I would have is what would the cost of such a system be?
thanks for sharing the knowledge, great content