Previous | Next --- Slide 17 of 41
Back to Lecture Thumbnails
ak47

Seeing the power cost makes you wonder about the economic issues of something like Folding@Home. My understanding is that it could be the case that the research is worth the energy cost, and the labs don't do it themselves because of the hardware cost and the difficulty of getting grants. But it could also be the case that the benefits are small compared to the energy footprint.

I would guess the guys at Stanford are pretty thoughtful on this kind of thing (and so far it's produced 128 papers, which sounds like a lot), but it's interesting to ponder.

HLAHat

There are lots of economic considerations with something like Folding@Home. First, the energy cost. Lots of homes running their computers and internet components do indeed use energy, but I think it's a fair assumption that many people who are folding would have been online doing something else if not contributing to this project. Therefore, it seems reasonable to assume that no extra energy past the baseline is being used. However, I think it's also reasonable to assume that most of the work done for Folding@Home is not useful. That is, most people working on it don't really contribute much to the solution, leading to inefficient work. This is the ultimate in brute force, only shipped out across to individuals instead of Processors.

There is also an opportunity cost to consider. The researchers could've spent lots of time researching and studying and programming to try and solve these folding problems. But why waste time on that when they can just have people solve the problems through trial and error? This leaves the researchers time to devote to other problems.

In light of this, I think Folding@Home wins out on a lot of economic considerations. It takes energy being used for the internet and converts it to work (albeit inefficiently) and gives researchers more time to focus on problems that do need specialized knowledge or equipment to solve.

lament

You failed to consider the energy necessary to run the humans. Consider all the energy costs necessary to raise a human to a mature-enough state to understand how to solve problems and go on to use the game. Humans can't even run for 24-hour stretches. These meat-machines seem very inefficient when it comes to energy use.

ak47

I don't understand the assumption that people would have been online doing something else if not contributing to the project.E.g. back in ~2008, I ran folding@home on my PS3 while I was at school; otherwise it would have just been turned off.

grose

I don't know if Folding@Home saved energy. Just because your computer was on and idling, doesn't mean it costs 0 energy to run Fold@Home. Fold@Home probably used more processing power and therefore more energy. Additionally, there's also the hidden cost that your PS3 can only do so much in its lifetime, and using it will wear it down. In other words, perhaps people who ran Fold@Home had their PS3's stop working earlier than otherwise.

That being said, whether it makes sense depends very much on the specific numbers.

Also, from what I've heard, protein folding is like, NP-complete. So I don't really blame them for using brute force, especially if it might like, cure cancer. (But I don't know much about this)